diff --git a/.gitignore b/.gitignore index 46d0f92..de62e8f 100644 --- a/.gitignore +++ b/.gitignore @@ -1,2 +1 @@ /coreutils-[0-9.]*.tar.xz -/coreutils-[0-9.]*.tar.xz.sig diff --git a/coreutils-4.5.3-langinfo.patch b/coreutils-4.5.3-langinfo.patch new file mode 100644 index 0000000..a8af3bd --- /dev/null +++ b/coreutils-4.5.3-langinfo.patch @@ -0,0 +1,20 @@ +diff --git a/src/date.c b/src/date.c +index ddb011e..619a72b 100644 +--- a/src/date.c ++++ b/src/date.c +@@ -490,14 +490,7 @@ main (int argc, char **argv) + format = DATE_FMT_LANGINFO (); + if (! *format) + { +- /* Do not wrap the following literal format string with _(...). +- For example, suppose LC_ALL is unset, LC_TIME=POSIX, +- and LANG="ko_KR". In that case, POSIX says that LC_TIME +- determines the format and contents of date and time strings +- written by date, which means "date" must generate output +- using the POSIX locale; but adding _() would cause "date" +- to use a Korean translation of the format. */ +- format = "%a %b %e %H:%M:%S %Z %Y"; ++ format = dcgettext(NULL, N_("%a %b %e %H:%M:%S %Z %Y"), LC_TIME); + } + } + diff --git a/coreutils-8.2-uname-processortype.patch b/coreutils-8.2-uname-processortype.patch new file mode 100644 index 0000000..ed01ab8 --- /dev/null +++ b/coreutils-8.2-uname-processortype.patch @@ -0,0 +1,49 @@ +diff --git a/src/uname.c b/src/uname.c +index 6371ca2..1ad8fd7 100644 +--- a/src/uname.c ++++ b/src/uname.c +@@ -300,13 +300,19 @@ main (int argc, char **argv) + + if (toprint & PRINT_PROCESSOR) + { +- char const *element = unknown; ++ char *element = unknown; + #if HAVE_SYSINFO && defined SI_ARCHITECTURE + { + static char processor[257]; + if (0 <= sysinfo (SI_ARCHITECTURE, processor, sizeof processor)) + element = processor; + } ++#else ++ { ++ static struct utsname u; ++ uname(&u); ++ element = u.machine; ++ } + #endif + #ifdef UNAME_PROCESSOR + if (element == unknown) +@@ -344,7 +350,7 @@ main (int argc, char **argv) + + if (toprint & PRINT_HARDWARE_PLATFORM) + { +- char const *element = unknown; ++ char *element = unknown; + #if HAVE_SYSINFO && defined SI_PLATFORM + { + static char hardware_platform[257]; +@@ -352,6 +358,14 @@ main (int argc, char **argv) + hardware_platform, sizeof hardware_platform)) + element = hardware_platform; + } ++#else ++ { ++ static struct utsname u; ++ uname(&u); ++ element = u.machine; ++ if(strlen(element)==4 && element[0]=='i' && element[2]=='8' && element[3]=='6') ++ element[1]='3'; ++ } + #endif + #ifdef UNAME_HARDWARE_PLATFORM + if (element == unknown) diff --git a/coreutils-8.32-DIR_COLORS.patch b/coreutils-8.32-DIR_COLORS.patch index 37ce3e6..9d422f4 100644 --- a/coreutils-8.32-DIR_COLORS.patch +++ b/coreutils-8.32-DIR_COLORS.patch @@ -1,4 +1,4 @@ -From bca11e30e8a6281a8cbddc9fb196dd86ab09c955 Mon Sep 17 00:00:00 2001 +From 81e25c8521937ecf7f444bab11fddaaf81cc3efd Mon Sep 17 00:00:00 2001 From: Kamil Dudka Date: Fri, 17 Jun 2016 16:58:18 +0200 Subject: [PATCH] downstream changes to default DIR_COLORS @@ -9,61 +9,61 @@ Subject: [PATCH] downstream changes to default DIR_COLORS 2 files changed, 23 insertions(+), 7 deletions(-) diff --git a/DIR_COLORS b/DIR_COLORS -index 540f6cd..b4785b6 100644 +index bd5df23..84f2417 100644 --- a/DIR_COLORS +++ b/DIR_COLORS @@ -1,3 +1,7 @@ +# This file goes in the /etc directory, and must be world readable. -+# You can override the system defaults by making a copy of this file -+# as ~/.dir_colors ++# You can copy this file to .dir_colors in your $HOME directory to override ++# the system defaults. + # Configuration file for dircolors, a utility to help you set the # LS_COLORS environment variable used by GNU ls with the --color option. -@@ -11,6 +15,9 @@ - - # Global config options can be specified before TERM or COLORTERM entries +@@ -8,6 +12,9 @@ + # The keywords COLOR, OPTIONS, and EIGHTBIT (honored by the + # slackware version of dircolors) are recognized but ignored. +# For compatibility, the pattern "^COLOR.*none" is recognized as a way to +# disable colorization. See https://bugzilla.redhat.com/1349579 for details. + - # =================================================================== - # Terminal filters - # =================================================================== -@@ -70,7 +77,7 @@ DOOR 01;35 # door + # Below are TERM entries, which can be a glob patterns, to match + # against the TERM environment variable to determine if it is colorizable. + TERM Eterm +@@ -58,7 +65,7 @@ DOOR 01;35 # door BLK 40;33;01 # block device driver CHR 40;33;01 # character device driver ORPHAN 40;31;01 # symlink to nonexistent file, or non-stat'able file ... -MISSING 00 # ... and the files they point to +MISSING 01;37;41 # ... and the files they point to - SETUID 37;41 # regular file that is setuid (u+s) - SETGID 30;43 # regular file that is setgid (g+s) - CAPABILITY 00 # regular file with capability (very expensive to lookup) + SETUID 37;41 # file that is setuid (u+s) + SETGID 30;43 # file that is setgid (g+s) + CAPABILITY 30;41 # file with capability diff --git a/DIR_COLORS.lightbgcolor b/DIR_COLORS.lightbgcolor -index e3b0ec3..39a0a4c 100644 +index 4316832..6402854 100644 --- a/DIR_COLORS.lightbgcolor +++ b/DIR_COLORS.lightbgcolor @@ -1,3 +1,9 @@ +# Configuration file for the color ls utility - modified for lighter backgrounds + +# This file goes in the /etc directory, and must be world readable. -+# You can override the system defaults by making a copy of this file -+# as ~/.dir_colors ++# You can copy this file to .dir_colors in your $HOME directory to override ++# the system defaults. + # Configuration file for dircolors, a utility to help you set the # LS_COLORS environment variable used by GNU ls with the --color option. -@@ -11,6 +17,9 @@ - - # Global config options can be specified before TERM or COLORTERM entries +@@ -8,6 +14,9 @@ + # The keywords COLOR, OPTIONS, and EIGHTBIT (honored by the + # slackware version of dircolors) are recognized but ignored. +# For compatibility, the pattern "^COLOR.*none" is recognized as a way to +# disable colorization. See https://bugzilla.redhat.com/1349579 for details. + - # =================================================================== - # Terminal filters - # =================================================================== -@@ -60,17 +69,17 @@ TERM xterm* + # Below are TERM entries, which can be a glob patterns, to match + # against the TERM environment variable to determine if it is colorizable. + TERM Eterm +@@ -48,17 +57,17 @@ TERM xterm* #NORMAL 00 # no color code at all #FILE 00 # regular file: use no color at all RESET 0 # reset to "normal" color @@ -83,18 +83,18 @@ index e3b0ec3..39a0a4c 100644 ORPHAN 40;31;01 # symlink to nonexistent file, or non-stat'able file ... -MISSING 00 # ... and the files they point to +MISSING 01;37;41 # ... and the files they point to - SETUID 37;41 # regular file that is setuid (u+s) - SETGID 30;43 # regular file that is setgid (g+s) - CAPABILITY 00 # regular file with capability (very expensive to lookup) -@@ -79,7 +88,7 @@ OTHER_WRITABLE 34;42 # dir that is other-writable (o+w) and not sticky + SETUID 37;41 # file that is setuid (u+s) + SETGID 30;43 # file that is setgid (g+s) + CAPABILITY 30;41 # file with capability +@@ -67,7 +76,7 @@ OTHER_WRITABLE 34;42 # dir that is other-writable (o+w) and not sticky STICKY 37;44 # dir with the sticky bit set (+t) and not other-writable - # This is for regular files with execute permission: + # This is for files with execute permission: -EXEC 01;32 +EXEC 00;32 - # =================================================================== - # File extension attributes + # List any file extensions like '.gz' or '.tar' that you would like ls + # to colorize below. Put the extension, a space, and the color init string. -- -2.49.0 +2.21.1 diff --git a/coreutils-8.32-coverity-utimens.patch b/coreutils-8.32-coverity-utimens.patch new file mode 100644 index 0000000..5bbbb6e --- /dev/null +++ b/coreutils-8.32-coverity-utimens.patch @@ -0,0 +1,52 @@ +From 51b9a8ba0974d262e0b0f81a2078b3c7907b25ed Mon Sep 17 00:00:00 2001 +From: Paul Eggert +Date: Wed, 7 Apr 2021 17:29:59 -0700 +Subject: [PATCH] utimens: fix confusing arg type in internal func +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +Although the old code was technically correct, this was accidental +and it understandably confused Coverity. Reported by Ondrej Dubaj in: +https://lists.gnu.org/r/bug-tar/2021-04/msg00000.html +* lib/utimens.c (update_timespec): Change arg type from ‘struct +timespec *[2]’ (pointer to array of 2 pointers to timespecs) to +‘struct timespec **’ (pointer to pointer to the first timespec in +an array of 2 timespecs). Although the old code happened to be +technically correct, it was misleading and confused Coverity. +And though the type ‘struct timespec (**)[2]’ (pointer to pointer +to array of 2 timespecs) would perhaps be more technically +correct, it would be almost as confusing and would require changes +elsewhere in this file; let’s quit while we’re ahead. + +Upstream-commit: a3a946f670718d0dee5a7425ad5ac0a29fb46ea1 +Signed-off-by: Kamil Dudka +--- + lib/utimens.c | 6 +++--- + 1 file changed, 3 insertions(+), 3 deletions(-) + +diff --git a/lib/utimens.c b/lib/utimens.c +index 3f53942..ea8c672 100644 +--- a/lib/utimens.c ++++ b/lib/utimens.c +@@ -123,14 +123,14 @@ validate_timespec (struct timespec timespec[2]) + return result + (utime_omit_count == 1); + } + +-/* Normalize any UTIME_NOW or UTIME_OMIT values in *TS, using stat +- buffer STATBUF to obtain the current timestamps of the file. If ++/* Normalize any UTIME_NOW or UTIME_OMIT values in (*TS)[0] and (*TS)[1], ++ using STATBUF to obtain the current timestamps of the file. If + both times are UTIME_NOW, set *TS to NULL (as this can avoid some + permissions issues). If both times are UTIME_OMIT, return true + (nothing further beyond the prior collection of STATBUF is + necessary); otherwise return false. */ + static bool +-update_timespec (struct stat const *statbuf, struct timespec *ts[2]) ++update_timespec (struct stat const *statbuf, struct timespec **ts) + { + struct timespec *timespec = *ts; + if (timespec[0].tv_nsec == UTIME_OMIT +-- +2.26.3 + diff --git a/coreutils-8.32-cp-file-range.patch b/coreutils-8.32-cp-file-range.patch new file mode 100644 index 0000000..6d3f651 --- /dev/null +++ b/coreutils-8.32-cp-file-range.patch @@ -0,0 +1,1339 @@ +From 5f2dac18054d9d9b3d84e7fba8c2a6e750d2c245 Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?P=C3=A1draig=20Brady?= +Date: Wed, 1 Apr 2020 12:51:34 +0100 +Subject: [PATCH 01/12] cp: ensure --attributes-only doesn't remove files + +* src/copy.c (copy_internal): Ensure we don't unlink the destination +unless explicitly requested. +* tests/cp/attr-existing.sh: Add test cases. +* NEWS: Mention the bug fix. +Fixes https://bugs.gnu.org/40352 + +Upstream-commit: 7b5f0fa47cd04c84975250d5b5da7c98e097e99f +Signed-off-by: Kamil Dudka +--- + src/copy.c | 9 +++++---- + tests/cp/attr-existing.sh | 21 ++++++++++++++++++--- + 2 files changed, 23 insertions(+), 7 deletions(-) + +diff --git a/src/copy.c b/src/copy.c +index 6e5efc7..54601ce 100644 +--- a/src/copy.c ++++ b/src/copy.c +@@ -2211,10 +2211,11 @@ copy_internal (char const *src_name, char const *dst_name, + /* Never unlink dst_name when in move mode. */ + && ! x->move_mode + && (x->unlink_dest_before_opening +- || (x->preserve_links && 1 < dst_sb.st_nlink) +- || (x->dereference == DEREF_NEVER +- && ! S_ISREG (src_sb.st_mode)) +- )) ++ || (x->data_copy_required ++ && ((x->preserve_links && 1 < dst_sb.st_nlink) ++ || (x->dereference == DEREF_NEVER ++ && ! S_ISREG (src_sb.st_mode)))) ++ )) + { + if (unlink (dst_name) != 0 && errno != ENOENT) + { +diff --git a/tests/cp/attr-existing.sh b/tests/cp/attr-existing.sh +index 59ce641..14fc844 100755 +--- a/tests/cp/attr-existing.sh ++++ b/tests/cp/attr-existing.sh +@@ -19,11 +19,26 @@ + . "${srcdir=.}/tests/init.sh"; path_prepend_ ./src + print_ver_ cp + +-printf '1' > file1 +-printf '2' > file2 +-printf '2' > file2.exp ++printf '1' > file1 || framework_failure_ ++printf '2' > file2 || framework_failure_ ++printf '2' > file2.exp || framework_failure_ + + cp --attributes-only file1 file2 || fail=1 + cmp file2 file2.exp || fail=1 + ++# coreutils v8.32 and before would remove destination files ++# if hardlinked or the source was not a regular file. ++ln file2 link2 || framework_failure_ ++cp -a --attributes-only file1 file2 || fail=1 ++cmp file2 file2.exp || fail=1 ++ ++ln -s file1 sym1 || framework_failure_ ++returns_ 1 cp -a --attributes-only sym1 file2 || fail=1 ++cmp file2 file2.exp || fail=1 ++ ++# One can still force removal though ++cp -a --remove-destination --attributes-only sym1 file2 || fail=1 ++test -L file2 || fail=1 ++cmp file1 file2 || fail=1 ++ + Exit $fail +-- +2.26.3 + + +From c728747b06e71894c96d1f27434f2484af992c75 Mon Sep 17 00:00:00 2001 +From: Paul Eggert +Date: Tue, 23 Jun 2020 19:18:04 -0700 +Subject: [PATCH 02/12] cp: refactor extent_copy + +* src/copy.c (extent_copy): New arg SCAN, replacing +REQUIRE_NORMAL_COPY. All callers changed. +(enum scantype): New type. +(infer_scantype): Rename from is_probably_sparse and return +the new type. Add args FD and SCAN. All callers changed. + +Upstream-commit: 761ba28400a04ee24eefe9cd4973ec8850cd7a52 +Signed-off-by: Kamil Dudka +--- + src/copy.c | 119 +++++++++++++++++++++++++---------------------------- + 1 file changed, 55 insertions(+), 64 deletions(-) + +diff --git a/src/copy.c b/src/copy.c +index 54601ce..f694f91 100644 +--- a/src/copy.c ++++ b/src/copy.c +@@ -422,9 +422,8 @@ extent_copy (int src_fd, int dest_fd, char *buf, size_t buf_size, + size_t hole_size, off_t src_total_size, + enum Sparse_type sparse_mode, + char const *src_name, char const *dst_name, +- bool *require_normal_copy) ++ struct extent_scan *scan) + { +- struct extent_scan scan; + off_t last_ext_start = 0; + off_t last_ext_len = 0; + +@@ -432,45 +431,25 @@ extent_copy (int src_fd, int dest_fd, char *buf, size_t buf_size, + We may need this at the end, for a final ftruncate. */ + off_t dest_pos = 0; + +- extent_scan_init (src_fd, &scan); +- +- *require_normal_copy = false; + bool wrote_hole_at_eof = true; +- do ++ while (true) + { +- bool ok = extent_scan_read (&scan); +- if (! ok) +- { +- if (scan.hit_final_extent) +- break; +- +- if (scan.initial_scan_failed) +- { +- *require_normal_copy = true; +- return false; +- } +- +- error (0, errno, _("%s: failed to get extents info"), +- quotef (src_name)); +- return false; +- } +- + bool empty_extent = false; +- for (unsigned int i = 0; i < scan.ei_count || empty_extent; i++) ++ for (unsigned int i = 0; i < scan->ei_count || empty_extent; i++) + { + off_t ext_start; + off_t ext_len; + off_t ext_hole_size; + +- if (i < scan.ei_count) ++ if (i < scan->ei_count) + { +- ext_start = scan.ext_info[i].ext_logical; +- ext_len = scan.ext_info[i].ext_length; ++ ext_start = scan->ext_info[i].ext_logical; ++ ext_len = scan->ext_info[i].ext_length; + } + else /* empty extent at EOF. */ + { + i--; +- ext_start = last_ext_start + scan.ext_info[i].ext_length; ++ ext_start = last_ext_start + scan->ext_info[i].ext_length; + ext_len = 0; + } + +@@ -498,7 +477,7 @@ extent_copy (int src_fd, int dest_fd, char *buf, size_t buf_size, + { + error (0, errno, _("cannot lseek %s"), quoteaf (src_name)); + fail: +- extent_scan_free (&scan); ++ extent_scan_free (scan); + return false; + } + +@@ -539,7 +518,7 @@ extent_copy (int src_fd, int dest_fd, char *buf, size_t buf_size, + /* For now, do not treat FIEMAP_EXTENT_UNWRITTEN specially, + because that (in combination with no sync) would lead to data + loss at least on XFS and ext4 when using 2.6.39-rc3 kernels. */ +- if (0 && (scan.ext_info[i].ext_flags & FIEMAP_EXTENT_UNWRITTEN)) ++ if (0 && (scan->ext_info[i].ext_flags & FIEMAP_EXTENT_UNWRITTEN)) + { + empty_extent = true; + last_ext_len = 0; +@@ -571,16 +550,23 @@ extent_copy (int src_fd, int dest_fd, char *buf, size_t buf_size, + extents beyond the apparent size. */ + if (dest_pos == src_total_size) + { +- scan.hit_final_extent = true; ++ scan->hit_final_extent = true; + break; + } + } + + /* Release the space allocated to scan->ext_info. */ +- extent_scan_free (&scan); ++ extent_scan_free (scan); + ++ if (scan->hit_final_extent) ++ break; ++ if (! extent_scan_read (scan) && ! scan->hit_final_extent) ++ { ++ error (0, errno, _("%s: failed to get extents info"), ++ quotef (src_name)); ++ return false; ++ } + } +- while (! scan.hit_final_extent); + + /* When the source file ends with a hole, we have to do a little more work, + since the above copied only up to and including the final extent. +@@ -1021,16 +1007,35 @@ fchmod_or_lchmod (int desc, char const *name, mode_t mode) + # define HAVE_STRUCT_STAT_ST_BLOCKS 0 + #endif + ++/* Type of scan being done on the input when looking for sparseness. */ ++enum scantype ++ { ++ /* No fancy scanning; just read and write. */ ++ PLAIN_SCANTYPE, ++ ++ /* Read and examine data looking for zero blocks; useful when ++ attempting to create sparse output. */ ++ ZERO_SCANTYPE, ++ ++ /* Extent information is available. */ ++ EXTENT_SCANTYPE ++ }; ++ + /* Use a heuristic to determine whether stat buffer SB comes from a file + with sparse blocks. If the file has fewer blocks than would normally + be needed for a file of its size, then at least one of the blocks in + the file is a hole. In that case, return true. */ +-static bool +-is_probably_sparse (struct stat const *sb) ++static enum scantype ++infer_scantype (int fd, struct stat const *sb, struct extent_scan *scan) + { +- return (HAVE_STRUCT_STAT_ST_BLOCKS +- && S_ISREG (sb->st_mode) +- && ST_NBLOCKS (*sb) < sb->st_size / ST_NBLOCKSIZE); ++ if (! (HAVE_STRUCT_STAT_ST_BLOCKS ++ && S_ISREG (sb->st_mode) ++ && ST_NBLOCKS (*sb) < sb->st_size / ST_NBLOCKSIZE)) ++ return PLAIN_SCANTYPE; ++ ++ extent_scan_init (fd, scan); ++ extent_scan_read (scan); ++ return scan->initial_scan_failed ? ZERO_SCANTYPE : EXTENT_SCANTYPE; + } + + +@@ -1061,6 +1066,7 @@ copy_reg (char const *src_name, char const *dst_name, + mode_t src_mode = src_sb->st_mode; + struct stat sb; + struct stat src_open_sb; ++ struct extent_scan scan; + bool return_val = true; + bool data_copy_required = x->data_copy_required; + +@@ -1260,23 +1266,13 @@ copy_reg (char const *src_name, char const *dst_name, + fdadvise (source_desc, 0, 0, FADVISE_SEQUENTIAL); + + /* Deal with sparse files. */ +- bool make_holes = false; +- bool sparse_src = is_probably_sparse (&src_open_sb); +- +- if (S_ISREG (sb.st_mode)) +- { +- /* Even with --sparse=always, try to create holes only +- if the destination is a regular file. */ +- if (x->sparse_mode == SPARSE_ALWAYS) +- make_holes = true; +- +- /* Use a heuristic to determine whether SRC_NAME contains any sparse +- blocks. If the file has fewer blocks than would normally be +- needed for a file of its size, then at least one of the blocks in +- the file is a hole. */ +- if (x->sparse_mode == SPARSE_AUTO && sparse_src) +- make_holes = true; +- } ++ enum scantype scantype = infer_scantype (source_desc, &src_open_sb, ++ &scan); ++ bool make_holes ++ = (S_ISREG (sb.st_mode) ++ && (x->sparse_mode == SPARSE_ALWAYS ++ || (x->sparse_mode == SPARSE_AUTO ++ && scantype != PLAIN_SCANTYPE))); + + /* If not making a sparse file, try to use a more-efficient + buffer size. */ +@@ -1305,10 +1301,8 @@ copy_reg (char const *src_name, char const *dst_name, + buf_alloc = xmalloc (buf_size + buf_alignment); + buf = ptr_align (buf_alloc, buf_alignment); + +- if (sparse_src) ++ if (scantype == EXTENT_SCANTYPE) + { +- bool normal_copy_required; +- + /* Perform an efficient extent-based copy, falling back to the + standard copy only if the initial extent scan fails. If the + '--sparse=never' option is specified, write all data but use +@@ -1316,14 +1310,11 @@ copy_reg (char const *src_name, char const *dst_name, + if (extent_copy (source_desc, dest_desc, buf, buf_size, hole_size, + src_open_sb.st_size, + make_holes ? x->sparse_mode : SPARSE_NEVER, +- src_name, dst_name, &normal_copy_required)) ++ src_name, dst_name, &scan)) + goto preserve_metadata; + +- if (! normal_copy_required) +- { +- return_val = false; +- goto close_src_and_dst_desc; +- } ++ return_val = false; ++ goto close_src_and_dst_desc; + } + + off_t n_read; +-- +2.26.3 + + +From ed7ff81de507bef46991f4caac550f41ab65e3ed Mon Sep 17 00:00:00 2001 +From: Paul Eggert +Date: Wed, 24 Jun 2020 17:05:20 -0700 +Subject: [PATCH 03/12] cp: avoid copy_reg goto + +* src/copy.c (copy_reg): Redo to avoid label and goto. + +Upstream-commit: 2fcd0f3328f5181a2986905fa5469a0152c67279 +Signed-off-by: Kamil Dudka +--- + src/copy.c | 34 +++++++++++----------------------- + 1 file changed, 11 insertions(+), 23 deletions(-) + +diff --git a/src/copy.c b/src/copy.c +index f694f91..b382cfa 100644 +--- a/src/copy.c ++++ b/src/copy.c +@@ -1301,29 +1301,18 @@ copy_reg (char const *src_name, char const *dst_name, + buf_alloc = xmalloc (buf_size + buf_alignment); + buf = ptr_align (buf_alloc, buf_alignment); + +- if (scantype == EXTENT_SCANTYPE) +- { +- /* Perform an efficient extent-based copy, falling back to the +- standard copy only if the initial extent scan fails. If the +- '--sparse=never' option is specified, write all data but use +- any extents to read more efficiently. */ +- if (extent_copy (source_desc, dest_desc, buf, buf_size, hole_size, +- src_open_sb.st_size, +- make_holes ? x->sparse_mode : SPARSE_NEVER, +- src_name, dst_name, &scan)) +- goto preserve_metadata; +- +- return_val = false; +- goto close_src_and_dst_desc; +- } +- + off_t n_read; +- bool wrote_hole_at_eof; +- if (! sparse_copy (source_desc, dest_desc, buf, buf_size, +- make_holes ? hole_size : 0, +- x->sparse_mode == SPARSE_ALWAYS, src_name, dst_name, +- UINTMAX_MAX, &n_read, +- &wrote_hole_at_eof)) ++ bool wrote_hole_at_eof = false; ++ if (! (scantype == EXTENT_SCANTYPE ++ ? extent_copy (source_desc, dest_desc, buf, buf_size, hole_size, ++ src_open_sb.st_size, ++ make_holes ? x->sparse_mode : SPARSE_NEVER, ++ src_name, dst_name, &scan) ++ : sparse_copy (source_desc, dest_desc, buf, buf_size, ++ make_holes ? hole_size : 0, ++ x->sparse_mode == SPARSE_ALWAYS, ++ src_name, dst_name, UINTMAX_MAX, &n_read, ++ &wrote_hole_at_eof))) + { + return_val = false; + goto close_src_and_dst_desc; +@@ -1336,7 +1325,6 @@ copy_reg (char const *src_name, char const *dst_name, + } + } + +-preserve_metadata: + if (x->preserve_timestamps) + { + struct timespec timespec[2]; +-- +2.26.3 + + +From 5631bded3a385ca0bbd77456b50767fe5580240c Mon Sep 17 00:00:00 2001 +From: Paul Eggert +Date: Thu, 25 Jun 2020 16:31:44 -0700 +Subject: [PATCH 04/12] cp: use SEEK_DATA/SEEK_HOLE if available + +If it works, prefer lseek with SEEK_DATA and SEEK_HOLE to FIEMAP, +as lseek is simpler and more portable (will be in next POSIX). +Problem reported in 2011 by Jeff Liu (Bug#8061). +* NEWS: Mention this. +* src/copy.c (lseek_copy) [SEEK_HOLE]: New function. +(enum scantype): New constants ERROR_SCANTYPE, LSEEK_SCANTYPE. +(union scan_inference): New type. +(infer_scantype): Last arg is now union scan_inference *, +not struct extent_scan *. All callers changed. +Prefer SEEK_HOLE to FIEMAP if both work, since +SEEK_HOLE is simpler and more portable. +(copy_reg): Do the fdadvise after initial scan, in case the scan +fails. Report an error if the initial scan fails. +(copy_reg) [SEEK_HOLE]: Use lseek_copy if scantype says so. + +Upstream-commit: a6eaee501f6ec0c152abe88640203a64c390993e +Signed-off-by: Kamil Dudka +--- + src/copy.c | 209 ++++++++++++++++++++++++++++++++++++++++++++++++++--- + 1 file changed, 198 insertions(+), 11 deletions(-) + +diff --git a/src/copy.c b/src/copy.c +index b382cfa..d88f8cf 100644 +--- a/src/copy.c ++++ b/src/copy.c +@@ -416,7 +416,12 @@ write_zeros (int fd, off_t n_bytes) + Upon a successful copy, return true. If the initial extent scan + fails, set *NORMAL_COPY_REQUIRED to true and return false. + Upon any other failure, set *NORMAL_COPY_REQUIRED to false and +- return false. */ ++ return false. ++ ++ FIXME: Once we no longer need to support Linux kernel versions ++ before 3.1 (2011), this function can be retired as it is superseded ++ by lseek_copy. That is, we no longer need extent-scan.h and can ++ remove any of the code that uses it. */ + static bool + extent_copy (int src_fd, int dest_fd, char *buf, size_t buf_size, + size_t hole_size, off_t src_total_size, +@@ -595,6 +600,150 @@ extent_copy (int src_fd, int dest_fd, char *buf, size_t buf_size, + return true; + } + ++#ifdef SEEK_HOLE ++/* Perform an efficient extent copy, if possible. This avoids ++ the overhead of detecting holes in hole-introducing/preserving ++ copy, and thus makes copying sparse files much more efficient. ++ Copy from SRC_FD to DEST_FD, using BUF (of size BUF_SIZE) for a buffer. ++ Look for holes of size HOLE_SIZE in the input. ++ The input file is of size SRC_TOTAL_SIZE. ++ Use SPARSE_MODE to determine whether to create holes in the output. ++ SRC_NAME and DST_NAME are the input and output file names. ++ Return true if successful, false (with a diagnostic) otherwise. */ ++ ++static bool ++lseek_copy (int src_fd, int dest_fd, char *buf, size_t buf_size, ++ size_t hole_size, off_t ext_start, off_t src_total_size, ++ enum Sparse_type sparse_mode, ++ char const *src_name, char const *dst_name) ++{ ++ off_t last_ext_start = 0; ++ off_t last_ext_len = 0; ++ off_t dest_pos = 0; ++ bool wrote_hole_at_eof = true; ++ ++ while (0 <= ext_start) ++ { ++ off_t ext_end = lseek (src_fd, ext_start, SEEK_HOLE); ++ if (ext_end < 0) ++ { ++ if (errno != ENXIO) ++ goto cannot_lseek; ++ ext_end = src_total_size; ++ if (ext_end <= ext_start) ++ { ++ /* The input file grew; get its current size. */ ++ src_total_size = lseek (src_fd, 0, SEEK_END); ++ if (src_total_size < 0) ++ goto cannot_lseek; ++ ++ /* If the input file shrank after growing, stop copying. */ ++ if (src_total_size <= ext_start) ++ break; ++ ++ ext_end = src_total_size; ++ } ++ } ++ /* If the input file must have grown, increase its measured size. */ ++ if (src_total_size < ext_end) ++ src_total_size = ext_end; ++ ++ if (lseek (src_fd, ext_start, SEEK_SET) < 0) ++ goto cannot_lseek; ++ ++ wrote_hole_at_eof = false; ++ off_t ext_hole_size = ext_start - last_ext_start - last_ext_len; ++ ++ if (ext_hole_size) ++ { ++ if (sparse_mode != SPARSE_NEVER) ++ { ++ if (! create_hole (dest_fd, dst_name, ++ sparse_mode == SPARSE_ALWAYS, ++ ext_hole_size)) ++ return false; ++ wrote_hole_at_eof = true; ++ } ++ else ++ { ++ /* When not inducing holes and when there is a hole between ++ the end of the previous extent and the beginning of the ++ current one, write zeros to the destination file. */ ++ if (! write_zeros (dest_fd, ext_hole_size)) ++ { ++ error (0, errno, _("%s: write failed"), ++ quotef (dst_name)); ++ return false; ++ } ++ } ++ } ++ ++ off_t ext_len = ext_end - ext_start; ++ last_ext_start = ext_start; ++ last_ext_len = ext_len; ++ ++ /* Copy this extent, looking for further opportunities to not ++ bother to write zeros unless --sparse=never, since SEEK_HOLE ++ is conservative and may miss some holes. */ ++ off_t n_read; ++ bool read_hole; ++ if ( ! sparse_copy (src_fd, dest_fd, buf, buf_size, ++ sparse_mode == SPARSE_NEVER ? 0 : hole_size, ++ true, src_name, dst_name, ext_len, &n_read, ++ &read_hole)) ++ return false; ++ ++ dest_pos = ext_start + n_read; ++ if (n_read) ++ wrote_hole_at_eof = read_hole; ++ if (n_read < ext_len) ++ { ++ /* The input file shrank. */ ++ src_total_size = dest_pos; ++ break; ++ } ++ ++ ext_start = lseek (src_fd, dest_pos, SEEK_DATA); ++ if (ext_start < 0) ++ { ++ if (errno != ENXIO) ++ goto cannot_lseek; ++ break; ++ } ++ } ++ ++ /* When the source file ends with a hole, we have to do a little more work, ++ since the above copied only up to and including the final extent. ++ In order to complete the copy, we may have to insert a hole or write ++ zeros in the destination corresponding to the source file's hole-at-EOF. ++ ++ In addition, if the final extent was a block of zeros at EOF and we've ++ just converted them to a hole in the destination, we must call ftruncate ++ here in order to record the proper length in the destination. */ ++ if ((dest_pos < src_total_size || wrote_hole_at_eof) ++ && ! (sparse_mode == SPARSE_NEVER ++ ? write_zeros (dest_fd, src_total_size - dest_pos) ++ : ftruncate (dest_fd, src_total_size) == 0)) ++ { ++ error (0, errno, _("failed to extend %s"), quoteaf (dst_name)); ++ return false; ++ } ++ ++ if (sparse_mode == SPARSE_ALWAYS && dest_pos < src_total_size ++ && punch_hole (dest_fd, dest_pos, src_total_size - dest_pos) < 0) ++ { ++ error (0, errno, _("error deallocating %s"), quoteaf (dst_name)); ++ return false; ++ } ++ ++ return true; ++ ++ cannot_lseek: ++ error (0, errno, _("cannot lseek %s"), quoteaf (src_name)); ++ return false; ++} ++#endif ++ + /* FIXME: describe */ + /* FIXME: rewrite this to use a hash table so we avoid the quadratic + performance hit that's probably noticeable only on trees deeper +@@ -1010,6 +1159,9 @@ fchmod_or_lchmod (int desc, char const *name, mode_t mode) + /* Type of scan being done on the input when looking for sparseness. */ + enum scantype + { ++ /* An error was found when determining scantype. */ ++ ERROR_SCANTYPE, ++ + /* No fancy scanning; just read and write. */ + PLAIN_SCANTYPE, + +@@ -1017,22 +1169,44 @@ enum scantype + attempting to create sparse output. */ + ZERO_SCANTYPE, + ++ /* lseek information is available. */ ++ LSEEK_SCANTYPE, ++ + /* Extent information is available. */ + EXTENT_SCANTYPE + }; + +-/* Use a heuristic to determine whether stat buffer SB comes from a file +- with sparse blocks. If the file has fewer blocks than would normally +- be needed for a file of its size, then at least one of the blocks in +- the file is a hole. In that case, return true. */ ++/* Result of infer_scantype. */ ++union scan_inference ++{ ++ /* Used if infer_scantype returns LSEEK_SCANTYPE. This is the ++ offset of the first data block, or -1 if the file has no data. */ ++ off_t ext_start; ++ ++ /* Used if infer_scantype returns EXTENT_SCANTYPE. */ ++ struct extent_scan extent_scan; ++}; ++ ++/* Return how to scan a file with descriptor FD and stat buffer SB. ++ Store any information gathered into *SCAN. */ + static enum scantype +-infer_scantype (int fd, struct stat const *sb, struct extent_scan *scan) ++infer_scantype (int fd, struct stat const *sb, ++ union scan_inference *scan_inference) + { + if (! (HAVE_STRUCT_STAT_ST_BLOCKS + && S_ISREG (sb->st_mode) + && ST_NBLOCKS (*sb) < sb->st_size / ST_NBLOCKSIZE)) + return PLAIN_SCANTYPE; + ++#ifdef SEEK_HOLE ++ scan_inference->ext_start = lseek (fd, 0, SEEK_DATA); ++ if (0 <= scan_inference->ext_start) ++ return LSEEK_SCANTYPE; ++ else if (errno != EINVAL && errno != ENOTSUP) ++ return errno == ENXIO ? LSEEK_SCANTYPE : ERROR_SCANTYPE; ++#endif ++ ++ struct extent_scan *scan = &scan_inference->extent_scan; + extent_scan_init (fd, scan); + extent_scan_read (scan); + return scan->initial_scan_failed ? ZERO_SCANTYPE : EXTENT_SCANTYPE; +@@ -1066,7 +1240,7 @@ copy_reg (char const *src_name, char const *dst_name, + mode_t src_mode = src_sb->st_mode; + struct stat sb; + struct stat src_open_sb; +- struct extent_scan scan; ++ union scan_inference scan_inference; + bool return_val = true; + bool data_copy_required = x->data_copy_required; + +@@ -1263,17 +1437,23 @@ copy_reg (char const *src_name, char const *dst_name, + size_t buf_size = io_blksize (sb); + size_t hole_size = ST_BLKSIZE (sb); + +- fdadvise (source_desc, 0, 0, FADVISE_SEQUENTIAL); +- + /* Deal with sparse files. */ + enum scantype scantype = infer_scantype (source_desc, &src_open_sb, +- &scan); ++ &scan_inference); ++ if (scantype == ERROR_SCANTYPE) ++ { ++ error (0, errno, _("cannot lseek %s"), quoteaf (src_name)); ++ return_val = false; ++ goto close_src_and_dst_desc; ++ } + bool make_holes + = (S_ISREG (sb.st_mode) + && (x->sparse_mode == SPARSE_ALWAYS + || (x->sparse_mode == SPARSE_AUTO + && scantype != PLAIN_SCANTYPE))); + ++ fdadvise (source_desc, 0, 0, FADVISE_SEQUENTIAL); ++ + /* If not making a sparse file, try to use a more-efficient + buffer size. */ + if (! make_holes) +@@ -1307,7 +1487,14 @@ copy_reg (char const *src_name, char const *dst_name, + ? extent_copy (source_desc, dest_desc, buf, buf_size, hole_size, + src_open_sb.st_size, + make_holes ? x->sparse_mode : SPARSE_NEVER, +- src_name, dst_name, &scan) ++ src_name, dst_name, &scan_inference.extent_scan) ++#ifdef SEEK_HOLE ++ : scantype == LSEEK_SCANTYPE ++ ? lseek_copy (source_desc, dest_desc, buf, buf_size, hole_size, ++ scan_inference.ext_start, src_open_sb.st_size, ++ make_holes ? x->sparse_mode : SPARSE_NEVER, ++ src_name, dst_name) ++#endif + : sparse_copy (source_desc, dest_desc, buf, buf_size, + make_holes ? hole_size : 0, + x->sparse_mode == SPARSE_ALWAYS, +-- +2.26.3 + + +From be7466be92d779cfbece418d4de33191ae52ab4a Mon Sep 17 00:00:00 2001 +From: Kamil Dudka +Date: Wed, 24 Mar 2021 16:06:53 +0100 +Subject: [PATCH 05/12] import the copy-file-range module from gnulib + +--- + aclocal.m4 | 1 + + lib/config.hin | 3 +++ + lib/copy-file-range.c | 33 +++++++++++++++++++++++++++++++++ + lib/gnulib.mk | 10 ++++++++++ + m4/copy-file-range.m4 | 36 ++++++++++++++++++++++++++++++++++++ + m4/gnulib-comp.m4 | 8 ++++++++ + 6 files changed, 91 insertions(+) + create mode 100644 lib/copy-file-range.c + create mode 100644 m4/copy-file-range.m4 + +diff --git a/aclocal.m4 b/aclocal.m4 +index 713f7c5..09a7ea8 100644 +--- a/aclocal.m4 ++++ b/aclocal.m4 +@@ -1165,6 +1165,7 @@ m4_include([m4/closedir.m4]) + m4_include([m4/codeset.m4]) + m4_include([m4/config-h.m4]) + m4_include([m4/configmake.m4]) ++m4_include([m4/copy-file-range.m4]) + m4_include([m4/ctype.m4]) + m4_include([m4/cycle-check.m4]) + m4_include([m4/d-ino.m4]) +diff --git a/lib/config.hin b/lib/config.hin +index 9769c39..bf9f9f8 100644 +--- a/lib/config.hin ++++ b/lib/config.hin +@@ -370,6 +370,9 @@ + /* Define to 1 when the gnulib module connect should be tested. */ + #undef GNULIB_TEST_CONNECT + ++/* Define to 1 when the gnulib module copy-file-range should be tested. */ ++#undef GNULIB_TEST_COPY_FILE_RANGE ++ + /* Define to 1 when the gnulib module dirfd should be tested. */ + #undef GNULIB_TEST_DIRFD + +diff --git a/lib/copy-file-range.c b/lib/copy-file-range.c +new file mode 100644 +index 0000000..069f144 +--- /dev/null ++++ b/lib/copy-file-range.c +@@ -0,0 +1,33 @@ ++/* Stub for copy_file_range ++ Copyright 2019-2020 Free Software Foundation, Inc. ++ ++ This program is free software: you can redistribute it and/or modify ++ it under the terms of the GNU General Public License as published by ++ the Free Software Foundation; either version 3 of the License, or ++ (at your option) any later version. ++ ++ This program is distributed in the hope that it will be useful, ++ but WITHOUT ANY WARRANTY; without even the implied warranty of ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ GNU General Public License for more details. ++ ++ You should have received a copy of the GNU General Public License ++ along with this program. If not, see . */ ++ ++#include ++ ++#include ++ ++#include ++ ++ssize_t ++copy_file_range (int infd, off_t *pinoff, ++ int outfd, off_t *poutoff, ++ size_t length, unsigned int flags) ++{ ++ /* There is little need to emulate copy_file_range with read+write, ++ since programs that use copy_file_range must fall back on ++ read+write anyway. */ ++ errno = ENOSYS; ++ return -1; ++} +diff --git a/lib/gnulib.mk b/lib/gnulib.mk +index b3633b8..86829f3 100644 +--- a/lib/gnulib.mk ++++ b/lib/gnulib.mk +@@ -65,6 +65,7 @@ + # closeout \ + # config-h \ + # configmake \ ++# copy-file-range \ + # crypto/md5 \ + # crypto/sha1 \ + # crypto/sha256 \ +@@ -800,6 +801,15 @@ CLEANFILES += lib/configmake.h lib/configmake.h-t + + ## end gnulib module configmake + ++## begin gnulib module copy-file-range ++ ++ ++EXTRA_DIST += lib/copy-file-range.c ++ ++EXTRA_lib_libcoreutils_a_SOURCES += lib/copy-file-range.c ++ ++## end gnulib module copy-file-range ++ + ## begin gnulib module count-leading-zeros + + lib_libcoreutils_a_SOURCES += lib/count-leading-zeros.c +diff --git a/m4/copy-file-range.m4 b/m4/copy-file-range.m4 +new file mode 100644 +index 0000000..5c5a274 +--- /dev/null ++++ b/m4/copy-file-range.m4 +@@ -0,0 +1,36 @@ ++# copy-file-range.m4 ++dnl Copyright 2019-2020 Free Software Foundation, Inc. ++dnl This file is free software; the Free Software Foundation ++dnl gives unlimited permission to copy and/or distribute it, ++dnl with or without modifications, as long as this notice is preserved. ++ ++AC_DEFUN([gl_FUNC_COPY_FILE_RANGE], ++[ ++ AC_REQUIRE([gl_UNISTD_H_DEFAULTS]) ++ ++ dnl Persuade glibc to declare copy_file_range. ++ AC_REQUIRE([AC_USE_SYSTEM_EXTENSIONS]) ++ ++ dnl Use AC_LINK_IFELSE, rather than AC_CHECK_FUNCS or a variant, ++ dnl since we don't want AC_CHECK_FUNCS's checks for glibc stubs. ++ dnl Programs that use copy_file_range must fall back on read+write ++ dnl anyway, and there's little point to substituting the Gnulib stub ++ dnl for a glibc stub. ++ AC_CACHE_CHECK([for copy_file_range], [gl_cv_func_copy_file_range], ++ [AC_LINK_IFELSE( ++ [AC_LANG_PROGRAM( ++ [[#include ++ ]], ++ [[ssize_t (*func) (int, off_t *, int, off_t, size_t, unsigned) ++ = copy_file_range; ++ return func (0, 0, 0, 0, 0, 0) & 127; ++ ]]) ++ ], ++ [gl_cv_func_copy_file_range=yes], ++ [gl_cv_func_copy_file_range=no]) ++ ]) ++ ++ if test "$gl_cv_func_copy_file_range" != yes; then ++ HAVE_COPY_FILE_RANGE=0 ++ fi ++]) +diff --git a/m4/gnulib-comp.m4 b/m4/gnulib-comp.m4 +index dead90e..953e7f0 100644 +--- a/m4/gnulib-comp.m4 ++++ b/m4/gnulib-comp.m4 +@@ -129,6 +129,7 @@ AC_DEFUN([gl_EARLY], + # Code from module configmake: + # Code from module connect: + # Code from module connect-tests: ++ # Code from module copy-file-range: + # Code from module count-leading-zeros: + # Code from module count-leading-zeros-tests: + # Code from module crypto/af_alg: +@@ -977,6 +978,11 @@ AC_DEFUN([gl_INIT], + gl_DIRENT_MODULE_INDICATOR([closedir]) + gl_CONFIG_H + gl_CONFIGMAKE_PREP ++ gl_FUNC_COPY_FILE_RANGE ++ if test $HAVE_COPY_FILE_RANGE = 0; then ++ AC_LIBOBJ([copy-file-range]) ++ fi ++ gl_UNISTD_MODULE_INDICATOR([copy-file-range]) + gl_AF_ALG + AC_DEFINE([GL_COMPILE_CRYPTO_STREAM], 1, [Compile Gnulib crypto stream ops.]) + AC_REQUIRE([AC_C_RESTRICT]) +@@ -2746,6 +2752,7 @@ AC_DEFUN([gl_FILE_LIST], [ + lib/closeout.c + lib/closeout.h + lib/copy-acl.c ++ lib/copy-file-range.c + lib/count-leading-zeros.c + lib/count-leading-zeros.h + lib/creat-safer.c +@@ -3438,6 +3445,7 @@ AC_DEFUN([gl_FILE_LIST], [ + m4/codeset.m4 + m4/config-h.m4 + m4/configmake.m4 ++ m4/copy-file-range.m4 + m4/ctype.m4 + m4/cycle-check.m4 + m4/d-ino.m4 +-- +2.26.3 + + +From 48370c95bcf7c25ce021fbd2145062d3d29ae6d5 Mon Sep 17 00:00:00 2001 +From: Paul Eggert +Date: Thu, 25 Jun 2020 17:34:23 -0700 +Subject: [PATCH 06/12] cp: use copy_file_range if available + +* NEWS: Mention this. +* bootstrap.conf (gnulib_modules): Add copy-file-range. +* src/copy.c (sparse_copy): Try copy_file_range if not +looking for holes. + +Upstream-commit: 4b04a0c3b792d27909670a81d21f2a3b3e0ea563 +Signed-off-by: Kamil Dudka +--- + bootstrap.conf | 1 + + src/copy.c | 40 ++++++++++++++++++++++++++++++++++++++++ + 2 files changed, 41 insertions(+) + +diff --git a/bootstrap.conf b/bootstrap.conf +index 2a342c1..7d53e28 100644 +--- a/bootstrap.conf ++++ b/bootstrap.conf +@@ -54,6 +54,7 @@ gnulib_modules=" + closeout + config-h + configmake ++ copy-file-range + crypto/md5 + crypto/sha1 + crypto/sha256 +diff --git a/src/copy.c b/src/copy.c +index d88f8cf..4050f69 100644 +--- a/src/copy.c ++++ b/src/copy.c +@@ -265,6 +265,46 @@ sparse_copy (int src_fd, int dest_fd, char *buf, size_t buf_size, + { + *last_write_made_hole = false; + *total_n_read = 0; ++ ++ /* If not looking for holes, use copy_file_range if available. */ ++ if (!hole_size) ++ while (max_n_read) ++ { ++ /* Copy at most COPY_MAX bytes at a time; this is min ++ (PTRDIFF_MAX, SIZE_MAX) truncated to a value that is ++ surely aligned well. */ ++ ssize_t ssize_max = TYPE_MAXIMUM (ssize_t); ++ ptrdiff_t copy_max = MIN (ssize_max, SIZE_MAX) >> 30 << 30; ++ ssize_t n_copied = copy_file_range (src_fd, NULL, dest_fd, NULL, ++ MIN (max_n_read, copy_max), 0); ++ if (n_copied == 0) ++ { ++ /* copy_file_range incorrectly returns 0 when reading from ++ the proc file system on the Linux kernel through at ++ least 5.6.19 (2020), so fall back on 'read' if the ++ input file seems empty. */ ++ if (*total_n_read == 0) ++ break; ++ return true; ++ } ++ if (n_copied < 0) ++ { ++ if (errno == ENOSYS || errno == EINVAL ++ || errno == EBADF || errno == EXDEV) ++ break; ++ if (errno == EINTR) ++ n_copied = 0; ++ else ++ { ++ error (0, errno, _("error copying %s to %s"), ++ quoteaf_n (0, src_name), quoteaf_n (1, dst_name)); ++ return false; ++ } ++ } ++ max_n_read -= n_copied; ++ *total_n_read += n_copied; ++ } ++ + bool make_hole = false; + off_t psize = 0; + +-- +2.26.3 + + +From 23ea1ba463d33e268f35847059e637a5935e4581 Mon Sep 17 00:00:00 2001 +From: Zorro Lang +Date: Mon, 26 Apr 2021 17:25:18 +0200 +Subject: [PATCH 07/12] copy: do not refuse to copy a swap file + +* src/copy.c (sparse_copy): Fallback to read() if copy_file_range() +fails with ETXTBSY. Otherwise it would be impossible to copy files +that are being used as swap. This used to work before introducing +the support for copy_file_range() in coreutils. (Bug#48036) + +Upstream-commit: 785478013b416cde50794be35475c0c4fdbb48b4 +Signed-off-by: Kamil Dudka +--- + src/copy.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/src/copy.c b/src/copy.c +index 4050f69..1798bb7 100644 +--- a/src/copy.c ++++ b/src/copy.c +@@ -290,7 +290,7 @@ sparse_copy (int src_fd, int dest_fd, char *buf, size_t buf_size, + if (n_copied < 0) + { + if (errno == ENOSYS || errno == EINVAL +- || errno == EBADF || errno == EXDEV) ++ || errno == EBADF || errno == EXDEV || errno == ETXTBSY) + break; + if (errno == EINTR) + n_copied = 0; +-- +2.31.1 + + +From cd7c7a6b5ad89ef0a61722552d532901fc1bed05 Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?P=C3=A1draig=20Brady?= +Date: Sun, 2 May 2021 21:27:17 +0100 +Subject: [PATCH 08/12] copy: ensure we enforce --reflink=never + +* src/copy.c (sparse_copy): Don't use copy_file_range() +with --reflink=never as copy_file_range() may implicitly +use acceleration techniques like reflinking. +(extent_copy): Pass through whether we allow reflinking. +(lseek_copy): Likewise. +Fixes https://bugs.gnu.org/48164 + +Upstream-commit: ea9af99234031ab8d5169c8a669434e2a6b4f864 +Signed-off-by: Kamil Dudka +--- + src/copy.c | 20 +++++++++++++------- + 1 file changed, 13 insertions(+), 7 deletions(-) + +diff --git a/src/copy.c b/src/copy.c +index 4050f69..0337538 100644 +--- a/src/copy.c ++++ b/src/copy.c +@@ -258,7 +258,7 @@ create_hole (int fd, char const *name, bool punch_holes, off_t size) + bytes read. */ + static bool + sparse_copy (int src_fd, int dest_fd, char *buf, size_t buf_size, +- size_t hole_size, bool punch_holes, ++ size_t hole_size, bool punch_holes, bool allow_reflink, + char const *src_name, char const *dst_name, + uintmax_t max_n_read, off_t *total_n_read, + bool *last_write_made_hole) +@@ -266,8 +266,9 @@ sparse_copy (int src_fd, int dest_fd, char *buf, size_t buf_size, + *last_write_made_hole = false; + *total_n_read = 0; + +- /* If not looking for holes, use copy_file_range if available. */ +- if (!hole_size) ++ /* If not looking for holes, use copy_file_range if available, ++ but don't use if reflink disallowed as that may be implicit. */ ++ if ((! hole_size) && allow_reflink) + while (max_n_read) + { + /* Copy at most COPY_MAX bytes at a time; this is min +@@ -466,6 +467,7 @@ static bool + extent_copy (int src_fd, int dest_fd, char *buf, size_t buf_size, + size_t hole_size, off_t src_total_size, + enum Sparse_type sparse_mode, ++ bool allow_reflink, + char const *src_name, char const *dst_name, + struct extent_scan *scan) + { +@@ -579,8 +581,8 @@ extent_copy (int src_fd, int dest_fd, char *buf, size_t buf_size, + + if ( ! sparse_copy (src_fd, dest_fd, buf, buf_size, + sparse_mode == SPARSE_ALWAYS ? hole_size: 0, +- true, src_name, dst_name, ext_len, &n_read, +- &read_hole)) ++ true, allow_reflink, src_name, dst_name, ++ ext_len, &n_read, &read_hole)) + goto fail; + + dest_pos = ext_start + n_read; +@@ -655,6 +657,7 @@ static bool + lseek_copy (int src_fd, int dest_fd, char *buf, size_t buf_size, + size_t hole_size, off_t ext_start, off_t src_total_size, + enum Sparse_type sparse_mode, ++ bool allow_reflink, + char const *src_name, char const *dst_name) + { + off_t last_ext_start = 0; +@@ -729,8 +732,8 @@ lseek_copy (int src_fd, int dest_fd, char *buf, size_t buf_size, + bool read_hole; + if ( ! sparse_copy (src_fd, dest_fd, buf, buf_size, + sparse_mode == SPARSE_NEVER ? 0 : hole_size, +- true, src_name, dst_name, ext_len, &n_read, +- &read_hole)) ++ true, allow_reflink, src_name, dst_name, ++ ext_len, &n_read, &read_hole)) + return false; + + dest_pos = ext_start + n_read; +@@ -1527,17 +1530,20 @@ copy_reg (char const *src_name, char const *dst_name, + ? extent_copy (source_desc, dest_desc, buf, buf_size, hole_size, + src_open_sb.st_size, + make_holes ? x->sparse_mode : SPARSE_NEVER, ++ x->reflink_mode != REFLINK_NEVER, + src_name, dst_name, &scan_inference.extent_scan) + #ifdef SEEK_HOLE + : scantype == LSEEK_SCANTYPE + ? lseek_copy (source_desc, dest_desc, buf, buf_size, hole_size, + scan_inference.ext_start, src_open_sb.st_size, + make_holes ? x->sparse_mode : SPARSE_NEVER, ++ x->reflink_mode != REFLINK_NEVER, + src_name, dst_name) + #endif + : sparse_copy (source_desc, dest_desc, buf, buf_size, + make_holes ? hole_size : 0, + x->sparse_mode == SPARSE_ALWAYS, ++ x->reflink_mode != REFLINK_NEVER, + src_name, dst_name, UINTMAX_MAX, &n_read, + &wrote_hole_at_eof))) + { +-- +2.30.2 + + +From 7978f1de88dcdb17b67db9268038930e9c71154f Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?P=C3=A1draig=20Brady?= +Date: Sat, 8 May 2021 17:18:54 +0100 +Subject: [PATCH 09/12] copy: handle ENOTSUP from copy_file_range() + +* src/copy.c (sparse_copy): Ensure we fall back to +a standard copy if copy_file_range() returns ENOTSUP. +This generally is best checked when checking ENOSYS, +but it also seems to be a practical concern on Centos 7, +as a quick search gave https://bugzilla.redhat.com/1840284 + +Upstream-commit: 8ec0d1799e19a079b8a661c6bb69f6c58e52f1aa +Signed-off-by: Kamil Dudka +--- + src/copy.c | 5 +++-- + 1 file changed, 3 insertions(+), 2 deletions(-) + +diff --git a/src/copy.c b/src/copy.c +index 9977193..e3977cd 100644 +--- a/src/copy.c ++++ b/src/copy.c +@@ -290,8 +290,9 @@ sparse_copy (int src_fd, int dest_fd, char *buf, size_t buf_size, + } + if (n_copied < 0) + { +- if (errno == ENOSYS || errno == EINVAL +- || errno == EBADF || errno == EXDEV || errno == ETXTBSY) ++ if (errno == ENOSYS || is_ENOTSUP (errno) ++ || errno == EINVAL || errno == EBADF ++ || errno == EXDEV || errno == ETXTBSY) + break; + if (errno == EINTR) + n_copied = 0; +-- +2.31.1 + + +From d8d3edbfc13ff13c185f23436209b788f906aa41 Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?P=C3=A1draig=20Brady?= +Date: Sun, 9 May 2021 21:55:22 +0100 +Subject: [PATCH 10/12] copy: handle EOPNOTSUPP from SEEK_DATA + +* src/copy.c (infer_scantype): Ensure we don't error out +if SEEK_DATA returns EOPNOTSUPP, on systems where this value +is distinct from ENOTSUP. Generally both of these should be checked. + +Upstream-commit: 017877bd088284d515753d78b81ca6e6a88c1350 +Signed-off-by: Kamil Dudka +--- + src/copy.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/src/copy.c b/src/copy.c +index e3977cd..de8030d 100644 +--- a/src/copy.c ++++ b/src/copy.c +@@ -1246,7 +1246,7 @@ infer_scantype (int fd, struct stat const *sb, + scan_inference->ext_start = lseek (fd, 0, SEEK_DATA); + if (0 <= scan_inference->ext_start) + return LSEEK_SCANTYPE; +- else if (errno != EINVAL && errno != ENOTSUP) ++ else if (errno != EINVAL && !is_ENOTSUP (errno)) + return errno == ENXIO ? LSEEK_SCANTYPE : ERROR_SCANTYPE; + #endif + +-- +2.31.1 + + +From 1daf8c0fc9a5766c22b7ea84bea8c88c86a0c495 Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?P=C3=A1draig=20Brady?= +Date: Sat, 8 May 2021 19:23:20 +0100 +Subject: [PATCH 11/12] copy: handle system security config issues with + copy_file_range() + +* src/copy.c (sparse_copy): Upon EPERM from copy_file_range(), +fall back to a standard copy, which will give a more accurate +error as to whether the issue is with the source or destination. +Also this will avoid the issue where seccomp or apparmor are +not configured to handle copy_file_range(), in which case +the fall back standard copy would succeed without issue. +This specific issue with seccomp was noticed for example in: +https://github.com/golang/go/issues/40900 + +Upstream-commit: 2e66e1732fced7af20fa76c60e636d39a1767d48 +Signed-off-by: Kamil Dudka +--- + src/copy.c | 9 +++++++++ + 1 file changed, 9 insertions(+) + +diff --git a/src/copy.c b/src/copy.c +index de8030d..62eec7b 100644 +--- a/src/copy.c ++++ b/src/copy.c +@@ -294,6 +294,15 @@ sparse_copy (int src_fd, int dest_fd, char *buf, size_t buf_size, + || errno == EINVAL || errno == EBADF + || errno == EXDEV || errno == ETXTBSY) + break; ++ ++ /* copy_file_range might not be enabled in seccomp filters, ++ so retry with a standard copy. EPERM can also occur ++ for immutable files, but that would only be in the edge case ++ where the file is made immutable after creating/truncating, ++ in which case the (more accurate) error is still shown. */ ++ if (errno == EPERM && *total_n_read == 0) ++ break; ++ + if (errno == EINTR) + n_copied = 0; + else +-- +2.31.1 + + +From 42c9e598f61ba6bc27a615e39e40023a676a523b Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?P=C3=A1draig=20Brady?= +Date: Wed, 12 May 2021 23:47:38 +0100 +Subject: [PATCH 12/12] copy: disallow copy_file_range() on Linux kernels + before 5.3 + +copy_file_range() before Linux kernel release 5.3 had many issues, +as described at https://lwn.net/Articles/789527/, which was +referenced from https://lwn.net/Articles/846403/; a more general +article discussing the generality of copy_file_range(). +Linux kernel 5.3 was released in September 2019, which is new enough +that we need to actively avoid older kernels. + +* src/copy.c (functional_copy_file_range): A new function +that returns false for Linux kernels before version 5.3. +(sparse_copy): Call this new function to gate use of +copy_file_range(). + +Upstream-commit: ba5e6885d2c255648cddb87b4e795659c1990374 +Signed-off-by: Kamil Dudka +--- + src/copy.c | 47 +++++++++++++++++++++++++++++++++++++++++++++-- + 1 file changed, 45 insertions(+), 2 deletions(-) + +diff --git a/src/copy.c b/src/copy.c +index 62eec7b..2e1699b 100644 +--- a/src/copy.c ++++ b/src/copy.c +@@ -21,6 +21,7 @@ + #include + #include + #include ++#include + #include + + #if HAVE_HURD_H +@@ -64,6 +65,7 @@ + #include "write-any-file.h" + #include "areadlink.h" + #include "yesno.h" ++#include "xstrtol.h" + #include "selinux.h" + + #if USE_XATTR +@@ -244,6 +246,47 @@ create_hole (int fd, char const *name, bool punch_holes, off_t size) + return true; + } + ++/* copy_file_range() before Linux kernel release 5.3 had many issues, ++ as described at https://lwn.net/Articles/789527/, ++ so return FALSE for Linux kernels earlier than that. ++ This function can be removed when such kernels (released before Sep 2019) ++ are no longer a consideration. */ ++ ++static bool ++functional_copy_file_range (void) ++{ ++#ifdef __linux__ ++ static int version_allowed = -1; ++ ++ if (version_allowed == -1) ++ version_allowed = 0; ++ else ++ return version_allowed; ++ ++ struct utsname name; ++ if (uname (&name) == -1) ++ return version_allowed; ++ ++ char *p = name.release; ++ uintmax_t ver[2] = {0, 0}; ++ size_t iver = 0; ++ ++ do ++ { ++ strtol_error err = xstrtoumax (p, &p, 10, &ver[iver], NULL); ++ if (err != LONGINT_OK || *p++ != '.') ++ break; ++ } ++ while (++iver < ARRAY_CARDINALITY (ver)); ++ ++ version_allowed = (ver[0] > 5 || (ver[0] == 5 && ver[1] >= 3)); ++ ++ return version_allowed; ++#else ++ return true; ++#endif ++ ++} + + /* Copy the regular file open on SRC_FD/SRC_NAME to DST_FD/DST_NAME, + honoring the MAKE_HOLES setting and using the BUF_SIZE-byte buffer +@@ -266,9 +309,9 @@ sparse_copy (int src_fd, int dest_fd, char *buf, size_t buf_size, + *last_write_made_hole = false; + *total_n_read = 0; + +- /* If not looking for holes, use copy_file_range if available, ++ /* If not looking for holes, use copy_file_range if functional, + but don't use if reflink disallowed as that may be implicit. */ +- if ((! hole_size) && allow_reflink) ++ if ((! hole_size) && allow_reflink && functional_copy_file_range ()) + while (max_n_read) + { + /* Copy at most COPY_MAX bytes at a time; this is min +-- +2.31.1 + diff --git a/coreutils-8.32-cp-reflink-auto.patch b/coreutils-8.32-cp-reflink-auto.patch new file mode 100644 index 0000000..a36c89c --- /dev/null +++ b/coreutils-8.32-cp-reflink-auto.patch @@ -0,0 +1,119 @@ +From 76126e2831580d0df20530f4d6f72189bd4f0b9a Mon Sep 17 00:00:00 2001 +From: Paul Eggert +Date: Thu, 18 Jun 2020 22:16:24 -0700 +Subject: [PATCH] cp: default to COW +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +Likewise for ‘install’. Proposed in Bug#24400, and long past due. +* NEWS: +* doc/coreutils.texi (cp invocation): +* src/copy.h (enum Reflink_type): Document this. +* src/cp.c (cp_option_init): +* src/install.c (cp_option_init): Implement this. + +Upstream-commit: 25725f9d41735d176d73a757430739fb71c7d043 +Signed-off-by: Kamil Dudka +--- + doc/coreutils.texi | 19 ++++++++++++------- + src/copy.h | 4 ++-- + src/cp.c | 2 +- + src/install.c | 2 +- + 4 files changed, 16 insertions(+), 11 deletions(-) + +diff --git a/doc/coreutils.texi b/doc/coreutils.texi +index 02e0c1c..2382a16 100644 +--- a/doc/coreutils.texi ++++ b/doc/coreutils.texi +@@ -8854,12 +8854,14 @@ The @var{when} value can be one of the following: + + @table @samp + @item always +-The default behavior: if the copy-on-write operation is not supported ++If the copy-on-write operation is not supported + then report the failure for each file and exit with a failure status. ++Plain @option{--reflink} is equivalent to @option{--reflink=when}. + + @item auto + If the copy-on-write operation is not supported then fall back + to the standard copy behavior. ++This is the default if no @option{--reflink} option is given. + + @item never + Disable copy-on-write operation and use the standard copy behavior. +@@ -8868,12 +8870,6 @@ Disable copy-on-write operation and use the standard copy behavior. + This option is overridden by the @option{--link}, @option{--symbolic-link} + and @option{--attributes-only} options, thus allowing it to be used + to configure the default data copying behavior for @command{cp}. +-For example, with the following alias, @command{cp} will use the +-minimum amount of space supported by the file system. +- +-@example +-alias cp='cp --reflink=auto --sparse=always' +-@end example + + @item --remove-destination + @opindex --remove-destination +@@ -8918,6 +8914,15 @@ This is useful in creating a file for use with the @command{mkswap} command, + since such a file must not have any holes. + @end table + ++For example, with the following alias, @command{cp} will use the ++minimum amount of space supported by the file system. ++(Older versions of @command{cp} can also benefit from ++@option{--reflink=auto} here.) ++ ++@example ++alias cp='cp --sparse=always' ++@end example ++ + @optStripTrailingSlashes + + @item -s +diff --git a/src/copy.h b/src/copy.h +index 874d6f7..a0ad494 100644 +--- a/src/copy.h ++++ b/src/copy.h +@@ -46,10 +46,10 @@ enum Sparse_type + /* Control creation of COW files. */ + enum Reflink_type + { +- /* Default to a standard copy. */ ++ /* Do a standard copy. */ + REFLINK_NEVER, + +- /* Try a COW copy and fall back to a standard copy. */ ++ /* Try a COW copy and fall back to a standard copy; this is the default. */ + REFLINK_AUTO, + + /* Require a COW copy and fail if not available. */ +diff --git a/src/cp.c b/src/cp.c +index 0193df8..9e7ad14 100644 +--- a/src/cp.c ++++ b/src/cp.c +@@ -796,7 +796,7 @@ cp_option_init (struct cp_options *x) + x->move_mode = false; + x->install_mode = false; + x->one_file_system = false; +- x->reflink_mode = REFLINK_NEVER; ++ x->reflink_mode = REFLINK_AUTO; + + x->preserve_ownership = false; + x->preserve_links = false; +diff --git a/src/install.c b/src/install.c +index 4ab44a6..aef16ca 100644 +--- a/src/install.c ++++ b/src/install.c +@@ -264,7 +264,7 @@ cp_option_init (struct cp_options *x) + { + cp_options_default (x); + x->copy_as_regular = true; +- x->reflink_mode = REFLINK_NEVER; ++ x->reflink_mode = REFLINK_AUTO; + x->dereference = DEREF_ALWAYS; + x->unlink_dest_before_opening = true; + x->unlink_dest_after_failed_open = false; +-- +2.25.4 + diff --git a/coreutils-8.32-df-duplicated-entries.patch b/coreutils-8.32-df-duplicated-entries.patch new file mode 100644 index 0000000..5d39c1e --- /dev/null +++ b/coreutils-8.32-df-duplicated-entries.patch @@ -0,0 +1,72 @@ +From 0f053de4bc3ca0cfd88a42d236881dfdddb10ee9 Mon Sep 17 00:00:00 2001 +From: Kamil Dudka +Date: Wed, 30 Jun 2021 17:53:22 +0200 +Subject: [PATCH] df: fix duplicated remote entries due to bind mounts + +As originally reported in , +df invoked without -a printed duplicated entries for NFS mounts +of bind mounts. This is a regression from commit v8.25-54-g1c17f61ef99, +which introduced the use of a hash table. + +The proposed patch makes sure that the devlist entry seen the last time +is used for comparison when eliminating duplicated mount entries. This +way it worked before introducing the hash table. + +Patch co-authored by Roberto Bergantinos. + +* src/ls.c (struct devlist): Introduce the seen_last pointer. +(devlist_for_dev): Return the devlist entry seen the last time if found. +(filter_mount_list): Remember the devlist entry seen the last time for +each hashed item. +Fixes https://bugs.gnu.org/49298 + +Upstream-commit: d6125af095c9553f38cba0696f15158f5abe4ecc +Signed-off-by: Kamil Dudka +--- + src/df.c | 15 +++++++++++++-- + 1 file changed, 13 insertions(+), 2 deletions(-) + +diff --git a/src/df.c b/src/df.c +index 7e01839..3e9247f 100644 +--- a/src/df.c ++++ b/src/df.c +@@ -54,6 +54,7 @@ struct devlist + dev_t dev_num; + struct mount_entry *me; + struct devlist *next; ++ struct devlist *seen_last; /* valid for hashed devlist entries only */ + }; + + /* Filled with device numbers of examined file systems to avoid +@@ -689,7 +690,13 @@ devlist_for_dev (dev_t dev) + return NULL; + struct devlist dev_entry; + dev_entry.dev_num = dev; +- return hash_lookup (devlist_table, &dev_entry); ++ ++ struct devlist *found = hash_lookup (devlist_table, &dev_entry); ++ if (found == NULL) ++ return NULL; ++ ++ /* Return the last devlist entry we have seen with this dev_num */ ++ return found->seen_last; + } + + static void +@@ -807,8 +814,12 @@ filter_mount_list (bool devices_only) + devlist->dev_num = buf.st_dev; + devlist->next = device_list; + device_list = devlist; +- if (hash_insert (devlist_table, devlist) == NULL) ++ ++ struct devlist *hash_entry = hash_insert (devlist_table, devlist); ++ if (hash_entry == NULL) + xalloc_die (); ++ /* Ensure lookups use this latest devlist. */ ++ hash_entry->seen_last = devlist; + + me = me->me_next; + } +-- +2.31.1 + diff --git a/coreutils-8.32-expr-unmatched-par.patch b/coreutils-8.32-expr-unmatched-par.patch new file mode 100644 index 0000000..1a82384 --- /dev/null +++ b/coreutils-8.32-expr-unmatched-par.patch @@ -0,0 +1,81 @@ +From 9618fb718b75920f37e5be2049ad1d0bb5c4a28c Mon Sep 17 00:00:00 2001 +From: Paul Eggert +Date: Tue, 26 Jan 2021 09:23:54 -0800 +Subject: [PATCH] expr: fix bug with unmatched \(...\) + +Problem reported by Qiuhao Li. +* doc/coreutils.texi (String expressions): +Document the correct behavior, which POSIX requires. +* src/expr.c (docolon): Treat unmatched \(...\) as empty. +* tests/misc/expr.pl: New test. + +Upstream-commit: 735083ba24878075235007b4417982ad5700436d +Signed-off-by: Kamil Dudka +--- + doc/coreutils.texi | 14 ++++++++------ + src/expr.c | 9 +++++++-- + tests/misc/expr.pl | 3 +++ + 3 files changed, 18 insertions(+), 8 deletions(-) + +diff --git a/doc/coreutils.texi b/doc/coreutils.texi +index 2382a16..5b2bb2c 100644 +--- a/doc/coreutils.texi ++++ b/doc/coreutils.texi +@@ -13529,12 +13529,14 @@ second is considered to be a (basic, a la GNU @code{grep}) regular + expression, with a @code{^} implicitly prepended. The first argument is + then matched against this regular expression. + +-If the match succeeds and @var{regex} uses @samp{\(} and @samp{\)}, the +-@code{:} expression returns the part of @var{string} that matched the +-subexpression; otherwise, it returns the number of characters matched. +- +-If the match fails, the @code{:} operator returns the null string if +-@samp{\(} and @samp{\)} are used in @var{regex}, otherwise 0. ++If @var{regex} does not use @samp{\(} and @samp{\)}, the @code{:} ++expression returns the number of characters matched, or 0 if the match ++fails. ++ ++If @var{regex} uses @samp{\(} and @samp{\)}, the @code{:} expression ++returns the part of @var{string} that matched the subexpression, or ++the null string if the match failed or the subexpression did not ++contribute to the match. + + @kindex \( @r{regexp operator} + Only the first @samp{\( @dots{} \)} pair is relevant to the return +diff --git a/src/expr.c b/src/expr.c +index e134872..0616a42 100644 +--- a/src/expr.c ++++ b/src/expr.c +@@ -721,8 +721,13 @@ docolon (VALUE *sv, VALUE *pv) + /* Were \(...\) used? */ + if (re_buffer.re_nsub > 0) + { +- sv->u.s[re_regs.end[1]] = '\0'; +- v = str_value (sv->u.s + re_regs.start[1]); ++ if (re_regs.end[1] < 0) ++ v = str_value (""); ++ else ++ { ++ sv->u.s[re_regs.end[1]] = '\0'; ++ v = str_value (sv->u.s + re_regs.start[1]); ++ } + } + else + { +diff --git a/tests/misc/expr.pl b/tests/misc/expr.pl +index e45f8e7..e57f79d 100755 +--- a/tests/misc/expr.pl ++++ b/tests/misc/expr.pl +@@ -84,6 +84,9 @@ my @Tests = + # In 5.94 and earlier, anchors incorrectly matched newlines. + ['anchor', "'a\nb' : 'a\$'", {OUT => '0'}, {EXIT => 1}], + ++ # In 8.32, \( ... \) that did not match caused memory errors. ++ ['emptysub', '"a" : "\\(b\\)*"', {OUT => ''}, {EXIT => 1}], ++ + # These tests are taken from grep/tests/bre.tests. + ['bre1', '"abc" : "a\\(b\\)c"', {OUT => 'b'}], + ['bre2', '"a(" : "a("', {OUT => '2'}], +-- +2.26.2 + diff --git a/coreutils-8.32-fuse-portal.patch b/coreutils-8.32-fuse-portal.patch new file mode 100644 index 0000000..54b7975 --- /dev/null +++ b/coreutils-8.32-fuse-portal.patch @@ -0,0 +1,38 @@ +From 602fb566468d3837b7871c17a0fab1a20228d119 Mon Sep 17 00:00:00 2001 +From: Kamil Dudka +Date: Mon, 7 Jun 2021 14:43:03 +0200 +Subject: [PATCH] mountlist: recognize fuse.portal as dummy file system + +This was originally proposed at: + + https://lists.gnu.org/archive/html/bug-gnulib/2021-02/msg00053.html + +As the full review might take some time, would it be possible to apply +at least the part related to fuse.portal file systems? They started to +cause problems recently: + + https://bugs.launchpad.net/ubuntu/+source/xdg-desktop-portal/+bug/1905623 + https://github.com/muesli/duf/issues/35 + https://bugzilla.redhat.com/1913358 + +Upstream-commit: 9a38d499ca16f2f4304992eb1ab0894cd0b478e1 +Signed-off-by: Kamil Dudka +--- + lib/mountlist.c | 1 + + 1 file changed, 1 insertion(+) + +diff --git a/lib/mountlist.c b/lib/mountlist.c +index e0227b7..e5f6b07 100644 +--- a/lib/mountlist.c ++++ b/lib/mountlist.c +@@ -170,6 +170,7 @@ + || strcmp (Fs_type, "debugfs") == 0 \ + || strcmp (Fs_type, "devpts") == 0 \ + || strcmp (Fs_type, "fusectl") == 0 \ ++ || strcmp (Fs_type, "fuse.portal") == 0 \ + || strcmp (Fs_type, "mqueue") == 0 \ + || strcmp (Fs_type, "rpc_pipefs") == 0 \ + || strcmp (Fs_type, "sysfs") == 0 \ +-- +2.31.1 + diff --git a/coreutils-8.32-gnulib-perror-test.patch b/coreutils-8.32-gnulib-perror-test.patch new file mode 100644 index 0000000..e545290 --- /dev/null +++ b/coreutils-8.32-gnulib-perror-test.patch @@ -0,0 +1,48 @@ +From f61085aaa37f169365c56e44f5129d0491913b6a Mon Sep 17 00:00:00 2001 +From: Paul Eggert +Date: Thu, 27 Aug 2020 17:52:58 -0700 +Subject: [PATCH] perror, strerror_r: remove unportable tests + +Problem reported by Florian Weimer in: +https://lists.gnu.org/r/bug-gnulib/2020-08/msg00220.html +* tests/test-perror2.c (main): +* tests/test-strerror_r.c (main): Omit unportable tests. + +Upstream-commit: 175e0bc72808d564074c4adcc72aeadb74adfcc6 +Signed-off-by: Kamil Dudka +--- + gnulib-tests/test-perror2.c | 3 --- + gnulib-tests/test-strerror_r.c | 3 --- + 2 files changed, 6 deletions(-) + +diff --git a/gnulib-tests/test-perror2.c b/gnulib-tests/test-perror2.c +index 1d14eda..c6214dd 100644 +--- a/gnulib-tests/test-perror2.c ++++ b/gnulib-tests/test-perror2.c +@@ -79,9 +79,6 @@ main (void) + errno = -5; + perror (""); + ASSERT (!ferror (stderr)); +- ASSERT (msg1 == msg2 || msg1 == msg4 || STREQ (msg1, str1)); +- ASSERT (msg2 == msg4 || STREQ (msg2, str2)); +- ASSERT (msg3 == msg4 || STREQ (msg3, str3)); + ASSERT (STREQ (msg4, str4)); + + free (str1); +diff --git a/gnulib-tests/test-strerror_r.c b/gnulib-tests/test-strerror_r.c +index b11d6fd..c1dbcf8 100644 +--- a/gnulib-tests/test-strerror_r.c ++++ b/gnulib-tests/test-strerror_r.c +@@ -165,9 +165,6 @@ main (void) + + strerror_r (EACCES, buf, sizeof buf); + strerror_r (-5, buf, sizeof buf); +- ASSERT (msg1 == msg2 || msg1 == msg4 || STREQ (msg1, str1)); +- ASSERT (msg2 == msg4 || STREQ (msg2, str2)); +- ASSERT (msg3 == msg4 || STREQ (msg3, str3)); + ASSERT (STREQ (msg4, str4)); + + free (str1); +-- +2.25.4 + diff --git a/coreutils-8.32-leaf-opt-xfs.patch b/coreutils-8.32-leaf-opt-xfs.patch new file mode 100644 index 0000000..c02db71 --- /dev/null +++ b/coreutils-8.32-leaf-opt-xfs.patch @@ -0,0 +1,164 @@ +From b9f9ed14bda93ecb407129b69e6476813c250046 Mon Sep 17 00:00:00 2001 +From: Paul Eggert +Date: Wed, 15 Apr 2020 20:50:32 -0700 +Subject: [PATCH] fts: remove NOSTAT_LEAF_OPTIMIZATION +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +It caused ‘find’ and ‘du’ to dump core, and it was useful +only for obsolescent Linux filesystems anyway. Problem reported in: +https://lists.gnu.org/r/bug-gnulib/2020-04/msg00068.html +Quite possibly there is still a serious underlying fts bug with +tight-loop-check and mutating file systems, but if so this patch +should cause the bug to be triggered less often. +* lib/fts.c (enum leaf_optimization): Remove +NOSTAT_LEAF_OPTIMIZATION, as it’s problematic. +(S_MAGIC_REISERFS, S_MAGIC_XFS): Remove; no longer needed. +(leaf_optimization): Remove special cases for ReiserFS and XFS. +(fts_read): Remove NOSTAT_LEAF_OPTIMIZATION code. +* lib/fts_.h (struct _ftsent.fts_n_dirs_remaining): +Remove. All uses removed. + +Upstream-commit: 47bf2cf3184027c1eb9c1dfeea5c5b8b2d69710d +Signed-off-by: Kamil Dudka +--- + lib/fts.c | 56 ++++++++------------------------------------------- + lib/fts_.h | 5 ----- + 2 files changed, 8 insertions(+), 53 deletions(-) + +diff --git a/lib/fts.c b/lib/fts.c +index 1093ce5..dfe3fef 100644 +--- a/lib/fts.c ++++ b/lib/fts.c +@@ -445,7 +445,6 @@ fts_open (char * const *argv, + if ((parent = fts_alloc(sp, "", 0)) == NULL) + goto mem2; + parent->fts_level = FTS_ROOTPARENTLEVEL; +- parent->fts_n_dirs_remaining = -1; + } + + /* The classic fts implementation would call fts_stat with +@@ -634,9 +633,8 @@ fts_close (FTS *sp) + } + + /* Minimum link count of a traditional Unix directory. When leaf +- optimization is OK and MIN_DIR_NLINK <= st_nlink, then st_nlink is +- an upper bound on the number of subdirectories (counting "." and +- ".."). */ ++ optimization is OK and a directory's st_nlink == MIN_DIR_NLINK, ++ then the directory has no subdirectories. */ + enum { MIN_DIR_NLINK = 2 }; + + /* Whether leaf optimization is OK for a directory. */ +@@ -645,12 +643,8 @@ enum leaf_optimization + /* st_nlink is not reliable for this directory's subdirectories. */ + NO_LEAF_OPTIMIZATION, + +- /* Leaf optimization is OK, but is not useful for avoiding stat calls. */ +- OK_LEAF_OPTIMIZATION, +- +- /* Leaf optimization is not only OK: it is useful for avoiding +- stat calls, because dirent.d_type does not work. */ +- NOSTAT_LEAF_OPTIMIZATION ++ /* st_nlink == 2 means the directory lacks subdirectories. */ ++ OK_LEAF_OPTIMIZATION + }; + + #if (defined __linux__ || defined __ANDROID__) \ +@@ -663,9 +657,7 @@ enum leaf_optimization + # define S_MAGIC_CIFS 0xFF534D42 + # define S_MAGIC_NFS 0x6969 + # define S_MAGIC_PROC 0x9FA0 +-# define S_MAGIC_REISERFS 0x52654973 + # define S_MAGIC_TMPFS 0x1021994 +-# define S_MAGIC_XFS 0x58465342 + + # ifdef HAVE___FSWORD_T + typedef __fsword_t fsword; +@@ -786,23 +778,15 @@ dirent_inode_sort_may_be_useful (FTSENT const *p, int dir_fd) + } + + /* Given an FTS entry P for a directory with descriptor DIR_FD, +- return true if it is both useful and valid to apply leaf optimization. +- The optimization is useful only for file systems that lack usable +- dirent.d_type info. The optimization is valid if an st_nlink value +- of at least MIN_DIR_NLINK is an upper bound on the number of +- subdirectories of D, counting "." and ".." as subdirectories. ++ return whether it is valid to apply leaf optimization. ++ The optimization is valid if a directory's st_nlink value equal ++ to MIN_DIR_NLINK means the directory has no subdirectories. + DIR_FD is negative if unavailable. */ + static enum leaf_optimization + leaf_optimization (FTSENT const *p, int dir_fd) + { + switch (filesystem_type (p, dir_fd)) + { +- /* List here the file system types that may lack usable dirent.d_type +- info, yet for which the optimization does apply. */ +- case S_MAGIC_REISERFS: +- case S_MAGIC_XFS: /* XFS lacked it until 2013-08-22 commit. */ +- return NOSTAT_LEAF_OPTIMIZATION; +- + case 0: + /* Leaf optimization is unsafe if the file system type is unknown. */ + FALLTHROUGH; +@@ -1027,26 +1011,7 @@ check_for_dir: + if (p->fts_info == FTS_NSOK) + { + if (p->fts_statp->st_size == FTS_STAT_REQUIRED) +- { +- FTSENT *parent = p->fts_parent; +- if (parent->fts_n_dirs_remaining == 0 +- && ISSET(FTS_NOSTAT) +- && ISSET(FTS_PHYSICAL) +- && (leaf_optimization (parent, sp->fts_cwd_fd) +- == NOSTAT_LEAF_OPTIMIZATION)) +- { +- /* nothing more needed */ +- } +- else +- { +- p->fts_info = fts_stat(sp, p, false); +- if (S_ISDIR(p->fts_statp->st_mode) +- && p->fts_level != FTS_ROOTLEVEL +- && 0 < parent->fts_n_dirs_remaining +- && parent->fts_n_dirs_remaining != (nlink_t) -1) +- parent->fts_n_dirs_remaining--; +- } +- } ++ p->fts_info = fts_stat(sp, p, false); + else + fts_assert (p->fts_statp->st_size == FTS_NO_STAT_REQUIRED); + } +@@ -1830,11 +1795,6 @@ err: memset(sbp, 0, sizeof(struct stat)); + } + + if (S_ISDIR(sbp->st_mode)) { +- p->fts_n_dirs_remaining +- = ((sbp->st_nlink < MIN_DIR_NLINK +- || p->fts_level <= FTS_ROOTLEVEL) +- ? -1 +- : sbp->st_nlink - (ISSET (FTS_SEEDOT) ? 0 : MIN_DIR_NLINK)); + if (ISDOT(p->fts_name)) { + /* Command-line "." and ".." are real directories. */ + return (p->fts_level == FTS_ROOTLEVEL ? FTS_D : FTS_DOT); +diff --git a/lib/fts_.h b/lib/fts_.h +index d40a116..2e76cc4 100644 +--- a/lib/fts_.h ++++ b/lib/fts_.h +@@ -227,11 +227,6 @@ typedef struct _ftsent { + + size_t fts_namelen; /* strlen(fts_name) */ + +- /* If not (nlink_t) -1, an upper bound on the number of +- remaining subdirectories of interest. If this becomes +- zero, some work can be avoided. */ +- nlink_t fts_n_dirs_remaining; +- + # define FTS_D 1 /* preorder directory */ + # define FTS_DC 2 /* directory that causes cycles */ + # define FTS_DEFAULT 3 /* none of the above */ +-- +2.21.1 + diff --git a/coreutils-8.32-ls-removed-dir.patch b/coreutils-8.32-ls-removed-dir.patch new file mode 100644 index 0000000..77dce89 --- /dev/null +++ b/coreutils-8.32-ls-removed-dir.patch @@ -0,0 +1,153 @@ +From 8c022656320592dbad146f5d3a3ae1875f419446 Mon Sep 17 00:00:00 2001 +From: Paul Eggert +Date: Thu, 5 Mar 2020 17:25:29 -0800 +Subject: [PATCH 1/2] ls: restore 8.31 behavior on removed directories + +* NEWS: Mention this. +* src/ls.c: Do not include +(print_dir): Don't worry about whether the directory is removed. +* tests/ls/removed-directory.sh: Adjust to match new (i.e., old) +behavior. + +Upstream-commit: 10fcb97bd728f09d4a027eddf8ad2900f0819b0a +Signed-off-by: Kamil Dudka +--- + src/ls.c | 22 ---------------------- + tests/ls/removed-directory.sh | 10 ++-------- + 2 files changed, 2 insertions(+), 30 deletions(-) + +diff --git a/src/ls.c b/src/ls.c +index 9d25f62..850ecc2 100644 +--- a/src/ls.c ++++ b/src/ls.c +@@ -49,10 +49,6 @@ + # include + #endif + +-#ifdef __linux__ +-# include +-#endif +- + #include + #include + #include +@@ -2896,7 +2892,6 @@ print_dir (char const *name, char const *realname, bool command_line_arg) + struct dirent *next; + uintmax_t total_blocks = 0; + static bool first = true; +- bool found_any_entries = false; + + errno = 0; + dirp = opendir (name); +@@ -2972,7 +2967,6 @@ print_dir (char const *name, char const *realname, bool command_line_arg) + next = readdir (dirp); + if (next) + { +- found_any_entries = true; + if (! file_ignored (next->d_name)) + { + enum filetype type = unknown; +@@ -3018,22 +3012,6 @@ print_dir (char const *name, char const *realname, bool command_line_arg) + if (errno != EOVERFLOW) + break; + } +-#ifdef __linux__ +- else if (! found_any_entries) +- { +- /* If readdir finds no directory entries at all, not even "." or +- "..", then double check that the directory exists. */ +- if (syscall (SYS_getdents, dirfd (dirp), NULL, 0) == -1 +- && errno != EINVAL) +- { +- /* We exclude EINVAL as that pertains to buffer handling, +- and we've passed NULL as the buffer for simplicity. +- ENOENT is returned if appropriate before buffer handling. */ +- file_failure (command_line_arg, _("reading directory %s"), name); +- } +- break; +- } +-#endif + else + break; + +diff --git a/tests/ls/removed-directory.sh b/tests/ls/removed-directory.sh +index e8c835d..fe8f929 100755 +--- a/tests/ls/removed-directory.sh ++++ b/tests/ls/removed-directory.sh +@@ -26,20 +26,14 @@ case $host_triplet in + *) skip_ 'non linux kernel' ;; + esac + +-LS_FAILURE=2 +- +-cat <<\EOF >exp-err || framework_failure_ +-ls: reading directory '.': No such file or directory +-EOF +- + cwd=$(pwd) + mkdir d || framework_failure_ + cd d || framework_failure_ + rmdir ../d || framework_failure_ + +-returns_ $LS_FAILURE ls >../out 2>../err || fail=1 ++ls >../out 2>../err || fail=1 + cd "$cwd" || framework_failure_ + compare /dev/null out || fail=1 +-compare exp-err err || fail=1 ++compare /dev/null err || fail=1 + + Exit $fail +-- +2.21.1 + + +From 847324a0debd9d12062c79e7a7a9d3d8ce76390d Mon Sep 17 00:00:00 2001 +From: Paul Eggert +Date: Sat, 7 Mar 2020 10:29:51 -0800 +Subject: [PATCH 2/2] ls: improve removed-directory test + +* tests/ls/removed-directory.sh: Remove host_triplet test. +Skip this test if one cannot remove the working directory. +From a suggestion by Bernhard Voelker (Bug#39929). + +Upstream-commit: 672819c73f2e94e61386dc0584bddf9da860cc26 +Signed-off-by: Kamil Dudka +--- + tests/ls/removed-directory.sh | 13 ++++--------- + 1 file changed, 4 insertions(+), 9 deletions(-) + +diff --git a/tests/ls/removed-directory.sh b/tests/ls/removed-directory.sh +index fe8f929..63b209d 100755 +--- a/tests/ls/removed-directory.sh ++++ b/tests/ls/removed-directory.sh +@@ -1,7 +1,7 @@ + #!/bin/sh +-# If ls is asked to list a removed directory (e.g. the parent process's +-# current working directory that has been removed by another process), it +-# emits an error message. ++# If ls is asked to list a removed directory (e.g., the parent process's ++# current working directory has been removed by another process), it ++# should not emit an error message merely because the directory is removed. + + # Copyright (C) 2020 Free Software Foundation, Inc. + +@@ -21,15 +21,10 @@ + . "${srcdir=.}/tests/init.sh"; path_prepend_ ./src + print_ver_ ls + +-case $host_triplet in +- *linux*) ;; +- *) skip_ 'non linux kernel' ;; +-esac +- + cwd=$(pwd) + mkdir d || framework_failure_ + cd d || framework_failure_ +-rmdir ../d || framework_failure_ ++rmdir ../d || skip_ "can't remove working directory on this platform" + + ls >../out 2>../err || fail=1 + cd "$cwd" || framework_failure_ +-- +2.21.1 + diff --git a/coreutils-8.32-ls-scontext-crash.patch b/coreutils-8.32-ls-scontext-crash.patch new file mode 100644 index 0000000..a3ddff9 --- /dev/null +++ b/coreutils-8.32-ls-scontext-crash.patch @@ -0,0 +1,85 @@ +From 53c6b01e8e3fd338d7f53e5ff817ef86f9efa852 Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?P=C3=A1draig=20Brady?= +Date: Wed, 11 Nov 2020 17:22:33 +0000 +Subject: [PATCH] ls: fix crash printing SELinux context for unstatable files + +This crash was identified by Cyber Independent Testing Lab: +https://cyber-itl.org/2020/10/28/citl-7000-defects.html +and was introduced with commit v6.9.90-11-g4245876e2 + +* src/ls.c (gobble_file): Ensure scontext is initialized +in the case where files are not statable. +* tests/ls/selinux-segfault.sh: Renamed from proc-selinux-segfault.sh, +and added test case for broken symlinks. +* tests/local.mk: Adjust for the renamed test. + +Upstream-commit: 6fc695cb4a26f09dfeef8b1c24895a707055334e +Signed-off-by: Kamil Dudka +--- + src/ls.c | 3 +++ + tests/local.mk | 2 +- + .../{proc-selinux-segfault.sh => selinux-segfault.sh} | 10 ++++++++-- + 3 files changed, 12 insertions(+), 3 deletions(-) + rename tests/ls/{proc-selinux-segfault.sh => selinux-segfault.sh} (77%) + +diff --git a/src/ls.c b/src/ls.c +index 4acf5f4..8eb483d 100644 +--- a/src/ls.c ++++ b/src/ls.c +@@ -3412,6 +3412,9 @@ gobble_file (char const *name, enum filetype type, ino_t inode, + provokes an exit status of 1. */ + file_failure (command_line_arg, + _("cannot access %s"), full_name); ++ ++ f->scontext = UNKNOWN_SECURITY_CONTEXT; ++ + if (command_line_arg) + return 0; + +diff --git a/tests/local.mk b/tests/local.mk +index 2aeff2b..2441fdc 100644 +--- a/tests/local.mk ++++ b/tests/local.mk +@@ -616,7 +616,7 @@ all_tests = \ + tests/ls/multihardlink.sh \ + tests/ls/no-arg.sh \ + tests/ls/no-cap.sh \ +- tests/ls/proc-selinux-segfault.sh \ ++ tests/ls/selinux-segfault.sh \ + tests/ls/quote-align.sh \ + tests/ls/readdir-mountpoint-inode.sh \ + tests/ls/recursive.sh \ +diff --git a/tests/ls/proc-selinux-segfault.sh b/tests/ls/selinux-segfault.sh +similarity index 77% +rename from tests/ls/proc-selinux-segfault.sh +rename to tests/ls/selinux-segfault.sh +index 831a00e..e2b7ef6 100755 +--- a/tests/ls/proc-selinux-segfault.sh ++++ b/tests/ls/selinux-segfault.sh +@@ -1,5 +1,5 @@ + #!/bin/sh +-# ls -l /proc/sys would segfault when built against libselinux1 2.0.15-2+b1 ++# Ensure we don't segfault in selinux handling + + # Copyright (C) 2008-2020 Free Software Foundation, Inc. + +@@ -19,9 +19,15 @@ + . "${srcdir=.}/tests/init.sh"; path_prepend_ ./src + print_ver_ ls + ++# ls -l /proc/sys would segfault when built against libselinux1 2.0.15-2+b1 + f=/proc/sys + test -r $f || f=. +- + ls -l $f > out || fail=1 + ++# ls <= 8.32 would segfault when printing ++# the security context of broken symlink targets ++mkdir sedir || framework_failure_ ++ln -sf missing sedir/broken || framework_failure_ ++returns_ 1 ls -L -R -Z -m sedir > out || fail=1 ++ + Exit $fail +-- +2.26.2 + diff --git a/coreutils-8.32-mem-leaks.patch b/coreutils-8.32-mem-leaks.patch new file mode 100644 index 0000000..78a90fc --- /dev/null +++ b/coreutils-8.32-mem-leaks.patch @@ -0,0 +1,186 @@ +From be77b4ab7cb68fd2daf9de90bd75d844392788ac Mon Sep 17 00:00:00 2001 +From: Kamil Dudka +Date: Thu, 25 Mar 2021 11:57:56 +0100 +Subject: [PATCH 1/4] ln: fix memory leaks in do_link + +* src/ln.c (do_link): Free memory allocated by convert_abs_rel +on all code paths (Bug#47373). + +Upstream-commit: 6e98f67758260579d7d44ea5f2df4c82d28c9f58 +Signed-off-by: Kamil Dudka +--- + src/ln.c | 19 +++++++++++++------ + 1 file changed, 13 insertions(+), 6 deletions(-) + +diff --git a/src/ln.c b/src/ln.c +index ffa278e..9b52602 100644 +--- a/src/ln.c ++++ b/src/ln.c +@@ -229,14 +229,14 @@ do_link (char const *source, int destdir_fd, char const *dest_base, + if (errno != ENOENT) + { + error (0, errno, _("failed to access %s"), quoteaf (dest)); +- return false; ++ goto fail; + } + force = false; + } + else if (S_ISDIR (dest_stats.st_mode)) + { + error (0, 0, _("%s: cannot overwrite directory"), quotef (dest)); +- return false; ++ goto fail; + } + else if (seen_file (dest_set, dest, &dest_stats)) + { +@@ -245,7 +245,7 @@ do_link (char const *source, int destdir_fd, char const *dest_base, + error (0, 0, + _("will not overwrite just-created %s with %s"), + quoteaf_n (0, dest), quoteaf_n (1, source)); +- return false; ++ goto fail; + } + else + { +@@ -274,7 +274,7 @@ do_link (char const *source, int destdir_fd, char const *dest_base, + { + error (0, 0, _("%s and %s are the same file"), + quoteaf_n (0, source), quoteaf_n (1, dest)); +- return false; ++ goto fail; + } + } + +@@ -285,7 +285,10 @@ do_link (char const *source, int destdir_fd, char const *dest_base, + fprintf (stderr, _("%s: replace %s? "), + program_name, quoteaf (dest)); + if (!yesno ()) +- return true; ++ { ++ free(rel_source); ++ return true; ++ } + } + + if (backup_type != no_backups) +@@ -304,7 +307,7 @@ do_link (char const *source, int destdir_fd, char const *dest_base, + { + error (0, rename_errno, _("cannot backup %s"), + quoteaf (dest)); +- return false; ++ goto fail; + } + force = false; + } +@@ -397,6 +400,10 @@ do_link (char const *source, int destdir_fd, char const *dest_base, + free (backup_base); + free (rel_source); + return link_errno <= 0; ++ ++fail: ++ free (rel_source); ++ return false; + } + + void +-- +2.26.3 + + +From c051578e69bd8acf8f8a469566ae34e855345532 Mon Sep 17 00:00:00 2001 +From: Paul Eggert +Date: Thu, 25 Mar 2021 09:15:50 -0700 +Subject: [PATCH 2/4] maint: indenting + +* src/ln.c: Fix indenting. + +Upstream-commit: 8980b7c898046d899646da01c296fd15f0cced21 +Signed-off-by: Kamil Dudka +--- + src/ln.c | 4 ++-- + 1 file changed, 2 insertions(+), 2 deletions(-) + +diff --git a/src/ln.c b/src/ln.c +index 9b52602..8881d6a 100644 +--- a/src/ln.c ++++ b/src/ln.c +@@ -286,7 +286,7 @@ do_link (char const *source, int destdir_fd, char const *dest_base, + program_name, quoteaf (dest)); + if (!yesno ()) + { +- free(rel_source); ++ free (rel_source); + return true; + } + } +@@ -304,7 +304,7 @@ do_link (char const *source, int destdir_fd, char const *dest_base, + free (backup_base); + backup_base = NULL; + if (rename_errno != ENOENT) +- { ++ { + error (0, rename_errno, _("cannot backup %s"), + quoteaf (dest)); + goto fail; +-- +2.26.3 + + +From 0d6a4afe5bee0e397fb2fc3b205a29b32a69af9d Mon Sep 17 00:00:00 2001 +From: Paul Eggert +Date: Thu, 25 Mar 2021 09:16:36 -0700 +Subject: [PATCH 3/4] hostname: use puts + +* src/hostname.c (main): Prefer puts to printf "%s\n". + +Upstream-commit: c7a588ac3632aae21642d4d568497177950d36bf +Signed-off-by: Kamil Dudka +--- + src/hostname.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/src/hostname.c b/src/hostname.c +index 0b5c0cf..62cc98c 100644 +--- a/src/hostname.c ++++ b/src/hostname.c +@@ -103,7 +103,7 @@ main (int argc, char **argv) + hostname = xgethostname (); + if (hostname == NULL) + die (EXIT_FAILURE, errno, _("cannot determine hostname")); +- printf ("%s\n", hostname); ++ puts (hostname); + } + + if (optind + 1 < argc) +-- +2.26.3 + + +From 19c98d2080251edbaad9fb271aa10ad34f953500 Mon Sep 17 00:00:00 2001 +From: Paul Eggert +Date: Thu, 25 Mar 2021 11:20:18 -0700 +Subject: [PATCH 4/4] hostname: pacify valgrind + +* src/hostname.c (main) [IF_LINT]: Free hostname (Bug#47384). + +Upstream-commit: 4698e284f37844bc9b9f63f00eb556ccaaed5030 +Signed-off-by: Kamil Dudka +--- + src/hostname.c | 1 + + 1 file changed, 1 insertion(+) + +diff --git a/src/hostname.c b/src/hostname.c +index 62cc98c..7210248 100644 +--- a/src/hostname.c ++++ b/src/hostname.c +@@ -104,6 +104,7 @@ main (int argc, char **argv) + if (hostname == NULL) + die (EXIT_FAILURE, errno, _("cannot determine hostname")); + puts (hostname); ++ IF_LINT (free (hostname)); + } + + if (optind + 1 < argc) +-- +2.26.3 + diff --git a/coreutils-8.32-new-fs-types.patch b/coreutils-8.32-new-fs-types.patch new file mode 100644 index 0000000..a8eb35c --- /dev/null +++ b/coreutils-8.32-new-fs-types.patch @@ -0,0 +1,104 @@ +From 09400b7f7f48d8eedc0df55de8073a43bc0aac96 Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?P=C3=A1draig=20Brady?= +Date: Tue, 27 Oct 2020 20:15:43 +0000 +Subject: [PATCH 1/2] stat,tail: sync file system constants from the linux + kernel + +* src/stat.c: Add magic constants for "devmem", and +"zonefs" file systems. +* NEWS: Mention the improvement. + +Upstream-commit: ff80b6b0a0507e24f39cc1aad09d147f5187430b +Signed-off-by: Kamil Dudka +--- + src/stat.c | 4 ++++ + 1 file changed, 4 insertions(+) + +diff --git a/src/stat.c b/src/stat.c +index 5012622..8cd69da 100644 +--- a/src/stat.c ++++ b/src/stat.c +@@ -347,6 +347,8 @@ human_fstype (STRUCT_STATVFS const *statfsbuf) + return "debugfs"; + case S_MAGIC_DEVFS: /* 0x1373 local */ + return "devfs"; ++ case S_MAGIC_DEVMEM: /* 0x454D444D local */ ++ return "devmem"; + case S_MAGIC_DEVPTS: /* 0x1CD1 local */ + return "devpts"; + case S_MAGIC_DMA_BUF: /* 0x444D4142 local */ +@@ -549,6 +551,8 @@ human_fstype (STRUCT_STATVFS const *statfsbuf) + return "z3fold"; + case S_MAGIC_ZFS: /* 0x2FC12FC1 local */ + return "zfs"; ++ case S_MAGIC_ZONEFS: /* 0x5A4F4653 local */ ++ return "zonefs"; + case S_MAGIC_ZSMALLOC: /* 0x58295829 local */ + return "zsmallocfs"; + +-- +2.25.4 + + +From d5948fd41013dfe4d2d10083111821667977c6d1 Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?P=C3=A1draig=20Brady?= +Date: Tue, 27 Oct 2020 21:04:14 +0000 +Subject: [PATCH 2/2] mountlist: recognize more file system types as remote + +Sync "remote" file systems from stat.c in coreutils. +Note we only consider file systems that do not use host:resource +mount source. I.e. those that don't generally use a colon when +mounting, as that case is already considered. Searching for +" /etc/fstab" was informative for identifying these. +The full list of "remote" file systems in coreutils is currently: + acfs afs ceph cifs coda fhgfs fuseblk fusectl + gfs gfs2 gpfs ibrix k-afs lustre novell nfs nfsd + ocfs2 panfs prl_fs smb smb2 snfs vboxsf vmhgfs vxfs +Note also we do not include virtual machine file systems, +as even though they're remote to the current kernel, +they are generally not distributed to separate hosts. + +* lib/mountlist.c (ME_REMOTE): Sync previously unconsidered +"remote" file systems from stat.c in coreutils. + +Upstream-commit: dd1fc46be12d671c1a9d9dc5a6fa8c766e99aa2f +Signed-off-by: Kamil Dudka +--- + lib/mountlist.c | 12 ++++++++++-- + 1 file changed, 10 insertions(+), 2 deletions(-) + +diff --git a/lib/mountlist.c b/lib/mountlist.c +index 7abe024..e0227b7 100644 +--- a/lib/mountlist.c ++++ b/lib/mountlist.c +@@ -221,8 +221,9 @@ me_remote (char const *fs_name, char const *fs_type _GL_UNUSED) + #ifndef ME_REMOTE + /* A file system is "remote" if its Fs_name contains a ':' + or if (it is of type (smbfs or cifs) and its Fs_name starts with '//') +- or if it is of type (afs or auristorfs) +- or Fs_name is equal to "-hosts" (used by autofs to mount remote fs). */ ++ or if it is of any other of the listed types ++ or Fs_name is equal to "-hosts" (used by autofs to mount remote fs). ++ "VM" file systems like prl_fs or vboxsf are not considered remote here. */ + # define ME_REMOTE(Fs_name, Fs_type) \ + (strchr (Fs_name, ':') != NULL \ + || ((Fs_name)[0] == '/' \ +@@ -230,8 +231,15 @@ me_remote (char const *fs_name, char const *fs_type _GL_UNUSED) + && (strcmp (Fs_type, "smbfs") == 0 \ + || strcmp (Fs_type, "smb3") == 0 \ + || strcmp (Fs_type, "cifs") == 0)) \ ++ || strcmp (Fs_type, "acfs") == 0 \ + || strcmp (Fs_type, "afs") == 0 \ ++ || strcmp (Fs_type, "coda") == 0 \ + || strcmp (Fs_type, "auristorfs") == 0 \ ++ || strcmp (Fs_type, "fhgfs") == 0 \ ++ || strcmp (Fs_type, "gpfs") == 0 \ ++ || strcmp (Fs_type, "ibrix") == 0 \ ++ || strcmp (Fs_type, "ocfs2") == 0 \ ++ || strcmp (Fs_type, "vxfs") == 0 \ + || strcmp ("-hosts", Fs_name) == 0) + #endif + +-- +2.25.4 + diff --git a/coreutils-8.32-rm-stray-skip.patch b/coreutils-8.32-rm-stray-skip.patch new file mode 100644 index 0000000..66a1efc --- /dev/null +++ b/coreutils-8.32-rm-stray-skip.patch @@ -0,0 +1,109 @@ +From 11b37b65d08c2a8b6d967fd866ebbdbe7e864949 Mon Sep 17 00:00:00 2001 +From: Nishant Nayan +Date: Thu, 26 Nov 2020 14:35:17 +0000 +Subject: [PATCH] rm: do not skip files upon failure to remove an empty dir + +When removing a directory fails for some reason, and that directory +is empty, the rm_fts code gets the return value of the excise call +confused with the return value of its earlier call to prompt, +causing fts_skip_tree to be called again and the next file +that rm would otherwise have deleted to survive. + +* src/remove.c (rm_fts): Ensure we only skip a single fts entry, +when processing empty dirs. I.e. only skip the entry +having successfully removed it. +* tests/rm/empty-immutable-skip.sh: New root-only test. +* tests/local.mk: Add it. +* NEWS: Mention the bug fix. +Fixes https://bugs.gnu.org/44883 + +Upstream-commit: 6bf108358a6104ec1c694c9530b3cd56b95f4b57 +Signed-off-by: Kamil Dudka +--- + src/remove.c | 3 ++- + tests/local.mk | 1 + + tests/rm/empty-immutable-skip.sh | 46 ++++++++++++++++++++++++++++++++ + 3 files changed, 49 insertions(+), 1 deletion(-) + create mode 100755 tests/rm/empty-immutable-skip.sh + +diff --git a/src/remove.c b/src/remove.c +index 2d40c55..adf9489 100644 +--- a/src/remove.c ++++ b/src/remove.c +@@ -506,7 +506,8 @@ rm_fts (FTS *fts, FTSENT *ent, struct rm_options const *x) + /* When we know (from prompt when in interactive mode) + that this is an empty directory, don't prompt twice. */ + s = excise (fts, ent, x, true); +- fts_skip_tree (fts, ent); ++ if (s == RM_OK) ++ fts_skip_tree (fts, ent); + } + + if (s != RM_OK) +diff --git a/tests/local.mk b/tests/local.mk +index 5f7f775..2aeff2b 100644 +--- a/tests/local.mk ++++ b/tests/local.mk +@@ -136,6 +136,7 @@ all_root_tests = \ + tests/rm/no-give-up.sh \ + tests/rm/one-file-system.sh \ + tests/rm/read-only.sh \ ++ tests/rm/empty-immutable-skip.sh \ + tests/tail-2/append-only.sh \ + tests/tail-2/end-of-device.sh \ + tests/touch/now-owned-by-other.sh +diff --git a/tests/rm/empty-immutable-skip.sh b/tests/rm/empty-immutable-skip.sh +new file mode 100755 +index 0000000..c91d8d4 +--- /dev/null ++++ b/tests/rm/empty-immutable-skip.sh +@@ -0,0 +1,46 @@ ++#!/bin/sh ++# Ensure that rm does not skip extra files after hitting an empty immutable dir. ++# Requires root access to do chattr +i, as well as an ext[23] or xfs file system ++ ++# Copyright (C) 2020 Free Software Foundation, Inc. ++ ++# This program is free software: you can redistribute it and/or modify ++# it under the terms of the GNU General Public License as published by ++# the Free Software Foundation, either version 3 of the License, or ++# (at your option) any later version. ++ ++# This program is distributed in the hope that it will be useful, ++# but WITHOUT ANY WARRANTY; without even the implied warranty of ++# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++# GNU General Public License for more details. ++ ++# You should have received a copy of the GNU General Public License ++# along with this program. If not, see . ++ ++. "${srcdir=.}/tests/init.sh"; path_prepend_ ./src ++print_ver_ rm ++require_root_ ++ ++# These simple one-file operations are expected to work even in the ++# presence of this bug, and we need them to set up the rest of the test. ++chattr_i_works=1 ++touch f ++chattr +i f 2>/dev/null || chattr_i_works=0 ++rm f 2>/dev/null ++test -f f || chattr_i_works=0 ++chattr -i f 2>/dev/null || chattr_i_works=0 ++rm f 2>/dev/null || chattr_i_works=0 ++test -f f && chattr_i_works=0 ++ ++if test $chattr_i_works = 0; then ++ skip_ "chattr +i doesn't work on this file system" ++fi ++ ++mkdir empty || framework_failure_ ++touch x y || framework_failure_ ++chattr +i empty || framework_failure_ ++rm -rf empty x y ++{ test -f x || test -f y; } && fail=1 ++chattr -i empty ++ ++Exit $fail +-- +2.26.2 + diff --git a/coreutils-8.32-split-number.patch b/coreutils-8.32-split-number.patch new file mode 100644 index 0000000..89541bf --- /dev/null +++ b/coreutils-8.32-split-number.patch @@ -0,0 +1,100 @@ +From bb0e7fabcaed9a7e71e30f05e638e9f243cdb13e Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?P=C3=A1draig=20Brady?= +Date: Mon, 25 Jan 2021 14:12:48 +0000 +Subject: [PATCH] split: fix --number=K/N to output correct part of file + +This functionality regressed with the adjustments +in commit v8.25-4-g62e7af032 + +* src/split.c (bytes_chunk_extract): Account for already read data +when seeking into the file. +* tests/split/b-chunk.sh: Use the hidden ---io-blksize option, +to test this functionality. +Fixes https://bugs.gnu.org/46048 + +Upstream-commit: bb21daa125aeb4e32546309d370918ca47e612db +Signed-off-by: Kamil Dudka +--- + src/split.c | 2 +- + tests/split/b-chunk.sh | 45 ++++++++++++++++++++++++------------------ + 2 files changed, 27 insertions(+), 20 deletions(-) + +diff --git a/src/split.c b/src/split.c +index 09e610b..19248f6 100644 +--- a/src/split.c ++++ b/src/split.c +@@ -1001,7 +1001,7 @@ bytes_chunk_extract (uintmax_t k, uintmax_t n, char *buf, size_t bufsize, + } + else + { +- if (lseek (STDIN_FILENO, start, SEEK_CUR) < 0) ++ if (lseek (STDIN_FILENO, start - initial_read, SEEK_CUR) < 0) + die (EXIT_FAILURE, errno, "%s", quotef (infile)); + initial_read = SIZE_MAX; + } +diff --git a/tests/split/b-chunk.sh b/tests/split/b-chunk.sh +index 864ce55..39a6799 100755 +--- a/tests/split/b-chunk.sh ++++ b/tests/split/b-chunk.sh +@@ -35,32 +35,39 @@ split -e -n 10 /dev/null || fail=1 + returns_ 1 stat x?? 2>/dev/null || fail=1 + + printf '1\n2\n3\n4\n5\n' > input || framework_failure_ ++printf '1\n2' > exp-1 || framework_failure_ ++printf '\n3\n' > exp-2 || framework_failure_ ++printf '4\n5\n' > exp-3 || framework_failure_ + + for file in input /proc/version /sys/kernel/profiling; do + test -f $file || continue + +- split -n 3 $file > out || fail=1 +- split -n 1/3 $file > b1 || fail=1 +- split -n 2/3 $file > b2 || fail=1 +- split -n 3/3 $file > b3 || fail=1 ++ for blksize in 1 2 4096; do ++ if ! test "$file" = 'input'; then ++ # For /proc like files we must be able to read all ++ # into the internal buffer to be able to determine size. ++ test "$blksize" = 4096 || continue ++ fi + +- case $file in +- input) +- printf '1\n2' > exp-1 +- printf '\n3\n' > exp-2 +- printf '4\n5\n' > exp-3 ++ split -n 3 ---io-blksize=$blksize $file > out || fail=1 ++ split -n 1/3 ---io-blksize=$blksize $file > b1 || fail=1 ++ split -n 2/3 ---io-blksize=$blksize $file > b2 || fail=1 ++ split -n 3/3 ---io-blksize=$blksize $file > b3 || fail=1 + +- compare exp-1 xaa || fail=1 +- compare exp-2 xab || fail=1 +- compare exp-3 xac || fail=1 +- ;; +- esac ++ case $file in ++ input) ++ compare exp-1 xaa || fail=1 ++ compare exp-2 xab || fail=1 ++ compare exp-3 xac || fail=1 ++ ;; ++ esac + +- compare xaa b1 || fail=1 +- compare xab b2 || fail=1 +- compare xac b3 || fail=1 +- cat xaa xab xac | compare - $file || fail=1 +- test -f xad && fail=1 ++ compare xaa b1 || fail=1 ++ compare xab b2 || fail=1 ++ compare xac b3 || fail=1 ++ cat xaa xab xac | compare - $file || fail=1 ++ test -f xad && fail=1 ++ done + done + + Exit $fail +-- +2.26.2 + diff --git a/coreutils-8.32-stat-exfat.patch b/coreutils-8.32-stat-exfat.patch new file mode 100644 index 0000000..ea8d0cd --- /dev/null +++ b/coreutils-8.32-stat-exfat.patch @@ -0,0 +1,32 @@ +From b87f944c87ffe04db6e5476b007a8e4979de933d Mon Sep 17 00:00:00 2001 +From: Kamil Dudka +Date: Thu, 18 Feb 2021 11:18:04 +0100 +Subject: [PATCH] stat,tail: add support for the exfat file system + +Bug: https://bugzilla.redhat.com/1921427 + +* src/stat.c (human_fstype): Add case for the 'exfat' file system type. +Fixes https://bugs.gnu.org/46613 + +Upstream-commit: a5e0d8f387e81e854427addbbaf2504541bbf4b9 +Signed-off-by: Kamil Dudka +--- + src/stat.c | 2 ++ + 1 file changed, 2 insertions(+) + +diff --git a/src/stat.c b/src/stat.c +index 8cd69da..4e1c8e3 100644 +--- a/src/stat.c ++++ b/src/stat.c +@@ -361,6 +361,8 @@ human_fstype (STRUCT_STATVFS const *statfsbuf) + return "efs"; + case S_MAGIC_EROFS_V1: /* 0xE0F5E1E2 local */ + return "erofs"; ++ case S_MAGIC_EXFAT: /* 0x2011BAB0 local */ ++ return "exfat"; + case S_MAGIC_EXFS: /* 0x45584653 local */ + return "exfs"; + case S_MAGIC_EXOFS: /* 0x5DF5 local */ +-- +2.26.2 + diff --git a/coreutils-8.32-tail-use-poll.patch b/coreutils-8.32-tail-use-poll.patch new file mode 100644 index 0000000..ed3a8f3 --- /dev/null +++ b/coreutils-8.32-tail-use-poll.patch @@ -0,0 +1,181 @@ +From c7a04cef4075da864a3468e63a5bb79334d8f556 Mon Sep 17 00:00:00 2001 +From: Paul Eggert +Date: Sat, 26 Jun 2021 18:23:52 -0700 +Subject: [PATCH] tail: use poll, not select +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +This fixes an unlikely stack out-of-bounds write reported by +Stepan Broz via Kamil Dudka (Bug#49209). +* src/tail.c: Do not include . +[!_AIX]: Include poll.h. +(check_output_alive) [!_AIX]: Use poll instead of select. +(tail_forever_inotify): Likewise. Simplify logic, as there is no +need for a ‘while (len <= evbuf_off)’ loop. + +Upstream-commit: da0d448bca62c6305fc432f67e2c5ccc2da75346 +Signed-off-by: Kamil Dudka +--- + src/tail.c | 100 +++++++++++++++++++++-------------------------------- + 1 file changed, 39 insertions(+), 61 deletions(-) + +diff --git a/src/tail.c b/src/tail.c +index 1c88723..5b4f21a 100644 +--- a/src/tail.c ++++ b/src/tail.c +@@ -28,12 +28,9 @@ + #include + #include + #include +-#include ++#include + #include + #include +-#ifdef _AIX +-# include +-#endif + + #include "system.h" + #include "argmatch.h" +@@ -351,27 +348,12 @@ check_output_alive (void) + if (! monitor_output) + return; + +-#ifdef _AIX +- /* select on AIX was seen to give a readable event immediately. */ + struct pollfd pfd; + pfd.fd = STDOUT_FILENO; + pfd.events = POLLERR; + + if (poll (&pfd, 1, 0) >= 0 && (pfd.revents & POLLERR)) + die_pipe (); +-#else +- struct timeval delay; +- delay.tv_sec = delay.tv_usec = 0; +- +- fd_set rfd; +- FD_ZERO (&rfd); +- FD_SET (STDOUT_FILENO, &rfd); +- +- /* readable event on STDOUT is equivalent to POLLERR, +- and implies an error condition on output like broken pipe. */ +- if (select (STDOUT_FILENO + 1, &rfd, NULL, NULL, &delay) == 1) +- die_pipe (); +-#endif + } + + static bool +@@ -1612,7 +1594,7 @@ tail_forever_inotify (int wd, struct File_spec *f, size_t n_files, + /* Wait for inotify events and handle them. Events on directories + ensure that watched files can be re-added when following by name. + This loop blocks on the 'safe_read' call until a new event is notified. +- But when --pid=P is specified, tail usually waits via the select. */ ++ But when --pid=P is specified, tail usually waits via poll. */ + while (1) + { + struct File_spec *fspec; +@@ -1629,54 +1611,51 @@ tail_forever_inotify (int wd, struct File_spec *f, size_t n_files, + return false; + } + +- /* When watching a PID, ensure that a read from WD will not block +- indefinitely. */ +- while (len <= evbuf_off) ++ if (len <= evbuf_off) + { +- struct timeval delay; /* how long to wait for file changes. */ ++ /* Poll for inotify events. When watching a PID, ensure ++ that a read from WD will not block indefinitely. ++ If MONITOR_OUTPUT, also poll for a broken output pipe. */ + +- if (pid) ++ int file_change; ++ struct pollfd pfd[2]; ++ do + { +- if (writer_is_dead) +- exit (EXIT_SUCCESS); ++ /* How many ms to wait for changes. -1 means wait forever. */ ++ int delay = -1; + +- writer_is_dead = (kill (pid, 0) != 0 && errno != EPERM); +- +- if (writer_is_dead) +- delay.tv_sec = delay.tv_usec = 0; +- else ++ if (pid) + { +- delay.tv_sec = (time_t) sleep_interval; +- delay.tv_usec = 1000000 * (sleep_interval - delay.tv_sec); ++ if (writer_is_dead) ++ exit (EXIT_SUCCESS); ++ ++ writer_is_dead = (kill (pid, 0) != 0 && errno != EPERM); ++ ++ if (writer_is_dead || sleep_interval <= 0) ++ delay = 0; ++ else if (sleep_interval < INT_MAX / 1000 - 1) ++ { ++ /* delay = ceil (sleep_interval * 1000), sans libm. */ ++ double ddelay = sleep_interval * 1000; ++ delay = ddelay; ++ delay += delay < ddelay; ++ } + } ++ ++ pfd[0].fd = wd; ++ pfd[0].events = POLLIN; ++ pfd[1].fd = STDOUT_FILENO; ++ pfd[1].events = pfd[1].revents = 0; ++ file_change = poll (pfd, monitor_output + 1, delay); + } ++ while (file_change == 0); + +- fd_set rfd; +- FD_ZERO (&rfd); +- FD_SET (wd, &rfd); +- if (monitor_output) +- FD_SET (STDOUT_FILENO, &rfd); +- +- int file_change = select (MAX (wd, STDOUT_FILENO) + 1, +- &rfd, NULL, NULL, pid ? &delay: NULL); +- +- if (file_change == 0) +- continue; +- else if (file_change == -1) +- die (EXIT_FAILURE, errno, +- _("error waiting for inotify and output events")); +- else if (FD_ISSET (STDOUT_FILENO, &rfd)) +- { +- /* readable event on STDOUT is equivalent to POLLERR, +- and implies an error on output like broken pipe. */ +- die_pipe (); +- } +- else +- break; +- } ++ if (file_change < 0) ++ die (EXIT_FAILURE, errno, ++ _("error waiting for inotify and output events")); ++ if (pfd[1].revents) ++ die_pipe (); + +- if (len <= evbuf_off) +- { + len = safe_read (wd, evbuf, evlen); + evbuf_off = 0; + +@@ -2437,8 +2416,7 @@ main (int argc, char **argv) + if (forever && ignore_fifo_and_pipe (F, n_files)) + { + /* If stdout is a fifo or pipe, then monitor it +- so that we exit if the reader goes away. +- Note select() on a regular file is always readable. */ ++ so that we exit if the reader goes away. */ + struct stat out_stat; + if (fstat (STDOUT_FILENO, &out_stat) < 0) + die (EXIT_FAILURE, errno, _("standard output")); +-- +2.31.1 + diff --git a/coreutils-8.32-tests-false-positives.patch b/coreutils-8.32-tests-false-positives.patch new file mode 100644 index 0000000..0b0a301 --- /dev/null +++ b/coreutils-8.32-tests-false-positives.patch @@ -0,0 +1,99 @@ +From fc6318841f008dadc1e7c93e539f10d24aa83e90 Mon Sep 17 00:00:00 2001 +From: Bernhard Voelker +Date: Wed, 21 Apr 2021 00:12:00 +0200 +Subject: [PATCH 1/2] tests: fix FP in ls/stat-free-color.sh + +On newer systems like Fedora 34 and openSUSE Tumbleweed, ls(1) calls +newfstatat(STDOUT_FILENO, ...), but only when there is something to +output. + +* tests/ls/stat-free-color.sh: Add -a option to the reference invocation +of ls, thus enforcing something gets output. + +Upstream-commit: b7091093bb6505c33279f9bc940b2e94763a6e5d +Signed-off-by: Kamil Dudka +--- + tests/ls/stat-free-color.sh | 10 ++++++---- + 1 file changed, 6 insertions(+), 4 deletions(-) + +diff --git a/tests/ls/stat-free-color.sh b/tests/ls/stat-free-color.sh +index 00942f7..87bed1c 100755 +--- a/tests/ls/stat-free-color.sh ++++ b/tests/ls/stat-free-color.sh +@@ -56,12 +56,14 @@ eval $(dircolors -b color-without-stat) + # The system may perform additional stat-like calls before main. + # Furthermore, underlying library functions may also implicitly + # add an extra stat call, e.g. opendir since glibc-2.21-360-g46f894d. +-# To avoid counting those, first get a baseline count for running +-# ls with one empty directory argument. Then, compare that with the +-# invocation under test. ++# Finally, ls(1) makes a stat call for stdout, but only in the case ++# when there is something to output. ++# To get the comparison right, first get a baseline count for running ++# 'ls -a' with one empty directory argument. Then, compare that with ++# the invocation under test. + mkdir d || framework_failure_ + +-strace -q -o log1 -e $stats ls --color=always d || fail=1 ++strace -q -o log1 -e $stats ls -a --color=always d || fail=1 + n_stat1=$(grep -vF '+++' log1 | wc -l) || framework_failure_ + + test $n_stat1 = 0 \ +-- +2.31.1 + + +From c16ca58f17a088e925c0d1c4015c48332c380a00 Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?P=C3=A1draig=20Brady?= +Date: Sun, 9 May 2021 23:41:00 +0100 +Subject: [PATCH 2/2] tests: fix tests/cp/sparse-2.sh false failure on some + systems + +* tests/cp/sparse-2.sh: Double check cp --sparse=always, +with dd conv=sparse, in the case where the former didn't +create a sparse file. Now that this test is being newly run +on macos, we're seeing a failure due to seek() not creating +holes on apfs unless the size is >= 16MiB. + +Upstream-commit: 6b499720fecae935dc00e236d6aefe94d9010482 +Signed-off-by: Kamil Dudka +--- + tests/cp/fiemap-2.sh | 13 ++++++++++--- + 1 file changed, 10 insertions(+), 3 deletions(-) + +diff --git a/tests/cp/fiemap-2.sh b/tests/cp/fiemap-2.sh +index 548a376..e20ce54 100755 +--- a/tests/cp/fiemap-2.sh ++++ b/tests/cp/fiemap-2.sh +@@ -17,7 +17,7 @@ + # along with this program. If not, see . + + . "${srcdir=.}/tests/init.sh"; path_prepend_ ./src +-print_ver_ cp ++print_ver_ cp stat dd + + # Require a fiemap-enabled FS. + touch fiemap_chk # check a file rather than current dir for best coverage +@@ -46,10 +46,17 @@ dd bs=1k seek=1 of=k count=255 < /dev/zero || framework_failure_ + + # cp should detect the all-zero blocks and convert some of them to holes. + # How many it detects/converts currently depends on io_blksize. +-# Currently, on my F14/ext4 desktop, this K starts off with size 256KiB, ++# Currently, on my F14/ext4 desktop, this K file starts off with size 256KiB, + # (note that the K in the preceding test starts off with size 4KiB). + # cp from coreutils-8.9 with --sparse=always reduces the size to 32KiB. + cp --sparse=always k k2 || fail=1 +-test $(stat -c %b k2) -lt $(stat -c %b k) || fail=1 ++if test $(stat -c %b k2) -ge $(stat -c %b k); then ++ # If not sparse, then double check by creating with dd ++ # as we're not guaranteed that seek will create a hole. ++ # apfs on darwin 19.2.0 for example was seen to not to create holes < 16MiB. ++ hole_size=$(stat -c %o k2) || framework_failure_ ++ dd if=k of=k2.dd bs=$hole_size conv=sparse || framework_failure_ ++ test $(stat -c %b k2) -eq $(stat -c %b k2.dd) || fail=1 ++fi + + Exit $fail +-- +2.31.1 + diff --git a/coreutils-8.32.tar.xz.sig b/coreutils-8.32.tar.xz.sig new file mode 100644 index 0000000..e1420fc --- /dev/null +++ b/coreutils-8.32.tar.xz.sig @@ -0,0 +1,16 @@ +-----BEGIN PGP SIGNATURE----- + +iQIzBAABCAAdFiEEbDfcEhIaUAa8HbgE32/ZcTBgN9kFAl5hC5MACgkQ32/ZcTBg +N9n92Q//Td2GE1f8AZKkxCNI76Q/TqbxAwhjbkR+KdzvsyMePmgHcMgHG6sO2MNF +g6DIBmHpO3vWGzvUxUZRRhuW5QBOnMxHb/WXZ0p/g45d5MQdn4i0dA0wUJgByOqn +/WVfygNg9mrWFx/uTeCdhrwL11m71C7j/eQVu7Wr5DIb20VJ8+nVC2IWW33ZvxRj +Goa0wwDpeeD9qYe/Y+E5ZyhDYHJGRmNAlS03SXLO3+RfsbZFwdQEtzvr+v1VN6/S +9OsoI/GLdRjY1tByppaoZ63ZybB6iF5zZfJiWDF7Nw4MduJpjZQDSywiNleJ9vOi +fwR1180PjMV6aTXvPwqbqQxZjDl7nqvO36ghlTvErJbqdJVIYxmUGNjeJyjqI85l +Lhckh0GWos9K/kl13Ry9KWsxNQgfjNhtgjXGh+W47ojrho2kCiK5BTwDFeVU0jtU +H/1EePSGAIUF/Sfjz3rmGgLaaBwPiRiyzEIuZMyd4NCJWwfOTqgOshOYw15GCWYq +wGesN/4LWzEja7Au5lHP7imXjP0bp4qE/sYrOb4WzVVLCn+z2hu6SEIzjJzSm+D+ +8Wv3Ia1/ypVpR+Z7gUt7VtEvI8zAwlySd/6Jw5U7TL0rzvZsTVWmCvEjPp+o3jCB +Fy/4ybao1gowBFtT1AtPMmxmiJ41KWCxLFrTuGJpFYCGvBH2y8s= +=yHV6 +-----END PGP SIGNATURE----- diff --git a/coreutils-8.4-mkdir-modenote.patch b/coreutils-8.4-mkdir-modenote.patch new file mode 100644 index 0000000..51a129e --- /dev/null +++ b/coreutils-8.4-mkdir-modenote.patch @@ -0,0 +1,13 @@ +diff --git a/doc/coreutils.texi b/doc/coreutils.texi +index 400e135..47e4480 100644 +--- a/doc/coreutils.texi ++++ b/doc/coreutils.texi +@@ -10074,6 +10074,8 @@ incorrect. @xref{Directory Setuid and Setgid}, for how the + set-user-ID and set-group-ID bits of directories are inherited unless + overridden in this way. + ++Note: The @option{--mode},@option{-m} option only applies to the right-most directories listed on the command line. When combined with @option{--parents}, @option{-p} option, any parent directories are created with @samp{u+wx} modified by umask. ++ + @item -p + @itemx --parents + @opindex -p diff --git a/coreutils-9.0-autofs-no-mount.patch b/coreutils-9.0-autofs-no-mount.patch new file mode 100644 index 0000000..b6d6523 --- /dev/null +++ b/coreutils-9.0-autofs-no-mount.patch @@ -0,0 +1,87 @@ +From f4422844dbcd839ce486bcbc15b7bd5b72c9198d Mon Sep 17 00:00:00 2001 +From: Rohan Sable +Date: Mon, 7 Mar 2022 14:14:13 +0000 +Subject: [PATCH 1/2] ls: avoid triggering automounts + +statx() has different defaults wrt automounting +compared to stat() or lstat(), so explicitly +set the AT_NO_AUTOMOUNT flag to suppress that behavior, +and avoid unintended operations or potential errors. + +* src/ls.c (do_statx): Pass AT_NO_AUTOMOUNT to avoid this behavior. +Fixes https://bugs.gnu.org/54286 + +Signed-off-by: Rohan Sable + +Upstream-commit: 85c975df2c25bd799370b04bb294e568e001102f +Signed-off-by: Kamil Dudka +--- + src/ls.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/src/ls.c b/src/ls.c +index 1047801..fe0e9f8 100644 +--- a/src/ls.c ++++ b/src/ls.c +@@ -1175,7 +1175,7 @@ do_statx (int fd, char const *name, struct stat *st, int flags, + { + struct statx stx; + bool want_btime = mask & STATX_BTIME; +- int ret = statx (fd, name, flags, mask, &stx); ++ int ret = statx (fd, name, flags | AT_NO_AUTOMOUNT, mask, &stx); + if (ret >= 0) + { + statx_to_stat (&stx, st); +-- +2.34.1 + + +From 3d227f9e4f3fe806064721e4b9451ee06526bc80 Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?P=C3=A1draig=20Brady?= +Date: Mon, 7 Mar 2022 23:29:20 +0000 +Subject: [PATCH 2/2] stat: only automount with --cached=never + +Revert to the default behavior before the introduction of statx(). + +* src/stat.c (do_stat): Set AT_NO_AUTOMOUNT without --cached=never. +* doc/coreutils.texi (stat invocation): Mention the automount +behavior with --cached=never. + +Fixes https://bugs.gnu.org/54287 + +Upstream-commit: 92cb8427c537f37edd43c5cef1909585201372ab +Signed-off-by: Kamil Dudka +--- + doc/coreutils.texi | 1 + + src/stat.c | 3 +++ + 2 files changed, 4 insertions(+) + +diff --git a/doc/coreutils.texi b/doc/coreutils.texi +index 19b535c..0f5c16a 100644 +--- a/doc/coreutils.texi ++++ b/doc/coreutils.texi +@@ -12564,6 +12564,7 @@ Always read the already cached attributes if available. + + @item never + Always sychronize with the latest file system attributes. ++This also mounts automounted files. + + @item default + Leave the caching behavior to the underlying file system. +diff --git a/src/stat.c b/src/stat.c +index 0c34501..803340a 100644 +--- a/src/stat.c ++++ b/src/stat.c +@@ -1381,6 +1381,9 @@ do_stat (char const *filename, char const *format, char const *format2) + else if (force_sync) + flags |= AT_STATX_FORCE_SYNC; + ++ if (! force_sync) ++ flags |= AT_NO_AUTOMOUNT; ++ + fd = statx (fd, pathname, flags, format_to_mask (format), &stx); + if (fd < 0) + { +-- +2.34.1 + diff --git a/coreutils-9.9-fix-cut-test-aarch64.patch b/coreutils-9.9-fix-cut-test-aarch64.patch deleted file mode 100644 index 600f87b..0000000 --- a/coreutils-9.9-fix-cut-test-aarch64.patch +++ /dev/null @@ -1,28 +0,0 @@ -From 95044cb5eaea83d02f768feb5ab79fcf5e6ad782 Mon Sep 17 00:00:00 2001 -From: =?UTF-8?q?P=C3=A1draig=20Brady?= -Date: Mon, 22 Dec 2025 17:12:48 +0000 -Subject: [PATCH] tests: avoid false failure due to ulimit on aarch64 - -* tests/cut/cut-huge-range.sh: Add an extra 1MiB headroom, -which was seen with aarch64. -Reported at https://bugzilla.redhat.com/2424302 - -Cherry-picked-by: Lukáš Zaoral -Upstream-commit: 95044cb5eaea83d02f768feb5ab79fcf5e6ad782 ---- - tests/cut/cut-huge-range.sh | 1 + - 1 file changed, 1 insertion(+) - -diff --git a/tests/cut/cut-huge-range.sh b/tests/cut/cut-huge-range.sh -index 4bd1b129d8..98d7e8f0b9 100755 ---- a/tests/cut/cut-huge-range.sh -+++ b/tests/cut/cut-huge-range.sh -@@ -22,6 +22,7 @@ getlimits_ - - vm=$(get_min_ulimit_v_ returns_ 0 cut -b1 /dev/null) \ - || skip_ 'shell lacks ulimit, or ASAN enabled' -+vm=$(($vm+1000)) # https://bugzilla.redhat.com/2424302 - - # Ensure we can cut up to our sentinel value. - # Don't use expr to subtract one, - diff --git a/coreutils-9.9-gnulib-c23.patch b/coreutils-9.9-gnulib-c23.patch deleted file mode 100644 index 82e3899..0000000 --- a/coreutils-9.9-gnulib-c23.patch +++ /dev/null @@ -1,169 +0,0 @@ -From 891761bca1aa78336e5b18c121075b6e4696c5d4 Mon Sep 17 00:00:00 2001 -From: Paul Eggert -Date: Sun, 23 Nov 2025 00:50:40 -0800 -Subject: [PATCH] Port to C23 qualifier-generic fns like strchr -MIME-Version: 1.0 -Content-Type: text/plain; charset=UTF-8 -Content-Transfer-Encoding: 8bit - -This ports Gnulib to strict C23 platforms that reject code -like ‘char *q = strchr (P, 'x');’ when P is a pointer to const, -because in C23 strchr is a qualifier-generic function so -strchr (P, 'x') returns char const *. -This patch does not attempt to do the following two things, -which might be useful in the future: -1. When compiling on non-C23 platforms, check user code for -portability to platforms that define qualifier-generic functions. -2. Port Gnulib to platforms that have qualifier-generic functions -not listed in the C23 standard, e.g., strchrnul. I don’t know -of any such platforms. -* lib/mbschr.c (mbschr): -* lib/memchr2.c (memchr2): -Port to C23, where functions like strchr are qualifier-generic. -* lib/c++defs.h (_GL_FUNCDECL_SYS_NAME): New macro. -* lib/c++defs.h (_GL_FUNCDECL_SYS): -* lib/stdlib.in.h (bsearch): -Use it, to prevent C23 names like strchr from acting like macros. -* lib/string.in.h (memchr, strchr, strpbrk, strrchr): -Do not #undef when GNULIB_POSIXCHECK is defined, as this could -cause conforming C23 code to fail to conform. It’s not clear why -_GL_WARN_ON_USE_CXX; perhaps it was needed but isn’t any more? -But for now, limit the removal of #undef to these four functions -where #undeffing is clearly undesirable in C23. -* lib/wchar.in.h (wmemchr): Parenthesize function name in decl, -to prevent it from acting like a macro. - -Cherry-picked-by: Lukáš Zaoral -Upstream-commit: df17f4f37ed3ca373d23ad42eae51122bdb96626 ---- - lib/c++defs.h | 12 +++++++++++- - lib/mbschr.c | 2 +- - lib/memchr2.c | 2 +- - lib/stdlib.in.h | 6 +++--- - lib/string.in.h | 4 ---- - lib/wchar.in.h | 2 +- - 6 files changed, 17 insertions(+), 11 deletions(-) - -diff --git a/lib/c++defs.h b/lib/c++defs.h -index b77979a..7384457 100644 ---- a/lib/c++defs.h -+++ b/lib/c++defs.h -@@ -127,6 +127,16 @@ - #define _GL_FUNCDECL_RPL_1(rpl_func,rettype,parameters,...) \ - _GL_EXTERN_C_FUNC __VA_ARGS__ rettype rpl_func parameters - -+/* _GL_FUNCDECL_SYS_NAME (func) expands to plain func if C++, and to -+ parenthsized func otherwise. Parenthesization is needed in C23 if -+ the function is like strchr and so is a qualifier-generic macro -+ that expands to something more complicated. */ -+#ifdef __cplusplus -+# define _GL_FUNCDECL_SYS_NAME(func) func -+#else -+# define _GL_FUNCDECL_SYS_NAME(func) (func) -+#endif -+ - /* _GL_FUNCDECL_SYS (func, rettype, parameters, [attributes]); - declares the system function, named func, with the given prototype, - consisting of return type, parameters, and attributes. -@@ -139,7 +149,7 @@ - _GL_FUNCDECL_SYS (posix_openpt, int, (int flags), _GL_ATTRIBUTE_NODISCARD); - */ - #define _GL_FUNCDECL_SYS(func,rettype,parameters,...) \ -- _GL_EXTERN_C_FUNC __VA_ARGS__ rettype func parameters -+ _GL_EXTERN_C_FUNC __VA_ARGS__ rettype _GL_FUNCDECL_SYS_NAME (func) parameters - - /* _GL_CXXALIAS_RPL (func, rettype, parameters); - declares a C++ alias called GNULIB_NAMESPACE::func -diff --git a/lib/mbschr.c b/lib/mbschr.c -index c9e14b5..6582134 100644 ---- a/lib/mbschr.c -+++ b/lib/mbschr.c -@@ -65,5 +65,5 @@ mbschr (const char *string, int c) - return NULL; - } - else -- return strchr (string, c); -+ return (char *) strchr (string, c); - } -diff --git a/lib/memchr2.c b/lib/memchr2.c -index 7493823..d7724ae 100644 ---- a/lib/memchr2.c -+++ b/lib/memchr2.c -@@ -55,7 +55,7 @@ memchr2 (void const *s, int c1_in, int c2_in, size_t n) - c2 = (unsigned char) c2_in; - - if (c1 == c2) -- return memchr (s, c1, n); -+ return (void *) memchr (s, c1, n); - - /* Handle the first few bytes by reading one byte at a time. - Do this until VOID_PTR is aligned on a longword boundary. */ -diff --git a/lib/stdlib.in.h b/lib/stdlib.in.h -index bef0aaa..fd0e1e0 100644 ---- a/lib/stdlib.in.h -+++ b/lib/stdlib.in.h -@@ -224,9 +224,9 @@ _GL_INLINE_HEADER_BEGIN - - /* Declarations for ISO C N3322. */ - #if defined __GNUC__ && __GNUC__ >= 15 && !defined __clang__ --_GL_EXTERN_C void *bsearch (const void *__key, -- const void *__base, size_t __nmemb, size_t __size, -- int (*__compare) (const void *, const void *)) -+_GL_EXTERN_C void *_GL_FUNCDECL_SYS_NAME (bsearch) -+ (const void *__key, const void *__base, size_t __nmemb, size_t __size, -+ int (*__compare) (const void *, const void *)) - _GL_ATTRIBUTE_NONNULL_IF_NONZERO (2, 3) _GL_ARG_NONNULL ((5)); - _GL_EXTERN_C void qsort (void *__base, size_t __nmemb, size_t __size, - int (*__compare) (const void *, const void *)) -diff --git a/lib/string.in.h b/lib/string.in.h -index fdcdd21..8b56acf 100644 ---- a/lib/string.in.h -+++ b/lib/string.in.h -@@ -409,7 +409,6 @@ _GL_CXXALIASWARN1 (memchr, void const *, - _GL_CXXALIASWARN (memchr); - # endif - #elif defined GNULIB_POSIXCHECK --# undef memchr - /* Assume memchr is always declared. */ - _GL_WARN_ON_USE (memchr, "memchr has platform-specific bugs - " - "use gnulib module memchr for portability" ); -@@ -674,7 +673,6 @@ _GL_WARN_ON_USE (stpncpy, "stpncpy is unportable - " - #if defined GNULIB_POSIXCHECK - /* strchr() does not work with multibyte strings if the locale encoding is - GB18030 and the character to be searched is a digit. */ --# undef strchr - /* Assume strchr is always declared. */ - _GL_WARN_ON_USE_CXX (strchr, - const char *, char *, (const char *, int), -@@ -981,7 +979,6 @@ _GL_CXXALIASWARN (strpbrk); - Even in this simple case, it does not work with multibyte strings if the - locale encoding is GB18030 and one of the characters to be searched is a - digit. */ --# undef strpbrk - _GL_WARN_ON_USE_CXX (strpbrk, - const char *, char *, (const char *, const char *), - "strpbrk cannot work correctly on character strings " -@@ -1011,7 +1008,6 @@ _GL_WARN_ON_USE (strspn, "strspn cannot work correctly on character strings " - #if defined GNULIB_POSIXCHECK - /* strrchr() does not work with multibyte strings if the locale encoding is - GB18030 and the character to be searched is a digit. */ --# undef strrchr - /* Assume strrchr is always declared. */ - _GL_WARN_ON_USE_CXX (strrchr, - const char *, char *, (const char *, int), -diff --git a/lib/wchar.in.h b/lib/wchar.in.h -index ab602a2..6be4515 100644 ---- a/lib/wchar.in.h -+++ b/lib/wchar.in.h -@@ -301,7 +301,7 @@ _GL_EXTERN_C int wcsncmp (const wchar_t *__s1, const wchar_t *__s2, size_t __n) - _GL_ATTRIBUTE_NONNULL_IF_NONZERO (1, 3) - _GL_ATTRIBUTE_NONNULL_IF_NONZERO (2, 3); - # ifndef __cplusplus --_GL_EXTERN_C wchar_t *wmemchr (const wchar_t *__s, wchar_t __wc, size_t __n) -+_GL_EXTERN_C wchar_t *(wmemchr) (const wchar_t *__s, wchar_t __wc, size_t __n) - _GL_ATTRIBUTE_NONNULL_IF_NONZERO (1, 3); - # endif - _GL_EXTERN_C wchar_t *wmemset (wchar_t *__s, wchar_t __wc, size_t __n) --- -2.52.0 - diff --git a/coreutils-colorls.csh b/coreutils-colorls.csh index 66ec2fa..8312dc9 100755 --- a/coreutils-colorls.csh +++ b/coreutils-colorls.csh @@ -15,6 +15,11 @@ alias l. 'ls -d .*' set COLORS=/etc/DIR_COLORS if ($?TERM) then + if ( -e "/etc/DIR_COLORS.256color" && -e "/usr/bin/tput" ) then + if ( "`/usr/bin/tput colors`" == "256" ) then + set COLORS=/etc/DIR_COLORS.256color + endif + endif if ( -e "/etc/DIR_COLORS.$TERM" ) then set COLORS="/etc/DIR_COLORS.$TERM" endif diff --git a/coreutils-colorls.sh b/coreutils-colorls.sh index 5162f1e..ac92268 100755 --- a/coreutils-colorls.sh +++ b/coreutils-colorls.sh @@ -22,6 +22,10 @@ if [ -z "$USER_LS_COLORS" ]; then [ -z "$COLORS" ] && [ -e "/etc/DIR_COLORS.$TERM" ] && \ COLORS="/etc/DIR_COLORS.$TERM" + [ -z "$COLORS" ] && [ -e "/etc/DIR_COLORS.256color" ] && \ + [ "x`/usr/bin/tty -s && /usr/bin/tput colors 2>/dev/null`" = "x256" ] && \ + COLORS="/etc/DIR_COLORS.256color" + [ -z "$COLORS" ] && [ -e "/etc/DIR_COLORS" ] && \ COLORS="/etc/DIR_COLORS" diff --git a/coreutils-df-direct.patch b/coreutils-df-direct.patch index 341ee2c..88015a1 100644 --- a/coreutils-df-direct.patch +++ b/coreutils-df-direct.patch @@ -1,4 +1,4 @@ -From 91be1a584108a6a3d96f64382bbf206c4213b3db Mon Sep 17 00:00:00 2001 +From 6e36198f10a2f63b89c89ebb5d5c185b20fb3a63 Mon Sep 17 00:00:00 2001 From: Kamil Dudka Date: Mon, 29 Mar 2010 17:20:34 +0000 Subject: [PATCH] coreutils-df-direct.patch @@ -11,10 +11,10 @@ Subject: [PATCH] coreutils-df-direct.patch create mode 100755 tests/df/direct.sh diff --git a/doc/coreutils.texi b/doc/coreutils.texi -index b420606..0ccb368 100644 +index 5b9a597..6810c15 100644 --- a/doc/coreutils.texi +++ b/doc/coreutils.texi -@@ -12597,6 +12597,13 @@ some systems (notably Solaris), doing this yields more up to date results, +@@ -11898,6 +11898,13 @@ some systems (notably SunOS), doing this yields more up to date results, but in general this option makes @command{df} much slower, especially when there are many or very busy file systems. @@ -27,12 +27,12 @@ index b420606..0ccb368 100644 + @item --total @opindex --total - @cindex grand total of file system size, usage and available space + @cindex grand total of disk size, usage and available space diff --git a/src/df.c b/src/df.c -index 75e638c..ef9f0a7 100644 +index 48025b9..c8efa5b 100644 --- a/src/df.c +++ b/src/df.c -@@ -121,6 +121,9 @@ static bool print_type; +@@ -125,6 +125,9 @@ static bool print_type; /* If true, print a grand total at the end. */ static bool print_grand_total; @@ -42,7 +42,7 @@ index 75e638c..ef9f0a7 100644 /* Grand total data. */ static struct fs_usage grand_fsu; -@@ -248,13 +251,15 @@ enum +@@ -252,13 +255,15 @@ enum NO_SYNC_OPTION = CHAR_MAX + 1, SYNC_OPTION, TOTAL_OPTION, @@ -53,16 +53,16 @@ index 75e638c..ef9f0a7 100644 static struct option const long_options[] = { - {"all", no_argument, nullptr, 'a'}, - {"block-size", required_argument, nullptr, 'B'}, -+ {"direct", no_argument, nullptr, DIRECT_OPTION}, - {"inodes", no_argument, nullptr, 'i'}, - {"human-readable", no_argument, nullptr, 'h'}, - {"si", no_argument, nullptr, 'H'}, -@@ -571,7 +576,10 @@ get_header (void) - for (idx_t col = 0; col < ncolumns; col++) + {"all", no_argument, NULL, 'a'}, + {"block-size", required_argument, NULL, 'B'}, ++ {"direct", no_argument, NULL, DIRECT_OPTION}, + {"inodes", no_argument, NULL, 'i'}, + {"human-readable", no_argument, NULL, 'h'}, + {"si", no_argument, NULL, 'H'}, +@@ -561,7 +566,10 @@ get_header (void) + for (col = 0; col < ncolumns; col++) { - char *cell; + char *cell = NULL; - char const *header = _(columns[col]->caption); + char const *header = (columns[col]->field == TARGET_FIELD + && direct_statfs)? @@ -71,7 +71,7 @@ index 75e638c..ef9f0a7 100644 if (columns[col]->field == SIZE_FIELD && (header_mode == DEFAULT_MODE -@@ -1446,6 +1454,17 @@ get_point (char const *point, const struct stat *statp) +@@ -1464,6 +1472,17 @@ get_point (const char *point, const struct stat *statp) static void get_entry (char const *name, struct stat const *statp) { @@ -79,17 +79,17 @@ index 75e638c..ef9f0a7 100644 + { + char *resolved = canonicalize_file_name (name); + if (resolved) -+ { -+ get_dev (NULL, resolved, name, NULL, NULL, false, false, NULL, false); -+ free (resolved); -+ return; -+ } ++ { ++ get_dev (NULL, resolved, name, NULL, NULL, false, false, NULL, false); ++ free (resolved); ++ return; ++ } + } + if ((S_ISBLK (statp->st_mode) || S_ISCHR (statp->st_mode)) - && get_device (name)) + && get_disk (name)) return; -@@ -1516,6 +1535,7 @@ or all file systems by default.\n\ +@@ -1534,6 +1553,7 @@ or all file systems by default.\n\ -B, --block-size=SIZE scale sizes by SIZE before printing them; e.g.,\n\ '-BM' prints sizes in units of 1,048,576 bytes;\n\ see SIZE format below\n\ @@ -97,7 +97,7 @@ index 75e638c..ef9f0a7 100644 -h, --human-readable print sizes in powers of 1024 (e.g., 1023M)\n\ -H, --si print sizes in powers of 1000 (e.g., 1.1G)\n\ "), stdout); -@@ -1610,6 +1630,9 @@ main (int argc, char **argv) +@@ -1624,6 +1644,9 @@ main (int argc, char **argv) xstrtol_fatal (e, oi, c, long_options, optarg); } break; @@ -107,14 +107,14 @@ index 75e638c..ef9f0a7 100644 case 'i': if (header_mode == OUTPUT_MODE) { -@@ -1706,6 +1729,13 @@ main (int argc, char **argv) +@@ -1720,6 +1743,13 @@ main (int argc, char **argv) } } + if (direct_statfs && show_local_fs) + { + error (0, 0, _("options --direct and --local (-l) are mutually " -+ "exclusive")); ++ "exclusive")); + usage (EXIT_FAILURE); + } + @@ -183,5 +183,5 @@ index 0000000..8e4cfb8 + +Exit $fail -- -2.52.0 +2.31.1 diff --git a/coreutils-getgrouplist.patch b/coreutils-getgrouplist.patch new file mode 100644 index 0000000..5349337 --- /dev/null +++ b/coreutils-getgrouplist.patch @@ -0,0 +1,94 @@ +diff --git a/lib/getugroups.c b/lib/getugroups.c +index 299bae6..8ece29b 100644 +--- a/lib/getugroups.c ++++ b/lib/getugroups.c +@@ -19,6 +19,9 @@ + + #include + ++/* We do not need this code if getgrouplist(3) is available. */ ++#ifndef HAVE_GETGROUPLIST ++ + #include "getugroups.h" + + #include +@@ -126,3 +129,4 @@ getugroups (int maxcount, gid_t *grouplist, char const *username, + } + + #endif /* HAVE_GRP_H */ ++#endif /* have getgrouplist */ +diff --git a/lib/mgetgroups.c b/lib/mgetgroups.c +index 76474c2..0a9d221 100644 +--- a/lib/mgetgroups.c ++++ b/lib/mgetgroups.c +@@ -31,6 +31,7 @@ + #endif + + #include "getugroups.h" ++#include "xalloc.h" + #include "xalloc-oversized.h" + + /* Work around an incompatibility of OS X 10.11: getgrouplist +@@ -121,9 +122,17 @@ mgetgroups (char const *username, gid_t gid, gid_t **groups) + /* else no username, so fall through and use getgroups. */ + #endif + +- max_n_groups = (username +- ? getugroups (0, NULL, username, gid) +- : getgroups (0, NULL)); ++ if (!username) ++ max_n_groups = getgroups(0, NULL); ++ else ++ { ++#ifdef HAVE_GETGROUPLIST ++ max_n_groups = 0; ++ getgrouplist (username, gid, NULL, &max_n_groups); ++#else ++ max_n_groups = getugroups (0, NULL, username, gid); ++#endif ++ } + + /* If we failed to count groups because there is no supplemental + group support, then return an array containing just GID. +@@ -145,10 +154,25 @@ mgetgroups (char const *username, gid_t gid, gid_t **groups) + if (g == NULL) + return -1; + +- ng = (username +- ? getugroups (max_n_groups, g, username, gid) +- : getgroups (max_n_groups - (gid != (gid_t) -1), +- g + (gid != (gid_t) -1))); ++ if (!username) ++ ng = getgroups (max_n_groups - (gid != (gid_t)-1), g + (gid != (gid_t)-1)); ++ else ++ { ++#ifdef HAVE_GETGROUPLIST ++ int e; ++ ng = max_n_groups; ++ while ((e = getgrouplist (username, gid, g, &ng)) == -1 ++ && ng > max_n_groups) ++ { ++ max_n_groups = ng; ++ g = xrealloc (g, max_n_groups * sizeof (GETGROUPS_T)); ++ } ++ if (e == -1) ++ ng = -1; ++#else ++ ng = getugroups (max_n_groups, g, username, gid); ++#endif ++ } + + if (ng < 0) + { +diff --git a/m4/jm-macros.m4 b/m4/jm-macros.m4 +index 62777c7..5180243 100644 +--- a/m4/jm-macros.m4 ++++ b/m4/jm-macros.m4 +@@ -82,6 +82,7 @@ AC_DEFUN([coreutils_MACROS], + fchown + fchmod + ftruncate ++ getgrouplist + iswspace + mkfifo + mbrlen diff --git a/coreutils-i18n-cut-old.patch b/coreutils-i18n-cut-old.patch new file mode 100644 index 0000000..757ee0f --- /dev/null +++ b/coreutils-i18n-cut-old.patch @@ -0,0 +1,565 @@ +diff --git a/src/cut.c b/src/cut.c +index 7ab6be4..022d0ad 100644 +--- a/src/cut.c ++++ b/src/cut.c +@@ -28,6 +28,11 @@ + #include + #include + #include ++ ++/* Get mbstate_t, mbrtowc(). */ ++#if HAVE_WCHAR_H ++# include ++#endif + #include "system.h" + + #include "error.h" +@@ -38,6 +43,18 @@ + + #include "set-fields.h" + ++/* MB_LEN_MAX is incorrectly defined to be 1 in at least one GCC ++ installation; work around this configuration error. */ ++#if !defined MB_LEN_MAX || MB_LEN_MAX < 2 ++# undef MB_LEN_MAX ++# define MB_LEN_MAX 16 ++#endif ++ ++/* Some systems, like BeOS, have multibyte encodings but lack mbstate_t. */ ++#if HAVE_MBRTOWC && defined mbstate_t ++# define mbrtowc(pwc, s, n, ps) (mbrtowc) (pwc, s, n, 0) ++#endif ++ + /* The official name of this program (e.g., no 'g' prefix). */ + #define PROGRAM_NAME "cut" + +@@ -54,6 +71,52 @@ + } \ + while (0) + ++/* Refill the buffer BUF to get a multibyte character. */ ++#define REFILL_BUFFER(BUF, BUFPOS, BUFLEN, STREAM) \ ++ do \ ++ { \ ++ if (BUFLEN < MB_LEN_MAX && !feof (STREAM) && !ferror (STREAM)) \ ++ { \ ++ memmove (BUF, BUFPOS, BUFLEN); \ ++ BUFLEN += fread (BUF + BUFLEN, sizeof(char), BUFSIZ, STREAM); \ ++ BUFPOS = BUF; \ ++ } \ ++ } \ ++ while (0) ++ ++/* Get wide character on BUFPOS. BUFPOS is not included after that. ++ If byte sequence is not valid as a character, CONVFAIL is true. Otherwise false. */ ++#define GET_NEXT_WC_FROM_BUFFER(WC, BUFPOS, BUFLEN, MBLENGTH, STATE, CONVFAIL) \ ++ do \ ++ { \ ++ mbstate_t state_bak; \ ++ \ ++ if (BUFLEN < 1) \ ++ { \ ++ WC = WEOF; \ ++ break; \ ++ } \ ++ \ ++ /* Get a wide character. */ \ ++ CONVFAIL = false; \ ++ state_bak = STATE; \ ++ MBLENGTH = mbrtowc ((wchar_t *)&WC, BUFPOS, BUFLEN, &STATE); \ ++ \ ++ switch (MBLENGTH) \ ++ { \ ++ case (size_t)-1: \ ++ case (size_t)-2: \ ++ CONVFAIL = true; \ ++ STATE = state_bak; \ ++ /* Fall througn. */ \ ++ \ ++ case 0: \ ++ MBLENGTH = 1; \ ++ break; \ ++ } \ ++ } \ ++ while (0) ++ + + /* Pointer inside RP. When checking if a byte or field is selected + by a finite range, we check if it is between CURRENT_RP.LO +@@ -61,6 +124,9 @@ + CURRENT_RP.HI then we make CURRENT_RP to point to the next range pair. */ + static struct field_range_pair *current_rp; + ++/* Length of the delimiter given as argument to -d. */ ++size_t delimlen; ++ + /* This buffer is used to support the semantics of the -s option + (or lack of same) when the specified field list includes (does + not include) the first field. In both of those cases, the entire +@@ -77,15 +143,25 @@ enum operating_mode + { + undefined_mode, + +- /* Output characters that are in the given bytes. */ ++ /* Output bytes that are at the given positions. */ + byte_mode, + ++ /* Output characters that are at the given positions. */ ++ character_mode, ++ + /* Output the given delimiter-separated fields. */ + field_mode + }; + + static enum operating_mode operating_mode; + ++/* If nonzero, when in byte mode, don't split multibyte characters. */ ++static int byte_mode_character_aware; ++ ++/* If nonzero, the function for single byte locale is work ++ if this program runs on multibyte locale. */ ++static int force_singlebyte_mode; ++ + /* If true do not output lines containing no delimiter characters. + Otherwise, all such lines are printed. This option is valid only + with field mode. */ +@@ -97,6 +173,9 @@ static bool complement; + + /* The delimiter character for field mode. */ + static unsigned char delim; ++#if HAVE_WCHAR_H ++static wchar_t wcdelim; ++#endif + + /* The delimiter for each line/record. */ + static unsigned char line_delim = '\n'; +@@ -164,7 +243,7 @@ Print selected parts of lines from each FILE to standard output.\n\ + -f, --fields=LIST select only these fields; also print any line\n\ + that contains no delimiter character, unless\n\ + the -s option is specified\n\ +- -n (ignored)\n\ ++ -n with -b: don't split multibyte characters\n\ + "), stdout); + fputs (_("\ + --complement complement the set of selected bytes, characters\n\ +@@ -280,6 +359,82 @@ cut_bytes (FILE *stream) + } + } + ++#if HAVE_MBRTOWC ++/* This function is in use for the following case. ++ ++ 1. Read from the stream STREAM, printing to standard output any selected ++ characters. ++ ++ 2. Read from stream STREAM, printing to standard output any selected bytes, ++ without splitting multibyte characters. */ ++ ++static void ++cut_characters_or_cut_bytes_no_split (FILE *stream) ++{ ++ uintmax_t idx; /* number of bytes or characters in the line so far. */ ++ char buf[MB_LEN_MAX + BUFSIZ]; /* For spooling a read byte sequence. */ ++ char *bufpos; /* Next read position of BUF. */ ++ size_t buflen; /* The length of the byte sequence in buf. */ ++ wint_t wc; /* A gotten wide character. */ ++ size_t mblength; /* The byte size of a multibyte character which shows ++ as same character as WC. */ ++ mbstate_t state; /* State of the stream. */ ++ bool convfail = false; /* true, when conversion failed. Otherwise false. */ ++ /* Whether to begin printing delimiters between ranges for the current line. ++ Set after we've begun printing data corresponding to the first range. */ ++ bool print_delimiter = false; ++ ++ idx = 0; ++ buflen = 0; ++ bufpos = buf; ++ memset (&state, '\0', sizeof(mbstate_t)); ++ ++ current_rp = frp; ++ ++ while (1) ++ { ++ REFILL_BUFFER (buf, bufpos, buflen, stream); ++ ++ GET_NEXT_WC_FROM_BUFFER (wc, bufpos, buflen, mblength, state, convfail); ++ (void) convfail; /* ignore unused */ ++ ++ if (wc == WEOF) ++ { ++ if (idx > 0) ++ putchar (line_delim); ++ break; ++ } ++ else if (wc == line_delim) ++ { ++ putchar (line_delim); ++ idx = 0; ++ print_delimiter = false; ++ current_rp = frp; ++ } ++ else ++ { ++ next_item (&idx); ++ if (print_kth (idx)) ++ { ++ if (output_delimiter_specified) ++ { ++ if (print_delimiter && is_range_start_index (idx)) ++ { ++ fwrite (output_delimiter_string, sizeof (char), ++ output_delimiter_length, stdout); ++ } ++ print_delimiter = true; ++ } ++ fwrite (bufpos, mblength, sizeof(char), stdout); ++ } ++ } ++ ++ buflen -= mblength; ++ bufpos += mblength; ++ } ++} ++#endif ++ + /* Read from stream STREAM, printing to standard output any selected fields. */ + + static void +@@ -425,13 +580,211 @@ cut_fields (FILE *stream) + } + } + ++#if HAVE_MBRTOWC ++static void ++cut_fields_mb (FILE *stream) ++{ ++ int c; ++ uintmax_t field_idx; ++ int found_any_selected_field; ++ int buffer_first_field; ++ int empty_input; ++ char buf[MB_LEN_MAX + BUFSIZ]; /* For spooling a read byte sequence. */ ++ char *bufpos; /* Next read position of BUF. */ ++ size_t buflen; /* The length of the byte sequence in buf. */ ++ wint_t wc = 0; /* A gotten wide character. */ ++ size_t mblength; /* The byte size of a multibyte character which shows ++ as same character as WC. */ ++ mbstate_t state; /* State of the stream. */ ++ bool convfail = false; /* true, when conversion failed. Otherwise false. */ ++ ++ current_rp = frp; ++ ++ found_any_selected_field = 0; ++ field_idx = 1; ++ bufpos = buf; ++ buflen = 0; ++ memset (&state, '\0', sizeof(mbstate_t)); ++ ++ c = getc (stream); ++ empty_input = (c == EOF); ++ if (c != EOF) ++ { ++ ungetc (c, stream); ++ wc = 0; ++ } ++ else ++ wc = WEOF; ++ ++ /* To support the semantics of the -s flag, we may have to buffer ++ all of the first field to determine whether it is `delimited.' ++ But that is unnecessary if all non-delimited lines must be printed ++ and the first field has been selected, or if non-delimited lines ++ must be suppressed and the first field has *not* been selected. ++ That is because a non-delimited line has exactly one field. */ ++ buffer_first_field = (suppress_non_delimited ^ !print_kth (1)); ++ ++ while (1) ++ { ++ if (field_idx == 1 && buffer_first_field) ++ { ++ int len = 0; ++ ++ while (1) ++ { ++ REFILL_BUFFER (buf, bufpos, buflen, stream); ++ ++ GET_NEXT_WC_FROM_BUFFER ++ (wc, bufpos, buflen, mblength, state, convfail); ++ ++ if (wc == WEOF) ++ break; ++ ++ field_1_buffer = xrealloc (field_1_buffer, len + mblength); ++ memcpy (field_1_buffer + len, bufpos, mblength); ++ len += mblength; ++ buflen -= mblength; ++ bufpos += mblength; ++ ++ if (!convfail && (wc == line_delim || wc == wcdelim)) ++ break; ++ } ++ ++ if (len <= 0 && wc == WEOF) ++ break; ++ ++ /* If the first field extends to the end of line (it is not ++ delimited) and we are printing all non-delimited lines, ++ print this one. */ ++ if (convfail || (!convfail && wc != wcdelim)) ++ { ++ if (suppress_non_delimited) ++ { ++ /* Empty. */ ++ } ++ else ++ { ++ fwrite (field_1_buffer, sizeof (char), len, stdout); ++ /* Make sure the output line is newline terminated. */ ++ if (convfail || (!convfail && wc != line_delim)) ++ putchar (line_delim); ++ } ++ continue; ++ } ++ ++ if (print_kth (1)) ++ { ++ /* Print the field, but not the trailing delimiter. */ ++ fwrite (field_1_buffer, sizeof (char), len - 1, stdout); ++ found_any_selected_field = 1; ++ } ++ next_item (&field_idx); ++ } ++ ++ if (wc != WEOF) ++ { ++ if (print_kth (field_idx)) ++ { ++ if (found_any_selected_field) ++ { ++ fwrite (output_delimiter_string, sizeof (char), ++ output_delimiter_length, stdout); ++ } ++ found_any_selected_field = 1; ++ } ++ ++ while (1) ++ { ++ REFILL_BUFFER (buf, bufpos, buflen, stream); ++ ++ GET_NEXT_WC_FROM_BUFFER ++ (wc, bufpos, buflen, mblength, state, convfail); ++ ++ if (wc == WEOF) ++ break; ++ else if (!convfail && (wc == wcdelim || wc == line_delim)) ++ { ++ buflen -= mblength; ++ bufpos += mblength; ++ break; ++ } ++ ++ if (print_kth (field_idx)) ++ fwrite (bufpos, mblength, sizeof(char), stdout); ++ ++ buflen -= mblength; ++ bufpos += mblength; ++ } ++ } ++ ++ if ((!convfail || wc == line_delim) && buflen < 1) ++ wc = WEOF; ++ ++ if (!convfail && wc == wcdelim) ++ next_item (&field_idx); ++ else if (wc == WEOF || (!convfail && wc == line_delim)) ++ { ++ if (found_any_selected_field ++ || (!empty_input && !(suppress_non_delimited && field_idx == 1))) ++ putchar (line_delim); ++ if (wc == WEOF) ++ break; ++ field_idx = 1; ++ current_rp = frp; ++ found_any_selected_field = 0; ++ } ++ } ++} ++#endif ++ + static void + cut_stream (FILE *stream) + { +- if (operating_mode == byte_mode) +- cut_bytes (stream); ++#if HAVE_MBRTOWC ++ if (MB_CUR_MAX > 1 && !force_singlebyte_mode) ++ { ++ switch (operating_mode) ++ { ++ case byte_mode: ++ if (byte_mode_character_aware) ++ cut_characters_or_cut_bytes_no_split (stream); ++ else ++ cut_bytes (stream); ++ break; ++ ++ case character_mode: ++ cut_characters_or_cut_bytes_no_split (stream); ++ break; ++ ++ case field_mode: ++ if (delimlen == 1) ++ { ++ /* Check if we have utf8 multibyte locale, so we can use this ++ optimization because of uniqueness of characters, which is ++ not true for e.g. SJIS */ ++ char * loc = setlocale(LC_CTYPE, NULL); ++ if (loc && (strstr (loc, "UTF-8") || strstr (loc, "utf-8") || ++ strstr (loc, "UTF8") || strstr (loc, "utf8"))) ++ { ++ cut_fields (stream); ++ break; ++ } ++ } ++ cut_fields_mb (stream); ++ break; ++ ++ default: ++ abort (); ++ } ++ } + else +- cut_fields (stream); ++#endif ++ { ++ if (operating_mode == field_mode) ++ cut_fields (stream); ++ else ++ cut_bytes (stream); ++ } + } + + /* Process file FILE to standard output. +@@ -483,6 +836,7 @@ main (int argc, char **argv) + bool ok; + bool delim_specified = false; + char *spec_list_string IF_LINT ( = NULL); ++ char mbdelim[MB_LEN_MAX + 1]; + + initialize_main (&argc, &argv); + set_program_name (argv[0]); +@@ -505,7 +859,6 @@ main (int argc, char **argv) + switch (optc) + { + case 'b': +- case 'c': + /* Build the byte list. */ + if (operating_mode != undefined_mode) + FATAL_ERROR (_("only one type of list may be specified")); +@@ -513,6 +866,14 @@ main (int argc, char **argv) + spec_list_string = optarg; + break; + ++ case 'c': ++ /* Build the character list. */ ++ if (operating_mode != undefined_mode) ++ FATAL_ERROR (_("only one type of list may be specified")); ++ operating_mode = character_mode; ++ spec_list_string = optarg; ++ break; ++ + case 'f': + /* Build the field list. */ + if (operating_mode != undefined_mode) +@@ -524,10 +885,38 @@ main (int argc, char **argv) + case 'd': + /* New delimiter. */ + /* Interpret -d '' to mean 'use the NUL byte as the delimiter.' */ +- if (optarg[0] != '\0' && optarg[1] != '\0') +- FATAL_ERROR (_("the delimiter must be a single character")); +- delim = optarg[0]; +- delim_specified = true; ++ { ++#if HAVE_MBRTOWC ++ if(MB_CUR_MAX > 1) ++ { ++ mbstate_t state; ++ ++ memset (&state, '\0', sizeof(mbstate_t)); ++ delimlen = mbrtowc (&wcdelim, optarg, strnlen(optarg, MB_LEN_MAX), &state); ++ ++ if (delimlen == (size_t)-1 || delimlen == (size_t)-2) ++ ++force_singlebyte_mode; ++ else ++ { ++ delimlen = (delimlen < 1) ? 1 : delimlen; ++ if (wcdelim != L'\0' && *(optarg + delimlen) != '\0') ++ FATAL_ERROR (_("the delimiter must be a single character")); ++ memcpy (mbdelim, optarg, delimlen); ++ mbdelim[delimlen] = '\0'; ++ if (delimlen == 1) ++ delim = *optarg; ++ } ++ } ++ ++ if (MB_CUR_MAX <= 1 || force_singlebyte_mode) ++#endif ++ { ++ if (optarg[0] != '\0' && optarg[1] != '\0') ++ FATAL_ERROR (_("the delimiter must be a single character")); ++ delim = (unsigned char) optarg[0]; ++ } ++ delim_specified = true; ++ } + break; + + case OUTPUT_DELIMITER_OPTION: +@@ -540,6 +929,7 @@ main (int argc, char **argv) + break; + + case 'n': ++ byte_mode_character_aware = 1; + break; + + case 's': +@@ -579,15 +969,34 @@ main (int argc, char **argv) + | (complement ? SETFLD_COMPLEMENT : 0) ); + + if (!delim_specified) +- delim = '\t'; ++ { ++ delim = '\t'; ++#ifdef HAVE_MBRTOWC ++ wcdelim = L'\t'; ++ mbdelim[0] = '\t'; ++ mbdelim[1] = '\0'; ++ delimlen = 1; ++#endif ++ } + + if (output_delimiter_string == NULL) + { +- static char dummy[2]; +- dummy[0] = delim; +- dummy[1] = '\0'; +- output_delimiter_string = dummy; +- output_delimiter_length = 1; ++#ifdef HAVE_MBRTOWC ++ if (MB_CUR_MAX > 1 && !force_singlebyte_mode) ++ { ++ output_delimiter_string = xstrdup(mbdelim); ++ output_delimiter_length = delimlen; ++ } ++ ++ if (MB_CUR_MAX <= 1 || force_singlebyte_mode) ++#endif ++ { ++ static char dummy[2]; ++ dummy[0] = delim; ++ dummy[1] = '\0'; ++ output_delimiter_string = dummy; ++ output_delimiter_length = 1; ++ } + } + + if (optind == argc) diff --git a/coreutils-i18n-expand-unexpand.patch b/coreutils-i18n-expand-unexpand.patch new file mode 100644 index 0000000..32089cf --- /dev/null +++ b/coreutils-i18n-expand-unexpand.patch @@ -0,0 +1,864 @@ +From e87ab5b991b08092a7e07af82b3ec822a8604151 Mon Sep 17 00:00:00 2001 +From: Ondrej Oprala +Date: Wed, 5 Aug 2015 09:15:09 +0200 +Subject: [PATCH] expand,unexpand: add multibyte support +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +* NEWS: Mention the changes. +* bootstrap.conf: Add mbfile to the list of modules. +* configure.ac: Properly initialize mbfile. +* src/expand.c (expand): Iterate over multibyte characters properly. +* src/unexpand.c (unexpand): Iterate over multibyte characters +properly. +* tests/local.mk: Add new tests. +* tests/{expand,unexpand}/mb.sh: New tests. + +Co-authored-by: Pádraig Brady +--- + bootstrap.conf | 1 + + configure.ac | 2 + + lib/mbfile.c | 3 + + lib/mbfile.h | 255 +++++++++++++++++++++++++++++++++++++++++++++++++++ + m4/mbfile.m4 | 14 +++ + src/expand.c | 43 +++++---- + src/local.mk | 4 +- + src/unexpand.c | 54 +++++++---- + tests/expand/mb.sh | 98 ++++++++++++++++++++ + tests/local.mk | 2 + + tests/unexpand/mb.sh | 97 ++++++++++++++++++++ + 10 files changed, 535 insertions(+), 34 deletions(-) + create mode 100644 lib/mbfile.c + create mode 100644 lib/mbfile.h + create mode 100644 m4/mbfile.m4 + create mode 100755 tests/expand/mb.sh + create mode 100755 tests/unexpand/mb.sh + +diff --git a/bootstrap.conf b/bootstrap.conf +index 8a0ff31..a1c78b2 100644 +--- a/bootstrap.conf ++++ b/bootstrap.conf +@@ -152,6 +152,7 @@ gnulib_modules=" + maintainer-makefile + malloc-gnu + manywarnings ++ mbfile + mbrlen + mbrtowc + mbsalign +diff --git a/configure.ac b/configure.ac +index 1e74b36..24c9725 100644 +--- a/configure.ac ++++ b/configure.ac +@@ -427,6 +427,8 @@ fi + # I'm leaving it here for now. This whole thing needs to be modernized... + gl_WINSIZE_IN_PTEM + ++gl_MBFILE ++ + gl_HEADER_TIOCGWINSZ_IN_TERMIOS_H + + if test $gl_cv_sys_tiocgwinsz_needs_termios_h = no && \ +diff --git a/lib/mbfile.c b/lib/mbfile.c +new file mode 100644 +index 0000000..b0a468e +--- /dev/null ++++ b/lib/mbfile.c +@@ -0,0 +1,3 @@ ++#include ++#define MBFILE_INLINE _GL_EXTERN_INLINE ++#include "mbfile.h" +diff --git a/lib/mbfile.h b/lib/mbfile.h +new file mode 100644 +index 0000000..11f1b12 +--- /dev/null ++++ b/lib/mbfile.h +@@ -0,0 +1,255 @@ ++/* Multibyte character I/O: macros for multi-byte encodings. ++ Copyright (C) 2001, 2005, 2009-2015 Free Software Foundation, Inc. ++ ++ This program is free software: you can redistribute it and/or modify ++ it under the terms of the GNU General Public License as published by ++ the Free Software Foundation; either version 3 of the License, or ++ (at your option) any later version. ++ ++ This program is distributed in the hope that it will be useful, ++ but WITHOUT ANY WARRANTY; without even the implied warranty of ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ GNU General Public License for more details. ++ ++ You should have received a copy of the GNU General Public License ++ along with this program. If not, see . */ ++ ++/* Written by Mitsuru Chinen ++ and Bruno Haible . */ ++ ++/* The macros in this file implement multi-byte character input from a ++ stream. ++ ++ mb_file_t ++ is the type for multibyte character input stream, usable for variable ++ declarations. ++ ++ mbf_char_t ++ is the type for multibyte character or EOF, usable for variable ++ declarations. ++ ++ mbf_init (mbf, stream) ++ initializes the MB_FILE for reading from stream. ++ ++ mbf_getc (mbc, mbf) ++ reads the next multibyte character from mbf and stores it in mbc. ++ ++ mb_iseof (mbc) ++ returns true if mbc represents the EOF value. ++ ++ Here are the function prototypes of the macros. ++ ++ extern void mbf_init (mb_file_t mbf, FILE *stream); ++ extern void mbf_getc (mbf_char_t mbc, mb_file_t mbf); ++ extern bool mb_iseof (const mbf_char_t mbc); ++ */ ++ ++#ifndef _MBFILE_H ++#define _MBFILE_H 1 ++ ++#include ++#include ++#include ++#include ++ ++/* Tru64 with Desktop Toolkit C has a bug: must be included before ++ . ++ BSD/OS 4.1 has a bug: and must be included before ++ . */ ++#include ++#include ++#include ++ ++#include "mbchar.h" ++ ++#ifndef _GL_INLINE_HEADER_BEGIN ++ #error "Please include config.h first." ++#endif ++_GL_INLINE_HEADER_BEGIN ++#ifndef MBFILE_INLINE ++# define MBFILE_INLINE _GL_INLINE ++#endif ++ ++struct mbfile_multi { ++ FILE *fp; ++ bool eof_seen; ++ bool have_pushback; ++ mbstate_t state; ++ unsigned int bufcount; ++ char buf[MBCHAR_BUF_SIZE]; ++ struct mbchar pushback; ++}; ++ ++MBFILE_INLINE void ++mbfile_multi_getc (struct mbchar *mbc, struct mbfile_multi *mbf) ++{ ++ size_t bytes; ++ ++ /* If EOF has already been seen, don't use getc. This matters if ++ mbf->fp is connected to an interactive tty. */ ++ if (mbf->eof_seen) ++ goto eof; ++ ++ /* Return character pushed back, if there is one. */ ++ if (mbf->have_pushback) ++ { ++ mb_copy (mbc, &mbf->pushback); ++ mbf->have_pushback = false; ++ return; ++ } ++ ++ /* Before using mbrtowc, we need at least one byte. */ ++ if (mbf->bufcount == 0) ++ { ++ int c = getc (mbf->fp); ++ if (c == EOF) ++ { ++ mbf->eof_seen = true; ++ goto eof; ++ } ++ mbf->buf[0] = (unsigned char) c; ++ mbf->bufcount++; ++ } ++ ++ /* Handle most ASCII characters quickly, without calling mbrtowc(). */ ++ if (mbf->bufcount == 1 && mbsinit (&mbf->state) && is_basic (mbf->buf[0])) ++ { ++ /* These characters are part of the basic character set. ISO C 99 ++ guarantees that their wide character code is identical to their ++ char code. */ ++ mbc->wc = mbc->buf[0] = mbf->buf[0]; ++ mbc->wc_valid = true; ++ mbc->ptr = &mbc->buf[0]; ++ mbc->bytes = 1; ++ mbf->bufcount = 0; ++ return; ++ } ++ ++ /* Use mbrtowc on an increasing number of bytes. Read only as many bytes ++ from mbf->fp as needed. This is needed to give reasonable interactive ++ behaviour when mbf->fp is connected to an interactive tty. */ ++ for (;;) ++ { ++ /* We don't know whether the 'mbrtowc' function updates the state when ++ it returns -2, - this is the ISO C 99 and glibc-2.2 behaviour - or ++ not - amended ANSI C, glibc-2.1 and Solaris 2.7 behaviour. We ++ don't have an autoconf test for this, yet. ++ The new behaviour would allow us to feed the bytes one by one into ++ mbrtowc. But the old behaviour forces us to feed all bytes since ++ the end of the last character into mbrtowc. Since we want to retry ++ with more bytes when mbrtowc returns -2, we must backup the state ++ before calling mbrtowc, because implementations with the new ++ behaviour will clobber it. */ ++ mbstate_t backup_state = mbf->state; ++ ++ bytes = mbrtowc (&mbc->wc, &mbf->buf[0], mbf->bufcount, &mbf->state); ++ ++ if (bytes == (size_t) -1) ++ { ++ /* An invalid multibyte sequence was encountered. */ ++ /* Return a single byte. */ ++ bytes = 1; ++ mbc->wc_valid = false; ++ break; ++ } ++ else if (bytes == (size_t) -2) ++ { ++ /* An incomplete multibyte character. */ ++ mbf->state = backup_state; ++ if (mbf->bufcount == MBCHAR_BUF_SIZE) ++ { ++ /* An overlong incomplete multibyte sequence was encountered. */ ++ /* Return a single byte. */ ++ bytes = 1; ++ mbc->wc_valid = false; ++ break; ++ } ++ else ++ { ++ /* Read one more byte and retry mbrtowc. */ ++ int c = getc (mbf->fp); ++ if (c == EOF) ++ { ++ /* An incomplete multibyte character at the end. */ ++ mbf->eof_seen = true; ++ bytes = mbf->bufcount; ++ mbc->wc_valid = false; ++ break; ++ } ++ mbf->buf[mbf->bufcount] = (unsigned char) c; ++ mbf->bufcount++; ++ } ++ } ++ else ++ { ++ if (bytes == 0) ++ { ++ /* A null wide character was encountered. */ ++ bytes = 1; ++ assert (mbf->buf[0] == '\0'); ++ assert (mbc->wc == 0); ++ } ++ mbc->wc_valid = true; ++ break; ++ } ++ } ++ ++ /* Return the multibyte sequence mbf->buf[0..bytes-1]. */ ++ mbc->ptr = &mbc->buf[0]; ++ memcpy (&mbc->buf[0], &mbf->buf[0], bytes); ++ mbc->bytes = bytes; ++ ++ mbf->bufcount -= bytes; ++ if (mbf->bufcount > 0) ++ { ++ /* It's not worth calling memmove() for so few bytes. */ ++ unsigned int count = mbf->bufcount; ++ char *p = &mbf->buf[0]; ++ ++ do ++ { ++ *p = *(p + bytes); ++ p++; ++ } ++ while (--count > 0); ++ } ++ return; ++ ++eof: ++ /* An mbchar_t with bytes == 0 is used to indicate EOF. */ ++ mbc->ptr = NULL; ++ mbc->bytes = 0; ++ mbc->wc_valid = false; ++ return; ++} ++ ++MBFILE_INLINE void ++mbfile_multi_ungetc (const struct mbchar *mbc, struct mbfile_multi *mbf) ++{ ++ mb_copy (&mbf->pushback, mbc); ++ mbf->have_pushback = true; ++} ++ ++typedef struct mbfile_multi mb_file_t; ++ ++typedef mbchar_t mbf_char_t; ++ ++#define mbf_init(mbf, stream) \ ++ ((mbf).fp = (stream), \ ++ (mbf).eof_seen = false, \ ++ (mbf).have_pushback = false, \ ++ memset (&(mbf).state, '\0', sizeof (mbstate_t)), \ ++ (mbf).bufcount = 0) ++ ++#define mbf_getc(mbc, mbf) mbfile_multi_getc (&(mbc), &(mbf)) ++ ++#define mbf_ungetc(mbc, mbf) mbfile_multi_ungetc (&(mbc), &(mbf)) ++ ++#define mb_iseof(mbc) ((mbc).bytes == 0) ++ ++#ifndef _GL_INLINE_HEADER_BEGIN ++ #error "Please include config.h first." ++#endif ++_GL_INLINE_HEADER_BEGIN ++ ++#endif /* _MBFILE_H */ +diff --git a/m4/mbfile.m4 b/m4/mbfile.m4 +new file mode 100644 +index 0000000..8589902 +--- /dev/null ++++ b/m4/mbfile.m4 +@@ -0,0 +1,14 @@ ++# mbfile.m4 serial 7 ++dnl Copyright (C) 2005, 2008-2015 Free Software Foundation, Inc. ++dnl This file is free software; the Free Software Foundation ++dnl gives unlimited permission to copy and/or distribute it, ++dnl with or without modifications, as long as this notice is preserved. ++ ++dnl autoconf tests required for use of mbfile.h ++dnl From Bruno Haible. ++ ++AC_DEFUN([gl_MBFILE], ++[ ++ AC_REQUIRE([AC_TYPE_MBSTATE_T]) ++ : ++]) +diff --git a/src/expand.c b/src/expand.c +index 9fa2e10..380e020 100644 +--- a/src/expand.c ++++ b/src/expand.c +@@ -37,6 +37,9 @@ + #include + #include + #include ++ ++#include ++ + #include "system.h" + #include "die.h" + #include "xstrndup.h" +@@ -100,19 +103,19 @@ expand (void) + { + /* Input stream. */ + FILE *fp = next_file (NULL); ++ mb_file_t mbf; ++ mbf_char_t c; + + if (!fp) + return; + ++ mbf_init (mbf, fp); ++ + while (true) + { +- /* Input character, or EOF. */ +- int c; +- + /* If true, perform translations. */ + bool convert = true; + +- + /* The following variables have valid values only when CONVERT + is true: */ + +@@ -122,17 +125,23 @@ expand (void) + /* Index in TAB_LIST of next tab stop to examine. */ + size_t tab_index = 0; + +- + /* Convert a line of text. */ + + do + { +- while ((c = getc (fp)) < 0 && (fp = next_file (fp))) +- continue; ++ do { ++ mbf_getc (c, mbf); ++ if (mb_iseof (c)) ++ { ++ mbf_init (mbf, fp = next_file (fp)); ++ continue; ++ } ++ } ++ while (false); + + if (convert) + { +- if (c == '\t') ++ if (mb_iseq (c, '\t')) + { + /* Column the next input tab stop is on. */ + uintmax_t next_tab_column; +@@ -151,32 +160,34 @@ expand (void) + if (putchar (' ') < 0) + die (EXIT_FAILURE, errno, _("write error")); + +- c = ' '; ++ mb_setascii (&c, ' '); + } +- else if (c == '\b') ++ else if (mb_iseq (c, '\b')) + { + /* Go back one column, and force recalculation of the + next tab stop. */ + column -= !!column; + tab_index -= !!tab_index; + } +- else ++ /* A leading control character could make us trip over. */ ++ else if (!mb_iscntrl (c)) + { +- column++; ++ column += mb_width (c); + if (!column) + die (EXIT_FAILURE, 0, _("input line is too long")); + } + +- convert &= convert_entire_line || !! isblank (c); ++ convert &= convert_entire_line || mb_isblank (c); + } + +- if (c < 0) ++ if (mb_iseof (c)) + return; + +- if (putchar (c) < 0) ++ mb_putc (c, stdout); ++ if (ferror (stdout)) + die (EXIT_FAILURE, errno, _("write error")); + } +- while (c != '\n'); ++ while (!mb_iseq (c, '\n')); + } + } + +diff --git a/src/local.mk b/src/local.mk +index 72db9c704..ef3bfa469 100644 +--- a/src/local.mk ++++ b/src/local.mk +@@ -415,8 +415,8 @@ src_basenc_CPPFLAGS = -DBASE_TYPE=42 $(AM_CPPFLAGS) + + src_ginstall_CPPFLAGS = -DENABLE_MATCHPATHCON=1 $(AM_CPPFLAGS) + +-src_expand_SOURCES = src/expand.c src/expand-common.c +-src_unexpand_SOURCES = src/unexpand.c src/expand-common.c ++src_expand_SOURCES = src/expand.c src/expand-common.c lib/mbfile.c ++src_unexpand_SOURCES = src/unexpand.c src/expand-common.c lib/mbfile.c + + # Ensure we don't link against libcoreutils.a as that lib is + # not compiled with -fPIC which causes issues on 64 bit at least +diff --git a/src/unexpand.c b/src/unexpand.c +index 7801274..569a7ee 100644 +--- a/src/unexpand.c ++++ b/src/unexpand.c +@@ -38,6 +38,9 @@ + #include + #include + #include ++ ++#include ++ + #include "system.h" + #include "die.h" + #include "xstrndup.h" +@@ -107,11 +110,12 @@ unexpand (void) + { + /* Input stream. */ + FILE *fp = next_file (NULL); ++ mb_file_t mbf; + + /* The array of pending blanks. In non-POSIX locales, blanks can + include characters other than spaces, so the blanks must be + stored, not merely counted. */ +- char *pending_blank; ++ mbf_char_t *pending_blank; + + if (!fp) + return; +@@ -119,12 +123,14 @@ unexpand (void) + /* The worst case is a non-blank character, then one blank, then a + tab stop, then MAX_COLUMN_WIDTH - 1 blanks, then a non-blank; so + allocate MAX_COLUMN_WIDTH bytes to store the blanks. */ +- pending_blank = xmalloc (max_column_width); ++ pending_blank = xmalloc (max_column_width * sizeof (mbf_char_t)); ++ ++ mbf_init (mbf, fp); + + while (true) + { + /* Input character, or EOF. */ +- int c; ++ mbf_char_t c; + + /* If true, perform translations. */ + bool convert = true; +@@ -158,12 +164,19 @@ unexpand (void) + + do + { +- while ((c = getc (fp)) < 0 && (fp = next_file (fp))) +- continue; ++ do { ++ mbf_getc (c, mbf); ++ if (mb_iseof (c)) ++ { ++ mbf_init (mbf, fp = next_file (fp)); ++ continue; ++ } ++ } ++ while (false); + + if (convert) + { +- bool blank = !! isblank (c); ++ bool blank = mb_isblank (c); + + if (blank) + { +@@ -180,16 +193,16 @@ unexpand (void) + if (next_tab_column < column) + die (EXIT_FAILURE, 0, _("input line is too long")); + +- if (c == '\t') ++ if (mb_iseq (c, '\t')) + { + column = next_tab_column; + + if (pending) +- pending_blank[0] = '\t'; ++ mb_setascii (&pending_blank[0], '\t'); + } + else + { +- column++; ++ column += mb_width (c); + + if (! (prev_blank && column == next_tab_column)) + { +@@ -197,13 +210,14 @@ unexpand (void) + will be replaced by tabs. */ + if (column == next_tab_column) + one_blank_before_tab_stop = true; +- pending_blank[pending++] = c; ++ mb_copy (&pending_blank[pending++], &c); + prev_blank = true; + continue; + } + + /* Replace the pending blanks by a tab or two. */ +- pending_blank[0] = c = '\t'; ++ mb_setascii (&c, '\t'); ++ mb_setascii (&pending_blank[0], '\t'); + } + + /* Discard pending blanks, unless it was a single +@@ -211,7 +225,7 @@ unexpand (void) + pending = one_blank_before_tab_stop; + } + } +- else if (c == '\b') ++ else if (mb_iseq (c, '\b')) + { + /* Go back one column, and force recalculation of the + next tab stop. */ +@@ -221,7 +235,7 @@ unexpand (void) + } + else + { +- column++; ++ column += mb_width (c); + if (!column) + die (EXIT_FAILURE, 0, _("input line is too long")); + } +@@ -229,8 +243,11 @@ unexpand (void) + if (pending) + { + if (pending > 1 && one_blank_before_tab_stop) +- pending_blank[0] = '\t'; +- if (fwrite (pending_blank, 1, pending, stdout) != pending) ++ mb_setascii (&pending_blank[0], '\t'); ++ ++ for (int n = 0; n < pending; ++n) ++ mb_putc (pending_blank[n], stdout); ++ if (ferror (stdout)) + die (EXIT_FAILURE, errno, _("write error")); + pending = 0; + one_blank_before_tab_stop = false; +@@ -240,16 +257,17 @@ unexpand (void) + convert &= convert_entire_line || blank; + } + +- if (c < 0) ++ if (mb_iseof (c)) + { + free (pending_blank); + return; + } + +- if (putchar (c) < 0) ++ mb_putc (c, stdout); ++ if (ferror (stdout)) + die (EXIT_FAILURE, errno, _("write error")); + } +- while (c != '\n'); ++ while (!mb_iseq (c, '\n')); + } + } + +diff --git a/tests/expand/mb.sh b/tests/expand/mb.sh +new file mode 100755 +index 0000000..7971e18 +--- /dev/null ++++ b/tests/expand/mb.sh +@@ -0,0 +1,98 @@ ++#!/bin/sh ++ ++# Copyright (C) 2012-2015 Free Software Foundation, Inc. ++ ++# This program is free software: you can redistribute it and/or modify ++# it under the terms of the GNU General Public License as published by ++# the Free Software Foundation, either version 3 of the License, or ++# (at your option) any later version. ++ ++# This program is distributed in the hope that it will be useful, ++# but WITHOUT ANY WARRANTY; without even the implied warranty of ++# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++# GNU General Public License for more details. ++ ++# You should have received a copy of the GNU General Public License ++# along with this program. If not, see . ++ ++. "${srcdir=.}/tests/init.sh"; path_prepend_ ./src ++print_ver_ expand ++ ++export LC_ALL=en_US.UTF-8 ++ ++#input containing multibyte characters ++cat <<\EOF > in || framework_failure_ ++1234567812345678123456781 ++. . . . ++a b c d ++. . . . ++ä ö ü ß ++. . . . ++EOF ++env printf ' äöü\t. öüä. \tä xx\n' >> in || framework_failure_ ++ ++cat <<\EOF > exp || framework_failure_ ++1234567812345678123456781 ++. . . . ++a b c d ++. . . . ++ä ö ü ß ++. . . . ++ äöü . öüä. ä xx ++EOF ++ ++expand < in > out || fail=1 ++compare exp out > /dev/null 2>&1 || fail=1 ++ ++#test characters with display widths != 1 ++env printf '12345678 ++e\t|ascii(1) ++\u00E9\t|composed(1) ++e\u0301\t|decomposed(1) ++\u3000\t|ideo-space(2) ++\uFF0D\t|full-hypen(2) ++' > in || framework_failure_ ++ ++env printf '12345678 ++e |ascii(1) ++\u00E9 |composed(1) ++e\u0301 |decomposed(1) ++\u3000 |ideo-space(2) ++\uFF0D |full-hypen(2) ++' > exp || framework_failure_ ++ ++expand < in > out || fail=1 ++compare exp out > /dev/null 2>&1 || fail=1 ++ ++#shouldn't fail with "input line too long" ++#when a line starts with a control character ++env printf '\n' > in || framework_failure_ ++ ++expand < in > out || fail=1 ++compare in out > /dev/null 2>&1 || fail=1 ++ ++#non-Unicode characters interspersed between Unicode ones ++env printf '12345678 ++\t\xFF| ++\xFF\t| ++\t\xFFä| ++ä\xFF\t| ++\tä\xFF| ++\xFF\tä| ++äbcdef\xFF\t| ++' > in || framework_failure_ ++ ++env printf '12345678 ++ \xFF| ++\xFF | ++ \xFFä| ++ä\xFF | ++ ä\xFF| ++\xFF ä| ++äbcdef\xFF | ++' > exp || framework_failure_ ++ ++expand < in > out || fail=1 ++compare exp out > /dev/null 2>&1 || fail=1 ++ ++exit $fail +diff --git a/tests/local.mk b/tests/local.mk +index 192f776..8053397 100644 +--- a/tests/local.mk ++++ b/tests/local.mk +@@ -544,6 +544,7 @@ all_tests = \ + tests/du/threshold.sh \ + tests/du/trailing-slash.sh \ + tests/du/two-args.sh \ ++ tests/expand/mb.sh \ + tests/id/gnu-zero-uids.sh \ + tests/id/no-context.sh \ + tests/id/context.sh \ +@@ -684,6 +685,7 @@ all_tests = \ + tests/touch/read-only.sh \ + tests/touch/relative.sh \ + tests/touch/trailing-slash.sh \ ++ tests/unexpand/mb.sh \ + $(all_root_tests) + + # See tests/factor/create-test.sh. +diff --git a/tests/unexpand/mb.sh b/tests/unexpand/mb.sh +new file mode 100755 +index 0000000..60d4c1a +--- /dev/null ++++ b/tests/unexpand/mb.sh +@@ -0,0 +1,97 @@ ++#!/bin/sh ++ ++# Copyright (C) 2012-2015 Free Software Foundation, Inc. ++ ++# This program is free software: you can redistribute it and/or modify ++# it under the terms of the GNU General Public License as published by ++# the Free Software Foundation, either version 3 of the License, or ++# (at your option) any later version. ++ ++# This program is distributed in the hope that it will be useful, ++# but WITHOUT ANY WARRANTY; without even the implied warranty of ++# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++# GNU General Public License for more details. ++ ++# You should have received a copy of the GNU General Public License ++# along with this program. If not, see . ++ ++. "${srcdir=.}/tests/init.sh"; path_prepend_ ./src ++print_ver_ unexpand ++ ++export LC_ALL=en_US.UTF-8 ++ ++#input containing multibyte characters ++cat > in <<\EOF ++1234567812345678123456781 ++. . . . ++a b c d ++. . . . ++ä ö ü ß ++. . . . ++ äöü . öüä. ä xx ++EOF ++ ++cat > exp <<\EOF ++1234567812345678123456781 ++. . . . ++a b c d ++. . . . ++ä ö ü ß ++. . . . ++ äöü . öüä. ä xx ++EOF ++ ++unexpand -a < in > out || fail=1 ++compare exp out > /dev/null 2>&1 || fail=1 ++ ++#test characters with a display width larger than 1 ++ ++env printf '12345678 ++e |ascii(1) ++\u00E9 |composed(1) ++e\u0301 |decomposed(1) ++\u3000 |ideo-space(2) ++\uFF0D |full-hypen(2) ++' > in || framework_failure_ ++ ++env printf '12345678 ++e\t|ascii(1) ++\u00E9\t|composed(1) ++e\u0301\t|decomposed(1) ++\u3000\t|ideo-space(2) ++\uFF0D\t|full-hypen(2) ++' > exp || framework_failure_ ++ ++unexpand -a < in > out || fail=1 ++compare exp out > /dev/null 2>&1 || fail=1 ++ ++#test input where a blank of width > 1 is not being substituted ++in="$(LC_ALL=en_US.UTF-8 printf ' \u3000 ö ü ß')" ++exp=' 　 ö ü ß' ++ ++unexpand -a < in > out || fail=1 ++compare exp out > /dev/null 2>&1 || fail=1 ++ ++#non-Unicode characters interspersed between Unicode ones ++env printf '12345678 ++ \xFF| ++\xFF | ++ \xFFä| ++ä\xFF | ++ ä\xFF| ++\xFF ä| ++äbcdef\xFF | ++' > in || framework_failure_ ++ ++env printf '12345678 ++\t\xFF| ++\xFF\t| ++\t\xFFä| ++ä\xFF\t| ++\tä\xFF| ++\xFF\tä| ++äbcdef\xFF\t| ++' > exp || framework_failure_ ++ ++unexpand -a < in > out || fail=1 ++compare exp out > /dev/null 2>&1 || fail=1 +-- +2.7.4 + diff --git a/coreutils-i18n-fix-unexpand.patch b/coreutils-i18n-fix-unexpand.patch new file mode 100644 index 0000000..f0c347c --- /dev/null +++ b/coreutils-i18n-fix-unexpand.patch @@ -0,0 +1,28 @@ +From 02424bfcd719bbaa695f4e1c3ef17ad91b0d23c0 Mon Sep 17 00:00:00 2001 +From: Lubomir Rintel +Date: Thu, 28 Jan 2016 20:57:22 +0100 +Subject: [PATCH] unexpand: fix blank line handling + + echo '' |./src/unexpand -a + +Really? +--- + src/unexpand.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/src/unexpand.c b/src/unexpand.c +index 569a7ee..3bbbd66 100644 +--- a/src/unexpand.c ++++ b/src/unexpand.c +@@ -233,7 +233,7 @@ unexpand (void) + next_tab_column = column; + tab_index -= !!tab_index; + } +- else ++ else if (!mb_iseq (c, '\n')) + { + column += mb_width (c); + if (!column) +-- +2.7.4 + diff --git a/coreutils-i18n-fix2-expand-unexpand.patch b/coreutils-i18n-fix2-expand-unexpand.patch new file mode 100644 index 0000000..b34d7b7 --- /dev/null +++ b/coreutils-i18n-fix2-expand-unexpand.patch @@ -0,0 +1,108 @@ +diff --git a/src/expand.c b/src/expand.c +index 380e020..310b349 100644 +--- a/src/expand.c ++++ b/src/expand.c +@@ -129,15 +129,19 @@ expand (void) + + do + { +- do { ++ while (true) { + mbf_getc (c, mbf); +- if (mb_iseof (c)) ++ if ((mb_iseof (c)) && (fp = next_file (fp))) + { +- mbf_init (mbf, fp = next_file (fp)); ++ mbf_init (mbf, fp); + continue; + } ++ else ++ { ++ break; ++ } + } +- while (false); ++ + + if (convert) + { +diff --git a/src/unexpand.c b/src/unexpand.c +index 3bbbd66..863a90a 100644 +--- a/src/unexpand.c ++++ b/src/unexpand.c +@@ -164,15 +164,19 @@ unexpand (void) + + do + { +- do { ++ while (true) { + mbf_getc (c, mbf); +- if (mb_iseof (c)) ++ if ((mb_iseof (c)) && (fp = next_file (fp))) + { +- mbf_init (mbf, fp = next_file (fp)); ++ mbf_init (mbf, fp); + continue; + } ++ else ++ { ++ break; ++ } + } +- while (false); ++ + + if (convert) + { +diff --git a/tests/expand/mb.sh b/tests/expand/mb.sh +index 7971e18..031be7a 100755 +--- a/tests/expand/mb.sh ++++ b/tests/expand/mb.sh +@@ -44,6 +44,20 @@ EOF + expand < in > out || fail=1 + compare exp out > /dev/null 2>&1 || fail=1 + ++#multiple files as an input ++cat <<\EOF >> exp || framework_failure_ ++1234567812345678123456781 ++. . . . ++a b c d ++. . . . ++ä ö ü ß ++. . . . ++ äöü . öüä. ä xx ++EOF ++ ++expand ./in ./in > out || fail=1 ++compare exp out > /dev/null 2>&1 || fail=1 ++ + #test characters with display widths != 1 + env printf '12345678 + e\t|ascii(1) +diff --git a/tests/unexpand/mb.sh b/tests/unexpand/mb.sh +index 60d4c1a..8d75652 100755 +--- a/tests/unexpand/mb.sh ++++ b/tests/unexpand/mb.sh +@@ -44,6 +44,22 @@ EOF + unexpand -a < in > out || fail=1 + compare exp out > /dev/null 2>&1 || fail=1 + ++ ++#multiple files as an input ++cat >> exp <<\EOF ++1234567812345678123456781 ++. . . . ++a b c d ++. . . . ++ä ö ü ß ++. . . . ++ äöü . öüä. ä xx ++EOF ++ ++ ++unexpand -a ./in ./in > out || fail=1 ++compare exp out > /dev/null 2>&1 || fail=1 ++ + #test characters with a display width larger than 1 + + env printf '12345678 diff --git a/coreutils-i18n-fold-newline.patch b/coreutils-i18n-fold-newline.patch new file mode 100644 index 0000000..f7286ef --- /dev/null +++ b/coreutils-i18n-fold-newline.patch @@ -0,0 +1,80 @@ +From ff424639fe863cbd6963add1a79b97290c1606c6 Mon Sep 17 00:00:00 2001 +From: rpm-build +Date: Fri, 3 Feb 2017 12:26:53 +0100 +Subject: [PATCH] fold.c: preserve new-lines in mutlibyte text + +--- + src/fold.c | 49 ++++++++++++++++++++++++------------------------- + 1 file changed, 24 insertions(+), 25 deletions(-) + +diff --git a/src/fold.c b/src/fold.c +index d23edd5..8c232a7 100644 +--- a/src/fold.c ++++ b/src/fold.c +@@ -342,39 +342,38 @@ fold_multibyte_text (FILE *istream, size_t width, int *saved_errno) + } + + rescan: +- if (operating_mode == byte_mode) /* byte mode */ ++ if (convfail) ++ increment = 1; ++ else if (wc == L'\n') ++ { ++ /* preserve newline */ ++ fwrite (line_out, sizeof(char), offset_out, stdout); ++ START_NEW_LINE; ++ continue; ++ } ++ else if (operating_mode == byte_mode) /* byte mode */ + increment = mblength; + else if (operating_mode == character_mode) /* character mode */ + increment = 1; +- else /* column mode */ ++ else /* column mode */ + { +- if (convfail) +- increment = 1; +- else ++ switch (wc) + { +- switch (wc) +- { +- case L'\n': +- fwrite (line_out, sizeof(char), offset_out, stdout); +- START_NEW_LINE; +- continue; ++ case L'\b': ++ increment = (column > 0) ? -1 : 0; ++ break; + +- case L'\b': +- increment = (column > 0) ? -1 : 0; +- break; ++ case L'\r': ++ increment = -1 * column; ++ break; + +- case L'\r': +- increment = -1 * column; +- break; ++ case L'\t': ++ increment = 8 - column % 8; ++ break; + +- case L'\t': +- increment = 8 - column % 8; +- break; +- +- default: +- increment = wcwidth (wc); +- increment = (increment < 0) ? 0 : increment; +- } ++ default: ++ increment = wcwidth (wc); ++ increment = (increment < 0) ? 0 : increment; + } + } + +-- +2.7.4 + diff --git a/coreutils-i18n-sort-human.patch b/coreutils-i18n-sort-human.patch new file mode 100644 index 0000000..6752493 --- /dev/null +++ b/coreutils-i18n-sort-human.patch @@ -0,0 +1,35 @@ +From 3976ef5a20369d8b490907ab2cba2d617305a5e0 Mon Sep 17 00:00:00 2001 +From: Kamil Dudka +Date: Mon, 30 May 2016 16:19:20 +0200 +Subject: [PATCH] sort: do not use static array 'blanks' in human_numcompare() + +... because the array is not initialized with MB locales. Note this is +rather a conservative fix. I plan to do more cleanup of the i18n patch +in Fedora to prevent mistakes like this in future updates of coreutils. +--- + src/sort.c | 8 +++----- + 1 file changed, 3 insertions(+), 5 deletions(-) + +diff --git a/src/sort.c b/src/sort.c +index 9e07ad8..e47b039 100644 +--- a/src/sort.c ++++ b/src/sort.c +@@ -2304,12 +2304,10 @@ find_unit_order (char const *number) + < K/k < M < G < T < P < E < Z < Y */ + + static int +-human_numcompare (char const *a, char const *b) ++human_numcompare (char *a, char *b) + { +- while (blanks[to_uchar (*a)]) +- a++; +- while (blanks[to_uchar (*b)]) +- b++; ++ skipblanks(&a, a + strlen(a)); ++ skipblanks(&b, b + strlen(b)); + + int diff = find_unit_order (a) - find_unit_order (b); + return (diff ? diff : strnumcmp (a, b, decimal_point, thousands_sep)); +-- +2.5.5 + diff --git a/coreutils-i18n-un-expand-BOM.patch b/coreutils-i18n-un-expand-BOM.patch new file mode 100644 index 0000000..6210ce7 --- /dev/null +++ b/coreutils-i18n-un-expand-BOM.patch @@ -0,0 +1,456 @@ +From 7a7c776a4e228d180e74614fd8c8afcad5d4bdf7 Mon Sep 17 00:00:00 2001 +From: Jakub Martisko +Date: Thu, 7 Jul 2016 12:53:26 +0200 +Subject: [PATCH] coreutils-i18n-un-expand-BOM.patch + +--- + src/expand-common.c | 114 +++++++++++++++++++++++++++++++++++++++++++++++++++ + src/expand-common.h | 12 ++++++ + src/expand.c | 45 +++++++++++++++++++- + src/unexpand.c | 43 ++++++++++++++++++- + tests/expand/mb.sh | 71 ++++++++++++++++++++++++++++++++ + tests/unexpand/mb.sh | 59 ++++++++++++++++++++++++++ + 6 files changed, 342 insertions(+), 2 deletions(-) + +diff --git a/src/expand-common.c b/src/expand-common.c +index 4657e46..97cbb09 100644 +--- a/src/expand-common.c ++++ b/src/expand-common.c +@@ -19,6 +19,7 @@ + #include + #include + #include ++#include + #include "system.h" + #include "die.h" + #include "error.h" +@@ -126,6 +127,119 @@ set_increment_size (uintmax_t tabval) + return ok; + } + ++extern int ++set_utf_locale (void) ++{ ++ /*try using some predefined locale */ ++ const char* predef_locales[] = {"C.UTF8","en_US.UTF8","en_GB.UTF8"}; ++ ++ const int predef_locales_count=3; ++ for (int i=0;ibufcount=0; ++ if (c == 0xEF) ++ { ++ c=fgetc(fp); ++ } ++ else ++ { ++ if (c != EOF) ++ { ++ ungetc(c,fp); ++ } ++ return false; ++ } ++ ++ if (c == 0xBB) ++ { ++ c=fgetc(fp); ++ } ++ else ++ { ++ if ( c!= EOF ) ++ { ++ mbf->buf[0]=(unsigned char) 0xEF; ++ mbf->bufcount=1; ++ ungetc(c,fp); ++ return false; ++ } ++ else ++ { ++ ungetc(0xEF,fp); ++ return false; ++ } ++ } ++ if (c == 0xBF) ++ { ++ mbf->bufcount=0; ++ return true; ++ } ++ else ++ { ++ if (c != EOF) ++ { ++ mbf->buf[0]=(unsigned char) 0xEF; ++ mbf->buf[1]=(unsigned char) 0xBB; ++ mbf->bufcount=2; ++ ungetc(c,fp); ++ return false; ++ } ++ else ++ { ++ mbf->buf[0]=(unsigned char) 0xEF; ++ mbf->bufcount=1; ++ ungetc(0xBB,fp); ++ return false; ++ } ++ } ++ return false; ++} ++ ++extern void ++print_bom(void) ++{ ++ putc (0xEF, stdout); ++ putc (0xBB, stdout); ++ putc (0xBF, stdout); ++} ++ + /* Add the comma or blank separated list of tab stops STOPS + to the list of tab stops. */ + extern void +diff --git a/src/expand-common.h b/src/expand-common.h +index 8cb2079..763bfda 100644 +--- a/src/expand-common.h ++++ b/src/expand-common.h +@@ -34,6 +34,18 @@ extern size_t max_column_width; + /* The desired exit status. */ + extern int exit_status; + ++extern int ++set_utf_locale (void); ++ ++extern bool ++check_utf_locale(void); ++ ++extern bool ++check_bom(FILE* fp, mb_file_t *mbf); ++ ++extern void ++print_bom(void); ++ + /* Add tab stop TABVAL to the end of 'tab_list'. */ + extern void + add_tab_stop (uintmax_t tabval); +diff --git a/src/expand.c b/src/expand.c +index 310b349..4136824 100644 +--- a/src/expand.c ++++ b/src/expand.c +@@ -103,11 +103,33 @@ expand (void) + FILE *fp = next_file (NULL); + mb_file_t mbf; + mbf_char_t c; ++ /* True if the starting locale is utf8. */ ++ bool using_utf_locale; ++ ++ /* True if the first file contains BOM header. */ ++ bool found_bom; ++ using_utf_locale=check_utf_locale(); + + if (!fp) + return; +- + mbf_init (mbf, fp); ++ found_bom=check_bom(fp,&mbf); ++ ++ if (using_utf_locale == false && found_bom == true) ++ { ++ /*try using some predefined locale */ ++ ++ if (set_utf_locale () != 0) ++ { ++ error (EXIT_FAILURE, errno, _("cannot set UTF-8 locale")); ++ } ++ } ++ ++ ++ if (found_bom == true) ++ { ++ print_bom(); ++ } + + while (true) + { +@@ -132,6 +154,27 @@ expand (void) + if ((mb_iseof (c)) && (fp = next_file (fp))) + { + mbf_init (mbf, fp); ++ if (fp!=NULL) ++ { ++ if (check_bom(fp,&mbf)==true) ++ { ++ /*Not the first file - check BOM header*/ ++ if (using_utf_locale==false && found_bom==false) ++ { ++ /*BOM header in subsequent file but not in the first one. */ ++ error (EXIT_FAILURE, errno, _("combination of files with and without BOM header")); ++ } ++ } ++ else ++ { ++ if(using_utf_locale==false && found_bom==true) ++ { ++ /*First file conatined BOM header - locale was switched to UTF ++ *all subsequent files should contain BOM. */ ++ error (EXIT_FAILURE, errno, _("combination of files with and without BOM header")); ++ } ++ } ++ } + continue; + } + else +diff --git a/src/unexpand.c b/src/unexpand.c +index 863a90a..5681b58 100644 +--- a/src/unexpand.c ++++ b/src/unexpand.c +@@ -116,16 +116,36 @@ unexpand (void) + include characters other than spaces, so the blanks must be + stored, not merely counted. */ + mbf_char_t *pending_blank; ++ /* True if the starting locale is utf8. */ ++ bool using_utf_locale; ++ ++ /* True if the first file contains BOM header. */ ++ bool found_bom; ++ using_utf_locale=check_utf_locale(); + + if (!fp) + return; ++ mbf_init (mbf, fp); ++ found_bom=check_bom(fp,&mbf); ++ ++ if (using_utf_locale == false && found_bom == true) ++ { ++ /*try using some predefined locale */ + ++ if (set_utf_locale () != 0) ++ { ++ error (EXIT_FAILURE, errno, _("cannot set UTF-8 locale")); ++ } ++ } + /* The worst case is a non-blank character, then one blank, then a + tab stop, then MAX_COLUMN_WIDTH - 1 blanks, then a non-blank; so + allocate MAX_COLUMN_WIDTH bytes to store the blanks. */ + pending_blank = xmalloc (max_column_width * sizeof (mbf_char_t)); + +- mbf_init (mbf, fp); ++ if (found_bom == true) ++ { ++ print_bom(); ++ } + + while (true) + { +@@ -169,6 +189,27 @@ unexpand (void) + if ((mb_iseof (c)) && (fp = next_file (fp))) + { + mbf_init (mbf, fp); ++ if (fp!=NULL) ++ { ++ if (check_bom(fp,&mbf)==true) ++ { ++ /*Not the first file - check BOM header*/ ++ if (using_utf_locale==false && found_bom==false) ++ { ++ /*BOM header in subsequent file but not in the first one. */ ++ error (EXIT_FAILURE, errno, _("combination of files with and without BOM header")); ++ } ++ } ++ else ++ { ++ if(using_utf_locale==false && found_bom==true) ++ { ++ /*First file conatined BOM header - locale was switched to UTF ++ *all subsequent files should contain BOM. */ ++ error (EXIT_FAILURE, errno, _("combination of files with and without BOM header")); ++ } ++ } ++ } + continue; + } + else +diff --git a/tests/expand/mb.sh b/tests/expand/mb.sh +index 031be7a..1621c84 100755 +--- a/tests/expand/mb.sh ++++ b/tests/expand/mb.sh +@@ -109,4 +109,75 @@ env printf '12345678 + expand < in > out || fail=1 + compare exp out > /dev/null 2>&1 || fail=1 + ++ ++ ++#BOM header test 1 ++printf "\xEF\xBB\xBF" > in; cat <<\EOF >> in || framework_failure_ ++1234567812345678123456781 ++. . . . ++a b c d ++. . . . ++ä ö ü ß ++. . . . ++EOF ++env printf ' äöü\t. öüä. \tä xx\n' >> in || framework_failure_ ++ ++printf "\xEF\xBB\xBF" > exp; cat <<\EOF >> exp || framework_failure_ ++1234567812345678123456781 ++. . . . ++a b c d ++. . . . ++ä ö ü ß ++. . . . ++ äöü . öüä. ä xx ++EOF ++ ++ ++expand < in > out || fail=1 ++compare exp out > /dev/null 2>&1 || fail=1 ++ ++LANG=C expand < in > out || fail=1 ++compare exp out > /dev/null 2>&1 || fail=1 ++ ++LC_ALL=C expand < in > out || fail=1 ++compare exp out > /dev/null 2>&1 || fail=1 ++ ++ ++printf '\xEF\xBB\xBF' > in1; cat <<\EOF >> in1 || framework_failure_ ++1234567812345678123456781 ++. . . . ++a b c d ++. . . . ++ä ö ü ß ++. . . . ++EOF ++env printf ' äöü\t. öüä. \tä xx\n' >> in1 || framework_failure_ ++ ++ ++printf '\xEF\xBB\xBF' > exp; cat <<\EOF >> exp || framework_failure_ ++1234567812345678123456781 ++. . . . ++a b c d ++. . . . ++ä ö ü ß ++. . . . ++ äöü . öüä. ä xx ++1234567812345678123456781 ++. . . . ++a b c d ++. . . . ++ä ö ü ß ++. . . . ++ äöü . öüä. ä xx ++EOF ++ ++expand in1 in1 > out || fail=1 ++compare exp out > /dev/null 2>&1 || fail=1 ++ ++LANG=C expand in1 in1 > out || fail=1 ++compare exp out > /dev/null 2>&1 || fail=1 ++ ++LC_ALL=C expand in1 in1 > out || fail=1 ++compare exp out > /dev/null 2>&1 || fail=1 ++ + exit $fail +diff --git a/tests/unexpand/mb.sh b/tests/unexpand/mb.sh +index 8d75652..9d4ee3e 100755 +--- a/tests/unexpand/mb.sh ++++ b/tests/unexpand/mb.sh +@@ -111,3 +111,62 @@ env printf '12345678 + + unexpand -a < in > out || fail=1 + compare exp out > /dev/null 2>&1 || fail=1 ++ ++#BOM header test 1 ++printf "\xEF\xBB\xBF" > in; cat <<\EOF >> in || framework_failure_ ++1234567812345678123456781 ++. . . . ++a b c d ++. . . . ++ä ö ü ß ++. . . . ++ äöü . öüä. ä xx ++EOF ++env printf ' äöü\t. öüä. \tä xx\n' >> in || framework_failure_ ++ ++printf "\xEF\xBB\xBF" > exp; cat <<\EOF >> exp || framework_failure_ ++1234567812345678123456781 ++. . . . ++a b c d ++. . . . ++ä ö ü ß ++. . . . ++ äöü . öüä. ä xx ++EOF ++ ++unexpand < in > out || fail=1 ++compare exp out > /dev/null 2>&1 || fail=1 ++ ++LANG=C unexpand < in > out || fail=1 ++compare exp out > /dev/null 2>&1 || fail=1 ++ ++LC_ALL=C unexpand < in > out || fail=1 ++compare exp out > /dev/null 2>&1 || fail=1 ++ ++ ++printf "\xEF\xBB\xBF" > exp; cat <<\EOF >> exp || framework_failure_ ++1234567812345678123456781 ++. . . . ++a b c d ++. . . . ++ä ö ü ß ++. . . . ++ äöü . öüä. ä xx ++1234567812345678123456781 ++. . . . ++a b c d ++. . . . ++ä ö ü ß ++. . . . ++ äöü . öüä. ä xx ++EOF ++ ++ ++unexpand in in > out || fail=1 ++compare exp out > /dev/null 2>&1 || fail=1 ++ ++LANG=C unexpand in in > out || fail=1 ++compare exp out > /dev/null 2>&1 || fail=1 ++ ++LC_ALL=C unexpand in in > out || fail=1 ++compare exp out > /dev/null 2>&1 || fail=1 +-- +2.9.3 + diff --git a/coreutils-i18n.patch b/coreutils-i18n.patch index 83579e9..ad2fa2e 100644 --- a/coreutils-i18n.patch +++ b/coreutils-i18n.patch @@ -1,80 +1,38 @@ -From a81b096084524e9aeef5e8b81fc829eb9efec581 Mon Sep 17 00:00:00 2001 -From: rpm-build -Date: Wed, 30 Aug 2023 17:19:58 +0200 +From 29117b2d07af00f4d4b87cf778e4294588ab1a83 Mon Sep 17 00:00:00 2001 +From: Kamil Dudka +Date: Thu, 1 Dec 2016 15:10:04 +0100 Subject: [PATCH] coreutils-i18n.patch +TODO: merge upstream --- - bootstrap.conf | 2 + - configure.ac | 6 + lib/linebuffer.h | 8 + - lib/mbfile.c | 20 + - lib/mbfile.h | 283 +++++++++++++ - m4/mbfile.m4 | 16 + - src/cut.c | 508 +++++++++++++++++++++-- - src/expand-common.c | 114 ++++++ - src/expand-common.h | 12 + - src/expand.c | 90 +++- - src/local.mk | 4 +- - src/pr.c | 443 ++++++++++++++++++-- - src/sort.c | 791 +++++++++++++++++++++++++++++++++--- - src/unexpand.c | 101 ++++- - tests/Coreutils.pm | 3 + - tests/expand/mb.sh | 183 +++++++++ + src/fold.c | 308 +++++++++++++-- + src/join.c | 359 ++++++++++++++--- + src/pr.c | 443 +++++++++++++++++++-- + src/sort.c | 764 ++++++++++++++++++++++++++++++++++-- + src/uniq.c | 119 +++++- tests/i18n/sort.sh | 29 ++ - tests/local.mk | 4 + + tests/local.mk | 2 + tests/misc/expand.pl | 42 ++ - tests/misc/sort-mb-tests.sh | 45 ++ + tests/misc/fold.pl | 50 ++- + tests/misc/join.pl | 50 +++ + tests/misc/sort-mb-tests.sh | 45 +++ + tests/misc/sort-merge.pl | 42 ++ + tests/misc/sort.pl | 40 +- tests/misc/unexpand.pl | 39 ++ + tests/misc/uniq.pl | 55 +++ tests/pr/pr-tests.pl | 49 +++ - tests/sort/sort-merge.pl | 42 ++ - tests/sort/sort.pl | 40 +- - tests/unexpand/mb.sh | 172 ++++++++ - 25 files changed, 2879 insertions(+), 167 deletions(-) - create mode 100644 lib/mbfile.c - create mode 100644 lib/mbfile.h - create mode 100644 m4/mbfile.m4 - create mode 100644 tests/expand/mb.sh - create mode 100644 tests/i18n/sort.sh - create mode 100644 tests/misc/sort-mb-tests.sh - create mode 100644 tests/unexpand/mb.sh + 17 files changed, 2290 insertions(+), 154 deletions(-) + create mode 100755 tests/i18n/sort.sh + create mode 100755 tests/misc/sort-mb-tests.sh -diff --git a/bootstrap.conf b/bootstrap.conf -index ec68ac8..ec2fbbe 100644 ---- a/bootstrap.conf -+++ b/bootstrap.conf -@@ -171,6 +171,8 @@ gnulib_modules=" - malloc-gnu - manywarnings - mbbuf -+ mbchar -+ mbfile - mbrlen - mbrtoc32 - mbrtowc -diff --git a/configure.ac b/configure.ac -index 5e99ef3..ac07577 100644 ---- a/configure.ac -+++ b/configure.ac -@@ -465,6 +465,12 @@ fi - # I'm leaving it here for now. This whole thing needs to be modernized... - gl_WINSIZE_IN_PTEM - -+gl_MBFILE -+dnl Do not use gl_MODULE_INDICATOR([mbfile]) here: we don't want 'struct mbchar' -+dnl to have a different size in lib/ than in tests/. -+AC_DEFINE([GNULIB_MBFILE], [1], -+ [Define to 1 if the gnulib module 'mbfile' is in use.]) -+ - gl_HEADER_TIOCGWINSZ_IN_TERMIOS_H - - if test $gl_cv_sys_tiocgwinsz_needs_termios_h = no && \ diff --git a/lib/linebuffer.h b/lib/linebuffer.h -index ca56f80..509b7e6 100644 +index 64181af..9b8fe5a 100644 --- a/lib/linebuffer.h +++ b/lib/linebuffer.h -@@ -27,6 +27,11 @@ extern "C" { - #endif +@@ -21,6 +21,11 @@ + # include +/* Get mbstate_t. */ +# if HAVE_WCHAR_H @@ -84,9 +42,9 @@ index ca56f80..509b7e6 100644 /* A 'struct linebuffer' holds a line of text. */ struct linebuffer -@@ -34,6 +39,9 @@ struct linebuffer - idx_t size; /* Allocated. */ - idx_t length; /* Used. */ +@@ -28,6 +33,9 @@ struct linebuffer + size_t size; /* Allocated. */ + size_t length; /* Used. */ char *buffer; +# if HAVE_WCHAR_H + mbstate_t state; @@ -94,365 +52,32 @@ index ca56f80..509b7e6 100644 }; /* Initialize linebuffer LINEBUFFER for use. */ -diff --git a/lib/mbfile.c b/lib/mbfile.c -new file mode 100644 -index 0000000..f4e3e77 ---- /dev/null -+++ b/lib/mbfile.c -@@ -0,0 +1,20 @@ -+/* Multibyte character I/O: macros for multi-byte encodings. -+ Copyright (C) 2012-2025 Free Software Foundation, Inc. -+ -+ This file is free software: you can redistribute it and/or modify -+ it under the terms of the GNU Lesser General Public License as -+ published by the Free Software Foundation, either version 3 of the -+ License, or (at your option) any later version. -+ -+ This file is distributed in the hope that it will be useful, -+ but WITHOUT ANY WARRANTY; without even the implied warranty of -+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -+ GNU Lesser General Public License for more details. -+ -+ You should have received a copy of the GNU Lesser General Public License -+ along with this program. If not, see . */ -+ -+#include -+ -+#define MBFILE_INLINE _GL_EXTERN_INLINE -+#include "mbfile.h" -diff --git a/lib/mbfile.h b/lib/mbfile.h -new file mode 100644 -index 0000000..c852f31 ---- /dev/null -+++ b/lib/mbfile.h -@@ -0,0 +1,283 @@ -+/* Multibyte character I/O: macros for multi-byte encodings. -+ Copyright (C) 2001, 2005, 2009-2025 Free Software Foundation, Inc. -+ -+ This file is free software: you can redistribute it and/or modify -+ it under the terms of the GNU Lesser General Public License as -+ published by the Free Software Foundation, either version 3 of the -+ License, or (at your option) any later version. -+ -+ This file is distributed in the hope that it will be useful, -+ but WITHOUT ANY WARRANTY; without even the implied warranty of -+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -+ GNU Lesser General Public License for more details. -+ -+ You should have received a copy of the GNU Lesser General Public License -+ along with this program. If not, see . */ -+ -+/* Written by Mitsuru Chinen -+ and Bruno Haible . */ -+ -+/* The macros in this file implement multi-byte character input from a -+ stream. -+ -+ mb_file_t -+ is the type for multibyte character input stream, usable for variable -+ declarations. -+ -+ mbf_char_t -+ is the type for multibyte character or EOF, usable for variable -+ declarations. -+ -+ mbf_init (mbf, stream) -+ initializes the MB_FILE for reading from stream. -+ -+ mbf_getc (mbc, mbf) -+ reads the next multibyte character from mbf and stores it in mbc. -+ -+ mb_iseof (mbc) -+ returns true if mbc represents the EOF value. -+ -+ Here are the function prototypes of the macros. -+ -+ extern void mbf_init (mb_file_t mbf, FILE *stream); -+ extern void mbf_getc (mbf_char_t mbc, mb_file_t mbf); -+ extern bool mb_iseof (const mbf_char_t mbc); -+ */ -+ -+#ifndef _MBFILE_H -+#define _MBFILE_H 1 -+ -+/* This file uses _GL_INLINE_HEADER_BEGIN, _GL_INLINE. */ -+#if !_GL_CONFIG_H_INCLUDED -+ #error "Please include config.h first." -+#endif -+ -+#include -+#include -+#include -+#include -+#include -+ -+#include "mbchar.h" -+ -+_GL_INLINE_HEADER_BEGIN -+#ifndef MBFILE_INLINE -+# define MBFILE_INLINE _GL_INLINE -+#endif -+ -+#ifdef __cplusplus -+extern "C" { -+#endif -+ -+ -+/* Guarantee two characters of pushback. */ -+#define MBFILE_MAX_PUSHBACK 2 -+ -+struct mbfile_multi { -+ FILE *fp; -+ bool eof_seen; -+ unsigned int pushback_count; /* <= MBFILE_MAX_PUSHBACK */ -+ mbstate_t state; -+ unsigned int bufcount; -+ char buf[MBCHAR_BUF_SIZE]; -+ struct mbchar pushback[MBFILE_MAX_PUSHBACK]; -+}; -+ -+MBFILE_INLINE void -+mbfile_multi_getc (struct mbchar *mbc, struct mbfile_multi *mbf) -+{ -+ unsigned int new_bufcount; -+ size_t bytes; -+ -+ /* Return character pushed back, if there is one. */ -+ if (mbf->pushback_count > 0) -+ { -+ mb_copy (mbc, &mbf->pushback[mbf->pushback_count - 1]); -+ mbf->pushback_count--; -+ return; -+ } -+ -+ /* If EOF has already been seen, don't use getc. This matters if -+ mbf->fp is connected to an interactive tty. */ -+ if (mbf->eof_seen) -+ goto eof; -+ -+ new_bufcount = mbf->bufcount; -+ -+ /* If mbf->state is not in an initial state, some more 32-bit wide character -+ may be hiding in the state. We need to call mbrtoc32 again. */ -+ #if GNULIB_MBRTOC32_REGULAR -+ assert (mbsinit (&mbf->state)); -+ #else -+ if (mbsinit (&mbf->state)) -+ #endif -+ { -+ /* Before using mbrtoc32, we need at least one byte. */ -+ if (new_bufcount == 0) -+ { -+ int c = getc (mbf->fp); -+ if (c == EOF) -+ { -+ mbf->eof_seen = true; -+ goto eof; -+ } -+ mbf->buf[0] = (unsigned char) c; -+ new_bufcount++; -+ } -+ -+ /* Handle most ASCII characters quickly, without calling mbrtoc32(). */ -+ if (new_bufcount == 1 && is_basic (mbf->buf[0])) -+ { -+ /* These characters are part of the POSIX portable character set. -+ For most of them, namely those in the ISO C basic character set, -+ ISO C 99 guarantees that their wide character code is identical to -+ their char code. For the few other ones, this is the case as well, -+ in all locale encodings that are in use. The 32-bit wide character -+ code is the same as well. */ -+ mbc->wc = mbc->buf[0] = mbf->buf[0]; -+ mbc->wc_valid = true; -+ mbc->ptr = &mbc->buf[0]; -+ mbc->bytes = 1; -+ mbf->bufcount = 0; -+ return; -+ } -+ } -+ -+ /* Use mbrtoc32 on an increasing number of bytes. Read only as many bytes -+ from mbf->fp as needed. This is needed to give reasonable interactive -+ behaviour when mbf->fp is connected to an interactive tty. */ -+ for (;;) -+ { -+ /* Feed the bytes one by one into mbrtoc32. */ -+ bytes = mbrtoc32 (&mbc->wc, &mbf->buf[mbf->bufcount], new_bufcount - mbf->bufcount, &mbf->state); -+ -+ if (bytes == (size_t) -1) -+ { -+ /* An invalid multibyte sequence was encountered. */ -+ mbf->bufcount = new_bufcount; -+ /* Return a single byte. */ -+ bytes = 1; -+ mbc->wc_valid = false; -+ /* Allow the next invocation to continue from a sane state. */ -+ mbszero (&mbf->state); -+ break; -+ } -+ else if (bytes == (size_t) -2) -+ { -+ /* An incomplete multibyte character. */ -+ mbf->bufcount = new_bufcount; -+ if (mbf->bufcount == MBCHAR_BUF_SIZE) -+ { -+ /* An overlong incomplete multibyte sequence was encountered. */ -+ /* Return a single byte. */ -+ bytes = 1; -+ mbc->wc_valid = false; -+ break; -+ } -+ else -+ { -+ /* Read one more byte and retry mbrtoc32. */ -+ int c = getc (mbf->fp); -+ if (c == EOF) -+ { -+ /* An incomplete multibyte character at the end. */ -+ mbf->eof_seen = true; -+ bytes = new_bufcount; -+ mbc->wc_valid = false; -+ break; -+ } -+ mbf->buf[new_bufcount] = (unsigned char) c; -+ new_bufcount++; -+ } -+ } -+ else -+ { -+ #if !GNULIB_MBRTOC32_REGULAR -+ if (bytes == (size_t) -3) -+ { -+ /* The previous multibyte sequence produced an additional 32-bit -+ wide character. */ -+ mbf->bufcount = new_bufcount; -+ bytes = 0; -+ } -+ else -+ #endif -+ { -+ bytes = mbf->bufcount + bytes; -+ mbf->bufcount = new_bufcount; -+ if (bytes == 0) -+ { -+ /* A null 32-bit wide character was encountered. */ -+ bytes = 1; -+ assert (mbf->buf[0] == '\0'); -+ assert (mbc->wc == 0); -+ } -+ } -+ mbc->wc_valid = true; -+ break; -+ } -+ } -+ -+ /* Return the multibyte sequence mbf->buf[0..bytes-1]. */ -+ mbc->ptr = &mbc->buf[0]; -+ memcpy (&mbc->buf[0], &mbf->buf[0], bytes); -+ mbc->bytes = bytes; -+ -+ mbf->bufcount -= bytes; -+ if (mbf->bufcount > 0) -+ { -+ /* It's not worth calling memmove() for so few bytes. */ -+ unsigned int count = mbf->bufcount; -+ char *p = &mbf->buf[0]; -+ -+ do -+ { -+ *p = *(p + bytes); -+ p++; -+ } -+ while (--count > 0); -+ } -+ return; -+ -+eof: -+ /* An mbchar_t with bytes == 0 is used to indicate EOF. */ -+ mbc->ptr = NULL; -+ mbc->bytes = 0; -+ mbc->wc_valid = false; -+ return; -+} -+ -+MBFILE_INLINE void -+mbfile_multi_ungetc (const struct mbchar *mbc, struct mbfile_multi *mbf) -+{ -+ if (mbf->pushback_count == MBFILE_MAX_PUSHBACK) -+ abort (); -+ mb_copy (&mbf->pushback[mbf->pushback_count], mbc); -+ mbf->pushback_count++; -+} -+ -+typedef struct mbfile_multi mb_file_t; -+ -+typedef mbchar_t mbf_char_t; -+ -+#define mbf_init(mbf, stream) \ -+ ((mbf).fp = (stream), \ -+ (mbf).eof_seen = false, \ -+ (mbf).pushback_count = 0, \ -+ mbszero (&(mbf).state), \ -+ (mbf).bufcount = 0) -+ -+#define mbf_getc(mbc, mbf) mbfile_multi_getc (&(mbc), &(mbf)) -+ -+#define mbf_ungetc(mbc, mbf) mbfile_multi_ungetc (&(mbc), &(mbf)) -+ -+#define mb_iseof(mbc) ((mbc).bytes == 0) -+ -+ -+#ifdef __cplusplus -+} -+#endif -+ -+_GL_INLINE_HEADER_END -+ -+#endif /* _MBFILE_H */ -diff --git a/m4/mbfile.m4 b/m4/mbfile.m4 -new file mode 100644 -index 0000000..1d126e0 ---- /dev/null -+++ b/m4/mbfile.m4 -@@ -0,0 +1,16 @@ -+# mbfile.m4 -+# serial 7 -+dnl Copyright (C) 2005, 2008-2025 Free Software Foundation, Inc. -+dnl This file is free software; the Free Software Foundation -+dnl gives unlimited permission to copy and/or distribute it, -+dnl with or without modifications, as long as this notice is preserved. -+dnl This file is offered as-is, without any warranty. -+ -+dnl autoconf tests required for use of mbfile.h -+dnl From Bruno Haible. -+ -+AC_DEFUN([gl_MBFILE], -+[ -+ AC_REQUIRE([AC_TYPE_MBSTATE_T]) -+ : -+]) -diff --git a/src/cut.c b/src/cut.c -index f0effb9..36479d6 100644 ---- a/src/cut.c -+++ b/src/cut.c -@@ -27,6 +27,11 @@ - #include +diff --git a/src/fold.c b/src/fold.c +index 8cd0d6b..d23edd5 100644 +--- a/src/fold.c ++++ b/src/fold.c +@@ -22,12 +22,34 @@ #include #include -+ -+/* Get mbstate_t, mbrtowc(). */ + ++/* Get mbstate_t, mbrtowc(), wcwidth(). */ +#if HAVE_WCHAR_H +# include +#endif ++ ++/* Get iswprint(), iswblank(), wcwidth(). */ ++#if HAVE_WCTYPE_H ++# include ++#endif ++ #include "system.h" - - #include "assure.h" -@@ -35,6 +40,18 @@ - - #include "set-fields.h" + #include "die.h" + #include "error.h" + #include "fadvise.h" + #include "xdectoint.h" +/* MB_LEN_MAX is incorrectly defined to be 1 in at least one GCC -+ installation; work around this configuration error. */ ++ installation; work around this configuration error. */ +#if !defined MB_LEN_MAX || MB_LEN_MAX < 2 +# undef MB_LEN_MAX +# define MB_LEN_MAX 16 @@ -463,965 +88,870 @@ index f0effb9..36479d6 100644 +# define mbrtowc(pwc, s, n, ps) (mbrtowc) (pwc, s, n, 0) +#endif + + #define TAB_WIDTH 8 + /* The official name of this program (e.g., no 'g' prefix). */ - #define PROGRAM_NAME "cut" +@@ -35,20 +57,41 @@ -@@ -51,6 +68,52 @@ - } \ - while (0) + #define AUTHORS proper_name ("David MacKenzie") -+/* Refill the buffer BUF to get a multibyte character. */ -+#define REFILL_BUFFER(BUF, BUFPOS, BUFLEN, STREAM) \ -+ do \ -+ { \ -+ if (BUFLEN < MB_LEN_MAX && !feof (STREAM) && !ferror (STREAM)) \ -+ { \ -+ memmove (BUF, BUFPOS, BUFLEN); \ -+ BUFLEN += fread (BUF + BUFLEN, sizeof(char), BUFSIZ, STREAM); \ -+ BUFPOS = BUF; \ -+ } \ -+ } \ ++#define FATAL_ERROR(Message) \ ++ do \ ++ { \ ++ error (0, 0, (Message)); \ ++ usage (2); \ ++ } \ + while (0) + -+/* Get wide character on BUFPOS. BUFPOS is not included after that. -+ If byte sequence is not valid as a character, CONVFAIL is true. Otherwise false. */ -+#define GET_NEXT_WC_FROM_BUFFER(WC, BUFPOS, BUFLEN, MBLENGTH, STATE, CONVFAIL) \ -+ do \ -+ { \ -+ mbstate_t state_bak; \ -+ \ -+ if (BUFLEN < 1) \ -+ { \ -+ WC = WEOF; \ -+ break; \ -+ } \ -+ \ -+ /* Get a wide character. */ \ -+ CONVFAIL = false; \ -+ state_bak = STATE; \ -+ MBLENGTH = mbrtowc ((wchar_t *)&WC, BUFPOS, BUFLEN, &STATE); \ -+ \ -+ switch (MBLENGTH) \ -+ { \ -+ case (size_t)-1: \ -+ case (size_t)-2: \ -+ CONVFAIL = true; \ -+ STATE = state_bak; \ -+ /* Fall througn. */ \ -+ \ -+ case 0: \ -+ MBLENGTH = 1; \ -+ break; \ -+ } \ -+ } \ -+ while (0) -+ - - /* Pointer inside RP. When checking if a byte or field is selected - by a finite range, we check if it is between CURRENT_RP.LO -@@ -58,6 +121,9 @@ - CURRENT_RP.HI then we make CURRENT_RP to point to the next range pair. */ - static struct field_range_pair *current_rp; - -+/* Length of the delimiter given as argument to -d. */ -+size_t delimlen; -+ - /* This buffer is used to support the semantics of the -s option - (or lack of same) when the specified field list includes (does - not include) the first field. In both of those cases, the entire -@@ -70,6 +136,29 @@ static char *field_1_buffer; - /* The number of bytes allocated for FIELD_1_BUFFER. */ - static size_t field_1_bufsize; - +enum operating_mode -+ { -+ undefined_mode, ++{ ++ /* Fold texts by columns that are at the given positions. */ ++ column_mode, + -+ /* Output bytes that are at the given positions. */ -+ byte_mode, ++ /* Fold texts by bytes that are at the given positions. */ ++ byte_mode, + -+ /* Output characters that are at the given positions. */ -+ character_mode, -+ -+ /* Output the given delimiter-separated fields. */ -+ field_mode -+ }; ++ /* Fold texts by characters that are at the given positions. */ ++ character_mode, ++}; + ++/* The argument shows current mode. (Default: column_mode) */ +static enum operating_mode operating_mode; + -+/* If nonzero, when in byte mode, don't split multibyte characters. */ -+static int byte_mode_character_aware; -+ -+/* If nonzero, the function for single byte locale is work -+ if this program runs on multibyte locale. */ -+static int force_singlebyte_mode; -+ - /* If true, do not output lines containing no delimiter characters. - Otherwise, all such lines are printed. This option is valid only - with field mode. */ -@@ -81,10 +170,16 @@ static bool complement; + /* If nonzero, try to break on whitespace. */ + static bool break_spaces; - /* The delimiter character for field mode. */ - static unsigned char delim; -+#if HAVE_WCHAR_H -+static wchar_t wcdelim; -+#endif - - /* The delimiter for each line/record. */ - static unsigned char line_delim = '\n'; - -+/* True if the --output-delimiter=STRING option was specified. */ -+static bool output_delimiter_specified; -+ - /* The length of output_delimiter_string. */ - static size_t output_delimiter_length; - -@@ -92,9 +187,6 @@ static size_t output_delimiter_length; - string consisting of the input delimiter. */ - static char *output_delimiter_string; - --/* The output delimiter string contents, if the default. */ --static char output_delimiter_default[1]; +-/* If nonzero, count bytes, not column positions. */ +-static bool count_bytes; - - /* True if we have ever read standard input. */ + /* If nonzero, at least one of the files we read was standard input. */ static bool have_read_stdin; -@@ -148,7 +240,7 @@ Print selected parts of lines from each FILE to standard output.\n\ - -f, --fields=LIST select only these fields; also print any line\n\ - that contains no delimiter character, unless\n\ - the -s option is specified\n\ -- -n (ignored)\n\ -+ -n with -b: don't split multibyte characters\n\ - "), stdout); - fputs (_("\ - --complement complement the set of selected bytes, characters\n\ -@@ -252,7 +344,7 @@ cut_bytes (FILE *stream) - next_item (&byte_idx); - if (print_kth (byte_idx)) - { -- if (output_delimiter_string != output_delimiter_default) -+ if (output_delimiter_specified) - { - if (print_delimiter && is_range_start_index (byte_idx)) - { -@@ -271,6 +363,82 @@ cut_bytes (FILE *stream) - } - } +-static char const shortopts[] = "bsw:0::1::2::3::4::5::6::7::8::9::"; ++static char const shortopts[] = "bcsw:0::1::2::3::4::5::6::7::8::9::"; -+#if HAVE_MBRTOWC -+/* This function is in use for the following case. -+ -+ 1. Read from the stream STREAM, printing to standard output any selected -+ characters. -+ -+ 2. Read from stream STREAM, printing to standard output any selected bytes, -+ without splitting multibyte characters. */ -+ + static struct option const longopts[] = + { + {"bytes", no_argument, NULL, 'b'}, ++ {"characters", no_argument, NULL, 'c'}, + {"spaces", no_argument, NULL, 's'}, + {"width", required_argument, NULL, 'w'}, + {GETOPT_HELP_OPTION_DECL}, +@@ -76,6 +119,7 @@ Wrap input lines in each FILE, writing to standard output.\n\ + + fputs (_("\ + -b, --bytes count bytes rather than columns\n\ ++ -c, --characters count characters rather than columns\n\ + -s, --spaces break at spaces\n\ + -w, --width=WIDTH use WIDTH columns instead of 80\n\ + "), stdout); +@@ -93,7 +137,7 @@ Wrap input lines in each FILE, writing to standard output.\n\ + static size_t + adjust_column (size_t column, char c) + { +- if (!count_bytes) ++ if (operating_mode != byte_mode) + { + if (c == '\b') + { +@@ -116,30 +160,14 @@ adjust_column (size_t column, char c) + to stdout, with maximum line length WIDTH. + Return true if successful. */ + +-static bool +-fold_file (char const *filename, size_t width) +static void -+cut_characters_or_cut_bytes_no_split (FILE *stream) ++fold_text (FILE *istream, size_t width, int *saved_errno) + { +- FILE *istream; + int c; + size_t column = 0; /* Screen column where next char will go. */ + size_t offset_out = 0; /* Index in 'line_out' for next char. */ + static char *line_out = NULL; + static size_t allocated_out = 0; +- int saved_errno; +- +- if (STREQ (filename, "-")) +- { +- istream = stdin; +- have_read_stdin = true; +- } +- else +- istream = fopen (filename, "r"); +- +- if (istream == NULL) +- { +- error (0, errno, "%s", quotef (filename)); +- return false; +- } + + fadvise (istream, FADVISE_SEQUENTIAL); + +@@ -169,6 +197,15 @@ fold_file (char const *filename, size_t width) + bool found_blank = false; + size_t logical_end = offset_out; + ++ /* If LINE_OUT has no wide character, ++ put a new wide character in LINE_OUT ++ if column is bigger than width. */ ++ if (offset_out == 0) ++ { ++ line_out[offset_out++] = c; ++ continue; ++ } ++ + /* Look for the last blank. */ + while (logical_end) + { +@@ -215,11 +252,221 @@ fold_file (char const *filename, size_t width) + line_out[offset_out++] = c; + } + +- saved_errno = errno; ++ *saved_errno = errno; + + if (offset_out) + fwrite (line_out, sizeof (char), (size_t) offset_out, stdout); + ++} ++ ++#if HAVE_MBRTOWC ++static void ++fold_multibyte_text (FILE *istream, size_t width, int *saved_errno) +{ -+ uintmax_t idx; /* number of bytes or characters in the line so far. */ + char buf[MB_LEN_MAX + BUFSIZ]; /* For spooling a read byte sequence. */ -+ char *bufpos; /* Next read position of BUF. */ -+ size_t buflen; /* The length of the byte sequence in buf. */ ++ size_t buflen = 0; /* The length of the byte sequence in buf. */ ++ char *bufpos = buf; /* Next read position of BUF. */ + wint_t wc; /* A gotten wide character. */ + size_t mblength; /* The byte size of a multibyte character which shows + as same character as WC. */ -+ mbstate_t state; /* State of the stream. */ -+ bool convfail = false; /* true, when conversion failed. Otherwise false. */ -+ /* Whether to begin printing delimiters between ranges for the current line. -+ Set after we've begun printing data corresponding to the first range. */ -+ bool print_delimiter = false; ++ mbstate_t state, state_bak; /* State of the stream. */ ++ int convfail = 0; /* 1, when conversion is failed. Otherwise 0. */ + -+ idx = 0; -+ buflen = 0; -+ bufpos = buf; ++ static char *line_out = NULL; ++ size_t offset_out = 0; /* Index in `line_out' for next char. */ ++ static size_t allocated_out = 0; ++ ++ int increment; ++ size_t column = 0; ++ ++ size_t last_blank_pos; ++ size_t last_blank_column; ++ int is_blank_seen; ++ int last_blank_increment = 0; ++ int is_bs_following_last_blank; ++ size_t bs_following_last_blank_num; ++ int is_cr_after_last_blank; ++ ++#define CLEAR_FLAGS \ ++ do \ ++ { \ ++ last_blank_pos = 0; \ ++ last_blank_column = 0; \ ++ is_blank_seen = 0; \ ++ is_bs_following_last_blank = 0; \ ++ bs_following_last_blank_num = 0; \ ++ is_cr_after_last_blank = 0; \ ++ } \ ++ while (0) ++ ++#define START_NEW_LINE \ ++ do \ ++ { \ ++ putchar ('\n'); \ ++ column = 0; \ ++ offset_out = 0; \ ++ CLEAR_FLAGS; \ ++ } \ ++ while (0) ++ ++ CLEAR_FLAGS; + memset (&state, '\0', sizeof(mbstate_t)); + -+ current_rp = frp; -+ -+ while (1) ++ for (;; bufpos += mblength, buflen -= mblength) + { -+ REFILL_BUFFER (buf, bufpos, buflen, stream); -+ -+ GET_NEXT_WC_FROM_BUFFER (wc, bufpos, buflen, mblength, state, convfail); -+ (void) convfail; /* ignore unused */ -+ -+ if (wc == WEOF) ++ if (buflen < MB_LEN_MAX && !feof (istream) && !ferror (istream)) + { -+ if (idx > 0) -+ putchar (line_delim); ++ memmove (buf, bufpos, buflen); ++ buflen += fread (buf + buflen, sizeof(char), BUFSIZ, istream); ++ bufpos = buf; ++ } ++ ++ if (buflen < 1) ++ break; ++ ++ /* Get a wide character. */ ++ state_bak = state; ++ mblength = mbrtowc ((wchar_t *)&wc, bufpos, buflen, &state); ++ ++ switch (mblength) ++ { ++ case (size_t)-1: ++ case (size_t)-2: ++ convfail++; ++ state = state_bak; ++ /* Fall through. */ ++ ++ case 0: ++ mblength = 1; + break; + } -+ else if (wc == line_delim) ++ ++rescan: ++ if (operating_mode == byte_mode) /* byte mode */ ++ increment = mblength; ++ else if (operating_mode == character_mode) /* character mode */ ++ increment = 1; ++ else /* column mode */ + { -+ putchar (line_delim); -+ idx = 0; -+ print_delimiter = false; -+ current_rp = frp; -+ } -+ else -+ { -+ next_item (&idx); -+ if (print_kth (idx)) -+ { -+ if (output_delimiter_specified) -+ { -+ if (print_delimiter && is_range_start_index (idx)) -+ { -+ fwrite (output_delimiter_string, sizeof (char), -+ output_delimiter_length, stdout); -+ } -+ print_delimiter = true; -+ } -+ fwrite (bufpos, mblength, sizeof(char), stdout); -+ } -+ } -+ -+ buflen -= mblength; -+ bufpos += mblength; -+ } -+} -+#endif -+ - /* Read from stream STREAM, printing to standard output any selected fields. */ - - static void -@@ -433,11 +601,218 @@ cut_fields (FILE *stream) - } - } - --/* Process file FILE to standard output, using CUT_STREAM. -+#if HAVE_MBRTOWC -+static void -+cut_fields_mb (FILE *stream) -+{ -+ int c; -+ uintmax_t field_idx; -+ int found_any_selected_field; -+ int buffer_first_field; -+ int empty_input; -+ char buf[MB_LEN_MAX + BUFSIZ]; /* For spooling a read byte sequence. */ -+ char *bufpos; /* Next read position of BUF. */ -+ size_t buflen; /* The length of the byte sequence in buf. */ -+ wint_t wc = 0; /* A gotten wide character. */ -+ size_t mblength; /* The byte size of a multibyte character which shows -+ as same character as WC. */ -+ mbstate_t state; /* State of the stream. */ -+ bool convfail = false; /* true, when conversion failed. Otherwise false. */ -+ -+ current_rp = frp; -+ -+ found_any_selected_field = 0; -+ field_idx = 1; -+ bufpos = buf; -+ buflen = 0; -+ memset (&state, '\0', sizeof(mbstate_t)); -+ -+ c = getc (stream); -+ empty_input = (c == EOF); -+ if (c != EOF) -+ { -+ ungetc (c, stream); -+ wc = 0; -+ } -+ else -+ wc = WEOF; -+ -+ /* To support the semantics of the -s flag, we may have to buffer -+ all of the first field to determine whether it is `delimited.' -+ But that is unnecessary if all non-delimited lines must be printed -+ and the first field has been selected, or if non-delimited lines -+ must be suppressed and the first field has *not* been selected. -+ That is because a non-delimited line has exactly one field. */ -+ buffer_first_field = (suppress_non_delimited ^ !print_kth (1)); -+ -+ while (1) -+ { -+ if (field_idx == 1 && buffer_first_field) -+ { -+ int len = 0; -+ -+ while (1) -+ { -+ REFILL_BUFFER (buf, bufpos, buflen, stream); -+ -+ GET_NEXT_WC_FROM_BUFFER -+ (wc, bufpos, buflen, mblength, state, convfail); -+ -+ if (wc == WEOF) -+ break; -+ -+ field_1_buffer = xrealloc (field_1_buffer, len + mblength); -+ memcpy (field_1_buffer + len, bufpos, mblength); -+ len += mblength; -+ buflen -= mblength; -+ bufpos += mblength; -+ -+ if (!convfail && (wc == line_delim || wc == wcdelim)) -+ break; -+ } -+ -+ if (len <= 0 && wc == WEOF) -+ break; -+ -+ /* If the first field extends to the end of line (it is not -+ delimited) and we are printing all non-delimited lines, -+ print this one. */ -+ if (convfail || (!convfail && wc != wcdelim)) -+ { -+ if (suppress_non_delimited) -+ { -+ /* Empty. */ -+ } -+ else -+ { -+ fwrite (field_1_buffer, sizeof (char), len, stdout); -+ /* Make sure the output line is newline terminated. */ -+ if (convfail || (!convfail && wc != line_delim)) -+ putchar (line_delim); -+ } -+ continue; -+ } -+ -+ if (print_kth (1)) -+ { -+ /* Print the field, but not the trailing delimiter. */ -+ fwrite (field_1_buffer, sizeof (char), len - 1, stdout); -+ found_any_selected_field = 1; -+ } -+ next_item (&field_idx); -+ } -+ -+ if (wc != WEOF) -+ { -+ if (print_kth (field_idx)) -+ { -+ if (found_any_selected_field) -+ { -+ fwrite (output_delimiter_string, sizeof (char), -+ output_delimiter_length, stdout); -+ } -+ found_any_selected_field = 1; -+ } -+ -+ while (1) -+ { -+ REFILL_BUFFER (buf, bufpos, buflen, stream); -+ -+ GET_NEXT_WC_FROM_BUFFER -+ (wc, bufpos, buflen, mblength, state, convfail); -+ -+ if (wc == WEOF) -+ break; -+ else if (!convfail && (wc == wcdelim || wc == line_delim)) -+ { -+ buflen -= mblength; -+ bufpos += mblength; -+ break; -+ } -+ -+ if (print_kth (field_idx)) -+ fwrite (bufpos, mblength, sizeof(char), stdout); -+ -+ buflen -= mblength; -+ bufpos += mblength; -+ } -+ } -+ -+ if ((!convfail || wc == line_delim) && buflen < 1) -+ wc = WEOF; -+ -+ if (!convfail && wc == wcdelim) -+ next_item (&field_idx); -+ else if (wc == WEOF || (!convfail && wc == line_delim)) -+ { -+ if (found_any_selected_field -+ || (!empty_input && !(suppress_non_delimited && field_idx == 1))) -+ putchar (line_delim); -+ if (wc == WEOF) -+ break; -+ field_idx = 1; -+ current_rp = frp; -+ found_any_selected_field = 0; -+ } -+ } -+} -+#endif -+ -+static void -+cut_stream (FILE *stream) -+{ -+#if HAVE_MBRTOWC -+ if (MB_CUR_MAX > 1 && !force_singlebyte_mode) -+ { -+ switch (operating_mode) -+ { -+ case byte_mode: -+ if (byte_mode_character_aware) -+ cut_characters_or_cut_bytes_no_split (stream); ++ if (convfail) ++ increment = 1; + else -+ cut_bytes (stream); -+ break; -+ -+ case character_mode: -+ cut_characters_or_cut_bytes_no_split (stream); -+ break; -+ -+ case field_mode: -+ if (delimlen == 1) + { -+ /* Check if we have utf8 multibyte locale, so we can use this -+ optimization because of uniqueness of characters, which is -+ not true for e.g. SJIS */ -+ char * loc = setlocale(LC_CTYPE, NULL); -+ if (loc && (strstr (loc, "UTF-8") || strstr (loc, "utf-8") || -+ strstr (loc, "UTF8") || strstr (loc, "utf8"))) ++ switch (wc) + { -+ cut_fields (stream); ++ case L'\n': ++ fwrite (line_out, sizeof(char), offset_out, stdout); ++ START_NEW_LINE; ++ continue; ++ ++ case L'\b': ++ increment = (column > 0) ? -1 : 0; + break; ++ ++ case L'\r': ++ increment = -1 * column; ++ break; ++ ++ case L'\t': ++ increment = 8 - column % 8; ++ break; ++ ++ default: ++ increment = wcwidth (wc); ++ increment = (increment < 0) ? 0 : increment; + } + } -+ cut_fields_mb (stream); -+ break; ++ } + -+ default: -+ abort (); ++ if (column + increment > width && break_spaces && last_blank_pos) ++ { ++ fwrite (line_out, sizeof(char), last_blank_pos, stdout); ++ putchar ('\n'); ++ ++ offset_out = offset_out - last_blank_pos; ++ column = column - last_blank_column + ((is_cr_after_last_blank) ++ ? last_blank_increment : bs_following_last_blank_num); ++ memmove (line_out, line_out + last_blank_pos, offset_out); ++ CLEAR_FLAGS; ++ goto rescan; ++ } ++ ++ if (column + increment > width && column != 0) ++ { ++ fwrite (line_out, sizeof(char), offset_out, stdout); ++ START_NEW_LINE; ++ goto rescan; ++ } ++ ++ if (allocated_out < offset_out + mblength) ++ { ++ line_out = X2REALLOC (line_out, &allocated_out); ++ } ++ ++ memcpy (line_out + offset_out, bufpos, mblength); ++ offset_out += mblength; ++ column += increment; ++ ++ if (is_blank_seen && !convfail && wc == L'\r') ++ is_cr_after_last_blank = 1; ++ ++ if (is_bs_following_last_blank && !convfail && wc == L'\b') ++ ++bs_following_last_blank_num; ++ else ++ is_bs_following_last_blank = 0; ++ ++ if (break_spaces && !convfail && iswblank (wc)) ++ { ++ last_blank_pos = offset_out; ++ last_blank_column = column; ++ is_blank_seen = 1; ++ last_blank_increment = increment; ++ is_bs_following_last_blank = 1; ++ bs_following_last_blank_num = 0; ++ is_cr_after_last_blank = 0; + } + } ++ ++ *saved_errno = errno; ++ ++ if (offset_out) ++ fwrite (line_out, sizeof (char), (size_t) offset_out, stdout); ++ ++} ++#endif ++ ++/* Fold file FILENAME, or standard input if FILENAME is "-", ++ to stdout, with maximum line length WIDTH. ++ Return 0 if successful, 1 if an error occurs. */ ++ ++static bool ++fold_file (char const *filename, size_t width) ++{ ++ FILE *istream; ++ int saved_errno; ++ ++ if (STREQ (filename, "-")) ++ { ++ istream = stdin; ++ have_read_stdin = 1; ++ } ++ else ++ istream = fopen (filename, "r"); ++ ++ if (istream == NULL) ++ { ++ error (0, errno, "%s", filename); ++ return 1; ++ } ++ ++ /* Define how ISTREAM is being folded. */ ++#if HAVE_MBRTOWC ++ if (MB_CUR_MAX > 1) ++ fold_multibyte_text (istream, width, &saved_errno); + else +#endif -+ { -+ if (operating_mode == field_mode) -+ cut_fields (stream); -+ else -+ cut_bytes (stream); -+ } -+} ++ fold_text (istream, width, &saved_errno); + -+/* Process file FILE to standard output. - Return true if successful. */ - - static bool --cut_file (char const *file, void (*cut_stream) (FILE *)) -+cut_file (char const *file) - { - FILE *stream; - -@@ -482,8 +857,8 @@ main (int argc, char **argv) - int optc; - bool ok; - bool delim_specified = false; -- bool byte_mode = false; -- char *spec_list_string = nullptr; -+ char *spec_list_string IF_LINT ( = nullptr); -+ char mbdelim[MB_LEN_MAX + 1]; - - initialize_main (&argc, &argv); - set_program_name (argv[0]); -@@ -493,6 +868,8 @@ main (int argc, char **argv) + if (ferror (istream)) + { + error (0, saved_errno, "%s", quotef (filename)); +@@ -252,7 +499,8 @@ main (int argc, char **argv) atexit (close_stdout); -+ operating_mode = undefined_mode; -+ - /* By default, all non-delimited lines are printed. */ - suppress_non_delimited = false; +- break_spaces = count_bytes = have_read_stdin = false; ++ operating_mode = column_mode; ++ break_spaces = have_read_stdin = false; -@@ -505,35 +882,77 @@ main (int argc, char **argv) + while ((optc = getopt_long (argc, argv, shortopts, longopts, NULL)) != -1) + { +@@ -261,7 +509,15 @@ main (int argc, char **argv) switch (optc) { - case 'b': -- case 'c': - /* Build the byte list. */ -- byte_mode = true; -- FALLTHROUGH; -+ if (operating_mode != undefined_mode) -+ FATAL_ERROR (_("only one type of list may be specified")); + case 'b': /* Count bytes rather than columns. */ +- count_bytes = true; ++ if (operating_mode != column_mode) ++ FATAL_ERROR (_("only one way of folding may be specified")); + operating_mode = byte_mode; -+ spec_list_string = optarg; + break; + + case 'c': -+ /* Build the character list. */ -+ if (operating_mode != undefined_mode) -+ FATAL_ERROR (_("only one type of list may be specified")); ++ if (operating_mode != column_mode) ++ FATAL_ERROR (_("only one way of folding may be specified")); + operating_mode = character_mode; -+ spec_list_string = optarg; -+ break; -+ - case 'f': - /* Build the field list. */ -- if (spec_list_string) -- FATAL_ERROR (_("only one list may be specified")); -+ if (operating_mode != undefined_mode) -+ FATAL_ERROR (_("only one type of list may be specified")); -+ operating_mode = field_mode; - spec_list_string = optarg; break; - case 'd': - /* New delimiter. */ - /* Interpret -d '' to mean 'use the NUL byte as the delimiter.' */ -- if (optarg[0] != '\0' && optarg[1] != '\0') -- FATAL_ERROR (_("the delimiter must be a single character")); -- delim = optarg[0]; -- delim_specified = true; -+ { -+#if HAVE_MBRTOWC -+ if(MB_CUR_MAX > 1) -+ { -+ mbstate_t state; -+ -+ memset (&state, '\0', sizeof(mbstate_t)); -+ delimlen = mbrtowc (&wcdelim, optarg, strnlen(optarg, MB_LEN_MAX), &state); -+ -+ if (delimlen == (size_t)-1 || delimlen == (size_t)-2) -+ ++force_singlebyte_mode; -+ else -+ { -+ delimlen = (delimlen < 1) ? 1 : delimlen; -+ if (wcdelim != L'\0' && *(optarg + delimlen) != '\0') -+ FATAL_ERROR (_("the delimiter must be a single character")); -+ memcpy (mbdelim, optarg, delimlen); -+ mbdelim[delimlen] = '\0'; -+ if (delimlen == 1) -+ delim = *optarg; -+ } -+ } -+ -+ if (MB_CUR_MAX <= 1 || force_singlebyte_mode) -+#endif -+ { -+ if (optarg[0] != '\0' && optarg[1] != '\0') -+ FATAL_ERROR (_("the delimiter must be a single character")); -+ delim = (unsigned char) optarg[0]; -+ } -+ delim_specified = true; -+ } - break; - - case OUTPUT_DELIMITER_OPTION: -+ output_delimiter_specified = true; - /* Interpret --output-delimiter='' to mean - 'use the NUL byte as the delimiter.' */ - output_delimiter_length = (optarg[0] == '\0' - ? 1 : strlen (optarg)); -- output_delimiter_string = optarg; -+ output_delimiter_string = xstrdup (optarg); - break; - - case 'n': -+ byte_mode_character_aware = 1; - break; - - case 's': -@@ -555,40 +974,57 @@ main (int argc, char **argv) - } - } - -- if (!spec_list_string) -+ if (operating_mode == undefined_mode) - FATAL_ERROR (_("you must specify a list of bytes, characters, or fields")); - -- if (byte_mode) -- { -- if (delim_specified) -- FATAL_ERROR (_("an input delimiter may be specified only\ -+ if (delim_specified && operating_mode != field_mode) -+ FATAL_ERROR (_("an input delimiter may be specified only\ - when operating on fields")); - -- if (suppress_non_delimited) -- FATAL_ERROR (_("suppressing non-delimited lines makes sense\n\ -+ if (suppress_non_delimited && operating_mode != field_mode) -+ FATAL_ERROR (_("suppressing non-delimited lines makes sense\n\ - \tonly when operating on fields")); -- } - - set_fields (spec_list_string, -- ((byte_mode ? SETFLD_ERRMSG_USE_POS : 0) -- | (complement ? SETFLD_COMPLEMENT : 0))); -+ ( (operating_mode == field_mode) ? 0 : SETFLD_ERRMSG_USE_POS) -+ | (complement ? SETFLD_COMPLEMENT : 0) ); - - if (!delim_specified) -- delim = '\t'; -+ { -+ delim = '\t'; -+#ifdef HAVE_MBRTOWC -+ wcdelim = L'\t'; -+ mbdelim[0] = '\t'; -+ mbdelim[1] = '\0'; -+ delimlen = 1; -+#endif -+ } - - if (output_delimiter_string == nullptr) - { -- output_delimiter_default[0] = delim; -- output_delimiter_string = output_delimiter_default; -- output_delimiter_length = 1; -+#ifdef HAVE_MBRTOWC -+ if (MB_CUR_MAX > 1 && !force_singlebyte_mode) -+ { -+ output_delimiter_string = xstrdup(mbdelim); -+ output_delimiter_length = delimlen; -+ } -+ -+ if (MB_CUR_MAX <= 1 || force_singlebyte_mode) -+#endif -+ { -+ static char dummy[2]; -+ dummy[0] = delim; -+ dummy[1] = '\0'; -+ output_delimiter_string = dummy; -+ output_delimiter_length = 1; -+ } - } - -- void (*cut_stream) (FILE *) = byte_mode ? cut_bytes : cut_fields; - if (optind == argc) -- ok = cut_file ("-", cut_stream); -+ ok = cut_file ("-"); - else - for (ok = true; optind < argc; optind++) -- ok &= cut_file (argv[optind], cut_stream); -+ ok &= cut_file (argv[optind]); - - - if (have_read_stdin && fclose (stdin) == EOF) -diff --git a/src/expand-common.c b/src/expand-common.c -index 14dd804..0d8eaaa 100644 ---- a/src/expand-common.c -+++ b/src/expand-common.c -@@ -19,6 +19,7 @@ - #include - #include + case 's': /* Break at word boundaries. */ +diff --git a/src/join.c b/src/join.c +index 98b461c..9990f38 100644 +--- a/src/join.c ++++ b/src/join.c +@@ -22,19 +22,33 @@ #include -+#include + #include + ++/* Get mbstate_t, mbrtowc(), mbrtowc(), wcwidth(). */ ++#if HAVE_WCHAR_H ++# include ++#endif ++ ++/* Get iswblank(), towupper. */ ++#if HAVE_WCTYPE_H ++# include ++#endif ++ #include "system.h" - #include "c-ctype.h" + #include "die.h" + #include "error.h" #include "fadvise.h" -@@ -132,6 +133,119 @@ set_increment_size (colno tabval) - return ok; + #include "hard-locale.h" + #include "linebuffer.h" +-#include "memcasecmp.h" + #include "quote.h" + #include "stdio--.h" + #include "xmemcoll.h" + #include "xstrtol.h" + #include "argmatch.h" + ++/* Some systems, like BeOS, have multibyte encodings but lack mbstate_t. */ ++#if HAVE_MBRTOWC && defined mbstate_t ++# define mbrtowc(pwc, s, n, ps) (mbrtowc) (pwc, s, n, 0) ++#endif ++ + /* The official name of this program (e.g., no 'g' prefix). */ + #define PROGRAM_NAME "join" + +@@ -136,10 +150,12 @@ static struct outlist outlist_head; + /* Last element in 'outlist', where a new element can be added. */ + static struct outlist *outlist_end = &outlist_head; + +-/* Tab character separating fields. If negative, fields are separated +- by any nonempty string of blanks, otherwise by exactly one +- tab character whose value (when cast to unsigned char) equals TAB. */ +-static int tab = -1; ++/* Tab character separating fields. If NULL, fields are separated ++ by any nonempty string of blanks. */ ++static char *tab = NULL; ++ ++/* The number of bytes used for tab. */ ++static size_t tablen = 0; + + /* If nonzero, check that the input is correctly ordered. */ + static enum +@@ -276,13 +292,14 @@ xfields (struct line *line) + if (ptr == lim) + return; + +- if (0 <= tab && tab != '\n') ++ if (tab != NULL) + { ++ unsigned char t = tab[0]; + char *sep; +- for (; (sep = memchr (ptr, tab, lim - ptr)) != NULL; ptr = sep + 1) ++ for (; (sep = memchr (ptr, t, lim - ptr)) != NULL; ptr = sep + 1) + extract_field (line, ptr, sep - ptr); + } +- else if (tab < 0) ++ else + { + /* Skip leading blanks before the first field. */ + while (field_sep (*ptr)) +@@ -306,6 +323,147 @@ xfields (struct line *line) + extract_field (line, ptr, lim - ptr); } -+extern int -+set_utf_locale (void) ++#if HAVE_MBRTOWC ++static void ++xfields_multibyte (struct line *line) +{ -+ /*try using some predefined locale */ -+ const char* predef_locales[] = {"C.UTF8","en_US.UTF8","en_GB.UTF8"}; ++ char *ptr = line->buf.buffer; ++ char const *lim = ptr + line->buf.length - 1; ++ wchar_t wc = 0; ++ size_t mblength = 1; ++ mbstate_t state, state_bak; + -+ const int predef_locales_count=3; -+ for (int i=0;i= lim) ++ return; ++ ++ if (tab != NULL) ++ { ++ char *sep = ptr; ++ for (; ptr < lim; ptr = sep + mblength) ++ { ++ sep = ptr; ++ while (sep < lim) ++ { ++ state_bak = state; ++ mblength = mbrtowc (&wc, sep, lim - sep + 1, &state); ++ ++ if (mblength == (size_t)-1 || mblength == (size_t)-2) ++ { ++ mblength = 1; ++ state = state_bak; ++ } ++ mblength = (mblength < 1) ? 1 : mblength; ++ ++ if (mblength == tablen && !memcmp (sep, tab, mblength)) ++ break; ++ else ++ { ++ sep += mblength; ++ continue; ++ } ++ } ++ ++ if (sep >= lim) ++ break; ++ ++ extract_field (line, ptr, sep - ptr); ++ } ++ } ++ else ++ { ++ /* Skip leading blanks before the first field. */ ++ while(ptr < lim) ++ { ++ state_bak = state; ++ mblength = mbrtowc (&wc, ptr, lim - ptr + 1, &state); ++ ++ if (mblength == (size_t)-1 || mblength == (size_t)-2) + { ++ mblength = 1; ++ state = state_bak; + break; + } -+ else if (i==predef_locales_count-1) -+ { -+ return 1; -+ error (EXIT_FAILURE, errno, _("cannot set UTF-8 locale")); -+ } -+ } -+ return 0; ++ mblength = (mblength < 1) ? 1 : mblength; ++ ++ if (!iswblank(wc) && wc != '\n') ++ break; ++ ptr += mblength; ++ } ++ ++ do ++ { ++ char *sep; ++ state_bak = state; ++ mblength = mbrtowc (&wc, ptr, lim - ptr + 1, &state); ++ if (mblength == (size_t)-1 || mblength == (size_t)-2) ++ { ++ mblength = 1; ++ state = state_bak; ++ break; ++ } ++ mblength = (mblength < 1) ? 1 : mblength; ++ ++ sep = ptr + mblength; ++ while (sep < lim) ++ { ++ state_bak = state; ++ mblength = mbrtowc (&wc, sep, lim - sep + 1, &state); ++ if (mblength == (size_t)-1 || mblength == (size_t)-2) ++ { ++ mblength = 1; ++ state = state_bak; ++ break; ++ } ++ mblength = (mblength < 1) ? 1 : mblength; ++ ++ if (iswblank (wc) || wc == '\n') ++ break; ++ ++ sep += mblength; ++ } ++ ++ extract_field (line, ptr, sep - ptr); ++ if (sep >= lim) ++ return; ++ ++ state_bak = state; ++ mblength = mbrtowc (&wc, sep, lim - sep + 1, &state); ++ if (mblength == (size_t)-1 || mblength == (size_t)-2) ++ { ++ mblength = 1; ++ state = state_bak; ++ break; ++ } ++ mblength = (mblength < 1) ? 1 : mblength; ++ ++ ptr = sep + mblength; ++ while (ptr < lim) ++ { ++ state_bak = state; ++ mblength = mbrtowc (&wc, ptr, lim - ptr + 1, &state); ++ if (mblength == (size_t)-1 || mblength == (size_t)-2) ++ { ++ mblength = 1; ++ state = state_bak; ++ break; ++ } ++ mblength = (mblength < 1) ? 1 : mblength; ++ ++ if (!iswblank (wc) && wc != '\n') ++ break; ++ ++ ptr += mblength; ++ } ++ } ++ while (ptr < lim); ++ } ++ ++ extract_field (line, ptr, lim - ptr); +} ++#endif + -+extern bool -+check_utf_locale(void) -+{ -+ char* locale = setlocale (LC_CTYPE , NULL); -+ if (locale == NULL) -+ { -+ return false; -+ } -+ else if (strcasestr(locale, "utf8") == NULL && strcasestr(locale, "utf-8") == NULL) -+ { -+ return false; -+ } -+ return true; -+} -+ -+extern bool -+check_bom(FILE* fp, mb_file_t *mbf) -+{ -+ int c; -+ -+ -+ c=fgetc(fp); -+ -+ /*test BOM header of the first file */ -+ mbf->bufcount=0; -+ if (c == 0xEF) -+ { -+ c=fgetc(fp); -+ } -+ else -+ { -+ if (c != EOF) -+ { -+ ungetc(c,fp); -+ } -+ return false; -+ } -+ -+ if (c == 0xBB) -+ { -+ c=fgetc(fp); -+ } -+ else -+ { -+ if ( c!= EOF ) -+ { -+ mbf->buf[0]=(unsigned char) 0xEF; -+ mbf->bufcount=1; -+ ungetc(c,fp); -+ return false; -+ } -+ else -+ { -+ ungetc(0xEF,fp); -+ return false; -+ } -+ } -+ if (c == 0xBF) -+ { -+ mbf->bufcount=0; -+ return true; -+ } -+ else -+ { -+ if (c != EOF) -+ { -+ mbf->buf[0]=(unsigned char) 0xEF; -+ mbf->buf[1]=(unsigned char) 0xBB; -+ mbf->bufcount=2; -+ ungetc(c,fp); -+ return false; -+ } -+ else -+ { -+ mbf->buf[0]=(unsigned char) 0xEF; -+ mbf->bufcount=1; -+ ungetc(0xBB,fp); -+ return false; -+ } -+ } -+ return false; -+} -+ -+extern void -+print_bom(void) -+{ -+ putc (0xEF, stdout); -+ putc (0xBB, stdout); -+ putc (0xBF, stdout); -+} -+ - /* Add the comma or blank separated list of tab stops STOPS - to the list of tab stops. */ - extern void -diff --git a/src/expand-common.h b/src/expand-common.h -index 46ef4e3..e19469b 100644 ---- a/src/expand-common.h -+++ b/src/expand-common.h -@@ -29,6 +29,18 @@ extern idx_t max_column_width; - /* The desired exit status. */ - extern int exit_status; - -+extern int -+set_utf_locale (void); -+ -+extern bool -+check_utf_locale(void); -+ -+extern bool -+check_bom(FILE* fp, mb_file_t *mbf); -+ -+extern void -+print_bom(void); -+ - /* Add tab stop TABVAL to the end of 'tab_list'. */ - extern void - add_tab_stop (colno tabval); -diff --git a/src/expand.c b/src/expand.c -index 5ec7ce9..65ac315 100644 ---- a/src/expand.c -+++ b/src/expand.c -@@ -38,6 +38,9 @@ - #include - #include - #include -+ -+#include -+ - #include "system.h" - #include "expand-common.h" - -@@ -96,19 +99,41 @@ expand (void) + static void + freeline (struct line *line) { - /* Input stream. */ - FILE *fp = next_file (nullptr); -+ mb_file_t mbf; -+ mbf_char_t c; -+ /* True if the starting locale is utf8. */ -+ bool using_utf_locale; -+ -+ /* True if the first file contains BOM header. */ -+ bool found_bom; -+ using_utf_locale=check_utf_locale(); +@@ -327,56 +485,133 @@ keycmp (struct line const *line1, struct line const *line2, + size_t jf_1, size_t jf_2) + { + /* Start of field to compare in each file. */ +- char *beg1; +- char *beg2; +- +- size_t len1; +- size_t len2; /* Length of fields to compare. */ ++ char *beg[2]; ++ char *copy[2]; ++ size_t len[2]; /* Length of fields to compare. */ + int diff; ++ int i, j; ++ int mallocd = 0; - if (!fp) - return; -+ mbf_init (mbf, fp); -+ found_bom=check_bom(fp,&mbf); - -- while (true) -+ if (using_utf_locale == false && found_bom == true) -+ { -+ /*try using some predefined locale */ -+ -+ if (set_utf_locale () != 0) + if (jf_1 < line1->nfields) { -- /* Input character, or EOF. */ -- int c; -+ error (EXIT_FAILURE, errno, _("cannot set UTF-8 locale")); -+ } -+ } -+ -+ -+ if (found_bom == true) -+ { -+ print_bom(); -+ } - -+ while (true) -+ { - /* If true, perform translations. */ - bool convert = true; - -- - /* The following variables have valid values only when CONVERT - is true: */ - -@@ -118,17 +143,48 @@ expand (void) - /* Index in TAB_LIST of next tab stop to examine. */ - idx_t tab_index = 0; - -- - /* Convert a line of text. */ - - do - { -- while ((c = getc (fp)) < 0 && (fp = next_file (fp))) -- continue; -+ while (true) { -+ mbf_getc (c, mbf); -+ if ((mb_iseof (c)) && (fp = next_file (fp))) -+ { -+ mbf_init (mbf, fp); -+ if (fp!=NULL) -+ { -+ if (check_bom(fp,&mbf)==true) -+ { -+ /*Not the first file - check BOM header*/ -+ if (using_utf_locale==false && found_bom==false) -+ { -+ /*BOM header in subsequent file but not in the first one. */ -+ error (EXIT_FAILURE, errno, _("combination of files with and without BOM header")); -+ } -+ } -+ else -+ { -+ if(using_utf_locale==false && found_bom==true) -+ { -+ /*First file conatined BOM header - locale was switched to UTF -+ *all subsequent files should contain BOM. */ -+ error (EXIT_FAILURE, errno, _("combination of files with and without BOM header")); -+ } -+ } -+ } -+ continue; -+ } -+ else -+ { -+ break; -+ } -+ } -+ - - if (convert) - { -- if (c == '\t') -+ if (mb_iseq (c, '\t')) - { - /* Column the next input tab stop is on. */ - bool last_tab; -@@ -139,31 +195,33 @@ expand (void) - if (putchar (' ') < 0) - write_error (); - -- c = ' '; -+ mb_setascii (&c, ' '); - } -- else if (c == '\b') -+ else if (mb_iseq (c, '\b')) - { - /* Go back one column, and force recalculation of the - next tab stop. */ - column -= !!column; - tab_index -= !!tab_index; - } -- else -+ /* A leading control character could make us trip over. */ -+ else if (!mb_iscntrl (c)) - { -- if (ckd_add (&column, column, 1)) -+ if (ckd_add (&column, column, mb_width (c))) - error (EXIT_FAILURE, 0, _("input line is too long")); - } - -- convert &= convert_entire_line || !! isblank (c); -+ convert &= convert_entire_line || !! mb_isblank (c); - } - -- if (c < 0) -+ if (mb_iseof (c)) - return; - -- if (putchar (c) < 0) -+ mb_putc (c, stdout); -+ if (ferror (stdout)) - write_error (); - } -- while (c != '\n'); -+ while (!mb_iseq (c, '\n')); +- beg1 = line1->fields[jf_1].beg; +- len1 = line1->fields[jf_1].len; ++ beg[0] = line1->fields[jf_1].beg; ++ len[0] = line1->fields[jf_1].len; } + else + { +- beg1 = NULL; +- len1 = 0; ++ beg[0] = NULL; ++ len[0] = 0; + } + + if (jf_2 < line2->nfields) + { +- beg2 = line2->fields[jf_2].beg; +- len2 = line2->fields[jf_2].len; ++ beg[1] = line2->fields[jf_2].beg; ++ len[1] = line2->fields[jf_2].len; + } + else + { +- beg2 = NULL; +- len2 = 0; ++ beg[1] = NULL; ++ len[1] = 0; + } + +- if (len1 == 0) +- return len2 == 0 ? 0 : -1; +- if (len2 == 0) ++ if (len[0] == 0) ++ return len[1] == 0 ? 0 : -1; ++ if (len[1] == 0) + return 1; + + if (ignore_case) + { +- /* FIXME: ignore_case does not work with NLS (in particular, +- with multibyte chars). */ +- diff = memcasecmp (beg1, beg2, MIN (len1, len2)); ++#ifdef HAVE_MBRTOWC ++ if (MB_CUR_MAX > 1) ++ { ++ size_t mblength; ++ wchar_t wc, uwc; ++ mbstate_t state, state_bak; ++ ++ memset (&state, '\0', sizeof (mbstate_t)); ++ ++ for (i = 0; i < 2; i++) ++ { ++ mallocd = 1; ++ copy[i] = xmalloc (len[i] + 1); ++ memset (copy[i], '\0',len[i] + 1); ++ ++ for (j = 0; j < MIN (len[0], len[1]);) ++ { ++ state_bak = state; ++ mblength = mbrtowc (&wc, beg[i] + j, len[i] - j, &state); ++ ++ switch (mblength) ++ { ++ case (size_t) -1: ++ case (size_t) -2: ++ state = state_bak; ++ /* Fall through */ ++ case 0: ++ mblength = 1; ++ break; ++ ++ default: ++ uwc = towupper (wc); ++ ++ if (uwc != wc) ++ { ++ mbstate_t state_wc; ++ size_t mblen; ++ ++ memset (&state_wc, '\0', sizeof (mbstate_t)); ++ mblen = wcrtomb (copy[i] + j, uwc, &state_wc); ++ assert (mblen != (size_t)-1); ++ } ++ else ++ memcpy (copy[i] + j, beg[i] + j, mblength); ++ } ++ j += mblength; ++ } ++ copy[i][j] = '\0'; ++ } ++ } ++ else ++#endif ++ { ++ for (i = 0; i < 2; i++) ++ { ++ mallocd = 1; ++ copy[i] = xmalloc (len[i] + 1); ++ ++ for (j = 0; j < MIN (len[0], len[1]); j++) ++ copy[i][j] = toupper (beg[i][j]); ++ ++ copy[i][j] = '\0'; ++ } ++ } + } + else + { +- if (hard_LC_COLLATE) +- return xmemcoll (beg1, len1, beg2, len2); +- diff = memcmp (beg1, beg2, MIN (len1, len2)); ++ copy[0] = beg[0]; ++ copy[1] = beg[1]; + } + ++ if (hard_LC_COLLATE) ++ { ++ diff = xmemcoll ((char *) copy[0], len[0], (char *) copy[1], len[1]); ++ ++ if (mallocd) ++ for (i = 0; i < 2; i++) ++ free (copy[i]); ++ ++ return diff; ++ } ++ diff = memcmp (copy[0], copy[1], MIN (len[0], len[1])); ++ ++ if (mallocd) ++ for (i = 0; i < 2; i++) ++ free (copy[i]); ++ ++ + if (diff) + return diff; +- return len1 < len2 ? -1 : len1 != len2; ++ return len[0] - len[1]; } -diff --git a/src/local.mk b/src/local.mk -index a8ad6b4..b0e61ec 100644 ---- a/src/local.mk -+++ b/src/local.mk -@@ -490,8 +490,8 @@ src_base32_CPPFLAGS = -DBASE_TYPE=32 $(AM_CPPFLAGS) - src_basenc_SOURCES = src/basenc.c - src_basenc_CPPFLAGS = -DBASE_TYPE=42 $(AM_CPPFLAGS) + /* Check that successive input lines PREV and CURRENT from input file +@@ -468,6 +703,11 @@ get_line (FILE *fp, struct line **linep, int which) + } + ++line_no[which - 1]; --src_expand_SOURCES = src/expand.c src/expand-common.c --src_unexpand_SOURCES = src/unexpand.c src/expand-common.c -+src_expand_SOURCES = src/expand.c src/expand-common.c lib/mbfile.c lib/mbchar.c -+src_unexpand_SOURCES = src/unexpand.c src/expand-common.c lib/mbfile.c lib/mbchar.c ++#if HAVE_MBRTOWC ++ if (MB_CUR_MAX > 1) ++ xfields_multibyte (line); ++ else ++#endif + xfields (line); + + if (prevline[which - 1]) +@@ -563,21 +803,28 @@ prfield (size_t n, struct line const *line) + + /* Output all the fields in line, other than the join field. */ + ++#define PUT_TAB_CHAR \ ++ do \ ++ { \ ++ (tab != NULL) ? \ ++ fwrite(tab, sizeof(char), tablen, stdout) : putchar (' '); \ ++ } \ ++ while (0) ++ + static void + prfields (struct line const *line, size_t join_field, size_t autocount) + { + size_t i; + size_t nfields = autoformat ? autocount : line->nfields; +- char output_separator = tab < 0 ? ' ' : tab; + + for (i = 0; i < join_field && i < nfields; ++i) + { +- putchar (output_separator); ++ PUT_TAB_CHAR; + prfield (i, line); + } + for (i = join_field + 1; i < nfields; ++i) + { +- putchar (output_separator); ++ PUT_TAB_CHAR; + prfield (i, line); + } + } +@@ -588,7 +835,6 @@ static void + prjoin (struct line const *line1, struct line const *line2) + { + const struct outlist *outlist; +- char output_separator = tab < 0 ? ' ' : tab; + size_t field; + struct line const *line; + +@@ -622,7 +868,7 @@ prjoin (struct line const *line1, struct line const *line2) + o = o->next; + if (o == NULL) + break; +- putchar (output_separator); ++ PUT_TAB_CHAR; + } + putchar (eolchar); + } +@@ -1098,20 +1344,43 @@ main (int argc, char **argv) + + case 't': + { +- unsigned char newtab = optarg[0]; ++ char *newtab = NULL; ++ size_t newtablen; ++ newtab = xstrdup (optarg); ++#if HAVE_MBRTOWC ++ if (MB_CUR_MAX > 1) ++ { ++ mbstate_t state; ++ ++ memset (&state, 0, sizeof (mbstate_t)); ++ newtablen = mbrtowc (NULL, newtab, ++ strnlen (newtab, MB_LEN_MAX), ++ &state); ++ if (newtablen == (size_t) 0 ++ || newtablen == (size_t) -1 ++ || newtablen == (size_t) -2) ++ newtablen = 1; ++ } ++ else ++#endif ++ newtablen = 1; + if (! newtab) +- newtab = '\n'; /* '' => process the whole line. */ ++ newtab = (char*)"\n"; /* '' => process the whole line. */ + else if (optarg[1]) + { +- if (STREQ (optarg, "\\0")) +- newtab = '\0'; +- else +- die (EXIT_FAILURE, 0, _("multi-character tab %s"), +- quote (optarg)); ++ if (newtablen == 1 && newtab[1]) ++ { ++ if (STREQ (newtab, "\\0")) ++ newtab[0] = '\0'; ++ } ++ } ++ if (tab != NULL && strcmp (tab, newtab)) ++ { ++ free (newtab); ++ die (EXIT_FAILURE, 0, _("incompatible tabs")); + } +- if (0 <= tab && tab != newtab) +- die (EXIT_FAILURE, 0, _("incompatible tabs")); + tab = newtab; ++ tablen = newtablen; + } + break; - src_wc_SOURCES = src/wc.c - if USE_AVX512_WC_LINECOUNT diff --git a/src/pr.c b/src/pr.c -index 10b8c52..079c86c 100644 +index 26f221f..633f50e 100644 --- a/src/pr.c +++ b/src/pr.c -@@ -312,6 +312,24 @@ - #include +@@ -311,6 +311,24 @@ + #include #include + @@ -1443,8 +973,8 @@ index 10b8c52..079c86c 100644 +#endif + #include "system.h" - #include "c-ctype.h" - #include "fadvise.h" + #include "die.h" + #include "error.h" @@ -325,6 +343,18 @@ #include "xstrtol-error.h" #include "xdectoint.h" @@ -1486,23 +1016,23 @@ index 10b8c52..079c86c 100644 static bool read_line (COLUMN *p); static bool print_page (void); static bool print_stored (COLUMN *p); -@@ -428,6 +471,7 @@ static void pad_across_to (int position); - static void add_line_number (COLUMN *p); - static int getoptnum (char const *n_str, int min, char const *errfmt); +@@ -429,6 +472,7 @@ static void add_line_number (COLUMN *p); + static void getoptnum (const char *n_str, int min, int *num, + const char *errfmt); static void getoptarg (char *arg, char switch_char, char *character, + int *character_length, int *character_width, int *number); static void print_files (int number_of_files, char **av); static void init_parameters (int number_of_files); -@@ -441,7 +485,6 @@ static void store_char (char c); +@@ -442,7 +486,6 @@ static void store_char (char c); static void pad_down (unsigned int lines); static void read_rest_of_line (COLUMN *p); static void skip_read (COLUMN *p, int column_number); -static void print_char (char c); static void cleanup (void); static void print_sep_string (void); - static void separator_string (char const *optarg_S); -@@ -453,7 +496,7 @@ static COLUMN *column_vector; + static void separator_string (const char *optarg_S); +@@ -454,7 +497,7 @@ static COLUMN *column_vector; we store the leftmost columns contiguously in buff. To print a line from buff, get the index of the first character from line_vector[i], and print up to line_vector[i + 1]. */ @@ -1511,7 +1041,7 @@ index 10b8c52..079c86c 100644 /* Index of the position in buff where the next character will be stored. */ -@@ -557,7 +600,7 @@ static int chars_per_column; +@@ -558,7 +601,7 @@ static int chars_per_column; static bool untabify_input = false; /* (-e) The input tab character. */ @@ -1520,7 +1050,7 @@ index 10b8c52..079c86c 100644 /* (-e) Tabstops are at chars_per_tab, 2*chars_per_tab, 3*chars_per_tab, ... where the leftmost column is 1. */ -@@ -567,7 +610,10 @@ static int chars_per_input_tab = 8; +@@ -568,7 +611,10 @@ static int chars_per_input_tab = 8; static bool tabify_output = false; /* (-i) The output tab character. */ @@ -1532,7 +1062,7 @@ index 10b8c52..079c86c 100644 /* (-i) The width of the output tab. */ static int chars_per_output_tab = 8; -@@ -637,7 +683,13 @@ static int line_number; +@@ -638,7 +684,13 @@ static int line_number; static bool numbered_lines = false; /* (-n) Character which follows each line number. */ @@ -1547,7 +1077,7 @@ index 10b8c52..079c86c 100644 /* (-n) line counting starts with 1st line of input file (not with 1st line of 1st page printed). */ -@@ -690,6 +742,7 @@ static bool use_col_separator = false; +@@ -691,6 +743,7 @@ static bool use_col_separator = false; -a|COLUMN|-m is a 'space' and with the -J option a 'tab'. */ static char const *col_sep_string = ""; static int col_sep_length = 0; @@ -1555,7 +1085,7 @@ index 10b8c52..079c86c 100644 static char *column_separator = (char *) " "; static char *line_separator = (char *) "\t"; -@@ -852,6 +905,13 @@ separator_string (char const *optarg_S) +@@ -852,6 +905,13 @@ separator_string (const char *optarg_S) integer_overflow (); col_sep_length = len; col_sep_string = optarg_S; @@ -1621,7 +1151,7 @@ index 10b8c52..079c86c 100644 /* Could check tab width > 0. */ tabify_output = true; break; -@@ -986,8 +1069,8 @@ main (int argc, char **argv) +@@ -985,8 +1068,8 @@ main (int argc, char **argv) case 'n': numbered_lines = true; if (optarg) @@ -1632,7 +1162,7 @@ index 10b8c52..079c86c 100644 break; case 'N': skip_count = false; -@@ -1013,6 +1096,7 @@ main (int argc, char **argv) +@@ -1011,6 +1094,7 @@ main (int argc, char **argv) /* Reset an additional input of -s, -S dominates -s */ col_sep_string = ""; col_sep_length = 0; @@ -1640,7 +1170,7 @@ index 10b8c52..079c86c 100644 use_col_separator = true; if (optarg) separator_string (optarg); -@@ -1168,7 +1252,8 @@ getoptnum (char const *n_str, int min, char const *err) +@@ -1166,10 +1250,45 @@ getoptnum (const char *n_str, int min, int *num, const char *err) a number. */ static void @@ -1648,12 +1178,7 @@ index 10b8c52..079c86c 100644 +getoptarg (char *arg, char switch_char, char *character, int *character_length, + int *character_width, int *number) { - if (!*arg) - { -@@ -1177,7 +1262,41 @@ getoptarg (char *arg, char switch_char, char *character, int *number) - } - - if (!c_isdigit (*arg)) + if (!ISDIGIT (*arg)) - *character = *arg++; + { +#ifdef HAVE_MBRTOWC @@ -1693,7 +1218,7 @@ index 10b8c52..079c86c 100644 if (*arg) { long int tmp_long; -@@ -1206,6 +1325,11 @@ static void +@@ -1191,6 +1310,11 @@ static void init_parameters (int number_of_files) { int chars_used_by_number = 0; @@ -1705,7 +1230,7 @@ index 10b8c52..079c86c 100644 lines_per_body = lines_per_page - lines_per_header - lines_per_footer; if (lines_per_body <= 0) -@@ -1243,7 +1367,7 @@ init_parameters (int number_of_files) +@@ -1228,7 +1352,7 @@ init_parameters (int number_of_files) else col_sep_string = column_separator; @@ -1714,7 +1239,7 @@ index 10b8c52..079c86c 100644 use_col_separator = true; } /* It's rather pointless to define a TAB separator with column -@@ -1275,11 +1399,11 @@ init_parameters (int number_of_files) +@@ -1258,11 +1382,11 @@ init_parameters (int number_of_files) + TAB_WIDTH (chars_per_input_tab, chars_per_number); */ /* Estimate chars_per_text without any margin and keep it constant. */ @@ -1728,16 +1253,16 @@ index 10b8c52..079c86c 100644 /* The number is part of the column width unless we are printing files in parallel. */ -@@ -1288,7 +1412,7 @@ init_parameters (int number_of_files) +@@ -1271,7 +1395,7 @@ init_parameters (int number_of_files) } int sep_chars, useful_chars; -- if (ckd_mul (&sep_chars, columns - 1, col_sep_length)) -+ if (ckd_mul (&sep_chars, columns - 1, col_sep_width)) +- if (INT_MULTIPLY_WRAPV (columns - 1, col_sep_length, &sep_chars)) ++ if (INT_MULTIPLY_WRAPV (columns - 1, col_sep_width, &sep_chars)) sep_chars = INT_MAX; - if (ckd_sub (&useful_chars, chars_per_line - chars_used_by_number, - sep_chars)) -@@ -1311,7 +1435,7 @@ init_parameters (int number_of_files) + if (INT_SUBTRACT_WRAPV (chars_per_line - chars_used_by_number, sep_chars, + &useful_chars)) +@@ -1294,7 +1418,7 @@ init_parameters (int number_of_files) We've to use 8 as the lower limit, if we use chars_per_default_tab = 8 to expand a tab which is not an input_tab-char. */ free (clump_buff); @@ -1746,7 +1271,7 @@ index 10b8c52..079c86c 100644 } /* Open the necessary files, -@@ -1417,7 +1541,7 @@ init_funcs (void) +@@ -1400,7 +1524,7 @@ init_funcs (void) /* Enlarge p->start_position of first column to use the same form of padding_not_printed with all columns. */ @@ -1755,7 +1280,7 @@ index 10b8c52..079c86c 100644 /* This loop takes care of all but the rightmost column. */ -@@ -1451,7 +1575,7 @@ init_funcs (void) +@@ -1434,7 +1558,7 @@ init_funcs (void) } else { @@ -1764,7 +1289,7 @@ index 10b8c52..079c86c 100644 h_next = h + chars_per_column; } } -@@ -1751,9 +1875,9 @@ static void +@@ -1725,9 +1849,9 @@ static void align_column (COLUMN *p) { padding_not_printed = p->start_position; @@ -1776,9 +1301,9 @@ index 10b8c52..079c86c 100644 padding_not_printed = ANYWHERE; } -@@ -2030,13 +2154,13 @@ store_char (char c) +@@ -2002,13 +2126,13 @@ store_char (char c) /* May be too generous. */ - buff = xpalloc (buff, &buff_allocated, 1, -1, sizeof *buff); + buff = X2REALLOC (buff, &buff_allocated); } - buff[buff_current++] = c; + buff[buff_current++] = (unsigned char) c; @@ -1792,7 +1317,7 @@ index 10b8c52..079c86c 100644 char *s; int num_width; -@@ -2053,22 +2177,24 @@ add_line_number (COLUMN *p) +@@ -2025,22 +2149,24 @@ add_line_number (COLUMN *p) /* Tabification is assumed for multiple columns, also for n-separators, but 'default n-separator = TAB' hasn't been given priority over equal column_width also specified by POSIX. */ @@ -1821,7 +1346,7 @@ index 10b8c52..079c86c 100644 output_position = POS_AFTER_TAB (chars_per_output_tab, output_position); } -@@ -2227,7 +2353,7 @@ print_white_space (void) +@@ -2199,7 +2325,7 @@ print_white_space (void) while (goal - h_old > 1 && (h_new = POS_AFTER_TAB (chars_per_output_tab, h_old)) <= goal) { @@ -1830,7 +1355,7 @@ index 10b8c52..079c86c 100644 h_old = h_new; } while (++h_old <= goal) -@@ -2247,6 +2373,7 @@ print_sep_string (void) +@@ -2219,6 +2345,7 @@ print_sep_string (void) { char const *s = col_sep_string; int l = col_sep_length; @@ -1838,7 +1363,7 @@ index 10b8c52..079c86c 100644 if (separators_not_printed <= 0) { -@@ -2258,6 +2385,7 @@ print_sep_string (void) +@@ -2230,6 +2357,7 @@ print_sep_string (void) { for (; separators_not_printed > 0; --separators_not_printed) { @@ -1846,7 +1371,7 @@ index 10b8c52..079c86c 100644 while (l-- > 0) { /* 3 types of sep_strings: spaces only, spaces and chars, -@@ -2271,12 +2399,15 @@ print_sep_string (void) +@@ -2243,12 +2371,15 @@ print_sep_string (void) } else { @@ -1863,7 +1388,7 @@ index 10b8c52..079c86c 100644 /* sep_string ends with some spaces */ if (spaces_not_printed > 0) print_white_space (); -@@ -2307,7 +2438,7 @@ print_clump (COLUMN *p, int n, char *clump) +@@ -2276,7 +2407,7 @@ print_clump (COLUMN *p, int n, char *clump) required number of tabs and spaces. */ static void @@ -1872,7 +1397,7 @@ index 10b8c52..079c86c 100644 { if (tabify_output) { -@@ -2331,6 +2462,74 @@ print_char (char c) +@@ -2300,6 +2431,74 @@ print_char (char c) putchar (c); } @@ -1947,7 +1472,7 @@ index 10b8c52..079c86c 100644 /* Skip to page PAGE before printing. PAGE may be larger than total number of pages. */ -@@ -2507,9 +2706,9 @@ read_line (COLUMN *p) +@@ -2477,9 +2676,9 @@ read_line (COLUMN *p) align_empty_cols = false; } @@ -1959,7 +1484,7 @@ index 10b8c52..079c86c 100644 padding_not_printed = ANYWHERE; } -@@ -2578,7 +2777,7 @@ print_stored (COLUMN *p) +@@ -2548,7 +2747,7 @@ print_stored (COLUMN *p) COLUMN *q; int line = p->current_line++; @@ -1968,7 +1493,7 @@ index 10b8c52..079c86c 100644 /* FIXME UMR: Uninitialized memory read: * This is occurring while in: -@@ -2590,7 +2789,7 @@ print_stored (COLUMN *p) +@@ -2560,7 +2759,7 @@ print_stored (COLUMN *p) xmalloc [xmalloc.c:94] init_store_cols [pr.c:1648] */ @@ -1977,7 +1502,7 @@ index 10b8c52..079c86c 100644 pad_vertically = true; -@@ -2610,9 +2809,9 @@ print_stored (COLUMN *p) +@@ -2580,9 +2779,9 @@ print_stored (COLUMN *p) } } @@ -1989,7 +1514,7 @@ index 10b8c52..079c86c 100644 padding_not_printed = ANYWHERE; } -@@ -2625,8 +2824,8 @@ print_stored (COLUMN *p) +@@ -2595,8 +2794,8 @@ print_stored (COLUMN *p) if (spaces_not_printed == 0) { output_position = p->start_position + end_vector[line]; @@ -2000,7 +1525,7 @@ index 10b8c52..079c86c 100644 } return true; -@@ -2645,7 +2844,7 @@ print_stored (COLUMN *p) +@@ -2615,7 +2814,7 @@ print_stored (COLUMN *p) number of characters is 1.) */ static int @@ -2009,7 +1534,7 @@ index 10b8c52..079c86c 100644 { unsigned char uc = c; char *s = clump_buff; -@@ -2655,10 +2854,10 @@ char_to_clump (char c) +@@ -2625,10 +2824,10 @@ char_to_clump (char c) int chars; int chars_per_c = 8; @@ -2022,7 +1547,7 @@ index 10b8c52..079c86c 100644 { width = TAB_WIDTH (chars_per_c, input_position); -@@ -2739,6 +2938,164 @@ char_to_clump (char c) +@@ -2709,6 +2908,164 @@ char_to_clump (char c) return chars; } @@ -2188,14 +1713,13 @@ index 10b8c52..079c86c 100644 looking for more options and printing the next batch of files. diff --git a/src/sort.c b/src/sort.c -index 05d00cc..eb51f20 100644 +index 6d2eec5..f189a0d 100644 --- a/src/sort.c +++ b/src/sort.c -@@ -30,6 +30,15 @@ +@@ -29,6 +29,14 @@ #include #include - #include -+ + #include +#if HAVE_WCHAR_H +# include +#endif @@ -2206,10 +1730,10 @@ index 05d00cc..eb51f20 100644 + #include "system.h" #include "argmatch.h" - #include "assure.h" -@@ -160,14 +169,39 @@ static int thousands_sep; - /* We currently ignore multi-byte grouping chars. */ - static bool thousands_sep_ignored; + #include "die.h" +@@ -157,14 +165,39 @@ static int decimal_point; + /* Thousands separator; if -1, then there isn't one. */ + static int thousands_sep; +/* True if -f is specified. */ +static bool folding; @@ -2248,9 +1772,9 @@ index 05d00cc..eb51f20 100644 /* The kind of blanks for '-b' to skip in various options. */ enum blanktype { bl_start, bl_end, bl_both }; -@@ -344,13 +378,11 @@ static bool stable; - /* An int value outside char range. */ - enum { NON_CHAR = CHAR_MAX + 1 }; +@@ -338,13 +371,11 @@ static bool reverse; + they were read if all keys compare equal. */ + static bool stable; -/* If TAB has this value, blanks separate fields. */ -enum { TAB_DEFAULT = CHAR_MAX + 1 }; @@ -2265,9 +1789,9 @@ index 05d00cc..eb51f20 100644 /* Flag to remove consecutive duplicate lines from the output. Only the last of a sequence of equal lines will be output. */ -@@ -386,6 +418,46 @@ struct tempnode - static struct tempnode *volatile temphead; - static struct tempnode *volatile *temptail = &temphead; +@@ -802,6 +833,46 @@ reap_all (void) + reap (-1); + } +/* Function pointers. */ +static void @@ -2312,7 +1836,7 @@ index 05d00cc..eb51f20 100644 /* Clean up any remaining temporary files. */ static void -@@ -1343,7 +1415,7 @@ zaptemp (char const *name) +@@ -1270,7 +1341,7 @@ zaptemp (char const *name) free (node); } @@ -2321,7 +1845,7 @@ index 05d00cc..eb51f20 100644 static int struct_month_cmp (void const *m1, void const *m2) -@@ -1358,7 +1430,7 @@ struct_month_cmp (void const *m1, void const *m2) +@@ -1285,7 +1356,7 @@ struct_month_cmp (void const *m1, void const *m2) /* Initialize the character class tables. */ static void @@ -2330,7 +1854,7 @@ index 05d00cc..eb51f20 100644 { size_t i; -@@ -1370,7 +1442,7 @@ inittables (void) +@@ -1297,7 +1368,7 @@ inittables (void) fold_toupper[i] = toupper (i); } @@ -2339,7 +1863,7 @@ index 05d00cc..eb51f20 100644 /* If we're not in the "C" locale, read different names for months. */ if (hard_LC_TIME) { -@@ -1450,6 +1522,84 @@ specify_nmerge (int oi, char c, char const *s) +@@ -1379,6 +1450,84 @@ specify_nmerge (int oi, char c, char const *s) xstrtol_fatal (e, oi, c, long_options, s); } @@ -2424,7 +1948,7 @@ index 05d00cc..eb51f20 100644 /* Specify the amount of main memory to use when sorting. */ static void specify_sort_size (int oi, char c, char const *s) -@@ -1676,7 +1826,7 @@ buffer_linelim (struct buffer const *buf) +@@ -1610,7 +1759,7 @@ buffer_linelim (struct buffer const *buf) by KEY in LINE. */ static char * @@ -2433,7 +1957,7 @@ index 05d00cc..eb51f20 100644 { char *ptr = line->text, *lim = ptr + line->length - 1; size_t sword = key->sword; -@@ -1685,10 +1835,10 @@ begfield (struct line const *line, struct keyfield const *key) +@@ -1619,10 +1768,10 @@ begfield (struct line const *line, struct keyfield const *key) /* The leading field separator itself is included in a field when -t is absent. */ @@ -2446,7 +1970,7 @@ index 05d00cc..eb51f20 100644 ++ptr; if (ptr < lim) ++ptr; -@@ -1718,12 +1868,71 @@ begfield (struct line const *line, struct keyfield const *key) +@@ -1648,11 +1797,70 @@ begfield (struct line const *line, struct keyfield const *key) return ptr; } @@ -2512,14 +2036,13 @@ index 05d00cc..eb51f20 100644 /* Return the limit of (a pointer to the first character after) the field in LINE specified by KEY. */ - ATTRIBUTE_PURE static char * -limfield (struct line const *line, struct keyfield const *key) -+limfield_uni (struct line const *line, struct keyfield const *key) ++limfield_uni (const struct line *line, const struct keyfield *key) { char *ptr = line->text, *lim = ptr + line->length - 1; size_t eword = key->eword, echar = key->echar; -@@ -1738,10 +1947,10 @@ limfield (struct line const *line, struct keyfield const *key) +@@ -1667,10 +1875,10 @@ limfield (struct line const *line, struct keyfield const *key) 'beginning' is the first character following the delimiting TAB. Otherwise, leave PTR pointing at the first 'blank' character after the preceding field. */ @@ -2532,7 +2055,7 @@ index 05d00cc..eb51f20 100644 ++ptr; if (ptr < lim && (eword || echar)) ++ptr; -@@ -1787,10 +1996,10 @@ limfield (struct line const *line, struct keyfield const *key) +@@ -1716,10 +1924,10 @@ limfield (struct line const *line, struct keyfield const *key) */ /* Make LIM point to the end of (one byte past) the current field. */ @@ -2545,12 +2068,12 @@ index 05d00cc..eb51f20 100644 if (newlim) lim = newlim; } -@@ -1825,6 +2034,130 @@ limfield (struct line const *line, struct keyfield const *key) +@@ -1750,6 +1958,130 @@ limfield (struct line const *line, struct keyfield const *key) return ptr; } +#if HAVE_MBRTOWC -+static char * _GL_ATTRIBUTE_PURE ++static char * +limfield_mb (const struct line *line, const struct keyfield *key) +{ + char *ptr = line->text, *lim = ptr + line->length - 1; @@ -2676,7 +2199,7 @@ index 05d00cc..eb51f20 100644 /* Fill BUF reading from FP, moving buf->left bytes from the end of buf->buf to the beginning first. If EOF is reached and the file wasn't terminated by a newline, supply one. Set up BUF's line -@@ -1911,8 +2244,22 @@ fillbuf (struct buffer *buf, FILE *fp, char const *file) +@@ -1836,8 +2168,22 @@ fillbuf (struct buffer *buf, FILE *fp, char const *file) else { if (key->skipsblanks) @@ -2701,32 +2224,16 @@ index 05d00cc..eb51f20 100644 line->keybeg = line_start; } } -@@ -2050,12 +2397,10 @@ find_unit_order (char const *number) +@@ -1987,7 +2333,7 @@ human_numcompare (char const *a, char const *b) + hideously fast. */ - ATTRIBUTE_PURE - static int --human_numcompare (char const *a, char const *b) -+human_numcompare (char *a, char *b) - { -- while (blanks[to_uchar (*a)]) -- a++; -- while (blanks[to_uchar (*b)]) -- b++; -+ skipblanks(&a, a + strlen(a)); -+ skipblanks(&b, b + strlen(b)); - - int diff = find_unit_order (a) - find_unit_order (b); - return (diff ? diff : strnumcmp (a, b, decimal_point, thousands_sep)); -@@ -2067,7 +2412,7 @@ human_numcompare (char const *a, char const *b) - - ATTRIBUTE_PURE static int -numcompare (char const *a, char const *b) +numcompare_uni (const char *a, const char *b) { while (blanks[to_uchar (*a)]) a++; -@@ -2077,6 +2422,25 @@ numcompare (char const *a, char const *b) +@@ -1997,6 +2343,25 @@ numcompare (char const *a, char const *b) return strnumcmp (a, b, decimal_point, thousands_sep); } @@ -2749,10 +2256,10 @@ index 05d00cc..eb51f20 100644 +} +#endif /* HAV_EMBRTOWC */ + - static int - nan_compare (long double a, long double b) - { -@@ -2118,7 +2482,7 @@ general_numcompare (char const *sa, char const *sb) + /* Work around a problem whereby the long double value returned by glibc's + strtold ("NaN", ...) contains uninitialized bits: clear all bytes of + A and B before calling strtold. FIXME: remove this function if +@@ -2047,7 +2412,7 @@ general_numcompare (char const *sa, char const *sb) Return 0 if the name in S is not recognized. */ static int @@ -2761,7 +2268,7 @@ index 05d00cc..eb51f20 100644 { size_t lo = 0; size_t hi = MONTHS_PER_YEAR; -@@ -2457,15 +2821,14 @@ debug_key (struct line const *line, struct keyfield const *key) +@@ -2323,15 +2688,14 @@ debug_key (struct line const *line, struct keyfield const *key) char saved = *lim; *lim = '\0'; @@ -2779,7 +2286,7 @@ index 05d00cc..eb51f20 100644 else if (key->general_numeric) ignore_value (strtold (beg, &tighter_lim)); else if (key->numeric || key->human_numeric) -@@ -2611,7 +2974,7 @@ key_warnings (struct keyfield const *gkey, bool gkey_only) +@@ -2465,7 +2829,7 @@ key_warnings (struct keyfield const *gkey, bool gkey_only) /* Warn about significant leading blanks. */ bool implicit_skip = key_numeric (key) || key->month; bool line_offset = key->eword == 0 && key->echar != 0; /* -k1.x,1.y */ @@ -2788,58 +2295,8 @@ index 05d00cc..eb51f20 100644 && ((!key->skipsblanks && !implicit_skip) || (!key->skipsblanks && key->schar) || (!key->skipeblanks && key->echar))) -@@ -2659,9 +3022,9 @@ key_warnings (struct keyfield const *gkey, bool gkey_only) - bool number_locale_warned = false; - if (basic_numeric_field_span) - { -- if (tab == TAB_DEFAULT -- ? thousands_sep != NON_CHAR && (isblank (to_uchar (thousands_sep))) -- : tab == thousands_sep) -+ if (tab_length -+ ? tab[0] == thousands_sep -+ : thousands_sep != NON_CHAR && (isblank (to_uchar (thousands_sep)))) - { - error (0, 0, - _("field separator %s is treated as a " -@@ -2672,9 +3035,9 @@ key_warnings (struct keyfield const *gkey, bool gkey_only) - } - if (basic_numeric_field_span || general_numeric_field_span) - { -- if (tab == TAB_DEFAULT -- ? thousands_sep != NON_CHAR && (isblank (to_uchar (decimal_point))) -- : tab == decimal_point) -+ if (tab_length -+ ? tab[0] == decimal_point -+ : thousands_sep != NON_CHAR && (isblank (to_uchar (decimal_point)))) - { - error (0, 0, - _("field separator %s is treated as a " -@@ -2682,19 +3045,19 @@ key_warnings (struct keyfield const *gkey, bool gkey_only) - quote (((char []) {decimal_point, 0}))); - number_locale_warned = true; - } -- else if (tab == '-') -+ else if (tab_length && tab[0] == '-') - { - error (0, 0, - _("field separator %s is treated as a " - "minus sign in numbers"), -- quote (((char []) {tab, 0}))); -+ quote (((char []) {tab[0], 0}))); - } -- else if (general_numeric_field_span && tab == '+') -+ else if (general_numeric_field_span && tab_length && tab[0] == '+') - { - error (0, 0, - _("field separator %s is treated as a " - "plus sign in numbers"), -- quote (((char []) {tab, 0}))); -+ quote (((char []) {tab[0], 0}))); - } - } - -@@ -2746,11 +3109,87 @@ diff_reversed (int diff, bool reversed) - return reversed ? _GL_CMP (0, diff) : diff; +@@ -2523,11 +2887,87 @@ key_warnings (struct keyfield const *gkey, bool gkey_only) + error (0, 0, _("option '-r' only applies to last-resort comparison")); } +#if HAVE_MBRTOWC @@ -2927,17 +2384,17 @@ index 05d00cc..eb51f20 100644 { struct keyfield *key = keylist; -@@ -2831,7 +3270,7 @@ keycompare (struct line const *a, struct line const *b) +@@ -2612,7 +3052,7 @@ keycompare (struct line const *a, struct line const *b) else if (key->human_numeric) diff = human_numcompare (ta, tb); else if (key->month) -- diff = getmonth (ta, nullptr) - getmonth (tb, nullptr); -+ diff = getmonth (ta, tlena, nullptr) - getmonth (tb, tlenb, nullptr); +- diff = getmonth (ta, NULL) - getmonth (tb, NULL); ++ diff = getmonth (ta, tlena, NULL) - getmonth (tb, tlenb, NULL); else if (key->random) diff = compare_random (ta, tlena, tb, tlenb); else if (key->version) -@@ -2941,6 +3380,211 @@ keycompare (struct line const *a, struct line const *b) - return diff_reversed (diff, key->reverse); +@@ -2728,6 +3168,211 @@ keycompare (struct line const *a, struct line const *b) + return key->reverse ? -diff : diff; } +#if HAVE_MBRTOWC @@ -3148,7 +2605,7 @@ index 05d00cc..eb51f20 100644 /* Compare two lines A and B, returning negative, zero, or positive depending on whether A compares less than, equal to, or greater than B. */ -@@ -2968,7 +3612,7 @@ compare (struct line const *a, struct line const *b) +@@ -2755,7 +3400,7 @@ compare (struct line const *a, struct line const *b) diff = - NONZERO (blen); else if (blen == 0) diff = 1; @@ -3157,7 +2614,7 @@ index 05d00cc..eb51f20 100644 { /* xmemcoll0 is a performance enhancement as it will not unconditionally write '\0' after the -@@ -4340,6 +4984,7 @@ set_ordering (char const *s, struct keyfield *key, enum blanktype blanktype) +@@ -4145,6 +4790,7 @@ set_ordering (char const *s, struct keyfield *key, enum blanktype blanktype) break; case 'f': key->translate = fold_toupper; @@ -3165,7 +2622,7 @@ index 05d00cc..eb51f20 100644 break; case 'g': key->general_numeric = true; -@@ -4419,7 +5064,7 @@ main (int argc, char **argv) +@@ -4224,7 +4870,7 @@ main (int argc, char **argv) initialize_exit_failure (SORT_FAILURE); hard_LC_COLLATE = hard_locale (LC_COLLATE); @@ -3174,8 +2631,8 @@ index 05d00cc..eb51f20 100644 hard_LC_TIME = hard_locale (LC_TIME); #endif -@@ -4442,6 +5087,29 @@ main (int argc, char **argv) - thousands_sep = NON_CHAR; +@@ -4245,6 +4891,29 @@ main (int argc, char **argv) + thousands_sep = -1; } +#if HAVE_MBRTOWC @@ -3204,7 +2661,7 @@ index 05d00cc..eb51f20 100644 have_read_stdin = false; inittables (); -@@ -4717,13 +5385,34 @@ main (int argc, char **argv) +@@ -4519,13 +5188,34 @@ main (int argc, char **argv) case 't': { @@ -3214,7 +2671,7 @@ index 05d00cc..eb51f20 100644 + size_t newtab_length = 1; + strncpy (newtab, optarg, MB_LEN_MAX); + if (! newtab[0]) - error (SORT_FAILURE, 0, _("empty tab")); + die (SORT_FAILURE, 0, _("empty tab")); - if (optarg[1]) +#if HAVE_MBRTOWC + if (MB_CUR_MAX > 1) @@ -3237,435 +2694,206 @@ index 05d00cc..eb51f20 100644 +#endif + if (newtab_length == 1 && optarg[1]) { - if (streq (optarg, "\\0")) + if (STREQ (optarg, "\\0")) - newtab = '\0'; + newtab[0] = '\0'; else { /* Provoke with 'sort -txx'. Complain about -@@ -4734,9 +5423,11 @@ main (int argc, char **argv) - quote (optarg)); +@@ -4536,9 +5226,11 @@ main (int argc, char **argv) + quote (optarg)); } } - if (tab != TAB_DEFAULT && tab != newtab) + if (tab_length && (tab_length != newtab_length + || memcmp (tab, newtab, tab_length) != 0)) - error (SORT_FAILURE, 0, _("incompatible tabs")); + die (SORT_FAILURE, 0, _("incompatible tabs")); - tab = newtab; + memcpy (tab, newtab, newtab_length); + tab_length = newtab_length; } break; -diff --git a/src/unexpand.c b/src/unexpand.c -index ff234d7..7c36ef6 100644 ---- a/src/unexpand.c -+++ b/src/unexpand.c -@@ -39,6 +39,9 @@ - #include +@@ -4767,12 +5459,10 @@ main (int argc, char **argv) + sort (files, nfiles, outfile, nthreads); + } + +-#ifdef lint + if (files_from) + readtokens0_free (&tok); + else + free (files); +-#endif + + if (have_read_stdin && fclose (stdin) == EOF) + sort_die (_("close failed"), "-"); +diff --git a/src/uniq.c b/src/uniq.c +index 87a0c93..9f755d9 100644 +--- a/src/uniq.c ++++ b/src/uniq.c +@@ -21,6 +21,17 @@ #include #include + ++/* Get mbstate_t, mbrtowc(). */ ++#if HAVE_WCHAR_H ++# include ++#endif + -+#include ++/* Get isw* functions. */ ++#if HAVE_WCTYPE_H ++# include ++#endif ++#include + #include "system.h" - #include "expand-common.h" + #include "argmatch.h" + #include "linebuffer.h" +@@ -33,6 +44,18 @@ + #include "memcasecmp.h" + #include "quote.h" -@@ -105,24 +108,46 @@ unexpand (void) ++/* MB_LEN_MAX is incorrectly defined to be 1 in at least one GCC ++ installation; work around this configuration error. */ ++#if !defined MB_LEN_MAX || MB_LEN_MAX < 2 ++# define MB_LEN_MAX 16 ++#endif ++ ++/* Some systems, like BeOS, have multibyte encodings but lack mbstate_t. */ ++#if HAVE_MBRTOWC && defined mbstate_t ++# define mbrtowc(pwc, s, n, ps) (mbrtowc) (pwc, s, n, 0) ++#endif ++ ++ + /* The official name of this program (e.g., no 'g' prefix). */ + #define PROGRAM_NAME "uniq" + +@@ -139,6 +162,10 @@ enum + GROUP_OPTION = CHAR_MAX + 1 + }; + ++/* Function pointers. */ ++static char * ++(*find_field) (struct linebuffer *line); ++ + static struct option const longopts[] = { - /* Input stream. */ - FILE *fp = next_file (nullptr); -+ mb_file_t mbf; + {"count", no_argument, NULL, 'c'}, +@@ -253,7 +280,7 @@ size_opt (char const *opt, char const *msgid) + return a pointer to the beginning of the line's field to be compared. */ - /* The array of pending blanks. In non-POSIX locales, blanks can - include characters other than spaces, so the blanks must be - stored, not merely counted. */ -- char *pending_blank; -+ mbf_char_t *pending_blank; -+ /* True if the starting locale is utf8. */ -+ bool using_utf_locale; + static char * _GL_ATTRIBUTE_PURE +-find_field (struct linebuffer const *line) ++find_field_uni (struct linebuffer *line) + { + size_t count; + char const *lp = line->buffer; +@@ -273,6 +300,83 @@ find_field (struct linebuffer const *line) + return line->buffer + i; + } + ++#if HAVE_MBRTOWC + -+ /* True if the first file contains BOM header. */ -+ bool found_bom; -+ using_utf_locale=check_utf_locale(); - - if (!fp) - return; - -+ mbf_init (mbf, fp); -+ found_bom=check_bom(fp,&mbf); ++# define MBCHAR_TO_WCHAR(WC, MBLENGTH, LP, POS, SIZE, STATEP, CONVFAIL) \ ++ do \ ++ { \ ++ mbstate_t state_bak; \ ++ \ ++ CONVFAIL = 0; \ ++ state_bak = *STATEP; \ ++ \ ++ MBLENGTH = mbrtowc (&WC, LP + POS, SIZE - POS, STATEP); \ ++ \ ++ switch (MBLENGTH) \ ++ { \ ++ case (size_t)-2: \ ++ case (size_t)-1: \ ++ *STATEP = state_bak; \ ++ CONVFAIL++; \ ++ /* Fall through */ \ ++ case 0: \ ++ MBLENGTH = 1; \ ++ } \ ++ } \ ++ while (0) + -+ if (using_utf_locale == false && found_bom == true) ++static char * ++find_field_multi (struct linebuffer *line) ++{ ++ size_t count; ++ char *lp = line->buffer; ++ size_t size = line->length - 1; ++ size_t pos; ++ size_t mblength; ++ wchar_t wc; ++ mbstate_t *statep; ++ int convfail = 0; ++ ++ pos = 0; ++ statep = &(line->state); ++ ++ /* skip fields. */ ++ for (count = 0; count < skip_fields && pos < size; count++) + { -+ /* Try using some predefined locale */ -+ if (set_utf_locale () != 0) ++ while (pos < size) + { -+ error (EXIT_FAILURE, errno, _("cannot set UTF-8 locale")); ++ MBCHAR_TO_WCHAR (wc, mblength, lp, pos, size, statep, convfail); ++ ++ if (convfail || !(iswblank (wc) || wc == '\n')) ++ { ++ pos += mblength; ++ break; ++ } ++ pos += mblength; ++ } ++ ++ while (pos < size) ++ { ++ MBCHAR_TO_WCHAR (wc, mblength, lp, pos, size, statep, convfail); ++ ++ if (!convfail && (iswblank (wc) || wc == '\n')) ++ break; ++ ++ pos += mblength; + } + } + - /* The worst case is a non-blank character, then one blank, then a - tab stop, then MAX_COLUMN_WIDTH - 1 blanks, then a non-blank; so - allocate MAX_COLUMN_WIDTH bytes to store the blanks. */ -- pending_blank = ximalloc (max_column_width); -+ pending_blank = ximalloc (max_column_width * sizeof (mbf_char_t)); ++ /* skip fields. */ ++ for (count = 0; count < skip_chars && pos < size; count++) ++ { ++ MBCHAR_TO_WCHAR (wc, mblength, lp, pos, size, statep, convfail); ++ pos += mblength; ++ } + -+ if (found_bom == true) -+ print_bom(); ++ return lp + pos; ++} ++#endif ++ + /* Return false if two strings OLD and NEW match, true if not. + OLD and NEW point not to the beginnings of the lines + but rather to the beginnings of the fields to compare. +@@ -493,6 +597,19 @@ main (int argc, char **argv) - while (true) - { - /* Input character, or EOF. */ -- int c; -+ mbf_char_t c; + atexit (close_stdout); - /* If true, perform translations. */ - bool convert = true; -@@ -156,12 +181,44 @@ unexpand (void) - - do - { -- while ((c = getc (fp)) < 0 && (fp = next_file (fp))) -- continue; -+ while (true) { -+ mbf_getc (c, mbf); -+ if ((mb_iseof (c)) && (fp = next_file (fp))) -+ { -+ mbf_init (mbf, fp); -+ if (fp!=NULL) -+ { -+ if (check_bom(fp,&mbf)==true) -+ { -+ /*Not the first file - check BOM header*/ -+ if (using_utf_locale==false && found_bom==false) -+ { -+ /*BOM header in subsequent file but not in the first one. */ -+ error (EXIT_FAILURE, errno, _("combination of files with and without BOM header")); -+ } -+ } -+ else -+ { -+ if(using_utf_locale==false && found_bom==true) -+ { -+ /*First file conatined BOM header - locale was switched to UTF -+ *all subsequent files should contain BOM. */ -+ error (EXIT_FAILURE, errno, _("combination of files with and without BOM header")); -+ } -+ } -+ } -+ continue; -+ } -+ else -+ { -+ break; -+ } -+ } -+ - - if (convert) - { -- bool blank = !! isblank (c); -+ bool blank = mb_isblank (c); - - if (blank) - { -@@ -175,16 +232,16 @@ unexpand (void) - - if (convert) - { -- if (c == '\t') -+ if (mb_iseq (c, '\t')) - { - column = next_tab_column; - - if (pending) -- pending_blank[0] = '\t'; -+ mb_setascii (&pending_blank[0], '\t'); - } - else - { -- column++; -+ column += mb_width (c); - - if (! (prev_blank && column == next_tab_column)) - { -@@ -192,13 +249,14 @@ unexpand (void) - will be replaced by tabs. */ - if (column == next_tab_column) - one_blank_before_tab_stop = true; -- pending_blank[pending++] = c; -+ mb_copy (&pending_blank[pending++], &c); - prev_blank = true; - continue; - } - - /* Replace the pending blanks by a tab or two. */ -- pending_blank[0] = c = '\t'; -+ mb_setascii (&c, '\t'); -+ mb_setascii (&pending_blank[0], '\t'); - } - - /* Discard pending blanks, unless it was a single -@@ -206,7 +264,7 @@ unexpand (void) - pending = one_blank_before_tab_stop; - } - } -- else if (c == '\b') -+ else if (mb_iseq (c, '\b')) - { - /* Go back one column, and force recalculation of the - next tab stop. */ -@@ -216,16 +274,20 @@ unexpand (void) - } - else - { -- column++; -- if (!column) -+ const uintmax_t orig_column = column; -+ column += mb_width (c); -+ if (column < orig_column) - error (EXIT_FAILURE, 0, _("input line is too long")); - } - - if (pending) - { - if (pending > 1 && one_blank_before_tab_stop) -- pending_blank[0] = '\t'; -- if (fwrite (pending_blank, 1, pending, stdout) != pending) -+ mb_setascii (&pending_blank[0], '\t'); -+ -+ for (int n = 0; n < pending; ++n) -+ mb_putc (pending_blank[n], stdout); -+ if (ferror (stdout)) - write_error (); - pending = 0; - one_blank_before_tab_stop = false; -@@ -235,16 +297,17 @@ unexpand (void) - convert &= convert_entire_line || blank; - } - -- if (c < 0) -+ if (mb_iseof (c)) - { - free (pending_blank); - return; - } - -- if (putchar (c) < 0) -+ mb_putc (c, stdout); -+ if (ferror (stdout)) - write_error (); - } -- while (c != '\n'); -+ while (!mb_iseq (c, '\n')); - } - } - -diff --git a/tests/Coreutils.pm b/tests/Coreutils.pm -index b55fb9d..ac80f49 100644 ---- a/tests/Coreutils.pm -+++ b/tests/Coreutils.pm -@@ -269,6 +269,9 @@ sub run_tests ($$$$$) - # Yes, this is an arbitrary limit. If it causes trouble, - # consider removing it. - my $max = 30; -+ # The downstream i18n multi-byte tests have a "-mb" suffix. -+ # Therefore add 3 to the maximum test name length. -+ $max += 3; - if ($max < length $test_name) - { - warn "$program_name: $test_name: test name is too long (> $max)\n"; -diff --git a/tests/expand/mb.sh b/tests/expand/mb.sh -new file mode 100644 -index 0000000..dd6007c ---- /dev/null -+++ b/tests/expand/mb.sh -@@ -0,0 +1,183 @@ -+#!/bin/sh -+ -+# Copyright (C) 2012-2015 Free Software Foundation, Inc. -+ -+# This program is free software: you can redistribute it and/or modify -+# it under the terms of the GNU General Public License as published by -+# the Free Software Foundation, either version 3 of the License, or -+# (at your option) any later version. -+ -+# This program is distributed in the hope that it will be useful, -+# but WITHOUT ANY WARRANTY; without even the implied warranty of -+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -+# GNU General Public License for more details. -+ -+# You should have received a copy of the GNU General Public License -+# along with this program. If not, see . -+ -+. "${srcdir=.}/tests/init.sh"; path_prepend_ ./src -+print_ver_ expand -+ -+export LC_ALL=en_US.UTF-8 -+ -+#input containing multibyte characters -+cat <<\EOF > in || framework_failure_ -+1234567812345678123456781 -+. . . . -+a b c d -+. . . . -+ä ö ü ß -+. . . . -+EOF -+env printf ' äöü\t. öüä. \tä xx\n' >> in || framework_failure_ -+ -+cat <<\EOF > exp || framework_failure_ -+1234567812345678123456781 -+. . . . -+a b c d -+. . . . -+ä ö ü ß -+. . . . -+ äöü . öüä. ä xx -+EOF -+ -+expand < in > out || fail=1 -+compare exp out > /dev/null 2>&1 || fail=1 -+ -+#multiple files as an input -+cat <<\EOF >> exp || framework_failure_ -+1234567812345678123456781 -+. . . . -+a b c d -+. . . . -+ä ö ü ß -+. . . . -+ äöü . öüä. ä xx -+EOF -+ -+expand ./in ./in > out || fail=1 -+compare exp out > /dev/null 2>&1 || fail=1 -+ -+#test characters with display widths != 1 -+env printf '12345678 -+e\t|ascii(1) -+\u00E9\t|composed(1) -+e\u0301\t|decomposed(1) -+\u3000\t|ideo-space(2) -+\uFF0D\t|full-hypen(2) -+' > in || framework_failure_ -+ -+env printf '12345678 -+e |ascii(1) -+\u00E9 |composed(1) -+e\u0301 |decomposed(1) -+\u3000 |ideo-space(2) -+\uFF0D |full-hypen(2) -+' > exp || framework_failure_ -+ -+expand < in > out || fail=1 -+compare exp out > /dev/null 2>&1 || fail=1 -+ -+#shouldn't fail with "input line too long" -+#when a line starts with a control character -+env printf '\n' > in || framework_failure_ -+ -+expand < in > out || fail=1 -+compare in out > /dev/null 2>&1 || fail=1 -+ -+#non-Unicode characters interspersed between Unicode ones -+env printf '12345678 -+\t\xFF| -+\xFF\t| -+\t\xFFä| -+ä\xFF\t| -+\tä\xFF| -+\xFF\tä| -+äbcdef\xFF\t| -+' > in || framework_failure_ -+ -+env printf '12345678 -+ \xFF| -+\xFF | -+ \xFFä| -+ä\xFF | -+ ä\xFF| -+\xFF ä| -+äbcdef\xFF | -+' > exp || framework_failure_ -+ -+expand < in > out || fail=1 -+compare exp out > /dev/null 2>&1 || fail=1 ++#if HAVE_MBRTOWC ++ if (MB_CUR_MAX > 1) ++ { ++ find_field = find_field_multi; ++ } ++ else ++#endif ++ { ++ find_field = find_field_uni; ++ } + + + -+#BOM header test 1 -+printf "\xEF\xBB\xBF" > in; cat <<\EOF >> in || framework_failure_ -+1234567812345678123456781 -+. . . . -+a b c d -+. . . . -+ä ö ü ß -+. . . . -+EOF -+env printf ' äöü\t. öüä. \tä xx\n' >> in || framework_failure_ -+ -+printf "\xEF\xBB\xBF" > exp; cat <<\EOF >> exp || framework_failure_ -+1234567812345678123456781 -+. . . . -+a b c d -+. . . . -+ä ö ü ß -+. . . . -+ äöü . öüä. ä xx -+EOF -+ -+ -+expand < in > out || fail=1 -+compare exp out > /dev/null 2>&1 || fail=1 -+ -+LANG=C expand < in > out || fail=1 -+compare exp out > /dev/null 2>&1 || fail=1 -+ -+LC_ALL=C expand < in > out || fail=1 -+compare exp out > /dev/null 2>&1 || fail=1 -+ -+ -+printf '\xEF\xBB\xBF' > in1; cat <<\EOF >> in1 || framework_failure_ -+1234567812345678123456781 -+. . . . -+a b c d -+. . . . -+ä ö ü ß -+. . . . -+EOF -+env printf ' äöü\t. öüä. \tä xx\n' >> in1 || framework_failure_ -+ -+ -+printf '\xEF\xBB\xBF' > exp; cat <<\EOF >> exp || framework_failure_ -+1234567812345678123456781 -+. . . . -+a b c d -+. . . . -+ä ö ü ß -+. . . . -+ äöü . öüä. ä xx -+1234567812345678123456781 -+. . . . -+a b c d -+. . . . -+ä ö ü ß -+. . . . -+ äöü . öüä. ä xx -+EOF -+ -+expand in1 in1 > out || fail=1 -+compare exp out > /dev/null 2>&1 || fail=1 -+ -+LANG=C expand in1 in1 > out || fail=1 -+compare exp out > /dev/null 2>&1 || fail=1 -+ -+LC_ALL=C expand in1 in1 > out || fail=1 -+compare exp out > /dev/null 2>&1 || fail=1 -+ -+exit $fail + skip_chars = 0; + skip_fields = 0; + check_chars = SIZE_MAX; diff --git a/tests/i18n/sort.sh b/tests/i18n/sort.sh -new file mode 100644 +new file mode 100755 index 0000000..26c95de --- /dev/null +++ b/tests/i18n/sort.sh @@ -3700,36 +2928,20 @@ index 0000000..26c95de + +Exit $fail diff --git a/tests/local.mk b/tests/local.mk -index 53fc53e..0148422 100644 +index 568944e..192f776 100644 --- a/tests/local.mk +++ b/tests/local.mk -@@ -412,6 +412,8 @@ all_tests = \ - tests/sort/sort-field-limit.sh \ - tests/sort/sort-files0-from.pl \ - tests/sort/sort-float.sh \ +@@ -369,6 +369,8 @@ all_tests = \ + tests/misc/sort-discrim.sh \ + tests/misc/sort-files0-from.pl \ + tests/misc/sort-float.sh \ + tests/misc/sort-mb-tests.sh \ + tests/i18n/sort.sh \ - tests/sort/sort-h-thousands-sep.sh \ - tests/sort/sort-merge.pl \ - tests/sort/sort-merge-fdlimit.sh \ -@@ -618,6 +620,7 @@ all_tests = \ - tests/du/threshold.sh \ - tests/du/trailing-slash.sh \ - tests/du/two-args.sh \ -+ tests/expand/mb.sh \ - tests/id/gnu-zero-uids.sh \ - tests/id/no-context.sh \ - tests/id/context.sh \ -@@ -774,6 +777,7 @@ all_tests = \ - tests/touch/read-only.sh \ - tests/touch/relative.sh \ - tests/touch/trailing-slash.sh \ -+ tests/unexpand/mb.sh \ - $(all_root_tests) - - # See tests/factor/create-test.sh. + tests/misc/sort-h-thousands-sep.sh \ + tests/misc/sort-merge.pl \ + tests/misc/sort-merge-fdlimit.sh \ diff --git a/tests/misc/expand.pl b/tests/misc/expand.pl -index 4b07210..68b9ea1 100755 +index 8a9cad1..9293e39 100755 --- a/tests/misc/expand.pl +++ b/tests/misc/expand.pl @@ -27,6 +27,15 @@ my $prog = 'expand'; @@ -3795,8 +3007,151 @@ index 4b07210..68b9ea1 100755 my $save_temps = $ENV{DEBUG}; my $verbose = $ENV{VERBOSE}; +diff --git a/tests/misc/fold.pl b/tests/misc/fold.pl +index 7b192b4..76f073f 100755 +--- a/tests/misc/fold.pl ++++ b/tests/misc/fold.pl +@@ -20,9 +20,18 @@ use strict; + + (my $program_name = $0) =~ s|.*/||; + ++my $prog = 'fold'; ++my $try = "Try \`$prog --help' for more information.\n"; ++my $inval = "$prog: invalid byte, character or field list\n$try"; ++ + # Turn off localization of executable's output. + @ENV{qw(LANGUAGE LANG LC_ALL)} = ('C') x 3; + ++# uncommented to enable multibyte paths ++my $mb_locale = $ENV{LOCALE_FR_UTF8}; ++! defined $mb_locale || $mb_locale eq 'none' ++ and $mb_locale = 'C'; ++ + my @Tests = + ( + ['s1', '-w2 -s', {IN=>"a\t"}, {OUT=>"a\n\t"}], +@@ -31,9 +40,48 @@ my @Tests = + ['s4', '-w4 -s', {IN=>"abc ef\n"}, {OUT=>"abc \nef\n"}], + ); + ++# Add _POSIX2_VERSION=199209 to the environment of each test ++# that uses an old-style option like +1. ++if ($mb_locale ne 'C') ++ { ++ # Duplicate each test vector, appending "-mb" to the test name and ++ # inserting {ENV => "LC_ALL=$mb_locale"} in the copy, so that we ++ # provide coverage for the distro-added multi-byte code paths. ++ my @new; ++ foreach my $t (@Tests) ++ { ++ my @new_t = @$t; ++ my $test_name = shift @new_t; ++ ++ # Depending on whether fold is multi-byte-patched, ++ # it emits different diagnostics: ++ # non-MB: invalid byte or field list ++ # MB: invalid byte, character or field list ++ # Adjust the expected error output accordingly. ++ if (grep {ref $_ eq 'HASH' && exists $_->{ERR} && $_->{ERR} eq $inval} ++ (@new_t)) ++ { ++ my $sub = {ERR_SUBST => 's/, character//'}; ++ push @new_t, $sub; ++ push @$t, $sub; ++ } ++ push @new, ["$test_name-mb", @new_t, {ENV => "LC_ALL=$mb_locale"}]; ++ } ++ push @Tests, @new; ++ } ++ ++@Tests = triple_test \@Tests; ++ ++# Remember that triple_test creates from each test with exactly one "IN" ++# file two more tests (.p and .r suffix on name) corresponding to reading ++# input from a file and from a pipe. The pipe-reading test would fail ++# due to a race condition about 1 in 20 times. ++# Remove the IN_PIPE version of the "output-is-input" test above. ++# The others aren't susceptible because they have three inputs each. ++@Tests = grep {$_->[0] ne 'output-is-input.p'} @Tests; ++ + my $save_temps = $ENV{DEBUG}; + my $verbose = $ENV{VERBOSE}; + +-my $prog = 'fold'; + my $fail = run_tests ($program_name, $prog, \@Tests, $save_temps, $verbose); + exit $fail; +diff --git a/tests/misc/join.pl b/tests/misc/join.pl +index 4d399d8..07f2823 100755 +--- a/tests/misc/join.pl ++++ b/tests/misc/join.pl +@@ -25,6 +25,15 @@ my $limits = getlimits (); + + my $prog = 'join'; + ++my $try = "Try \`$prog --help' for more information.\n"; ++my $inval = "$prog: invalid byte, character or field list\n$try"; ++ ++my $mb_locale; ++#Comment out next line to disable multibyte tests ++$mb_locale = $ENV{LOCALE_FR_UTF8}; ++! defined $mb_locale || $mb_locale eq 'none' ++ and $mb_locale = 'C'; ++ + my $delim = chr 0247; + sub t_subst ($) + { +@@ -333,8 +342,49 @@ foreach my $t (@tv) + push @Tests, $new_ent; + } + ++# Add _POSIX2_VERSION=199209 to the environment of each test ++# that uses an old-style option like +1. ++if ($mb_locale ne 'C') ++ { ++ # Duplicate each test vector, appending "-mb" to the test name and ++ # inserting {ENV => "LC_ALL=$mb_locale"} in the copy, so that we ++ # provide coverage for the distro-added multi-byte code paths. ++ my @new; ++ foreach my $t (@Tests) ++ { ++ my @new_t = @$t; ++ my $test_name = shift @new_t; ++ ++ # Depending on whether join is multi-byte-patched, ++ # it emits different diagnostics: ++ # non-MB: invalid byte or field list ++ # MB: invalid byte, character or field list ++ # Adjust the expected error output accordingly. ++ if (grep {ref $_ eq 'HASH' && exists $_->{ERR} && $_->{ERR} eq $inval} ++ (@new_t)) ++ { ++ my $sub = {ERR_SUBST => 's/, character//'}; ++ push @new_t, $sub; ++ push @$t, $sub; ++ } ++ #Adjust the output some error messages including test_name for mb ++ if (grep {ref $_ eq 'HASH' && exists $_->{ERR}} ++ (@new_t)) ++ { ++ my $sub2 = {ERR_SUBST => "s/$test_name-mb/$test_name/"}; ++ push @new_t, $sub2; ++ push @$t, $sub2; ++ } ++ push @new, ["$test_name-mb", @new_t, {ENV => "LC_ALL=$mb_locale"}]; ++ } ++ push @Tests, @new; ++ } ++ + @Tests = triple_test \@Tests; + ++#skip invalid-j-mb test, it is failing because of the format ++@Tests = grep {$_->[0] ne 'invalid-j-mb'} @Tests; ++ + my $save_temps = $ENV{DEBUG}; + my $verbose = $ENV{VERBOSE}; + diff --git a/tests/misc/sort-mb-tests.sh b/tests/misc/sort-mb-tests.sh -new file mode 100644 +new file mode 100755 index 0000000..11836ba --- /dev/null +++ b/tests/misc/sort-mb-tests.sh @@ -3846,136 +3201,10 @@ index 0000000..11836ba +compare exp out || { fail=1; cat out; } + +Exit $fail -diff --git a/tests/misc/unexpand.pl b/tests/misc/unexpand.pl -index bb7469c..c1dec95 100755 ---- a/tests/misc/unexpand.pl -+++ b/tests/misc/unexpand.pl -@@ -27,6 +27,14 @@ my $limits = getlimits (); - - my $prog = 'unexpand'; - -+# comment out next line to disable multibyte tests -+my $mb_locale = $ENV{LOCALE_FR_UTF8}; -+! defined $mb_locale || $mb_locale eq 'none' -+ and $mb_locale = 'C'; -+ -+my $try = "Try \`$prog --help' for more information.\n"; -+my $inval = "$prog: invalid byte, character or field list\n$try"; -+ - my @Tests = - ( - ['a1', {IN=> ' 'x 1 ."y\n"}, {OUT=> ' 'x 1 ."y\n"}], -@@ -132,6 +140,37 @@ my @Tests = - ['ts2', '-t5,8', {IN=>"x\t \t y\n"}, {OUT=>"x\t\t y\n"}], - ); - -+if ($mb_locale ne 'C') -+ { -+ # Duplicate each test vector, appending "-mb" to the test name and -+ # inserting {ENV => "LC_ALL=$mb_locale"} in the copy, so that we -+ # provide coverage for the distro-added multi-byte code paths. -+ my @new; -+ foreach my $t (@Tests) -+ { -+ my @new_t = @$t; -+ my $test_name = shift @new_t; -+ -+ # Depending on whether unexpand is multi-byte-patched, -+ # it emits different diagnostics: -+ # non-MB: invalid byte or field list -+ # MB: invalid byte, character or field list -+ # Adjust the expected error output accordingly. -+ if (grep {ref $_ eq 'HASH' && exists $_->{ERR} && $_->{ERR} eq $inval} -+ (@new_t)) -+ { -+ my $sub = {ERR_SUBST => 's/, character//'}; -+ push @new_t, $sub; -+ push @$t, $sub; -+ } -+ next if ($test_name =~ 'b-1'); -+ push @new, ["$test_name-mb", @new_t, {ENV => "LC_ALL=$mb_locale"}]; -+ } -+ push @Tests, @new; -+ } -+ -+@Tests = triple_test \@Tests; -+ - my $save_temps = $ENV{DEBUG}; - my $verbose = $ENV{VERBOSE}; - -diff --git a/tests/pr/pr-tests.pl b/tests/pr/pr-tests.pl -index 60e6106..3c64a08 100755 ---- a/tests/pr/pr-tests.pl -+++ b/tests/pr/pr-tests.pl -@@ -24,6 +24,15 @@ use strict; - my $prog = 'pr'; - my $normalize_strerror = "s/': .*/'/"; - -+my $mb_locale; -+#Uncomment the following line to enable multibyte tests -+$mb_locale = $ENV{LOCALE_FR_UTF8}; -+! defined $mb_locale || $mb_locale eq 'none' -+ and $mb_locale = 'C'; -+ -+my $try = "Try \`$prog --help' for more information.\n"; -+my $inval = "$prog: invalid byte, character or field list\n$try"; -+ - my @tv = ( - - # -b option is no longer an official option. But it's still working to -@@ -515,8 +524,48 @@ push @Tests, - {IN=>"x\tx\tx\tx\tx\nx\tx\tx\tx\tx\n"}, - {OUT=>"x\tx\tx\tx\tx\tx\tx\tx\tx\tx\n"} ]; - -+# Add _POSIX2_VERSION=199209 to the environment of each test -+# that uses an old-style option like +1. -+if ($mb_locale ne 'C') -+ { -+ # Duplicate each test vector, appending "-mb" to the test name and -+ # inserting {ENV => "LC_ALL=$mb_locale"} in the copy, so that we -+ # provide coverage for the distro-added multi-byte code paths. -+ my @new; -+ foreach my $t (@Tests) -+ { -+ my @new_t = @$t; -+ my $test_name = shift @new_t; -+ -+ # Depending on whether pr is multi-byte-patched, -+ # it emits different diagnostics: -+ # non-MB: invalid byte or field list -+ # MB: invalid byte, character or field list -+ # Adjust the expected error output accordingly. -+ if (grep {ref $_ eq 'HASH' && exists $_->{ERR} && $_->{ERR} eq $inval} -+ (@new_t)) -+ { -+ my $sub = {ERR_SUBST => 's/, character//'}; -+ push @new_t, $sub; -+ push @$t, $sub; -+ } -+ #temporarily skip some failing tests -+ next if ($test_name =~ "col-0" or $test_name =~ "col-inval" or $test_name =~ "asan1"); -+ push @new, ["$test_name-mb", @new_t, {ENV => "LC_ALL=$mb_locale"}]; -+ } -+ push @Tests, @new; -+ } -+ - @Tests = triple_test \@Tests; - -+# Remember that triple_test creates from each test with exactly one "IN" -+# file two more tests (.p and .r suffix on name) corresponding to reading -+# input from a file and from a pipe. The pipe-reading test would fail -+# due to a race condition about 1 in 20 times. -+# Remove the IN_PIPE version of the "output-is-input" test above. -+# The others aren't susceptible because they have three inputs each. -+@Tests = grep {$_->[0] ne 'output-is-input.p'} @Tests; -+ - my $save_temps = $ENV{DEBUG}; - my $verbose = $ENV{VERBOSE}; - -diff --git a/tests/sort/sort-merge.pl b/tests/sort/sort-merge.pl -index a3204d3..40942a5 100755 ---- a/tests/sort/sort-merge.pl -+++ b/tests/sort/sort-merge.pl +diff --git a/tests/misc/sort-merge.pl b/tests/misc/sort-merge.pl +index 23f6ed2..402a987 100755 +--- a/tests/misc/sort-merge.pl ++++ b/tests/misc/sort-merge.pl @@ -26,6 +26,15 @@ my $prog = 'sort'; # Turn off localization of executable's output. @ENV{qw(LANGUAGE LANG LC_ALL)} = ('C') x 3; @@ -4032,10 +3261,10 @@ index a3204d3..40942a5 100755 my $save_temps = $ENV{DEBUG}; my $verbose = $ENV{VERBOSE}; -diff --git a/tests/sort/sort.pl b/tests/sort/sort.pl -index 5fa9d52..a66952a 100755 ---- a/tests/sort/sort.pl -+++ b/tests/sort/sort.pl +diff --git a/tests/misc/sort.pl b/tests/misc/sort.pl +index c3e7f8e..6ecd3ff 100755 +--- a/tests/misc/sort.pl ++++ b/tests/misc/sort.pl @@ -24,10 +24,15 @@ my $prog = 'sort'; # Turn off localization of executable's output. @ENV{qw(LANGUAGE LANG LC_ALL)} = ('C') x 3; @@ -4053,7 +3282,7 @@ index 5fa9d52..a66952a 100755 # Since each test is run with a file name and with redirected stdin, # the name in the diagnostic is either the file name or "-". # Normalize each diagnostic to use '-'. -@@ -428,6 +433,38 @@ foreach my $t (@Tests) +@@ -423,6 +428,38 @@ foreach my $t (@Tests) } } @@ -4092,7 +3321,7 @@ index 5fa9d52..a66952a 100755 @Tests = triple_test \@Tests; # Remember that triple_test creates from each test with exactly one "IN" -@@ -437,6 +474,7 @@ foreach my $t (@Tests) +@@ -432,6 +469,7 @@ foreach my $t (@Tests) # Remove the IN_PIPE version of the "output-is-input" test above. # The others aren't susceptible because they have three inputs each. @Tests = grep {$_->[0] ne 'output-is-input.p'} @Tests; @@ -4100,184 +3329,208 @@ index 5fa9d52..a66952a 100755 my $save_temps = $ENV{DEBUG}; my $verbose = $ENV{VERBOSE}; -diff --git a/tests/unexpand/mb.sh b/tests/unexpand/mb.sh -new file mode 100644 -index 0000000..8a82d74 ---- /dev/null -+++ b/tests/unexpand/mb.sh -@@ -0,0 +1,172 @@ -+#!/bin/sh +diff --git a/tests/misc/unexpand.pl b/tests/misc/unexpand.pl +index 6ba6d40..de86723 100755 +--- a/tests/misc/unexpand.pl ++++ b/tests/misc/unexpand.pl +@@ -27,6 +27,14 @@ my $limits = getlimits (); + + my $prog = 'unexpand'; + ++# comment out next line to disable multibyte tests ++my $mb_locale = $ENV{LOCALE_FR_UTF8}; ++! defined $mb_locale || $mb_locale eq 'none' ++ and $mb_locale = 'C'; + -+# Copyright (C) 2012-2015 Free Software Foundation, Inc. ++my $try = "Try \`$prog --help' for more information.\n"; ++my $inval = "$prog: invalid byte, character or field list\n$try"; + -+# This program is free software: you can redistribute it and/or modify -+# it under the terms of the GNU General Public License as published by -+# the Free Software Foundation, either version 3 of the License, or -+# (at your option) any later version. + my @Tests = + ( + ['a1', {IN=> ' 'x 1 ."y\n"}, {OUT=> ' 'x 1 ."y\n"}], +@@ -128,6 +136,37 @@ my @Tests = + ['ts2', '-t5,8', {IN=>"x\t \t y\n"}, {OUT=>"x\t\t y\n"}], + ); + ++if ($mb_locale ne 'C') ++ { ++ # Duplicate each test vector, appending "-mb" to the test name and ++ # inserting {ENV => "LC_ALL=$mb_locale"} in the copy, so that we ++ # provide coverage for the distro-added multi-byte code paths. ++ my @new; ++ foreach my $t (@Tests) ++ { ++ my @new_t = @$t; ++ my $test_name = shift @new_t; + -+# This program is distributed in the hope that it will be useful, -+# but WITHOUT ANY WARRANTY; without even the implied warranty of -+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -+# GNU General Public License for more details. ++ # Depending on whether unexpand is multi-byte-patched, ++ # it emits different diagnostics: ++ # non-MB: invalid byte or field list ++ # MB: invalid byte, character or field list ++ # Adjust the expected error output accordingly. ++ if (grep {ref $_ eq 'HASH' && exists $_->{ERR} && $_->{ERR} eq $inval} ++ (@new_t)) ++ { ++ my $sub = {ERR_SUBST => 's/, character//'}; ++ push @new_t, $sub; ++ push @$t, $sub; ++ } ++ next if ($test_name =~ 'b-1'); ++ push @new, ["$test_name-mb", @new_t, {ENV => "LC_ALL=$mb_locale"}]; ++ } ++ push @Tests, @new; ++ } + -+# You should have received a copy of the GNU General Public License -+# along with this program. If not, see . ++@Tests = triple_test \@Tests; + -+. "${srcdir=.}/tests/init.sh"; path_prepend_ ./src -+print_ver_ unexpand + my $save_temps = $ENV{DEBUG}; + my $verbose = $ENV{VERBOSE}; + +diff --git a/tests/misc/uniq.pl b/tests/misc/uniq.pl +index f028036..8eaf59a 100755 +--- a/tests/misc/uniq.pl ++++ b/tests/misc/uniq.pl +@@ -23,9 +23,17 @@ my $limits = getlimits (); + my $prog = 'uniq'; + my $try = "Try '$prog --help' for more information.\n"; + ++my $inval = "$prog: invalid byte, character or field list\n$try"; + -+export LC_ALL=en_US.UTF-8 + # Turn off localization of executable's output. + @ENV{qw(LANGUAGE LANG LC_ALL)} = ('C') x 3; + ++my $mb_locale; ++#Comment out next line to disable multibyte tests ++$mb_locale = $ENV{LOCALE_FR_UTF8}; ++! defined $mb_locale || $mb_locale eq 'none' ++ and $mb_locale = 'C'; + -+#input containing multibyte characters -+cat > in <<\EOF -+1234567812345678123456781 -+. . . . -+a b c d -+. . . . -+ä ö ü ß -+. . . . -+ äöü . öüä. ä xx -+EOF + # When possible, create a "-z"-testing variant of each test. + sub add_z_variants($) + { +@@ -262,6 +270,53 @@ foreach my $t (@Tests) + and push @$t, {ENV=>'_POSIX2_VERSION=199209'}; + } + ++if ($mb_locale ne 'C') ++ { ++ # Duplicate each test vector, appending "-mb" to the test name and ++ # inserting {ENV => "LC_ALL=$mb_locale"} in the copy, so that we ++ # provide coverage for the distro-added multi-byte code paths. ++ my @new; ++ foreach my $t (@Tests) ++ { ++ my @new_t = @$t; ++ my $test_name = shift @new_t; + -+cat > exp <<\EOF -+1234567812345678123456781 -+. . . . -+a b c d -+. . . . -+ä ö ü ß -+. . . . -+ äöü . öüä. ä xx -+EOF ++ # Depending on whether uniq is multi-byte-patched, ++ # it emits different diagnostics: ++ # non-MB: invalid byte or field list ++ # MB: invalid byte, character or field list ++ # Adjust the expected error output accordingly. ++ if (grep {ref $_ eq 'HASH' && exists $_->{ERR} && $_->{ERR} eq $inval} ++ (@new_t)) ++ { ++ my $sub = {ERR_SUBST => 's/, character//'}; ++ push @new_t, $sub; ++ push @$t, $sub; ++ } ++ # In test #145, replace the each ‘...’ by '...'. ++ if ($test_name =~ "145") ++ { ++ my $sub = { ERR_SUBST => "s/‘([^’]+)’/'\$1'/g"}; ++ push @new_t, $sub; ++ push @$t, $sub; ++ } ++ next if ( $test_name =~ "schar" ++ or $test_name =~ "^obs-plus" ++ or $test_name =~ "119"); ++ push @new, ["$test_name-mb", @new_t, {ENV => "LC_ALL=$mb_locale"}]; ++ } ++ push @Tests, @new; ++ } + -+unexpand -a < in > out || fail=1 -+compare exp out > /dev/null 2>&1 || fail=1 ++# Remember that triple_test creates from each test with exactly one "IN" ++# file two more tests (.p and .r suffix on name) corresponding to reading ++# input from a file and from a pipe. The pipe-reading test would fail ++# due to a race condition about 1 in 20 times. ++# Remove the IN_PIPE version of the "output-is-input" test above. ++# The others aren't susceptible because they have three inputs each. + ++@Tests = grep {$_->[0] ne 'output-is-input.p'} @Tests; + -+#multiple files as an input -+cat >> exp <<\EOF -+1234567812345678123456781 -+. . . . -+a b c d -+. . . . -+ä ö ü ß -+. . . . -+ äöü . öüä. ä xx -+EOF + @Tests = add_z_variants \@Tests; + @Tests = triple_test \@Tests; + +diff --git a/tests/pr/pr-tests.pl b/tests/pr/pr-tests.pl +index ec3980a..136657d 100755 +--- a/tests/pr/pr-tests.pl ++++ b/tests/pr/pr-tests.pl +@@ -24,6 +24,15 @@ use strict; + my $prog = 'pr'; + my $normalize_strerror = "s/': .*/'/"; + ++my $mb_locale; ++#Uncomment the following line to enable multibyte tests ++$mb_locale = $ENV{LOCALE_FR_UTF8}; ++! defined $mb_locale || $mb_locale eq 'none' ++ and $mb_locale = 'C'; + ++my $try = "Try \`$prog --help' for more information.\n"; ++my $inval = "$prog: invalid byte, character or field list\n$try"; + -+unexpand -a ./in ./in > out || fail=1 -+compare exp out > /dev/null 2>&1 || fail=1 + my @tv = ( + + # -b option is no longer an official option. But it's still working to +@@ -474,8 +483,48 @@ push @Tests, + {IN=>{2=>"a\n"}}, + {OUT=>"a\t\t\t\t \t\t\ta\n"} ]; + ++# Add _POSIX2_VERSION=199209 to the environment of each test ++# that uses an old-style option like +1. ++if ($mb_locale ne 'C') ++ { ++ # Duplicate each test vector, appending "-mb" to the test name and ++ # inserting {ENV => "LC_ALL=$mb_locale"} in the copy, so that we ++ # provide coverage for the distro-added multi-byte code paths. ++ my @new; ++ foreach my $t (@Tests) ++ { ++ my @new_t = @$t; ++ my $test_name = shift @new_t; + -+#test characters with a display width larger than 1 ++ # Depending on whether pr is multi-byte-patched, ++ # it emits different diagnostics: ++ # non-MB: invalid byte or field list ++ # MB: invalid byte, character or field list ++ # Adjust the expected error output accordingly. ++ if (grep {ref $_ eq 'HASH' && exists $_->{ERR} && $_->{ERR} eq $inval} ++ (@new_t)) ++ { ++ my $sub = {ERR_SUBST => 's/, character//'}; ++ push @new_t, $sub; ++ push @$t, $sub; ++ } ++ #temporarily skip some failing tests ++ next if ($test_name =~ "col-0" or $test_name =~ "col-inval" or $test_name =~ "asan1"); ++ push @new, ["$test_name-mb", @new_t, {ENV => "LC_ALL=$mb_locale"}]; ++ } ++ push @Tests, @new; ++ } + -+env printf '12345678 -+e |ascii(1) -+\u00E9 |composed(1) -+e\u0301 |decomposed(1) -+\u3000 |ideo-space(2) -+\uFF0D |full-hypen(2) -+' > in || framework_failure_ + @Tests = triple_test \@Tests; + ++# Remember that triple_test creates from each test with exactly one "IN" ++# file two more tests (.p and .r suffix on name) corresponding to reading ++# input from a file and from a pipe. The pipe-reading test would fail ++# due to a race condition about 1 in 20 times. ++# Remove the IN_PIPE version of the "output-is-input" test above. ++# The others aren't susceptible because they have three inputs each. ++@Tests = grep {$_->[0] ne 'output-is-input.p'} @Tests; + -+env printf '12345678 -+e\t|ascii(1) -+\u00E9\t|composed(1) -+e\u0301\t|decomposed(1) -+\u3000\t|ideo-space(2) -+\uFF0D\t|full-hypen(2) -+' > exp || framework_failure_ -+ -+unexpand -a < in > out || fail=1 -+compare exp out > /dev/null 2>&1 || fail=1 -+ -+#test input where a blank of width > 1 is not being substituted -+in="$(LC_ALL=en_US.UTF-8 printf ' \u3000 ö ü ß')" -+exp=' 　 ö ü ß' -+ -+unexpand -a < in > out || fail=1 -+compare exp out > /dev/null 2>&1 || fail=1 -+ -+#non-Unicode characters interspersed between Unicode ones -+env printf '12345678 -+ \xFF| -+\xFF | -+ \xFFä| -+ä\xFF | -+ ä\xFF| -+\xFF ä| -+äbcdef\xFF | -+' > in || framework_failure_ -+ -+env printf '12345678 -+\t\xFF| -+\xFF\t| -+\t\xFFä| -+ä\xFF\t| -+\tä\xFF| -+\xFF\tä| -+äbcdef\xFF\t| -+' > exp || framework_failure_ -+ -+unexpand -a < in > out || fail=1 -+compare exp out > /dev/null 2>&1 || fail=1 -+ -+#BOM header test 1 -+printf "\xEF\xBB\xBF" > in; cat <<\EOF >> in || framework_failure_ -+1234567812345678123456781 -+. . . . -+a b c d -+. . . . -+ä ö ü ß -+. . . . -+ äöü . öüä. ä xx -+EOF -+env printf ' äöü\t. öüä. \tä xx\n' >> in || framework_failure_ -+ -+printf "\xEF\xBB\xBF" > exp; cat <<\EOF >> exp || framework_failure_ -+1234567812345678123456781 -+. . . . -+a b c d -+. . . . -+ä ö ü ß -+. . . . -+ äöü . öüä. ä xx -+EOF -+ -+unexpand < in > out || fail=1 -+compare exp out > /dev/null 2>&1 || fail=1 -+ -+LANG=C unexpand < in > out || fail=1 -+compare exp out > /dev/null 2>&1 || fail=1 -+ -+LC_ALL=C unexpand < in > out || fail=1 -+compare exp out > /dev/null 2>&1 || fail=1 -+ -+ -+printf "\xEF\xBB\xBF" > exp; cat <<\EOF >> exp || framework_failure_ -+1234567812345678123456781 -+. . . . -+a b c d -+. . . . -+ä ö ü ß -+. . . . -+ äöü . öüä. ä xx -+1234567812345678123456781 -+. . . . -+a b c d -+. . . . -+ä ö ü ß -+. . . . -+ äöü . öüä. ä xx -+EOF -+ -+ -+unexpand in in > out || fail=1 -+compare exp out > /dev/null 2>&1 || fail=1 -+ -+LANG=C unexpand in in > out || fail=1 -+compare exp out > /dev/null 2>&1 || fail=1 -+ -+LC_ALL=C unexpand in in > out || fail=1 -+compare exp out > /dev/null 2>&1 || fail=1 + my $save_temps = $ENV{DEBUG}; + my $verbose = $ENV{VERBOSE}; + -- -2.52.0 +2.7.4 diff --git a/coreutils-keyring.gpg b/coreutils-keyring.gpg deleted file mode 100644 index 003a885..0000000 --- a/coreutils-keyring.gpg +++ /dev/null @@ -1,312 +0,0 @@ -Release GPG keyring of coreutils group. - ------BEGIN PGP PUBLIC KEY BLOCK----- - -mQINBE58fE4BEADGS6VzDkx2OOQMPQedsmBtRs3S5sz9tzO51EwkS779js3Sjt96 -KlQM0SbwtbUxOFor42LRXJKUU9T/Jl3v3+onASvoHAUcuAL15WAhnY9cuQeFOvZP -/iy0I1+bV0CILrz364T6vL614obnBBdTg8ZqSZM+csRlpGwXJiuY6mkrsPLXakxA -35n/nAgQOcQPj36CuuvpCH4JKPkzklwUMqueDzXkYMNSdWmVnI+ZSfDmeiwzAbFY -tE5uGW+c3DzD98RGCLt3FLr86n24IDlaTZSsaWbTJVsur9s4sbp6rST3pspDSQYF -ShhJ5aqqEYIvPp5kXj2CZJjOFBnIkn+0aDSps+XrnZjJn/f8f9lIAg0/0JjmytHY -yopo6HFZMdtOvklmnsIuJ/fdyk7761+necYHf5dopVuv29PSu62+A/gnKGfGaqtY -AjXFfsiLp/+iTQ+LNV4hWFbFKHHZOn4G194pWl6nY1gArwQKPZ5p6uy5EXgiNPRs -C1CcuVZNJp1RiayhTI68uuI+cldBU6N7+yZKGhjDUQKjIZ3eDB8X7vsCC9S1GgvX -Hcv8mjcMcHtnoC0w0FiW35JYtAu9mY4+uQhoRPTyPHh+ufX+OdKf7q5BKCppY1r7 -HF1VRFKjSybhEwMeGBdj1EEY413/A8ynpgpHLosPT36n8HtAWUGu+TadZQARAQAB -tCFQw6FkcmFpZyBCcmFkeSA8UEBkcmFpZ0JyYWR5LmNvbT6JAjsEEwECACUCGwMG -CwkIBwMCBhUIAgkKCwQWAgMBAh4BAheABQJOfIDXAhkBAAoJEN9v2XEwYDfZ4AEP -/jr6zmXUVhNiVCtqiHqc4jOs1OPC51iEcMUwpeaEEWHq17uMMIqz+nd8B7CAyjzw -FJIW4gtwPS3uTsXR2+KOl1VnMS5O/M9suyG5eM+fpCWkzyTC1He/1M9iaRMGY8u2 -wOjZoeY40QFN5fvL/BuC8GLBefI0rTzMaYO0WFlVWTpaemj4pL1Z4JoQdmR49H6O -qI155jfsXuv2VWjN1NoYT8w3FEugc7rdNWe4dmscU5H54JEQMuFd34X7Ja2S9YnQ -OdqO/nVQGm3te2X6ElOBoA68HyuXcEozf0KgKkcPrBEV/tjQrzn5Mc7jOgeCDDV3 -7MFwBZUi+z69jjOc85tNYf/FHRfUFnBLPC1HrOIlrraaqydPfvHBRTybTJVhXlQW -b9kqfrT1HU8UGfwP+5cwTy2WjZecxvozZakYBO4cdcmsSNE5jM8Tp7EU7ktxPXg1 -IQwZ8sEFJN6HRhRVmhK1FyR1hrwdcvfYrFmoYbyWUCW1RNuGw3RXdjXjGSl6VxzC -vrWXjeiMyLQQ7l7IneFaIPV22quPi/NVJbNeT5DqKa58kYgEVASfZVZkL7S3PJvj -fEqhw5jTi3l84AHtYNNo95UXWQQCWhpYjZ3q61satme++Eth552VAGP+JK4634mj -vVViYmWAnjs0efSN9yCOWKDKBONviW5WGZwi7MVtgF6uiQIcBBMBCAAGBQJOhIiA -AAoJEH/Z/MsAC+7uW3YP/RJlgRTkRa8t0t4oK06zg+jSMMQ3ZFsiipQEBMzJfCXy -C9pG+gU/mgcOoqnpxY6iA9ufY0dLOJYhMPsSLtrkjwMAIU54UY+WRpaTcXB+5Zma -1OoA/Oh6wcZHy61PEUkSfoiQ8vtXhzqQn7PAUbi7ds5ecn0hy8E6KKEEysFt+Say -zrINiCeO3wr6LUqUtpxdo8JGaHhdXGZsk5OMARnYlC/rzZxFKsie3+FKO7KNFoNr -edIElFKdx6b7r4CXqfK9XpZr8SaM+f3wh8mBCK4W2Re50/6inHAnTYwW0octwr0b -AtlHOY2myauBdj+19IWntZnhoKxuhVPLaEoG8j26k+LIP6h8fB8GoRh5oUarLiCk -fahDRNY/bPFtBnsE6Co5OTTy41CFkGX0JbguTpL0uPQxygIKz7x29P509fMpq5t8 -z0hcVYJ5/cXiNjFLid2JsWugKAWe5k53E7qQKR+jLSvPtZ2oOHAMUzu6hOnwDY4Q -5r+j6t81tFAlS6P4fcpVU+alUwvVNdXc6MSkfmK9ahumjYnLKy0uo242U1wuBZgN -adr6pFxKrMiC/0PVJz+ZQOZU4OUt/t4E9KpyUEasfOl1z4r+q+6dZffRbrP5CCIQ -M6A01GRbEufrcXjgYnmaDncV8JnmLbHOoZ8WF+xczywFg45ULSt0N5ZiS/BbatuO -iEYEExECAAYFAlE+Z4wACgkQFg9ft4s9SAbl2ACgqTFvvpXJzTpZrKrisKY2i9RR -dNsAoJt0xI/urG+JIn5kUJobcPsZtY62iQIcBBMBAgAGBQJRY/sdAAoJEGiHnitP -7eG+NrgP/iKO5+3ytwRYwwbtQNROUQSbLwpUN2N3S3XH9lRV2NGEx5nx9Yn0l22w -gRMbULeFk4S1Ak3mR16D/mlnfPMyVqrJotp/E4rkK5OzjIsy58vL6B4PLgut2Xlw -Clg+XklxkQDfT4m/QtLGJYOnx/AjBn6ABu2zD4FWsn7hIMdXDq9bpr9IT96iqd1P -MYogglYK5OBFIGefjf+Sut7i3vuwJcjuNdL79tIbn5yple96EOL6eoHSRv5ndcoS -mxHtmin0lVeQ1ajoBETNh/E6/yItZNtc7BKbttYF/tN0GYpQB+dcCeiXLAOJ9n/2 -ET0gAWMP+kHzdcl3mdfw+KiZOK0gak0cu6LW+3GGAXXZkFYh7I70Y7K0wYNNgy3W -rzlISbAeC0zKOzpJJ2eQWJAs13FyinUPEyKRorRSubajbhJzHa4t3SwevR9DExnG -DL23UWreDO4ElDbvT3MzMA7ifaVSLFR/Rxu/6xsK1lPs2NygmGpdDnPjJung9CTa -1yyadMi3Cfgggu5IuUaKtzW61lbD8sMXqLRoFIIkZjcQagS/ybGeKIAedCE6pqMh -MViSIVi2G/F1wVaahfhjvaj87yYPxUuHq4hHwx2RE1EjP66a1IoR4key/eZDWfHi -th2VeeaVnma12NTl1GzfaEig3mzgtB7lM50/qJ7ml+MeE5agga7oiQIcBBABCgAG -BQJVfds0AAoJECFMgsI2H9co0u0QAK+EhLBUwJJ1XfuheL8pXSJ7FxaicSPk9dKL -Fbhc2oIAItqGSAsBRncYH8jYAPSwtCq1whATbyPgoEDm5G8KQEdAZ4bA6mhXw7Nj -UgHtkbnm1bIbavM/lhZLdNi+H0ZV4w5G4e47/zMLbwK84ZhiArRdklq58200CmPB -qNnfaxRxxkJBA6Bn6Cnv0FUSRvHUlSXgOw+pZXGNFZpzi50d38L1na7iCxrfyxH1 -sS3Nhn3zTf8BCKDcCVyP8UeBP9Fb2+fYJ/f3/KN0C12Hnbqc9WsYvFSYWq6u+I8/ -GP0oQohe3Fv5S1VEHWB1feCNmvVtV50J7hukTBojERhC07Z+2T5G2aw1Cc4zxkOA -uwOBTzuij53sErn5o+hca/pJTlXz8jJ8OxSFY8FT1QGVlLAN10yl9mDsdnZb2VSn -bQdqYG/qfbdC2cm2rCrhcKFpierXURLr14UC1O7tbDmLFYXDxvQfjtj9GSo4NHrK -wlTvfHi+3x6fGyx+Auulcjt65A5kQ3mycOc5paZhTdIKhS5mFdQoKw1Sg6RPiRIy -OH5fx8ob6I7gN+bcuX3r8KCH1FcIiyZd3WsVQlI1EfnpujkFo1O2xDVpm/D9IjAO -MRqKTtnHbCjaPUrsRRysAharr47YuzQUYGaMmIJfSN3kP4U2OukYmVFSUpxrlNep -M7LGQsmdiQIzBBMBCAAdFiEEP0srMOiHNvRbOiwMIhOnPE4lafEFAlimlJcACgkQ -IhOnPE4lafEZRA//bTNw4mi9B04yacqaFlJ5f3i9v0fWnsSXNEkW5wslhbjoD8Ab -PpVNrrw7Jm0YTFNa5TtnNc7fcnHNNfsL0LbtCfdrZOTm7vZstFJrASHam7La5655 -RgvTbozSWuuYrfLyKAituRmhJyv3ntpP9K6yUAAuJjxR3ny1sn8KNIFX7g49emlC -k5eIEujcmaVJp1l8wbnf/jioKr6QeXz0cxWUUFXolR0AUt8Vy11V+qRQb9Iw902y -2gmbMun4HjYEtCtm+eY9TRD5jY7hCHTYTFEfWITnGIRDyHyLIS9a0xql3W3EyWO0 -JkHNIm1ajqbuUp4IIxoZmIxNmEmW3aS8rsuIY1P0zXjj6j+GcRX8lZJOIhieBfWk -ku6dZwwtv9wF3K1UQzFwRsERqiwj4CAwlUy4um1eLOcjx6ge7Ub67FQCihx0VEpO -jnngjQN4clth8YM5nst/+lOFsZb/k2SHqTnpdE9pFl39aij5Y7nAI6xZL8xyM+CG -9tcFMXBrmyZAmD83v0N2PyjWuLAyY2b2SSYhOXIVfonHYSfuGw27yvn6mQ0jICZu -vdeZQASYgBAohMDXOgDgKdL6g143d96tQST72RflXAxoKTbblKK0kqxZnfdRIja6 -MTLstYJUNwWQsSD7bwcY8wnTXPK5TpPtBH4q0sjkjd1ZNNAQvbbowTlZ/smJAjME -EwEKAB0WIQQSG9otSstjYWs2eg5Y4Rux5BTZrQUCWKj25AAKCRBY4Rux5BTZrY/1 -EACrnMsYUnN9sc9qhy67pAMPy5QaGsYY5IMOnQlTcjXYrBRBx0kEWhiMrX4USqRK -Yj51J5U/6MIyeFbmDMaGrUQ/Ba9GxxjOnYAUri5S7lvtuYZGYsQqQc4ORgNCSRAQ -GMiB5Q+3oWbkaoads3ezhcE+R7/0HrqgxgCRg5mzTx6up5vrkBN8kbI6BIpgoPBy -AzTOul/EIkJuBYHg4IPt9dWOmbFbJyxMJg7kNwTS65GypIEiMeQXK4VzcdB6jr2L -Ju77Ia+pWyzKpq323swdRZtM/hHrGJrwJDbdKMfWxoWf9e8cqvO5hIM2mzchHCQ8 -7OQSnb3JIsHQIPHCxeaxzMOS1smRNbYu4/yY/MRcWaNiScuoMJqI0gVWd+XIScwE -PSGyKlncV8moki4pFNkseaLw3MEQDoxqf9TtxXnEB7ZduvR/UcELUB85lVjNnoiy -GjrcagTZ4jDISxADvqBP+a02GsY28dLOk6smqPPwezbVWqV+ABPeQ+bgPd313MGl -a22s72O4/nXzzt0rNgmgEIqMy0OkgmxAUBCSfcQp88HEnk/roHsUV4iYwAks2cOp -CDriBnwjIywK+hVq0r8nuBNRQt0P/Yp75ZITffPRrOLVXvA8D7tV+kIm1GrjDbIZ -OFCNpAUCHXv/cXPeUvHsSd9hmyjGbNN3UzpxhykiDF9GWYkBHAQTAQgABgUCWKaB -cAAKCRCZRTN/KIg+y2JSCADAOSj2N7T8PriPsuGbRWehb2zvfjQ1C/IiDIWf6s7F -QuEjfg4NuWUJ2rPl2bYFey2yzSx7Ld0yNNdzSRxng6QADHUHYAneQi2WuGlyA06P -DDfFERlWRv6JZgnL9R7rWHB+RAa6DnPPgpxifABv9RR4caU+8uAP24KHRxCQXPx7 -LfB8hi+G8G3UYbuLnO5FTTuCObjjSh50h8qEt0f5y65R4kDDA40/L26POJNsHc5u -EE9rZlh0c4AqmakRSmH83+Q6XRWOtn/zPggj85ir0gsxLAezZG/OtuAyXW+rOC0L -RJDJ3JaiScUC3xewY5L/7jgg9aTcvuwxKoLBIaHOYtJQiQIzBBMBCgAdFiEE+ymK -u+HQChyPpNwfqLUfXoAyzOQFAlimepMACgkQqLUfXoAyzOQokg/8CqbMll42B+nG -VDdSNFCNjhjhKYctR/aZa2th7iDRwsTFuqSVHbywRL0XrkI0YOOJU57V56fBY7Uh -kfOKc6oeL7EXxpox8ehMToWMOcLSvi37EGMmlGLXokM9bN1gxfdFIrZr1Ji1kBYX -hvSj2Fxxi2NGRp0uy+IIOa0vB29u2xHi6GWk8U8MBMn0UcP6H053Kk6tMsMDEhF2 -rSYGpvKFSWywuFuELosSS6jG73+6pg9fMWBTDYQyWFH8YRA9AlpxWxT29gcKaftM -SBIz86Svh3PZ7qOEDVxh+yWAQTVUTVuGzUSleDDuJt75QLSt+ZERS9iezodB6EOb -AZr6canAJGmDwmjPTLwS0E3U197QW6encv3qUSA0Sb/QyAzr5007d2PzkIk6wJq3 -SxBdBRqCjAyR0VxZr2kE1Yr3t5rI3MOFsVWIKIpmkLzmCSPuUGFTvOZHlYVaTOKI -x5ge8d9smXdHjpSF0iGl45e0u1UMDsiU9dpo++ygdZWAnMI12Md5MO+K2uB4gLk8 -Njln1duZ0MQP9M6swkiIwH6jig3BkRCIAIWCNhbScBWJ79+HKD4Swk92+vTKDFRV -lrD8TQlQSbS69Lbon4/v+NwgcpHRTigY5TZZ6s4DXBUl8OIkXDs2LHeboTvm2Zu8 -gX+uWujFHr0nJmvwI1P/ih3kYoEFqLuJAjMEEwEKAB0WIQS7Pk4P+lsqogxkAaHa -lBBIg4QoJgUCWKeHnwAKCRDalBBIg4QoJv+gD/9AygNKRsaxJ19u0wyLifpGOsi2 -a6mlmwZkLLYhomeC82iV4+7EeI++QFhLc+KlRNZtkQld9rmihbcJo9UOfqTwwG/W -bzSF/Ed0GSFzPtS6HDjVPTn7qiKQoeat/e6g+VmYoK765wLknj75Tq0jPltX0/Yl -78s0ZwMI+HhirTBreOS6AVPlS8wFD4ywe64PN/YjgePAEfiIEiYICXmwGUHjPBgK -a50z9VuVs3TRLo+b00N73YDEW8tlpouhETQuL8hAYhjGgivHss0DRnuB5fNe6FgN -vwretguK3uknup1vrvVvDXOUOIdI1UksplrJvDbjYrFJB+L4VSbyGk7Kl6oSGKiz -YRF7gM4I+hpXlVWSKVxEdUlA9F6KPm3iqM5ld6K3Q6rDuppO/2BaqlBhinR+Z3bJ -TLtM1uKh0IgyGUstEkML/kjF9wJcCC+z7ZmW0k2CdA9JyMiHDQdVblxZpUI//Yge -gA4P32X1OofAFX2oXua88qehbEY2uYk3OFsR3bJwbTn40bJkxE8072IpBozYzskg -14Q/xnUxXkIL1wqLU1GPi9l+kbuh2+8yAdlz799x7De/uZhk8IwOOC5H+2oLp+vd -iRXDLKU1sDBiVFRJb9kosvUj7S/a15My1eqOSVP5Fa0GbXNw7ndvcpybMoFqbVSC -lzjlN2OgZuXYEl2PU4kCMwQQAQgAHRYhBH/Z1lK/X9LsXxORsHmPHjXLTTipBQJY -qndrAAoJEHmPHjXLTTip6uUP/j3RieBfyGnau1a4KClaXlPGHxlu9M1fFw+aRqV7 -r8ALWuQzsKlh8QlPEWhtqkty0BFXAhzRMYJd3G/5j9kaoS9NAeNpJpbZd9Gz25ZN -k+3PCkww4XthvKNY/ONwnwGuelLpIbwa25+f7Oct55tthkyM2TWXlwkRVNpeMNhk -uUkP4+gFnpvtzUTFqwYtaEtNY3UFw1CjmcA5xTGL6pIg2FKf6m1YyJJkDLpU2/pB -Ca8Mk/A9wQZ/9+M/l8goNq05vsQsp8nlh9zo1XpwWYBq3OwPQKDt4d6rAwU+zMHC -XI5MP5B2g2Pj+M5bQMNOxa4sLw71ALaCYETeHHi24Kp/ZhOWsUomwc+v7t5gApAk -6gjxbGklMWhdJuk2I+lv796J4cFI4VZpTXAygMSnnlo+GoMiqTz0C9eElZlp8z/Z -yy9g88Z8fBoAY1SmrroaxLOvlFKRG92xhd+JUh0kj72loB+Fozg5HV1OqkF6c2us -w3XCoIcht87TxmZWPTXqXdPXrStS74g59vrVyGvsNN2hG/l4dPGZSEV63Kn2eiti -Of3JPYJcy0iQpBBnhhKQwPVNgWso7NxsNsVYOUZCDeSoCFEvrdUFSr6q26IBBLcw -itnF/KEX3MyJLGr1BjDF9KqdP3+YL5Eqrq1Zn7LtyAbC2Odo4KY6vOT3SRrSkBRH -RRq4iQIcBBMBAgAGBQJYrLQ2AAoJEPaR/VwUa1eD1d8P/1qcubzbb/p4jpnrZsXW -i6+CAeJuA2f2qyBJtdVPhiz2swSHMNIlhVWh20w4892yv7Mgafj6i3Zoben088Bd -BTvCUOXRtkepCSTLTg1fTa/l3a2vNxLyK3LT6Xf8KuY5lXTH+XWn7vG/N4T6jyd2 -MQLP9VUltRkk7aNarIZvoYMd6/JVqKVhvxg42UZmcjke3PFKiHMIHBVSGBu3W1Mx -TDNgVZqTJlsqvfShwoBjPPYLBpSVZKHKgjirsDkZTS+ufpVmt2rzlujeVyC6y5f4 -subOde/pxGnTT+sMJENe/3uJxjUIy07xyXKBRnhpPxXbpTafZCcVc688er0CLRW2 -JsL9aEmEM0FV6HlnvW4ivoW1v9mSevAxe+KvgCO2cU2+HFqN/tCtxnr8rZ2HIpf8 -00cTpdvIn7wibGP9jfwMisD2Mugx28eLrZ+1sNaRLwVmroedjo9NJr2BiyPozOEN -lGX8V/RxQLaQfiHwyuKVpxA8rlx5evvtDE2d31ekVtdLXtN+GmCymnPhu1KbD5Mq -+Xk+yj1t8tdMD+SiFclz1uVeAOGpX5u7GMIsy4W8yoB5JlrwrsFot6UBaVZjAVHB -XTdMvBGsfxmimO7d0p2tBFJ1QV2lAafVhVIklCT8zXk4McqqtWxXIKWEB9dfIpbD -/A5MPtu7X91BTISC7SmRdBjViQGcBBMBCAAGBQJYrnXHAAoJEBzIdvEMrJ+JDgAM -AJyHN3j+g47bSERRxLevoRybp8/BoRfK/OjcLRxhOru4prOAiJEfNo77IbG9Quz3 -aBn7vRDh44BxXIR/NjI6kM3hsN40BBDVwfeFEFGKciV3cjCBqlqnhwt4MV6iDoGQ -1CkTm4LZQvtjQN26PAXUxxl/GO39vze3a8z3QP9BatZ+KrLOp2u7pOkwHNkY3Anb -/H0AUq0fH2Dq5omDJB8R54jlHc3/ZrLvujCVAmEuTPxK6LGl5xg4TaBtYeUgIki8 -A9iwrcFgh9OjgAuG1PFs+6RroE+nVPm/ZPDJ5l45ZHR4qQB52qp2lxf745PlSHj7 -23d6ASx/I8mDZ7bPqk2aCKXGQqkZ31b+I+Ut2ru2nEW6JAna26kgBMhNrINqLNxO -qPXjZHqZHG1amvlTAwGpAgeW5WBPvNjFn1WNPB1+9vCPTSwkWLR8dnzy46Rsfohk -RAGFtQjdccBxaikRHuUlIUI32M0WjKCP/sy5nVLQKrX9xqkOj+mSblmbS+u8cmIH -0rQiUMOhZHJhaWcgQnJhZHkgPHBicmFkeUByZWRoYXQuY29tPokCHwQwAQIACQUC -Vp+cpgIdIAAKCRDfb9lxMGA32UftD/9jYqsCfNAzb0vhDOaU1AchzaQa1pIKEjoL -6d4AMeXFSBpMi4nYJpN+rmM8DAzcbenBcSoIqecdfENp3mY+hI8mYdnMiVpldsro -EAl/SDxY6//pPd0Dnmoe6sNodBB2uwHxhQi9ubz72iWX5WiKP8+OUAj91cLMl9nK -IYfcHy3iinSRqT02JP33DGDwsHCoAMmp59g6AHnf0sjCtZEtK79MtKiKTkUdMazP -VGs81x1jCO2kvvmy0fDZxkGuyso0inae2hsaMSqqoga5lC0jQanFIXSEkLZgJglj -LmiWPO0IGHPFth/e//51atGUmpdd2ufQ/QVoxSnQKRFQ98eO/SQ75bO5vbE8dGv8 -oX9S3M9NjKOY9VnXC/JDyMXt2aMDs9tqSo22lJuT2Wq20wM5hlszxKI7c9QphnuH -yPNtszzXo1+5/UEuCNIQoe59MoscGmx7GN5WvlENaixGg7tzpZ+wdftN7BUcpJfC -gsEQurHzPEIomlszp3xraX0G84plClas13Ie8CIVM7UPbF0Cwx6XwhryjaDTkq3f -+mjQXtNJQk487q8cc9dxplslXiDqBYVngV+oDKOjrqc5PXSQm2M8EYRn9SXuFnT8 -iF5SkuFYtgOEj3KNZ04ZB1I7AQebylS2LGwWan8yWJSAs22eR9urWBVpmre6GUGZ -fo7YBdOvbLQiUMOhZHJhaWcgQnJhZHkgPHBpeGVsYmVhdEBnbnUub3JnPokCOAQT -AQIAIgUCTnyAtAIbAwYLCQgHAwIGFQgCCQoLBBYCAwECHgECF4AACgkQ32/ZcTBg -N9moHg/+Mjq/O1RnNg7kdUjRK1wOflym7itgE8kq4G55EJvLSxo6wIgd7ZKUj+cv -X+iXQpGRc3bicpNTsKcW6EjDtyg+VCSWD7qJ3EtwxVf9mN3bIqWSVwP0k8kc1N+t -p+L+/9jve+h7Hf7rXZoNo+l9h0/AIIr9YyM2r1VtiAsMNCfD/Ssvc5Yx4fZHR+2V -kOLeVb2lqdYVe7ZrXDt8qkdBHMCtxm+9jaY3pZVDFKk19NeI74Vzr9+mYn0I0OZS -0capUuG7+a+FGI1Dx2jn8uL+x4eLDdI3vvr/vGWparikBExGq1pAKWm5gBF10CDP -4nx9+5hzjPipvCuQerRnjL3FQyXa6E/GpCp4Mk7SdB4zML1CmnYUzz4n0TcV5aFi -yaMQPk5TByxzYXWUqjFJzFwmU0z8Oy/d64ZMGLyAxCly6gBc/AmXzsUhg2hJB3nG -3JRw2WmpOIeOdYn6S3onfAFT1tGo7kWNIWYxX5fT7qAHVlnAmgjz+zvfB8Hwq/B0 -FDZPzgwYX4LeRMCj9VDspoCVnaMd4rWqbH2lKUU/k0SFRt3iAqjvT6WPbJIDtEF0 -ifU2R79laaZZU5rbYWZC52AfO7NdLP+7uwxtPYyOdP/4s0HS0e8WDuykdZbTaC9K -HbIiKMW9YXQJRo6YupWJWOpFpPkvx9ttcQQ7C5s8YFjVR/96dYKJAhwEEwEIAAYF -Ak6EiIAACgkQf9n8ywAL7u5OexAAjb2+LR8Pa80t3ooladI3Q1icII1hvfb6C+KB -lzm0d8nMNqcjpPdkbppcVmjqbF3xw70uMnT9m1Y5NGMzJEZiNv6VWT3/m+VJ/aih -ci/lccUA46pL6Edxw2F/l6ftEOcPRAefNvszCQPNSVHPoQ1m+HuweVgYs9by6s6E -FFCG8BbqxpAxGxaByoq7ZtlxfMAjKHSPCJSJQTntj5dz79+K+eI8i6bMP8isPBvu -HAT8ZJ8mn2kQTEpuMIyCl6GTEigKimwq21tebB666Kv7wwS/nwCzceqyPshlrXQp -YYWoKfLd4SrC1z99//H93/IkN9dZfDJaWvMOlgO+/Tjnlr0tnVsIafnYaOA7Pb4c -QAx/tbPeiSH3QyRQw4wD2T1CaoLGPLaS4aOCjJXbNBPk+44suO0gUkU8duBwyK0p -fjFAJQJnJnVEsqWDh0KustQW3jdPvlqEe1eWmhnivXnmtvBk4U2BPkOG/NC9+r3n -xIWnrRhINWZLT85wM47WtQ2l5BRK71UKrgZDixOIbAR4H54FLa+vrwub6JjpOrQM -MYaxA/aiEv5byP38nWVvWRSaFDC+QpPyOHLnzBSoxaAHvm8bsNR+4KALcL4zyrUm -+qqQWbaOpikgBDhYyI/qteW6REZunofpkrpXZbyE+oFUxn7Vwz1ivEkiYPrEhTWr -HomA9eSIRgQTEQIABgUCUT5njAAKCRAWD1+3iz1IBic4AJ0VP0N+M3OHLK84zhnb -r7NV/OsepwCghhSEutr+LFoP8SIDFZGyGwWNZkWJAhwEEwECAAYFAlFj+x0ACgkQ -aIeeK0/t4b4XvBAAnQEaY8PFnZgegqdsNakq1gLr433h3WwQBGzba9CHhElS7VdF -c3+VnZ031zRXFFMWSFOovvQpyuRNsuGvgmvlr93+/OgP2jBZbgPFZy0B0KaTpvuE -3LD2XyPINajejIVJTMwNIuD2TTxz+zqRcdie4ExOdSmWHmjGNVCt2W7Xf2ZX18ex -FdH8jOVKtI2Hdm0YdfgNrfbcSLVgGr5MJMvaifsgGyQkPS/iDXVvLZxmSJiloupJ -ZfXrCcw5mzd8qodWwC7VJbZWIYkUBo5ir+tFAr1GuxD8D1l2U1RA3jRIgsmjd2CD -S6eKOmBXR3UVxFypOkHqfsHlST2vzTpvWGhzeQXAbo5ahjtI6m2c5mn6Tvb0V6BA -o6Fjw1id/iOWmfUSyI9byZkC7HJD/68jgvFha5eXixSo7v66MFptGl9B3sWG0gf5 -iSMbIj4EFzuBySv905kmXJ9VXnawQWalNC7n5JvJkIwAMC6bNU7aO84+9K7kh7bo -rGaBkiYfD5W09BgTipJAEgq5cVKLOGKaN47DhSszu3QAXl6Wk/VL/RTJfzWWGU13 -nZ7UY7f2uavA30mHOznAn+2v1GnMwq6ZhCQt2Y37YgDowBSR2PrFFZJOzZJhd6GN -5XWjq6A4QKZouIK19zRAA2Zsvi1TNDzUw01qGT4i+hsxPKXgEbp883D/ZMuJAhwE -EAEKAAYFAlV92zQACgkQIUyCwjYf1yjk7w/9FyPk/VEJsUYvG6Oap8Qh+bwCQRRm -vApZKVurqkMAXntN93GbCudWyPdt5igZDQf7CAHobvkUrn4fIGSMAUu6jmy7qoFf -AnGNKDMWLVYIUi5T/Sb5WCoV6DGpRJ46MjEkbplbnvQyemVsUVQLkB5GrnkO1WRV -UCk3vnsgSqrJ7B9HyLHAjsbEgm3L2OWe+1Nz9+Evg7etyVHyLLN5N2pMK3/ZMHKf -42p9SEh5x6JL9YlcxW6EseOnoy64MHDvVvnXnuUWxuTEsEpytvRXlJ7SkG+2lLcn -nbPNPY3zWfjOEI7j8RvXQJGU3FKt7NZNMGe/jdjq5nF1R5QtilRBnpVFboVmkWNN -/eiOT1Xy9/PEZKe0GUHLLh2t2ffI2du4FPKKmZ3i8sl2VTh70okKEO8zxTohs+7h -1bff6XgIFCqzpzVoiIMHLloN/Qxr27lywFzu42UISXIJBW34nSzJ1SUkaVdAdkE8 -TtXEk0xHmMkATTWTwDHoWmp0E2QcVugFpUlw85Dj1FkFNf6IvwaMw0wpE9aP9IKM -oPmKuarKzC1PUiGqm8o66Oh8I/ycQVv70VgvazeUY1f0GXV/49nT9tyj988/XKAT -T3OkyHmZIE0Q3HLlSK98bN1ddlVn35IABE7LSa9aJWN2QTOU0pw6D1Gc7pRD9smT -HAxs4LUP8TXOkGWJAjMEEwEIAB0WIQQ/Sysw6Ic29Fs6LAwiE6c8TiVp8QUCWKaU -mQAKCRAiE6c8TiVp8bY1D/476x3jkMpbkhg5wd6YlVH33kvxocqaMEdt9jIMj/Xr -xJbMZKQgHBAESf6XiIYqLRZOsIcdi8k/0goaqP+HENnUj/lK/vBii8P7Qtcct8F5 -55UEDC6GWCFaqEZn0l2qgbHjGWcwh8toq+NZ5VniPyhQErm3b7dToauqE7sOoibm -/RpkfwxNmtySd/nmmyanP4Q27AgZ7Csq8h68P1wiVTAnOuBJ28CW1z5XsJ5YRTq8 -ae/6kJs7g9eRoDtMkUr516EYmui10khYFUaZ87KjpsTXpgfiHUTtbbW73yGBdqwM -QD/s10UUCrlv7j8gJ1V8Z5NRSda8kwDyeorziwD+sfGIuxK86Q7NA5tjP9QY5tJA -m+yxgiwcv56XL12p7G82L3WRDujVm4pDs5NGFRGQNsmkb1T9DEFQMOnsBgVWH4sl -sjPsN51YNs/wHmu0jOv3CBbVDJAELxjqIroTZuT1yjG1xV5maPTqppMr3+gT16eR -SZ2nBy4ev8rYM4N4K9EEBjcbXWLNQOYeger1vz5S0bEzUuGeHY4ahMNB+dRTf1eN -UjRlhSzddlWpoNMbb1+PlYwtcTEIfh2vs2iJEbhZhuETVFye2RmBQ+MQ6oXDGEcg -fwbwrbyW0X+Z9KPIDhTRMrdXJiBui5RvQ1AxAuQ3sZglw/xySSvztqf5InRrt1lE -N4kCMwQTAQoAHRYhBBIb2i1Ky2NhazZ6DljhG7HkFNmtBQJYqPbkAAoJEFjhG7Hk -FNmt9cgP/2owqkabfUTz1Gf5BZn8cWlcZT0ePFJuXqceoyZfPj4VhwKSyRiXEu2U -LGi75TSUp1ESok6w3KG7chq2GoH2EITysqUhpcroOTAT5qjWGwf1WEP/zEYrmQb2 -ayFRBHdcoNNgRnJVd2HB7FlHF6fg5aPrM3P7o8ajYDneYSDwubajN2xnUsFV8yYv -liNV8DtFOsX2AHSf9ipsF3P5ArsjRJoMI6Z/PgZuECRiya8qzbxZoIZGgT+khcvC -SwxsX6YXmNImwhGpugUnSrjvBPdiyN4CilTOdaiEqPLwFDpFWEkByx3ewfJYfBfH -EOzTdSgtPEXJB0Xxb7ge5fpBtFunI45bPRwRIT6EM29WcYWwCp12HCTt4N2LHwu6 -h+JL3ikFucgtJsSO68h/oId7THD24ft4UpIfBR0zZ0/i+ier3SAB/gN0xE4Hpy4Q -YNcl9rkt/ApuHX2hQqcN8woUhGV9HV4n07Z6FIqs3qSj+o4w2hV5xaEqOiVoKdMC -p7DCECjR5ACmhvtLTI8ddS/2rXPK/8Kttg4e74LysK5WOSbCiX7M+GjNIuVh7aA8 -BVR7hLjQ1CRAu/c0/m6EsTEViHuNZjX4deJo/c70kWLbP5UVN9yXrJjwVXwBOz1O -XiZzzJCl+ICT2fu8K2P7nL4yqkSAuMZHz1sQxzIvLs8hjbMYbdqMiQIzBBMBCgAd -FiEE+ymKu+HQChyPpNwfqLUfXoAyzOQFAlimepUACgkQqLUfXoAyzOTxgA/7BZpk -HIlTGVobZ3drVXXLRVdydLbypAJ2d6KU37hY1xuCM1bQ36H/hQKtHBgdTVc87IvB -0iZTKfwxPHBloK4MBDl0zj0Nz+Y6OK5oaUccDCSIDyBuMnkwu+U3O94mWoftQZuB -FH9urpElmgLftknKE1PMsPT2PVEpLVKX82yXo28+aAhXjcO7W/FYQhHX5vqPZmjC -uJGS2DZJHUjF0Vem1Eh2a200+t47JfFSMRSFBm2S4Z1Bo5UUjSk83yJ66tCynJ3x -D5vhMTWJXRLhZd7DXjjPBMrp6MqDElL8tNF1w86Bk4kIwX/hMre6c0/+4b5rJBwF -mjRkAwfk/YLJPz5dfoMiTf0kqj93F86BKDnYZNQ4L0Yn1QWWKJSEf55ldmxvaxwz -ZE3NpOALdBwkG7yjhttsHIe5kpWgluxcuYTvWpX7KGIZnt1qSl+Cv0VAOY7eo2Zl -KTtpqAQeFqtCZL0tcLxj2Ce8LqViuraKGxNKi13FtpS74W4DgseXv8tnhdy0uqlP -gRQ2WCHpUhXUlU/KaXtXXmS6oRFnCzXikYXzH5ZFTHzNthg1gO5Fk3y2B+5eL71V -SHDeIpi0jTpiO3Mav6AtVlw6QRXXfn61cdF1M37k1XA5lFPb+ifkV0sF/rkFE5NL -r0NAFqcwB3CE7K0fgOidFhdnH9zi+qcfCiyMjpGJAjMEEwEKAB0WIQS7Pk4P+lsq -ogxkAaHalBBIg4QoJgUCWKeHoQAKCRDalBBIg4QoJnzFD/wNhcOyJURvQtQXcys2 -bSw93rubuZO2OUpIgs6CCcZgCKt4sES9Xv9Qt2qRdk6GBgnlsTdTfwDWEDla1NYf -+/894Kf+3dLhaiTmYkWVh3UhysE6rihKZ5SHeriNCFCyaOvflOfpGQn20TX17I04 -fLBPQ2tZLIZYELpHHr5OXDm0YiBR+1Rc7mj80cTdw2+1vNa1p2r72n2GzKz76Yl4 -BI7dWud4GAEW26yrwF3VtdGFacRcDsjSM5rR5pxREY2WGzONCCD1yuaJUqk8Q+QK -8g/2PybkAUJpzmNzWqgsn8FhPESfObl2FPuIbxIjR+N531QGeU4HcH62zJeJjCdR -XxJk+k1VRP7SIIg310q6J4WiHa6LU79BVTFEV/0gyHh8psLpySr6nJN9TAImdekd -2+BN6xdWcub6/JJTdJgg+g/VuD+2vUm9zPtcP7nnpadqen6k2pobiDfuGepa4k7s -1jdgSoyKdgntJNBEpBCCx/fQQeXlR6kcscjUP1aMa+XIgpeZhyKTWWcGfbzsf58u -YYOP2nMn1GvPvoKcW4AfbPui14eNh3m3hQ3numJKBZGLzBASJsdc10CkOJzLq448 -nhdJTp8ZLRGYl9mEbpVuHNrYQnxYe67OtGS9Nv/DlAJXR4fUiX/Yq/Z+w5zz7HMK -Pbu/XhRIZcfJWgmRkgr3DSwGdIkCMwQQAQgAHRYhBH/Z1lK/X9LsXxORsHmPHjXL -TTipBQJYqndrAAoJEHmPHjXLTTip1XkP/R2nPYovKt4/ytjOMRDfO7XSzpUn2d0g -hmKRn0MHe21n3IjBzDG+BapdTMQCOc4Ucs1UicPV2lxRD8TQa8hh1MYCp9gkZ0Hx -I9R4q8StipyPLq7B5TQJ4tsHqT4Vc8reuxRInV/2XZ6gdr412v9dsK08o9lYri59 -mv5YJaxZmdov5555oK0ieMAbIRXiSqSsONcA4ph/MPXpVRXZvmu8+IhKJZbAd0cw -iOhjTU8z0qCBcU4vYB0nxwp2AWbQG4QDpk5lTp40Tn7A1dL1XUbiXsK9h2jAF5zn -Rssb+drhNeafoqYfVRsB2ObZPhfqD9nq6isbj1ocDU1nQLOrFdYu1o9+JgMFs6F5 -NvavG7RY4RdVLlXDQuoMiX5e1PyGsBgLliy4Tz5gogKtqzm40nV0573pcEkUR7Um -rNEzEuPoC8PFr7W8lYEHx70Yhql6IZ7rGXKDQNBWLp4drmPKajhdH8xPOKn1Tocr -qdL8hkzWh6wqLcSzwhgR14/bjCZDj3AJr6bRdAAbcE1xqWt86XiJRM0upe0j/Q7E -s2eUybhUb/YPSe++llkUsePqvLGMSY2nUN1lwHST1/yI1gjJ0qTSdrHUDd/V1KkU -SY8CL122N61FoSAEy2Tk4hVNrSNEM0DUuYXaEFZFazJT5/QwfmCaE7lBFzHFRQFs -mbsPONxL6qjFiQIcBBMBAgAGBQJYrLRFAAoJEPaR/VwUa1eDkbsQAKFy6zUg6GQz -i3pSqoaWvwCh0rdQzlQJ0Rr+1k70AnGvGnPmtFpceT8AHsJkzfhH4AetZLYeuOpf -FlcMca9267VdyWgwInob8fcvAURW1ZN4qn8MvNPOBXudj5W5+8XowWmDES4qNr1/ -2Oj4IgHDlMRgUYhsql0ybYarpfZdRxxKKj3ZW4B55Qqds2mG1w40zTSeW9ErXQvJ -EYkqFsAhEme0Ii+tKP1oM/qRrHuCfKiQw8Zc99v0uU19KbdD2B8sCsBfgkIJpGny -6ne8BuNAJRDJa4JhzyRu0Aw7f+U/ewn4T+GYdzgsqnsqH6nEwEabeHUtEsChXxZp -7mu9nSww8fJUEgHuTonr/w8UcMtB7HVwhQ1/AuFxzaQx7uKkyU+uyJaElZ3LD30O -f20p9Z0v6LXpyiqxBUUytoPCsBtRi2aPQKvNmnkPbtH4P45nz0Nc7CVaWM3tvAGR -53WQMxowHcek/J3mtNVprhG1gn5V+NnF/a5cjVqGxQbs/G7lhqZXYBNeflW3mUFx -7DaO5C5KcqjJBN1h9W6a48qh0sqyIbuBFlWJFNdizV0eMi1ypsHXKSZcl7SZ7PB2 -QGMVLg6VW0RneL7zzpoaLHaey97bxeccP721rA0/6w5qCE+qlTUXgV5sZSXJkj9M -wwpVQWwqa6q/cG0G7iFCxbD+OPZ3/9jZiQGcBBMBCAAGBQJYrnXJAAoJEBzIdvEM -rJ+J7Q4MAKz8ITE6nKeltOLAJF3xHNNcvyIHFLcOF2BI4bJTinMS7hFwEM3tg8+s -fnClHe7Lu/YpJdtqJ+jz8+nZMEF9tpS49C4bA8sPDyBHVqBi75xivKDSchnogPql -jctZF6NWbOt3Bf21DqbJFnrrtg/aEDrHQIDdXZUIKM66artlELC1XmFUnzfUBYNB -vksoPD+ehG9Im4ugC5kQCGxMNDjHXGyw+DzSm5n+hyEtkjnOPq1x2uPaELrHweMZ -c06ivHndOBp3vU2EgkyuCvRebc4OLu94RSel/ANv2VRdt3ryRQrW5tqxQJhEwPLW -fWqNTmR3vZhuUrY5Bk/R6Spn+iNJE1qCUutbB89aIeT9KMV5Dl0Zes4gdK7PLnLJ -5rjEMoVvJ1Tdl4LUToKZk+7el+2jEMZpkv5jnXkeRqMZAB584wWVOA7+7pKDUGL4 -r7RJByaBz6wRFCGmSJ5DIZQ8HKSeF1ikCwUHqVDYfAWmlyR/t7ZH3ZgUT1ezi67/ -PLQrM9JPbLkCDQROfHxOARAA5hb6RwSG2oH8LMWk6rmPthWH5IBE8yw4InTPpsA8 -V7LyFlNUOH+BuHI8mTpTHk4aRfg3h8wxqw9VfnncWN/H69Y6bhgYp8XZ37esQjPr -kujaQ7QaLp9EB++96AvF+5pTvf1eBlkhprMXUolw/D3UpGnC6uXW2iCjKEjt4HGU -G/nJQum9U9fcmZJWrtKFOW8NK/DVJ3iIdh2RmR+DceBDXUJF2qL9DEQvhEDAO5uY -glC8CwYdHwbdQaWjgLyDMWjr65SQZGbYJ1e+ZxPGGpucfQR89lylNaZwIg/HkFgU -bIvGnezleSwfO93ayQ34HVtpecr14TMG/jouh85xCsbsX7znnTLtCKzti+EkWRXa -NV0D+FvaPKo4jv440vgQZajcPzD9tbYWUfylpg83URVaQqZZglg1gLPU166vkB4V -/ov6nBjQ+Z6YxJsGvgPVhfBZth8IrckFUINyH5JKAAcwPZBtKR0QfUSHW+SxHer4 -DMLHpsjO39wHO9CIk4EcbLYUJwoEYlFpcnNWNYBwjLqAWXuMA+mE2fX/+NoMY1/c -rOZ46y3dLq0zJfD+LBgORx10j1fFaAj9j36pg43DUewZSwLtBhlYJ/SExW0Rz0xU -MU+C/4EJjy7+3ycLV+M8gnJGVwp2+z1H1ESe5bH6hSgARqQ4pOfP9sbM7sNX/y17 -KMEAEQEAAYkCHwQYAQIACQUCTnx8TgIbDAAKCRDfb9lxMGA32aBIEADAGhbCehSj -Wv8SEw9gUpN+slmIDBnZ7uqQgXjWO5OnG2TrSJyPNAwfk6ESY6JeoGuiASL3EpqD -vRTVsIvDzzqhNBwVa+mi/q3lof9yNs74dmJYsH0P20+9lVzNfWATWUDA4cVYBvON -BloCK1cVvn9zqFvfjFBcRbZskcvMBVPxO2Fv4xAzX+omPDfCnweY8G7i71Z8Nnl/ -HVkSZMI9uXrtcde00oISHf5xUebJdx96dxnUCDLPUwPiIxxYN44KvIl3cnIB5qwu -BV8F2XXUtBdxZDJexqsCIoAD3rhRoWq6E2fRJKeqt/4TmxwjsJ8ZODp+ilXhqRe/ -shHttoOvbo5QBZNZMujxkqxXeu+j2E3Ry5mSiGX1SewwbT1iUppwGI15Uwhthhrc -PwbtWxxIyzPBU6awwlrTrYxNTB1n7WM99gcQctLWZpWnEaoAnEmIEcPjnM+c2NRw -UJmE/C5h9intY4fOa2a8hpUPx6UbMkfPl0bkIA2cduvQtAFKy/G/Jm4H+0trSmrD -c+o+rl7v9sMJ9wKkMUdAcqUgNP0TEHzDPbzvztcKBCLnNLoUTKNIN4eNJjMGk8Si -/OgiN1NKkuVz7I3i916mVxxlFjKEyLYU4tYYXsbB+ZJy4dTP/YWHbQulJYLgju6Z -ELphkzjc6eM3CaOZ73u4GVXotheeUabUHQ== -=drvb ------END PGP PUBLIC KEY BLOCK----- \ No newline at end of file diff --git a/coreutils-provides.inc b/coreutils-provides.inc new file mode 100644 index 0000000..8314a4e --- /dev/null +++ b/coreutils-provides.inc @@ -0,0 +1,9 @@ +Provides: bundled(gnulib) + +# make it possible to install the latest available Adobe Reader RPM for Linux +Provides: /bin/cat +Provides: /bin/chmod +Provides: /bin/echo +Provides: /bin/ln +Provides: /bin/rm +Provides: /bin/touch diff --git a/coreutils-python3.patch b/coreutils-python3.patch deleted file mode 100644 index 447fdbc..0000000 --- a/coreutils-python3.patch +++ /dev/null @@ -1,65 +0,0 @@ -From 8927d505ecb5334f09c48ef98ef1f464f581d0f7 Mon Sep 17 00:00:00 2001 -From: rpm-build -Date: Tue, 2 Apr 2024 14:11:26 +0100 -Subject: [PATCH] coreutils-python3.patch - ---- - init.cfg | 4 ++-- - tests/d_type-check | 2 +- - tests/du/move-dir-while-traversing.sh | 6 +++--- - 3 files changed, 6 insertions(+), 6 deletions(-) - -diff --git a/init.cfg b/init.cfg -index ac05f7b..26d9516 100644 ---- a/init.cfg -+++ b/init.cfg -@@ -601,10 +601,10 @@ seek_data_capable_() - # Skip the current test if "." lacks d_type support. - require_dirent_d_type_() - { -- python < /dev/null \ -+ python3 < /dev/null \ - || skip_ python missing: assuming no d_type support - -- python "$abs_srcdir"/tests/d_type-check \ -+ python3 "$abs_srcdir"/tests/d_type-check \ - || skip_ requires d_type support - } - -diff --git a/tests/d_type-check b/tests/d_type-check -index 1a2f76f..42d3924 100644 ---- a/tests/d_type-check -+++ b/tests/d_type-check -@@ -1,4 +1,4 @@ --#!/usr/bin/python -+#!/usr/bin/python3 - # Exit 0 if "." and "./tempfile" have useful d_type information, else 1. - # Intended to exit 0 only on Linux/GNU systems. - import os -diff --git a/tests/du/move-dir-while-traversing.sh b/tests/du/move-dir-while-traversing.sh -index adf482b..cf9214a 100755 ---- a/tests/du/move-dir-while-traversing.sh -+++ b/tests/du/move-dir-while-traversing.sh -@@ -21,8 +21,8 @@ print_ver_ du - require_trap_signame_ - - # We use a python-inotify script, so... --python -m pyinotify -h > /dev/null \ -- || skip_ 'python inotify package not installed' -+python3 -m pyinotify -h > /dev/null \ -+ || skip_ 'python3 inotify package not installed' - - # Move a directory "up" while du is processing its sub-directories. - # While du is processing a hierarchy .../B/C/D/... this script -@@ -33,7 +33,7 @@ python -m pyinotify -h > /dev/null \ - # rename syscall before du finishes processing the subtree under D/. - - cat <<'EOF' > inotify-watch-for-dir-access.py --#!/usr/bin/env python -+#!/usr/bin/env python3 - import pyinotify as pn - import os,sys - --- -2.51.0 - diff --git a/coreutils-selinux.patch b/coreutils-selinux.patch new file mode 100644 index 0000000..8274a23 --- /dev/null +++ b/coreutils-selinux.patch @@ -0,0 +1,120 @@ +diff --git a/man/chcon.x b/man/chcon.x +index 8c1ff6f..c84fb96 100644 +--- a/man/chcon.x ++++ b/man/chcon.x +@@ -1,4 +1,4 @@ + [NAME] +-chcon \- change file security context ++chcon \- change file SELinux security context + [DESCRIPTION] + .\" Add any additional description here +diff --git a/man/runcon.x b/man/runcon.x +index d2df13e..5c5f5d8 100644 +--- a/man/runcon.x ++++ b/man/runcon.x +@@ -1,5 +1,5 @@ + [NAME] +-runcon \- run command with specified security context ++runcon \- run command with specified SELinux security context + [DESCRIPTION] + Run COMMAND with completely-specified CONTEXT, or with current or + transitioned security context modified by one or more of LEVEL, +diff --git a/src/cp.c b/src/cp.c +index 1b528c6..25dbb88 100644 +--- a/src/cp.c ++++ b/src/cp.c +@@ -203,6 +203,9 @@ Copy SOURCE to DEST, or multiple SOURCE(s) to DIRECTORY.\n\ + all\n\ + "), stdout); + fputs (_("\ ++ -c deprecated, same as --preserve=context\n\ ++"), stdout); ++ fputs (_("\ + --no-preserve=ATTR_LIST don't preserve the specified attributes\n\ + --parents use full source file name under DIRECTORY\n\ + "), stdout); +@@ -929,7 +932,7 @@ main (int argc, char **argv) + selinux_enabled = (0 < is_selinux_enabled ()); + cp_option_init (&x); + +- while ((c = getopt_long (argc, argv, "abdfHilLnprst:uvxPRS:TZ", ++ while ((c = getopt_long (argc, argv, "abcdfHilLnprst:uvxPRS:TZ", + long_opts, NULL)) + != -1) + { +@@ -977,6 +980,17 @@ main (int argc, char **argv) + copy_contents = true; + break; + ++ case 'c': ++ fprintf (stderr, "%s: warning: option '-c' is deprecated, please use '--preserve=context' instead\n", argv[0]); ++ if ( x.set_security_context ) { ++ (void) fprintf(stderr, "%s: cannot force target context and preserve it\n", argv[0]); ++ exit( 1 ); ++ } ++ else if (selinux_enabled) { ++ x.preserve_security_context = true; ++ x.require_preserve_context = true; ++ } ++ break; + case 'd': + x.preserve_links = true; + x.dereference = DEREF_NEVER; +diff --git a/doc/coreutils.texi b/doc/coreutils.texi +index 47e4480..cff2ead 100644 +--- a/doc/coreutils.texi ++++ b/doc/coreutils.texi +@@ -8083,6 +8083,11 @@ done + exit $fail + @end example + ++@item -c ++@cindex SELinux security context information, preserving ++Preserve SELinux security context of the original files if possible. ++Some file systems don't support storing of SELinux security context. ++ + @item --copy-contents + @cindex directories, copying recursively + @cindex copying directories recursively +diff --git a/src/install.c b/src/install.c +index d79d597..437889a 100644 +--- a/src/install.c ++++ b/src/install.c +@@ -673,7 +673,7 @@ In the 4th form, create all components of the given DIRECTORY(ies).\n\ + -v, --verbose print the name of each directory as it is created\n\ + "), stdout); + fputs (_("\ +- --preserve-context preserve SELinux security context\n\ ++ -P, --preserve-context preserve SELinux security context (-P deprecated)\n\ + -Z set SELinux security context of destination\n\ + file and each created directory to default type\n\ + --context[=CTX] like -Z, or if CTX is specified then set the\n\ +@@ -824,7 +824,7 @@ main (int argc, char **argv) + dir_arg = false; + umask (0); + +- while ((optc = getopt_long (argc, argv, "bcCsDdg:m:o:pt:TvS:Z", long_options, ++ while ((optc = getopt_long (argc, argv, "bcCsDdg:m:o:pPt:TvS:Z", long_options, + NULL)) != -1) + { + switch (optc) +@@ -885,6 +885,8 @@ main (int argc, char **argv) + no_target_directory = true; + break; + ++ case 'P': ++ fprintf (stderr, "%s: warning: option '-P' is deprecated, please use '--preserve-context' instead\n", argv[0]); + case PRESERVE_CONTEXT_OPTION: + if (! selinux_enabled) + { +@@ -892,6 +894,10 @@ main (int argc, char **argv) + "this kernel is not SELinux-enabled")); + break; + } ++ if ( x.set_security_context ) { ++ (void) fprintf(stderr, "%s: cannot force target context and preserve it\n", argv[0]); ++ exit( 1 ); ++ } + x.preserve_security_context = true; + use_default_selinux_context = false; + break; diff --git a/coreutils.spec b/coreutils.spec index 6712263..ed38c4c 100644 --- a/coreutils.spec +++ b/coreutils.spec @@ -1,50 +1,117 @@ Summary: A set of basic GNU tools commonly used in shell scripts Name: coreutils -Version: 9.9 -Release: 2%{?dist} -# some used parts of gnulib are under various variants of LGPL -License: GPL-3.0-or-later AND GFDL-1.3-no-invariants-or-later AND LGPL-2.1-or-later AND LGPL-3.0-or-later +Version: 8.32 +Release: 32%{?dist} +License: GPLv3+ Url: https://www.gnu.org/software/coreutils/ Source0: https://ftp.gnu.org/gnu/%{name}/%{name}-%{version}.tar.xz -Source1: https://ftp.gnu.org/gnu/%{name}/%{name}-%{version}.tar.xz.sig -# From https://savannah.gnu.org/project/release-gpgkeys.php?group=coreutils&download=1 -# which is linked as project keyring on https://savannah.gnu.org/projects/coreutils -Source2: coreutils-keyring.gpg Source50: supported_utils +Source51: coreutils-provides.inc Source105: coreutils-colorls.sh Source106: coreutils-colorls.csh # do not make coreutils-single depend on /usr/bin/coreutils %global __requires_exclude ^%{_bindir}/coreutils$ +# ls: restore 8.31 behavior on removed directories +Patch1: coreutils-8.32-ls-removed-dir.patch + +# du: simplify leaf optimization for XFS (#1823247) +Patch2: coreutils-8.32-leaf-opt-xfs.patch + +# cp: default to --reflink=auto (#1861108) +Patch3: coreutils-8.32-cp-reflink-auto.patch + +# eliminate unportable gnulib tests +Patch4: coreutils-8.32-gnulib-perror-test.patch + +# df,stat,tail: recognize more file system types +Patch5: coreutils-8.32-new-fs-types.patch + +# rm: do not skip files upon failure to remove an empty dir (#1905481) +Patch6: coreutils-8.32-rm-stray-skip.patch + +# expr: fix invalid read with unmatched \(...\) (#1919775) +Patch7: coreutils-8.32-expr-unmatched-par.patch + +# split: fix --number=K/N to output correct part of file (#1921246) +Patch8: coreutils-8.32-split-number.patch + +# ls: fix crash printing SELinux context for unstatable files (#1921249) +Patch9: coreutils-8.32-ls-scontext-crash.patch + +# stat: add support for the exfat file system (#1921427) +Patch10: coreutils-8.32-stat-exfat.patch + +# cp: use copy_file_range if available +Patch11: coreutils-8.32-cp-file-range.patch + +# hostname,ln: fix memory leaks detected by Coverity +Patch12: coreutils-8.32-mem-leaks.patch + +# utimens: fix confusing arg type in internal func +Patch13: coreutils-8.32-coverity-utimens.patch + +# fix false positives in the upstrem test-suite (#1960792) +Patch14: coreutils-8.32-tests-false-positives.patch + +# mountlist: recognize fuse.portal as dummy file system (#1913358) +Patch15: coreutils-8.32-fuse-portal.patch + +# tail: fix stack out-of-bounds write with --follow +Patch16: coreutils-8.32-tail-use-poll.patch + +# df: fix duplicated remote entries due to bind mounts (#1979814) +Patch17: coreutils-8.32-df-duplicated-entries.patch + +# ls, stat: avoid triggering automounts (#2044981) +Patch18: coreutils-9.0-autofs-no-mount.patch + # disable the test-lock gnulib test prone to deadlock Patch100: coreutils-8.26-test-lock.patch # require_selinux_(): use selinuxenabled(8) if available -Patch101: coreutils-8.26-selinuxenable.patch +Patch105: coreutils-8.26-selinuxenable.patch # downstream changes to default DIR_COLORS Patch102: coreutils-8.32-DIR_COLORS.patch - -# use python3 in tests -Patch103: coreutils-python3.patch - -# df --direct +#do display processor type for uname -p/-i based on uname(2) syscall +Patch103: coreutils-8.2-uname-processortype.patch +#df --direct Patch104: coreutils-df-direct.patch +#add note about mkdir --mode behaviour into info documentation(#610559) +Patch107: coreutils-8.4-mkdir-modenote.patch -# gnulib C23 support -# https://github.com/coreutils/gnulib/commit/df17f4f37ed3ca373d23ad42eae51122bdb96626 -Patch105: coreutils-9.9-gnulib-c23.patch - -# fix cut test failure on aarch64 rawhide (rhbz#2424302) -# https://github.com/coreutils/coreutils/commit/95044cb5eaea83d02f768feb5ab79fcf5e6ad782 -Patch106: coreutils-9.9-fix-cut-test-aarch64.patch +# sh-utils +#add info about TZ envvar to date manpage +Patch703: sh-utils-2.0.11-dateman.patch +Patch713: coreutils-4.5.3-langinfo.patch # (sb) lin18nux/lsb compliance - multibyte functionality patch Patch800: coreutils-i18n.patch +# (sb) lin18nux/lsb compliance - expand/unexpand +Patch801: coreutils-i18n-expand-unexpand.patch +# i18n patch for cut - old version - used +Patch804: coreutils-i18n-cut-old.patch +# The unexpand patch above is not correct. Sent to the patch authors +Patch803: coreutils-i18n-fix-unexpand.patch +#(un)expand - allow multiple files on input - broken by patch 801 +Patch805: coreutils-i18n-fix2-expand-unexpand.patch +#(un)expand - test BOM headers +Patch806: coreutils-i18n-un-expand-BOM.patch +# make 'sort -h' work for arbitrary column even when using UTF-8 locales +Patch807: coreutils-i18n-sort-human.patch +# fold: preserve new-lines in mutlibyte text (#1418505) +Patch808: coreutils-i18n-fold-newline.patch + +#getgrouplist() patch from Ulrich Drepper. +Patch908: coreutils-getgrouplist.patch + +#SELINUX Patch - implements Redhat changes +#(upstream did some SELinux implementation unlike with RedHat patch) +Patch950: coreutils-selinux.patch Conflicts: filesystem < 3 - # To avoid clobbering installs Conflicts: coreutils-single @@ -63,40 +130,22 @@ BuildRequires: libselinux-utils BuildRequires: make BuildRequires: openssl-devel BuildRequires: strace -BuildRequires: systemd-devel BuildRequires: texinfo -# For gpg verification of source tarball -BuildRequires: gnupg2 - # test-only dependencies -BuildRequires: acl -BuildRequires: gdb BuildRequires: perl-interpreter BuildRequires: perl(FileHandle) -BuildRequires: python3 -BuildRequires: tzdata -%ifarch %valgrind_arches -BuildRequires: valgrind -%endif - -%if 0%{?fedora} -BuildRequires: perl(Expect) -BuildRequires: python3-inotify -%endif - %if 23 < 0%{?fedora} || 7 < 0%{?rhel} # needed by i18n test-cases BuildRequires: glibc-langpack-en BuildRequires: glibc-langpack-fr BuildRequires: glibc-langpack-ko -BuildRequires: glibc-langpack-sv %endif Requires: %{name}-common = %{version}-%{release} Provides: coreutils-full = %{version}-%{release} -Provides: bundled(gnulib) +%include %{SOURCE51} Obsoletes: %{name} < 8.24-100 %description @@ -108,7 +157,7 @@ Summary: coreutils multicall binary Suggests: coreutils-common Provides: coreutils = %{version}-%{release} Provides: coreutils%{?_isa} = %{version}-%{release} -Provides: bundled(gnulib) +%include %{SOURCE51} # To avoid clobbering installs Conflicts: coreutils < 8.24-100 # Note RPM doesn't support separate buildroots for %files @@ -125,19 +174,16 @@ packaged as a single multicall binary. # https://bugzilla.redhat.com/show_bug.cgi?id=1107973#c7 Obsoletes: %{name} < 8.24-100 -# Gnulib translations are maintained seprately since coreutils 9.6 (#2393892) -Requires: gnulib-l10n - # info doc refers to "Specifying the Time Zone" from glibc-doc (#959597) -Suggests: glibc-doc +Recommends: glibc-doc +Recommends: ncurses Summary: coreutils common optional components %description common Optional though recommended components, including documentation and translations. %prep -%{gpgverify} --keyring='%{SOURCE2}' --signature='%{SOURCE1}' --data='%{SOURCE0}' %autosetup -N # will be regenerated in the build directories @@ -162,15 +208,16 @@ sed src/dircolors.hin \ find tests -name '*.sh' -perm 0644 -print -exec chmod 0755 '{}' '+' (echo "<<< done") 2>/dev/null -# FIXME: Force a newer gettext version to workaround `autoreconf -i` errors -# with coreutils 9.6 and bundled gettext 0.19.2 from gettext-common-devel. -sed -i "s/0.19.2/$(rpm -q --queryformat '%%{VERSION}\n' gettext-devel)/" bootstrap.conf configure.ac - autoreconf -fiv %build export CFLAGS="$RPM_OPT_FLAGS -fno-strict-aliasing -fpic" +# disable -flto on ppc64le to make test-float pass (#1789115) +%ifarch ppc64le +CFLAGS="$CFLAGS -fno-lto" +%endif + # Upstream suggests to build with -Dlint for static analyzers: # https://lists.gnu.org/archive/html/coreutils/2018-06/msg00110.html # ... and even for production binary RPMs: @@ -191,7 +238,7 @@ for type in separate single; do if test $type = 'single'; then config_single='--enable-single-binary' config_single="$config_single --without-openssl" # smaller/slower sha*sum - config_single="$config_single --without-libgmp" # expr/factor machine ints + config_single="$config_single --without-gmp" # expr/factor machine ints else config_single='--with-openssl' # faster sha*sum fi @@ -199,7 +246,6 @@ for type in separate single; do --cache-file=../config.cache \ --enable-install-program=arch \ --enable-no-install-program=kill,uptime \ - --enable-systemd \ --with-tty-group \ DEFAULT_POSIX2_VERSION=200112 alternative=199209 || : %make_build all V=1 @@ -226,20 +272,14 @@ for type in separate single; do fi (cd $type && make DESTDIR=$RPM_BUILD_ROOT/$subdir $install) -%if "%{_sbindir}" != "%{_bindir}" # chroot was in /usr/sbin : - mkdir -p $RPM_BUILD_ROOT/$subdir/%_sbindir + mkdir -p $RPM_BUILD_ROOT/$subdir/{%{_bindir},%{_sbindir}} mv $RPM_BUILD_ROOT/$subdir/{%_bindir,%_sbindir}/chroot -%endif # Move multicall variants to *.single. # RemovePathPostfixes will strip that later. if test $type = 'single'; then - for dir in %{_bindir} \ -%if "%{_sbindir}" != "%{_bindir}" -%{_sbindir} \ -%endif -%{_libexecdir}/%{name}; do + for dir in %{_bindir} %{_sbindir} %{_libexecdir}/%{name}; do for bin in $RPM_BUILD_ROOT/%{_libexecdir}/%{name}/$dir/*; do basebin=$(basename $bin) mv $bin $RPM_BUILD_ROOT/$dir/$basebin.single @@ -267,9 +307,7 @@ rm -f $RPM_BUILD_ROOT%{_infodir}/dir %files single %{_bindir}/*.single -%if "%{_sbindir}" != "%{_bindir}" %{_sbindir}/chroot.single -%endif %dir %{_libexecdir}/coreutils %{_libexecdir}/coreutils/*.so.single # duplicate the license because coreutils-common does not need to be installed @@ -286,180 +324,12 @@ rm -f $RPM_BUILD_ROOT%{_infodir}/dir %license COPYING %changelog -* Tue Jan 13 2026 Lukáš Zaoral - 9.9-2 -- fix cut test failure on aarch64 rawhide (rhbz#2424302) - -* Wed Nov 26 2025 Lukáš Zaoral - 9.9-1 -- rebase to latest upstream release (rhbz#2413803) - -* Mon Sep 29 2025 Lukáš Zaoral - 9.8-3 -- require gnulib-l10n for translations of gnulib messages (rhbz#2393892) - -* Thu Sep 25 2025 Lukáš Zaoral - 9.8-2 -- tail: fix tailing larger number of lines in regular files (rhbz#2398008) - -* Wed Sep 24 2025 Lukáš Zaoral - 9.8-1 -- rebase to latest upstream release (rhbz#2397467) -- remove downstream patch for selinux options deprecated since 2009 - -* Wed Jul 23 2025 Fedora Release Engineering - 9.7-5 -- Rebuilt for https://fedoraproject.org/wiki/Fedora_43_Mass_Rebuild - -* Mon Jun 30 2025 Lukáš Zaoral - 9.7-4 -- stty: add support for arbitrary baud rates (rhbz#2375439) - -* Wed May 28 2025 Lukáš Zaoral - 9.7-3 -- sort: fix buffer under-read (CVE-2025-5278) - -* Mon May 19 2025 Lukáš Zaoral - 9.7-2 -- cp/mv: do not fail when copying of trivial NFSv4 ACLs fails (rhbz#2363149) - -* Wed Apr 09 2025 Lukáš Zaoral - 9.7-1 -- rebase to latest upstream release (rhbz#2358624) - -* Tue Feb 25 2025 Lukáš Zaoral - 9.6-2 -- fix 'who -m' with guessed tty names (rhbz#2343998) - -* Mon Jan 20 2025 Lukáš Zaoral - 9.6-1 -- rebase to latest upstream version (rhbz#2338620) -- sync i18n patch with SUSE (Kudos to Berny Völker!) - -* Thu Jan 16 2025 Fedora Release Engineering - 9.5-13 -- Rebuilt for https://fedoraproject.org/wiki/Fedora_42_Mass_Rebuild - -* Sun Jan 12 2025 Zbigniew Jędrzejewski-Szmek - 9.5-12 -- Rebuilt for the bin-sbin merge (2nd attempt) - -* Wed Nov 13 2024 Florian Weimer - 9.5-11 -- Affinity mask handling in nproc for large CPU counts (rhbz#2325167) - -* Fri Sep 27 2024 Lukáš Zaoral - 9.5-10 -- fix fold -b with UTF8 locale (RHEL-60295) - -* Tue Aug 27 2024 Lukáš Zaoral - 9.5-9 -- show web sessions in who output (rhbz#2307847) - -* Wed Aug 21 2024 Lukáš Zaoral - 9.5-8 -- add missing systemd-devel buildrequires - -* Wed Jul 17 2024 Fedora Release Engineering - 9.5-7 -- Rebuilt for https://fedoraproject.org/wiki/Fedora_41_Mass_Rebuild - -* Mon Jul 15 2024 Lukáš Zaoral - 9.5-6 -- Rebuilt for the bin-sbin merge - -* Mon Jul 15 2024 Sohum Mendon - 9.5-5 -- fix incorrect exit status when fold is called with a non-existent file - -* Tue Jul 09 2024 Zbigniew Jędrzejewski-Szmek - 9.5-4 -- Rebuilt for the bin-sbin merge - -* Thu Jul 04 2024 Lukáš Zaoral - 9.5-3 -- do not buildrequire perl(Expect) on ELN - -* Tue Jun 04 2024 Lukáš Zaoral - 9.5-2 -- enable LTO on ppc64le - -* Tue Apr 02 2024 Lukáš Zaoral - 9.5-1 -- rebase to latest upstream version (rhbz#2272063) -- sync i18n patch with SUSE (Kudos to Berny Völker!) -- add some test dependencies to execute additional part of the upstream test-suite - -* Mon Jan 29 2024 Lukáš Zaoral - 9.4-6 -- fix tail on kernels with 64k page sizes (RHEL-22866) - -* Wed Jan 24 2024 Fedora Release Engineering - 9.4-5 -- Rebuilt for https://fedoraproject.org/wiki/Fedora_40_Mass_Rebuild - -* Fri Jan 19 2024 Fedora Release Engineering - 9.4-4 -- Rebuilt for https://fedoraproject.org/wiki/Fedora_40_Mass_Rebuild - -* Thu Jan 18 2024 Lukáš Zaoral - 9.4-3 -- fix compilation on i686 - -* Thu Jan 18 2024 Lukáš Zaoral - 9.4-2 -- fix buffer overflow in split (CVE-2024-0684) - -* Fri Sep 15 2023 Lukáš Zaoral - 9.4-1 -- new upstream release 9.4 (#2235759) -- enable integration with systemd -- fix the license field - -* Wed Jul 19 2023 Fedora Release Engineering - 9.3-2 -- Rebuilt for https://fedoraproject.org/wiki/Fedora_39_Mass_Rebuild - -* Tue Apr 18 2023 Kamil Dudka - 9.3-1 -- remove obsolete Provides for absolute paths -- new upstream release 9.3 - -* Tue Apr 11 2023 Lukáš Zaoral - 9.2-4 -- migrate to SPDX license format - -* Fri Mar 24 2023 Kamil Dudka - 9.2-3 -- copy: fix --reflink=auto to fallback in more cases (#2180056) -- cksum: fix reporting of failed checks (#2180056) - -* Wed Mar 22 2023 Kamil Dudka - 9.2-2 -- coreutils-getgrouplist.patch: drop a patch no longer needed - -* Wed Mar 22 2023 Kamil Dudka - 9.2-1 -- new upstream release 9.2 - -* Thu Jan 19 2023 Fedora Release Engineering -- Rebuilt for https://fedoraproject.org/wiki/Fedora_38_Mass_Rebuild - -* Mon Jan 02 2023 Kamil Dudka - 9.1-10 -- drop obsolete downstream-only extension of date(1) man page -- undocument downstream SELinux options deprecated since 2009 - -* Mon Jan 02 2023 Kamil Dudka - 9.1-9 -- basic support for checking NFSv4 ACLs (#2137866) - -* Mon Sep 19 2022 Kamil Dudka - 9.1-8 -- remove obsolete extension of mkdir(1) info documentation - -* Tue Aug 23 2022 Kamil Dudka - 9.1-7 -- remove non-upstream patch for uname -i/-p (#548834) - -* Mon Aug 08 2022 Kamil Dudka - 9.1-6 -- improve wording of a comment in /etc/DIR_COLORS (#2112593) - -* Mon Aug 08 2022 Kamil Dudka - 9.1-5 -- improve handling of control characters in unexpand (#2112870) - -* Mon Aug 01 2022 Kamil Dudka - 9.1-4 -- prevent unexpand from failing on control characters (#2112870) - -* Wed Jul 20 2022 Fedora Release Engineering -- Rebuilt for https://fedoraproject.org/wiki/Fedora_37_Mass_Rebuild - -* Sat Apr 23 2022 Pádraig Brady - 9.1-2 -- make simple backups in correct dir; broken in 9.1 - -* Tue Apr 19 2022 Kamil Dudka - 9.1-1 -- new upstream release 9.1 - -* Mon Mar 21 2022 Kamil Dudka - 9.0-5 +* Mon Mar 21 2022 Kamil Dudka - 8.32-32 - ls, stat: avoid triggering automounts (#2044981) -* Tue Mar 01 2022 Kamil Dudka - 9.0-4 +* Tue Mar 01 2022 Kamil Dudka - 8.32-31 - make `df --direct` work again (#2058686) -* Wed Jan 19 2022 Fedora Release Engineering -- Rebuilt for https://fedoraproject.org/wiki/Fedora_36_Mass_Rebuild - -* Mon Oct 04 2021 Kamil Dudka - 9.0-2 -- chmod: fix exit status when ignoring symlinks - -* Sun Sep 26 2021 Kamil Dudka - 9.0-1 -- new upstream release 9.0 - -* Tue Sep 14 2021 Sahana Prasad -- Rebuilt with OpenSSL 3.0.0 - -* Wed Jul 21 2021 Fedora Release Engineering -- Rebuilt for https://fedoraproject.org/wiki/Fedora_35_Mass_Rebuild - * Wed Jul 07 2021 Kamil Dudka - 8.32-30 - df: fix duplicated remote entries due to bind mounts (#1979814) @@ -479,8 +349,6 @@ rm -f $RPM_BUILD_ROOT%{_infodir}/dir - copy: do not refuse to copy a swap file * Fri Apr 09 2021 Kamil Dudka - 8.32-22 -- weaken the dependency on glibc-doc to reduce minimal installations -- drop the last use of ncurses no longer needed (#1830318) - utimens: fix confusing arg type in internal func * Fri Mar 26 2021 Kamil Dudka - 8.32-21 diff --git a/rh_i18n_wip.tar.gz b/rh_i18n_wip.tar.gz new file mode 100644 index 0000000..24b7cdc Binary files /dev/null and b/rh_i18n_wip.tar.gz differ diff --git a/sh-utils-2.0.11-dateman.patch b/sh-utils-2.0.11-dateman.patch new file mode 100644 index 0000000..60cdaa6 --- /dev/null +++ b/sh-utils-2.0.11-dateman.patch @@ -0,0 +1,12 @@ +diff -urNp coreutils-5.97-orig/man/date.x coreutils-5.97/man/date.x +--- coreutils-5.97-orig/man/date.x 1999-11-02 15:07:36.000000000 +0100 ++++ coreutils-5.97/man/date.x 2008-10-15 10:13:31.000000000 +0200 +@@ -11,3 +11,8 @@ calendar date, time of day, time zone, day of week, relative time, + relative date, and numbers. An empty string indicates the beginning + of the day. The date string format is more complex than is easily + documented here but is fully described in the info documentation. ++[ENVIRONMENT] ++.TP ++TZ ++Specifies the timezone, unless overridden by command line parameters. ++If neither is specified, the setting from /etc/localtime is used. diff --git a/sources b/sources index 0952ab1..97fe361 100644 --- a/sources +++ b/sources @@ -1,2 +1 @@ -SHA512 (coreutils-9.9.tar.xz.sig) = 0a3dfdfa6b4234e2e1d42142269f959bdf3cf8f6605a50270a27eff84dd22588f182121f7dd3eeb04be45f5109d02690215065b3d3b43882874d0e165a1435d0 -SHA512 (coreutils-9.9.tar.xz) = e7b0e59f7732d2c098ea4934014f470248bd5c4764210e9200a698010a8e3b95bbb26e543f0cd73ed5a4b8e1f8cda932c73f39954d68175e4deaa47526610c65 +SHA512 (coreutils-8.32.tar.xz) = 1c8f3584efd61b4b02e7ac5db8e103b63cfb2063432caaf1e64cb2dcc56d8c657d1133bbf10bd41468d6a1f31142e6caa81d16ae68fa3e6e84075c253613a145 diff --git a/tests/readlink-cannot-handle-recursive-symlinks/Makefile b/tests/readlink-cannot-handle-recursive-symlinks/Makefile new file mode 100644 index 0000000..49d37a7 --- /dev/null +++ b/tests/readlink-cannot-handle-recursive-symlinks/Makefile @@ -0,0 +1,63 @@ +# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +# +# Makefile of /CoreOS/coreutils/readlink-cannot-handle-recursive-symlink s +# Description: Test for readlink cannot handle recursive symlinks +# Author: Jan Scotka +# +# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +# +# Copyright (c) 2010 Red Hat, Inc. All rights reserved. +# +# This copyrighted material is made available to anyone wishing +# to use, modify, copy, or redistribute it subject to the terms +# and conditions of the GNU General Public License version 2. +# +# This program is distributed in the hope that it will be +# useful, but WITHOUT ANY WARRANTY; without even the implied +# warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR +# PURPOSE. See the GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public +# License along with this program; if not, write to the Free +# Software Foundation, Inc., 51 Franklin Street, Fifth Floor, +# Boston, MA 02110-1301, USA. +# +# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +export TEST=/CoreOS/coreutils/readlink-cannot-handle-recursive-symlinks +export TESTVERSION=1.0 + +BUILT_FILES= + +FILES=$(METADATA) runtest.sh Makefile PURPOSE + +.PHONY: all install download clean + +run: $(FILES) build + ./runtest.sh + +build: $(BUILT_FILES) + chmod a+x runtest.sh + +clean: + rm -f *~ $(BUILT_FILES) + + +include /usr/share/rhts/lib/rhts-make.include + +$(METADATA): Makefile + @echo "Owner: Jan Scotka " > $(METADATA) + @echo "Name: $(TEST)" >> $(METADATA) + @echo "TestVersion: $(TESTVERSION)" >> $(METADATA) + @echo "Path: $(TEST_DIR)" >> $(METADATA) + @echo "Description: Test for readlink cannot handle recursive symlinks" >> $(METADATA) + @echo "Type: Sanity" >> $(METADATA) + @echo "TestTime: 5m" >> $(METADATA) + @echo "RunFor: coreutils" >> $(METADATA) + @echo "Requires: coreutils" >> $(METADATA) + @echo "Priority: Normal" >> $(METADATA) + @echo "License: GPLv2" >> $(METADATA) + @echo "Confidential: no" >> $(METADATA) + @echo "Destructive: no" >> $(METADATA) + + rhts-lint $(METADATA) diff --git a/tests/readlink-cannot-handle-recursive-symlinks/PURPOSE b/tests/readlink-cannot-handle-recursive-symlinks/PURPOSE new file mode 100644 index 0000000..b9fd740 --- /dev/null +++ b/tests/readlink-cannot-handle-recursive-symlinks/PURPOSE @@ -0,0 +1,54 @@ +PURPOSE of /CoreOS/coreutils/readlink-cannot-handle-recursive-symlinks +Description: Test for readlink cannot handle recursive symlinks +Author: Jan Scotka +Bug summary: readlink cannot handle recursive symlinks + +Description: + +Description of problem: +The readlink command fails with an error "Too many levels of symbolic links", even if there are only 2 levels. + +The readlink command from RHEL 3 and RHEL 4 and from Fedora 11 all work fine. + +Among other changes between RHEL 4 and RHEL 5, RHEL 5's coreutils added calls to cycle_check() in lib/canonicalize.c, but in upstream canonicalize.c (now in gnulib instead of coreutils), we see the comment: + /* Detect loops. We cannot use the cycle-check module here, + since it's actually possible to encounter the same symlink + more than once in a given traversal. However, encountering + the same symlink,NAME pair twice does indicate a loop. */ + +http://git.savannah.gnu.org/gitweb/?p=gnulib.git;a=blob;f=lib/canonicalize.c;h=4f348398fd69ae516396313d18ac294a4ca3dde3;hb=b653eda3ac4864de205419d9f41eec267cb89eeb#l262 + +The latest canonicalize.c uses seen_triple() instead of cycle_check(). + + +Version-Release number of selected component (if applicable): +coreutils-5.97-19.el5 + +How reproducible: +every time + +Steps to Reproduce: +1. Create a directory with a symlink to itself + mkdir /tmp/dir + cd /tmp/dir + ln -s ../dir dirlink + +2. Run readlink using the 'dirlink' symlink recursively + readlink -v -f dirlink + readlink -v -f dirlink/dirlink + readlink -v -f dirlink/dirlink/dirlink + +Actual results: +The first readlink command on just dirlink succeeds, but the 2nd and 3rd commands fail with + readlink: dirlink/dirlink: Too many levels of symbolic links + +Expected results: +$ readlink -v -f dirlink +/tmp/dir +$ readlink -v -f dirlink/dirlink +/tmp/dir +$ readlink -v -f dirlink/dirlink/dirlink +/tmp/dir + + +Additional info: diff --git a/tests/readlink-cannot-handle-recursive-symlinks/runtest.sh b/tests/readlink-cannot-handle-recursive-symlinks/runtest.sh new file mode 100755 index 0000000..6ee251f --- /dev/null +++ b/tests/readlink-cannot-handle-recursive-symlinks/runtest.sh @@ -0,0 +1,60 @@ +#!/bin/bash +# vim: dict=/usr/share/beakerlib/dictionary.vim cpt=.,w,b,u,t,i,k +# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +# +# runtest.sh of /CoreOS/coreutils/readlink-cannot-handle-recursive-symlinks +# Description: Test for readlink cannot handle recursive symlinks +# Author: Jan Scotka +# +# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +# +# Copyright (c) 2010 Red Hat, Inc. All rights reserved. +# +# This copyrighted material is made available to anyone wishing +# to use, modify, copy, or redistribute it subject to the terms +# and conditions of the GNU General Public License version 2. +# +# This program is distributed in the hope that it will be +# useful, but WITHOUT ANY WARRANTY; without even the implied +# warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR +# PURPOSE. See the GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public +# License along with this program; if not, write to the Free +# Software Foundation, Inc., 51 Franklin Street, Fifth Floor, +# Boston, MA 02110-1301, USA. +# +# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +# Include rhts environment +. /usr/bin/rhts-environment.sh +. /usr/lib/beakerlib/beakerlib.sh + +PACKAGE="coreutils" + +rlJournalStart + rlPhaseStartSetup + rlAssertRpm $PACKAGE + rlRun "TmpDir=\`mktemp -d\`" 0 "Creating tmp directory" + rlRun "pushd $TmpDir" + rlRun "mkdir link" 0 "Creating test directory" + rlRun "cd link" 0 "cd to this dir" + rlRun "ln -s ../link link" 0 "creating symlink to ../link -> link" + rlPhaseEnd + + rlPhaseStartTest + rlLog "Test of readlink up to 20 iteration" + export TMPVAR="link" + for foo in `seq 20` + do echo $TMPVAR + rlRun "readlink -v -f $TMPVAR" 0 "readlink of $TMPVAR" + TMPVAR="$TMPVAR/link" + done + rlPhaseEnd + + rlPhaseStartCleanup + rlRun "popd" + rlRun "rm -r $TmpDir" 0 "Removing tmp directory" + rlPhaseEnd +rlJournalPrintText +rlJournalEnd diff --git a/tests/test-basics b/tests/test-basics new file mode 100755 index 0000000..7324553 --- /dev/null +++ b/tests/test-basics @@ -0,0 +1,39 @@ +#!/bin/sh + +# Checks that touch ls rm and foo work +# https://www.mankier.com/1/beakerlib#Examples +. /usr/share/beakerlib/beakerlib.sh + +# Set the full test name +TEST="/examples/beakerlib/Sanity/phases" + +# Package being tested +PACKAGE="coreutils" + +rlJournalStart + # Setup phase: Prepare test directory + rlPhaseStartSetup + rlAssertRpm $PACKAGE + rlRun 'TmpDir=$(mktemp -d)' 0 'Creating tmp directory' # no-reboot + rlRun "pushd $TmpDir" + rlPhaseEnd + + # Test phase: Testing touch, ls and rm commands + rlPhaseStartTest + rlRun "touch foo" 0 "Creating the foo test file" + rlAssertExists "foo" + rlRun "ls -l foo" 0 "Listing the foo test file" + rlRun "rm foo" 0 "Removing the foo test file" + rlAssertNotExists "foo" + rlRun "ls -l foo" 2 "Listing foo should now report an error" + rlPhaseEnd + + # Cleanup phase: Remove test directory + rlPhaseStartCleanup + rlRun "popd" + rlRun "rm -r $TmpDir" 0 "Removing tmp directory" + rlPhaseEnd +rlJournalEnd + +# Print the test report +rlJournalPrintText diff --git a/tests/test_basics.yml b/tests/test_basics.yml new file mode 100644 index 0000000..d5727cf --- /dev/null +++ b/tests/test_basics.yml @@ -0,0 +1,9 @@ +--- +# This first play always runs on the local staging system +- hosts: localhost + tags: + - atomic + - classic + - container + roles: + - { role: standard-test-beakerlib, tests: [ test-basics, readlink-cannot-handle-recursive-symlinks ] } diff --git a/tests/tests.yml b/tests/tests.yml new file mode 100644 index 0000000..529263d --- /dev/null +++ b/tests/tests.yml @@ -0,0 +1,2 @@ +# Fedora Continuous Integration: https://fedoraproject.org/wiki/CI +- include: test_basics.yml diff --git a/upstream-key.gpg b/upstream-key.gpg new file mode 100644 index 0000000..dc6dc1d --- /dev/null +++ b/upstream-key.gpg @@ -0,0 +1,123 @@ +-----BEGIN PGP PUBLIC KEY BLOCK----- +Version: GnuPG v1.4.6 (GNU/Linux) + +mQGiBDftyYoRBACvICTt5AWe7kdbRtJ37IZ+ED5tBA/IbISfqUPO+HmL/J9JSfkV +QHbdQR5dj5mrU6BY5YOY7L4KOS6lH3AgvsZ/NhkDBraBPgnMkpDqFb7z4keCIebb +AmlcBL2VQNTo0Lczo319YoZ+UaNH53OddlBY944qBTa0AlcJuS1SgEp7pwCg+CUj +4SjVzqZh5lgPTS0bnYvF/n0D/iItZ7WAm37KW+9UjArWZD6NO+mVMNq4GWmhcSBD +uyJOZFxFQWXdFRdM9sNO7lkWYVCxpXyFzmQcBzdrAt+zx/3QadEbduGAqEKAROQU +gSDlMITWGK97/Cadn1YRSDcGKNlJX9jlJvt5Q/xh+CnJ8HTwO0PF9A5N/phFuMMB +UH0pA/0e5eIBsr2Wvxy39+nGnNv5b+5tHkGXSSHKyI7+zOdIBTtRQO7lwTG9ioKg +/yMqb9NCSf4GdyZiFJsQ+TWoSyk1bvFHt7YUOhTeii7Zgbk7Due2q+b9KzzyH/r2 +kf+fLh0lgiy/LfBhvsfO8M9dji3XDyZpBLRO6gda9M9NqzEfgbQfSmltIE1leWVy +aW5nIDxqaW1AbWV5ZXJpbmcubmV0PohGBBARAgAGBQI9TvsUAAoJENoowjp5/0R0 +NTIAn2qpRF9QVupw/gz4UN5d5MKurlOMAKDNXKfXzWClHRq5ufCdwZead3WMMYhG +BBARAgAGBQJCk1gpAAoJEIvYLm8wuUtcqlIAn0KbOC5YSkgqhfhM1uRlHnvHB74A +AJ4qbzrkw7iitd1CH1eoMoFiP5CI14hGBBARAgAGBQJDYmg2AAoJELk/YMa1xM4T +ct0AoJIkdqI6dhTUDOVwiZRxaCKVYaoNAJsG8I+OPhhRhe7ZgN5iN3xlRfkhTohG +BBARAgAGBQJECHuEAAoJEFQUZr6xLcGbUyQAnRmg070gGrZ5E4ZPJRqL/DUoB7hN +AKCj7uAIpcRdrBAQW8PKiOWcPRvxjohGBBIRAgAGBQI/bJ2IAAoJEA6nVrUUSEP1 +QXoAoJ6dMlvbJUep2l5N8G0XFmRyxTrIAJ0bn5IYu7RMxqI0vv6DHn2VgEQLeohG +BBIRAgAGBQI/vFVMAAoJENKUXDvBNlC2gtYAn1zlWvzZaC2lxRXuW7fMWpB/5uVJ +AJ9RFEFFzl8BktsnskYJUIvrx5zVL4hGBBMRAgAGBQI/UFjyAAoJEDhZwDsuI25H +z80An0G2Xm22lMc7ThGGgKeovGP0GzPIAKCHFH2aY2Dv6XOYomNB1yvW7MU0ZIhG +BBMRAgAGBQI/cfsiAAoJEA3cqjJ41SZOmcoAoKulkHQ6TUVORoSN77UYtrdCKy0I +AKC5qT7peM0Jd6I9wPLwc7Fc65xraIhGBBMRAgAGBQJAmOELAAoJEAu1FKXQbtaf +ysgAoL7Zl3BSH+/F9ouPCXkduzIywdx9AJ9OevRoJwxpER+SwSiLnw9Q7fVmcYhX +BBMRAgAXBQI66oJOBQsHCgMEAxUDAgMWAgECF4AACgkQ/dLerNMzy6HlawCg5UXJ +LGWj9P0SuJKcGm+mqKb1J2MAn3YrgB3duqFNs/yS4mvxM74TzI5miFoEExECABoF +CwcKAwQDFQMCAxYCAQIXgAIZAQUCOuqCTwAKCRD90t6s0zPLoaVVAJ0UZOyi+B+q +cNTEDSDrc3Oc1MzZrQCg0UONeu4Dv4N5ZLI6lZBMZETaCmKIXwQTEQIAFwUCOuqC +TgULBwoDBAMVAwIDFgIBAheAABIJEP3S3qzTM8uhB2VHUEcAAQHlawCg5UXJLGWj +9P0SuJKcGm+mqKb1J2MAn3YrgB3duqFNs/yS4mvxM74TzI5miGIEExECABoFCwcK +AwQDFQMCAxYCAQIXgAIZAQUCOuqCTwASCRD90t6s0zPLoQdlR1BHAAEBpVUAnRRk +7KL4H6pw1MQNIOtzc5zUzNmtAKDRQ4167gO/g3lksjqVkExkRNoKYrQfSmltIE1l +eWVyaW5nIDxtZXllcmluZ0BnbnUub3JnPohGBBARAgAGBQJCk1gsAAoJEIvYLm8w +uUtcHS0AoIO9LsaLdn6aH3fskRVZ4qhpRBXbAJ0drV2s3abBKhkhUui7kpF87MTD ++4hGBBARAgAGBQJDYmg8AAoJELk/YMa1xM4TdT4Ani/0ORxwCzqGT0+BG2thzbO7 +aFkuAKCoKP+u6WhYYOBdEcaM6T5QLN56H4hGBBARAgAGBQJECHuHAAoJEFQUZr6x +LcGbrKEAoLef0BqLLpNGhAFJKSAvWEWOiGcxAJ9w7F7MtsDoegKeQ44yYiPX5jEu +5ohGBBIRAgAGBQI/bJ2IAAoJEA6nVrUUSEP13sUAn3IWX1RWnH50v+DZKcqzCaSA +oqHbAKCVvtirU/A3FJLnuyIBv+lguddi2IhGBBIRAgAGBQI/vFVRAAoJENKUXDvB +NlC2D68AnAzm1iw0YSQ1GuPaU3lG8n72p5EBAJ4pNBP+RFWjvZSfcUYhZAFhq8CB +QYhGBBMRAgAGBQI/cfslAAoJEA3cqjJ41SZO8asAnRsJcSER+vpIIzM/et8PakIC +ZJxsAJ9LjdnHkb+Zr9YDXzKXu6OTiJvIh4hGBBMRAgAGBQJAmOEOAAoJEAu1FKXQ +btafLL8AoJask7aB+OfOQgS/kMlKXAA25Hl3AKC/3XJeRRR0ze508VcIhx7EhYVV +84heBBMRAgAeBQI/UFjBAhsDBgsJCAcDAgMVAgMDFgIBAh4BAheAAAoJEP3S3qzT +M8uh8gwAoLfqQt7QgzavHlD44LxmAXovm5t0AJ4m8EQC+N9oJyODmpLbfQKNL6pq +zohmBBMRAgAeBQI/UFjBAhsDBgsJCAcDAgMVAgMDFgIBAh4BAheAABIJEP3S3qzT +M8uhB2VHUEcAAQHyDACgt+pC3tCDNq8eUPjgvGYBei+bm3QAnibwRAL432gnI4Oa +ktt9Ao0vqmrOtCFKaW0gTWV5ZXJpbmcgPG1leWVyaW5nQHBvYm94LmNvbT6IRgQQ +EQIABgUCQpNYLAAKCRCL2C5vMLlLXP7FAKCodISH72q8e30TxLwdoOh7hDjehACf +U97FCEOWICQaEI2BvOzGzn6yrO6IRgQQEQIABgUCQ2JoPAAKCRC5P2DGtcTOE0Gk +AJ43felw+/nxzJ7DVJYZ0tbASZ3BcACeNf2nXMkqkwrBZZ9DDMUGQ6tIB3GIRgQQ +EQIABgUCRAh7hwAKCRBUFGa+sS3Bm1nUAJ0foaMmGWqugETz37RZ2XpCfdQIlQCe +N50WxYPBxrGGmhhGOVbji1uhVSmIRgQSEQIABgUCP2ydiAAKCRAOp1a1FEhD9T73 +AJ4/51C6L0lHrX77DFXVJrB02yybsACgi/9TewF7HaF3x8fdMEZxsRK1HR+IRgQS +EQIABgUCP7xVUQAKCRDSlFw7wTZQtvjnAJ9FM83LyrTs2Dk/T7kOcSFTfjXqegCe +OlpOQ/sB4EtoHxrTSCy3OhToVsmIRgQTEQIABgUCP1BY+wAKCRA4WcA7LiNuR5yI +AJ9F3RsjjwtYX2rSx+j5o4+y4Dyl9wCfVR9uTBDLDP3kOaDrTT/H9XHTf6uIRgQT +EQIABgUCP3H7JQAKCRAN3KoyeNUmTv4eAJ9rCBUUXWYFUrjUayOenPULMW1BhACg +ncwdeTN+SGy8lX3zoo1vdNv+vTKIRgQTEQIABgUCQJjhDgAKCRALtRSl0G7WnyNP +AJ9Gn9yRup0zePUPMex36fX94o+i8wCggdDgtpKjzcaQ83o8VBiemFeiss+IXAQT +EQIAHAUCPjpzhwIbAwQLBwMCAxUCAwMWAgECHgECF4AACgkQ/dLerNMzy6FG5gCg +99D5pDqSRuZP2QJAT8LNiCZlRGgAn25OTXbNlHkM+gYFj0fyo+Ikj+T5iGQEExEC +ABwFAj46c4cCGwMECwcDAgMVAgMDFgIBAh4BAheAABIJEP3S3qzTM8uhB2VHUEcA +AQFG5gCg99D5pDqSRuZP2QJAT8LNiCZlRGgAn25OTXbNlHkM+gYFj0fyo+Ikj+T5 +tCJKaW0gTWV5ZXJpbmcgPG1leWVyaW5nQGFzY2VuZC5jb20+iEYEEBECAAYFAkKT +WCwACgkQi9gubzC5S1zwAQCgnPUCCl1g6eJdI5ZViACDiaaULAAAn19sIyQmkiaU +45QVcDtYuQTNSh/QiEYEEBECAAYFAkNiaDwACgkQuT9gxrXEzhP+igCfc526l8n/ +q8zVhIe9NonG+jVlrEoAnRXKebriKwmvVSdqbY8khlbJjB/ziEYEEBECAAYFAkQI +e4cACgkQVBRmvrEtwZs2owCgwzEOLdyXa2JGA/xkpBluqa8/UyMAnjZyxESMAj/A +2rUg3IvgtBmaetE4iEYEEhECAAYFAj+8VVEACgkQ0pRcO8E2ULaqIQCfQlbRoDOL +Hv+9YVxPgD8yhwFB850AnRTmAG4Z57YD92s4o1ne9sgaufmdiEYEExECAAYFAj9Q +WPsACgkQOFnAOy4jbkfOoQCgwfC1mkANwR+vv9TVlYkmoZ6wNL8An0dql+uy5ic1 +YpyKfV7g7MMuEMDwiEYEExECAAYFAj9x+yUACgkQDdyqMnjVJk6QCwCglS7PPvFR +HoOZxl7XgpVbAK6vZQgAniVxncBgSu06lmsDNHiJpiDMIZkkiEYEExECAAYFAkCY +4Q4ACgkQC7UUpdBu1p+QqwCeNzsozeUjiCFQBBiR+gCBnvZhQqgAnj4ImXyp45hs +fc3dZHP3qB1Ws5UjiFUEExECABUFAjftyYoDCwoDAxUDAgMWAgECF4AACgkQ/dLe +rNMzy6HnugCePkbs7JcEo0837WNqdoGf2WXL3vIAoK0cStFCa4zj4FV/SoG9cDZP +JOzfiF0EExECABUFAjftyYoDCwoDAxUDAgMWAgECF4AAEgkQ/dLerNMzy6EHZUdQ +RwABAee6AJ4+RuzslwSjTzftY2p2gZ/ZZcve8gCgrRxK0UJrjOPgVX9Kgb1wNk8k +7N+0IkppbSBNZXllcmluZyA8bWV5ZXJpbmdAbHVjZW50LmNvbT6IRQQTEQIABgUC +QJjhDgAKCRALtRSl0G7Wn/YLAJdAhf8twtaImmHzRT7eaUIf0b4+AJ9hRfAjWrRp +UF5cW5AzZsVwEW7Vc4hGBBARAgAGBQJCk1gsAAoJEIvYLm8wuUtceyMAoJGYrqPm +T+ThNBRLt5aIq/p3yBHmAJ0V0tEMjdIafWlY6IDZkst2VXBPFohGBBARAgAGBQJD +Ymg8AAoJELk/YMa1xM4TTxEAnAtkRTdyDNdPn5kW3HMKcQp9S02vAJ9wiBJbBeaB +jGcQ4zoafo0vw8ZMi4hGBBARAgAGBQJECHuHAAoJEFQUZr6xLcGbZi4AoK2Th3Pi +pC+CWdYDCA9qNa+uUkHsAKCHUU/oOSEqvjEHoYs22RZzVGbbVohGBBIRAgAGBQI/ +vFVRAAoJENKUXDvBNlC2qQ0An3hiEeuqRgzbuY6YLqiA9FH0GHEEAJ4j2O8AjZFq +Vc8RL32KA6nuwfJ28ohGBBMRAgAGBQI/UFj7AAoJEDhZwDsuI25HPicAoJOlcGaT +t5dvksbBg00BNCyZl8odAJ0UCIFlFzzB/x050scZKMrvquc2T4hGBBMRAgAGBQI/ +cfslAAoJEA3cqjJ41SZO5mQAoLTvGtjJxspvgEg3z3T/q6iI/FdxAJ4wgnqQjRvm +AHAWMibcDupPA10u+ohVBBMRAgAVBQI37e/HAwsKAwMVAwIDFgIBAheAAAoJEP3S +3qzTM8uh8vAAn23cUtWPdFr4wIwUNo9bsY1CUHMNAKCoHS3nayqM/WUfihcZJoOs +kQA22ohdBBMRAgAVBQI37e/HAwsKAwMVAwIDFgIBAheAABIJEP3S3qzTM8uhB2VH +UEcAAQHy8ACfbdxS1Y90WvjAjBQ2j1uxjUJQcw0AoKgdLedrKoz9ZR+KFxkmg6yR +ADbatCdKaW0gTWV5ZXJpbmcgPG1leWVyaW5nQG5hLW5ldC5vcm5sLmdvdj6IRgQQ +EQIABgUCPU77FAAKCRDaKMI6ef9EdBjQAJ41hqQaE3W2dHgN9otb7fL0n6U1YACg +kI9DvFQ1YmpLI8jdGwbDxDodAeOIRgQQEQIABgUCQpNYLAAKCRCL2C5vMLlLXMrg +AJ90LwV+nd+U4GEvzYixFvksHvtFGgCggD3NDeGXlgUhPB+nqyBq2QKfZxKIRgQQ +EQIABgUCQ2JoPAAKCRC5P2DGtcTOE4WfAJ4uxTyLyO4NCBk/IlTM0NAKLFHJgwCc +DP0YQC0oDm5uJ8/ZIkl0MUrzKXGIRgQQEQIABgUCRAh7hwAKCRBUFGa+sS3BmyTW +AJ4+X1CGNorq+Nme5tTIVskgYKH7wQCcD7UpPt2+r+NcGSYftkKk3O8R8TKIRgQS +EQIABgUCP7xVUQAKCRDSlFw7wTZQtolWAJ98yLyyC6jzrF/YG5kqeGqHSNdKtQCd +EdCDkGG09QJX8gFfZ/r8lWlflj+IRgQTEQIABgUCP1BY+wAKCRA4WcA7LiNuR4mz +AKC/1XBB9cBCs8X/KvoLLQP75q0i2QCbBb0UoVSUYgsdETzujbTwg+0HLseIRgQT +EQIABgUCP3H7JQAKCRAN3KoyeNUmTql1AJsEhcfoOC2U4JjHR6rWzqinaIxcNgCg +lmdHMQ3L8zCfNzD7lehquPy2P0eIRgQTEQIABgUCQJjhDgAKCRALtRSl0G7Wn+1r +AJ4nUVrAEtL+XBp2UU1QmVCxa7lcSwCfT8ds7xZ++aZomPK2Xvz230WnUsGIVQQT +EQIAFQUCN+3v9gMLCgMDFQMCAxYCAQIXgAAKCRD90t6s0zPLocAwAKCJ4wBEND4W +mzs6Sp47mWBsp96HRACfTH+SGkDfLqgkZ7JgEgzSDKGl4TyIXQQTEQIAFQUCN+3v +9gMLCgMDFQMCAxYCAQIXgAASCRD90t6s0zPLoQdlR1BHAAEBwDAAoInjAEQ0Phab +OzpKnjuZYGyn3odEAJ9Mf5IaQN8uqCRnsmASDNIMoaXhPLkBDQQ37cmSEAQAx3xz +BZlJikWJaiZGru3cEKYYnRFp8No2b4jhBwY9nKn8UIxuY5aQN4ka/k81wqjlC6cT +wn5R7kg2ha8eGXpwYhKGwn5MGvIxqfoj2tsQ76uluTowHA4seoavi7RGEDzm4Vpt +8Nua8krrZ2QPtLA86gkzL1QG5Bbv/o2Ldx8HHNcAAwcEAKcK2tj2X8RPgUarczXv +rdXMteeSFnI7fagbLpEfaTI2xa1ADLg5UO4M9Erz9m6k6xV6loxcBB9H5Ljm9GWf +el4T4p1lwzi3Lu5hKzIiFs+5vsy+fyEai4e5f6v9Ww3Q3Ec6UZpPZGyN+PDPlZxe +rf3ZIMogSGrrEBhprhLHReudiE4EGBECAAYFAjftyZIAEgkQ/dLerNMzy6EHZUdQ +RwABAQXiAKCilmALgD6mhccl4ISaUB5LfW74BQCgqd7wIfbV2+NKqf1Yuj75sryW +Ke4= +=zRdO +-----END PGP PUBLIC KEY BLOCK-----