commit 3543ae123352653f2a63267fbc4bae8eca1590a7 Author: overweight <5324761+overweight@user.noreply.gitee.com> Date: Mon Sep 30 10:35:48 2019 -0400 Package init diff --git a/bugfix-dummy_help2man.patch b/bugfix-dummy_help2man.patch new file mode 100644 index 0000000..c4fc4f0 --- /dev/null +++ b/bugfix-dummy_help2man.patch @@ -0,0 +1,20 @@ +diff --git a/man/local.mk b/man/local.mk +index 05d4401..601f459 100644 +--- a/man/local.mk ++++ b/man/local.mk +@@ -23,11 +23,11 @@ if CROSS_COMPILING + run_help2man = $(SHELL) $(srcdir)/man/dummy-man + else + ## Graceful degradation for systems lacking perl. +-if HAVE_PERL +-run_help2man = $(PERL) -- $(srcdir)/man/help2man +-else ++#if HAVE_PERL ++#run_help2man = $(PERL) -- $(srcdir)/man/help2man ++#else + run_help2man = $(SHELL) $(srcdir)/man/dummy-man +-endif ++#endif + endif + + man1_MANS = @man1_MANS@ diff --git a/bugfix-remove-usr-local-lib-from-m4.patch b/bugfix-remove-usr-local-lib-from-m4.patch new file mode 100644 index 0000000..8537a91 --- /dev/null +++ b/bugfix-remove-usr-local-lib-from-m4.patch @@ -0,0 +1,21 @@ +diff --git a/m4/getloadavg.m4 b/m4/getloadavg.m4 +index c9f5a6d..e8f94dd 100644 +--- a/m4/getloadavg.m4 ++++ b/m4/getloadavg.m4 +@@ -41,16 +41,6 @@ AC_CHECK_FUNC([getloadavg], [], + [LIBS="-lutil $LIBS" gl_func_getloadavg_done=yes]) + fi + +- if test $gl_func_getloadavg_done = no; then +- # There is a commonly available library for RS/6000 AIX. +- # Since it is not a standard part of AIX, it might be installed locally. +- gl_getloadavg_LIBS=$LIBS +- LIBS="-L/usr/local/lib $LIBS" +- AC_CHECK_LIB([getloadavg], [getloadavg], +- [LIBS="-lgetloadavg $LIBS" gl_func_getloadavg_done=yes], +- [LIBS=$gl_getloadavg_LIBS]) +- fi +- + # Set up the replacement function if necessary. + if test $gl_func_getloadavg_done = no; then + HAVE_GETLOADAVG=0 diff --git a/bugfix-selinux-flask.patch b/bugfix-selinux-flask.patch new file mode 100644 index 0000000..9f82f4d --- /dev/null +++ b/bugfix-selinux-flask.patch @@ -0,0 +1,17 @@ +diff --git a/m4/gnulib-comp.m4 b/m4/gnulib-comp.m4 +index 1401e52..e98db6f 100644 +--- a/m4/gnulib-comp.m4 ++++ b/m4/gnulib-comp.m4 +@@ -1751,11 +1751,11 @@ AC_DEFUN([gl_INIT], + AC_LIBOBJ([select]) + fi + gl_SYS_SELECT_MODULE_INDICATOR([select]) +- AC_CHECK_HEADERS([selinux/flask.h]) + AC_LIBOBJ([selinux-at]) + gl_HEADERS_SELINUX_SELINUX_H + gl_HEADERS_SELINUX_CONTEXT_H + if test "$with_selinux" != no && test "$ac_cv_header_selinux_selinux_h" = yes; then ++ AC_CHECK_HEADERS([selinux/flask.h]) + AC_LIBOBJ([getfilecon]) + fi + gl_SERVENT 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-6.10-manpages.patch b/coreutils-6.10-manpages.patch new file mode 100644 index 0000000..5aacff7 --- /dev/null +++ b/coreutils-6.10-manpages.patch @@ -0,0 +1,14 @@ +diff --git a/src/md5sum.c b/src/md5sum.c +index 8e21609..a857d62 100644 +--- a/src/md5sum.c ++++ b/src/md5sum.c +@@ -265,6 +265,9 @@ Print or check %s (%d-bit) checksums.\n\ + else + fputs (_("\ + -t, --text read in text mode (default)\n\ ++"), stdout); ++ fputs (_("\ ++ Note: There is no difference between binary and text mode option on GNU system.\n\ + "), stdout); + fputs (_("\ + -z, --zero end each output line with NUL, not newline,\n\ 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-8.25-DIR_COLORS.patch b/coreutils-8.25-DIR_COLORS.patch new file mode 100644 index 0000000..5490258 --- /dev/null +++ b/coreutils-8.25-DIR_COLORS.patch @@ -0,0 +1,686 @@ +From a13bc34f1eeebdf8b87e4b5a570341bb77a62f76 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 + +--- + DIR_COLORS | 41 ++++--- + DIR_COLORS.256color | 300 ++++++++++++++++++++++++------------------------ + DIR_COLORS.lightbgcolor | 211 ++++++++++++++++++---------------- + 3 files changed, 283 insertions(+), 269 deletions(-) + +diff --git a/DIR_COLORS b/DIR_COLORS +index d2ea453..27af9d7 100644 +--- a/DIR_COLORS ++++ b/DIR_COLORS +@@ -1,6 +1,10 @@ + # Configuration file for dircolors, a utility to help you set the + # LS_COLORS environment variable used by GNU ls with the --color option. + ++# This file goes in the /etc directory, and must be world readable. ++# You can copy this file to .dir_colors in your $HOME directory to override ++# the system defaults. ++ + # Copyright (C) 1996-2018 Free Software Foundation, Inc. + # Copying and distribution of this file, with or without modification, + # are permitted provided the copyright notice and this notice are preserved. +@@ -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. ++ + # 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 +@@ -56,7 +63,7 @@ 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;05;37;41 # ... and the files they point to + SETUID 37;41 # file that is setuid (u+s) + SETGID 30;43 # file that is setgid (g+s) + CAPABILITY 30;41 # file with capability +@@ -184,21 +191,21 @@ EXEC 01;32 + .ogx 01;35 + + # audio formats +-.aac 00;36 +-.au 00;36 +-.flac 00;36 +-.m4a 00;36 +-.mid 00;36 +-.midi 00;36 +-.mka 00;36 +-.mp3 00;36 +-.mpc 00;36 +-.ogg 00;36 +-.ra 00;36 +-.wav 00;36 ++.aac 01;36 ++.au 01;36 ++.flac 01;36 ++.m4a 01;36 ++.mid 01;36 ++.midi 01;36 ++.mka 01;36 ++.mp3 01;36 ++.mpc 01;36 ++.ogg 01;36 ++.ra 01;36 ++.wav 01;36 + + # https://wiki.xiph.org/MIME_Types_and_File_Extensions +-.oga 00;36 +-.opus 00;36 +-.spx 00;36 +-.xspf 00;36 ++.oga 01;36 ++.opus 01;36 ++.spx 01;36 ++.xspf 01;36 +diff --git a/DIR_COLORS.256color b/DIR_COLORS.256color +index d2ea453..74c34ba 100644 +--- a/DIR_COLORS.256color ++++ b/DIR_COLORS.256color +@@ -1,3 +1,9 @@ ++# Configuration file for the 256color ls utility ++ ++# This file goes in the /etc directory, and must be world readable. ++# You can copy this file to .dir_colors in your $HOME directory to override ++# the system defaults. ++ + # Configuration file for dircolors, a utility to help you set the + # LS_COLORS environment variable used by GNU ls with the --color option. + +@@ -8,32 +14,13 @@ + # The keywords COLOR, OPTIONS, and EIGHTBIT (honored by the + # slackware version of dircolors) are recognized but ignored. + ++# For compatibility, the pattern "^COLOR.*none" is recognized as a way to ++# disable colorization. See https://bugzilla.redhat.com/1349579 for details. ++ + # Below are TERM entries, which can be a glob patterns, to match + # against the TERM environment variable to determine if it is colorizable. +-TERM Eterm +-TERM ansi +-TERM *color* +-TERM con[0-9]*x[0-9]* +-TERM cons25 +-TERM console +-TERM cygwin +-TERM dtterm +-TERM gnome +-TERM hurd +-TERM jfbterm +-TERM konsole +-TERM kterm +-TERM linux +-TERM linux-c +-TERM mlterm +-TERM putty +-TERM rxvt* +-TERM screen* +-TERM st +-TERM terminator +-TERM tmux* +-TERM vt100 +-TERM xterm* ++TERM *256color* ++TERM rxvt-unicode256 + + # Below are the color init strings for the basic file types. A color init + # string consists of one or more of the following numeric codes: +@@ -43,29 +30,40 @@ TERM xterm* + # 30=black 31=red 32=green 33=yellow 34=blue 35=magenta 36=cyan 37=white + # Background color codes: + # 40=black 41=red 42=green 43=yellow 44=blue 45=magenta 46=cyan 47=white ++# Text color(256 colors mode) codes: ++# Valid syntax for text 256color is 38;5; , where color number ++# is number between 0 and 255. ++# You may find following command useful to search the best one for you: ++# for ((x=0; x<=255; x++));do echo -e "${x}:\033[38;5;${x}mcolor\033[000m";done ++# Background color(256 colors mode) codes: ++# Valid syntax for background 256color is 48;5; , where ++# color number is number between 0 and 255. ++# You may find following command useful to search the best one for you: ++# for ((x=0; x<=255; x++));do echo -e "${x}:\033[48;5;${x}mcolor\033[000m";done ++ + #NORMAL 00 # no color code at all + #FILE 00 # regular file: use no color at all + RESET 0 # reset to "normal" color +-DIR 01;34 # directory +-LINK 01;36 # symbolic link. (If you set this to 'target' instead of a ++DIR 38;5;33 # directory ++LINK 38;5;51 # symbolic link. (If you set this to 'target' instead of a + # numerical value, the color is as for the file pointed to.) + MULTIHARDLINK 00 # regular file with more than one link +-FIFO 40;33 # pipe +-SOCK 01;35 # socket +-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 +-SETUID 37;41 # file that is setuid (u+s) +-SETGID 30;43 # file that is setgid (g+s) +-CAPABILITY 30;41 # file with capability +-STICKY_OTHER_WRITABLE 30;42 # dir that is sticky and other-writable (+t,o+w) +-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 ++FIFO 40;38;5;11 # pipe ++SOCK 38;5;13 # socket ++DOOR 38;5;5 # door ++BLK 48;5;232;38;5;11 # block device driver ++CHR 48;5;232;38;5;3 # character device driver ++ORPHAN 48;5;232;38;5;9 # symlink to nonexistent file, or non-stat'able file ... ++MISSING 01;05;37;41 # ... and the files they point to ++SETUID 48;5;196;38;5;15 # file that is setuid (u+s) ++SETGID 48;5;11;38;5;16 # file that is setgid (g+s) ++CAPABILITY 48;5;196;38;5;226 # file with capability ++STICKY_OTHER_WRITABLE 48;5;10;38;5;16 # dir that is sticky and other-writable (+t,o+w) ++OTHER_WRITABLE 48;5;10;38;5;21 # dir that is other-writable (o+w) and not sticky ++STICKY 48;5;21;38;5;15 # dir with the sticky bit set (+t) and not other-writable + + # This is for files with execute permission: +-EXEC 01;32 ++EXEC 38;5;40 + + # 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. +@@ -83,122 +81,122 @@ EXEC 01;32 + #.csh 01;32 + + # archives or compressed (bright red) +-.tar 01;31 +-.tgz 01;31 +-.arc 01;31 +-.arj 01;31 +-.taz 01;31 +-.lha 01;31 +-.lz4 01;31 +-.lzh 01;31 +-.lzma 01;31 +-.tlz 01;31 +-.txz 01;31 +-.tzo 01;31 +-.t7z 01;31 +-.zip 01;31 +-.z 01;31 +-.dz 01;31 +-.gz 01;31 +-.lrz 01;31 +-.lz 01;31 +-.lzo 01;31 +-.xz 01;31 +-.zst 01;31 +-.tzst 01;31 +-.bz2 01;31 +-.bz 01;31 +-.tbz 01;31 +-.tbz2 01;31 +-.tz 01;31 +-.deb 01;31 +-.rpm 01;31 +-.jar 01;31 +-.war 01;31 +-.ear 01;31 +-.sar 01;31 +-.rar 01;31 +-.alz 01;31 +-.ace 01;31 +-.zoo 01;31 +-.cpio 01;31 +-.7z 01;31 +-.rz 01;31 +-.cab 01;31 +-.wim 01;31 +-.swm 01;31 +-.dwm 01;31 +-.esd 01;31 ++.tar 38;5;9 ++.tgz 38;5;9 ++.arc 38;5;9 ++.arj 38;5;9 ++.taz 38;5;9 ++.lha 38;5;9 ++.lz4 38;5;9 ++.lzh 38;5;9 ++.lzma 38;5;9 ++.tlz 38;5;9 ++.txz 38;5;9 ++.tzo 38;5;9 ++.t7z 38;5;9 ++.zip 38;5;9 ++.z 38;5;9 ++.dz 38;5;9 ++.gz 38;5;9 ++.lrz 38;5;9 ++.lz 38;5;9 ++.lzo 38;5;9 ++.xz 38;5;9 ++.zst 38;5;9 ++.tzst 38;5;9 ++.bz2 38;5;9 ++.bz 38;5;9 ++.tbz 38;5;9 ++.tbz2 38;5;9 ++.tz 38;5;9 ++.deb 38;5;9 ++.rpm 38;5;9 ++.jar 38;5;9 ++.war 38;5;9 ++.ear 38;5;9 ++.sar 38;5;9 ++.rar 38;5;9 ++.alz 38;5;9 ++.ace 38;5;9 ++.zoo 38;5;9 ++.cpio 38;5;9 ++.7z 38;5;9 ++.rz 38;5;9 ++.cab 38;5;9 ++.wim 38;5;9 ++.swm 38;5;9 ++.dwm 38;5;9 ++.esd 38;5;9 + + # image formats +-.jpg 01;35 +-.jpeg 01;35 +-.mjpg 01;35 +-.mjpeg 01;35 +-.gif 01;35 +-.bmp 01;35 +-.pbm 01;35 +-.pgm 01;35 +-.ppm 01;35 +-.tga 01;35 +-.xbm 01;35 +-.xpm 01;35 +-.tif 01;35 +-.tiff 01;35 +-.png 01;35 +-.svg 01;35 +-.svgz 01;35 +-.mng 01;35 +-.pcx 01;35 +-.mov 01;35 +-.mpg 01;35 +-.mpeg 01;35 +-.m2v 01;35 +-.mkv 01;35 +-.webm 01;35 +-.ogm 01;35 +-.mp4 01;35 +-.m4v 01;35 +-.mp4v 01;35 +-.vob 01;35 +-.qt 01;35 +-.nuv 01;35 +-.wmv 01;35 +-.asf 01;35 +-.rm 01;35 +-.rmvb 01;35 +-.flc 01;35 +-.avi 01;35 +-.fli 01;35 +-.flv 01;35 +-.gl 01;35 +-.dl 01;35 +-.xcf 01;35 +-.xwd 01;35 +-.yuv 01;35 +-.cgm 01;35 +-.emf 01;35 ++.jpg 38;5;13 ++.jpeg 38;5;13 ++.mjpg 38;5;13 ++.mjpeg 38;5;13 ++.gif 38;5;13 ++.bmp 38;5;13 ++.pbm 38;5;13 ++.pgm 38;5;13 ++.ppm 38;5;13 ++.tga 38;5;13 ++.xbm 38;5;13 ++.xpm 38;5;13 ++.tif 38;5;13 ++.tiff 38;5;13 ++.png 38;5;13 ++.svg 38;5;13 ++.svgz 38;5;13 ++.mng 38;5;13 ++.pcx 38;5;13 ++.mov 38;5;13 ++.mpg 38;5;13 ++.mpeg 38;5;13 ++.m2v 38;5;13 ++.mkv 38;5;13 ++.webm 38;5;13 ++.ogm 38;5;13 ++.mp4 38;5;13 ++.m4v 38;5;13 ++.mp4v 38;5;13 ++.vob 38;5;13 ++.qt 38;5;13 ++.nuv 38;5;13 ++.wmv 38;5;13 ++.asf 38;5;13 ++.rm 38;5;13 ++.rmvb 38;5;13 ++.flc 38;5;13 ++.avi 38;5;13 ++.fli 38;5;13 ++.flv 38;5;13 ++.gl 38;5;13 ++.dl 38;5;13 ++.xcf 38;5;13 ++.xwd 38;5;13 ++.yuv 38;5;13 ++.cgm 38;5;13 ++.emf 38;5;13 + + # https://wiki.xiph.org/MIME_Types_and_File_Extensions +-.ogv 01;35 +-.ogx 01;35 ++.ogv 38;5;13 ++.ogx 38;5;13 + + # audio formats +-.aac 00;36 +-.au 00;36 +-.flac 00;36 +-.m4a 00;36 +-.mid 00;36 +-.midi 00;36 +-.mka 00;36 +-.mp3 00;36 +-.mpc 00;36 +-.ogg 00;36 +-.ra 00;36 +-.wav 00;36 ++.aac 38;5;45 ++.au 38;5;45 ++.flac 38;5;45 ++.m4a 38;5;45 ++.mid 38;5;45 ++.midi 38;5;45 ++.mka 38;5;45 ++.mp3 38;5;45 ++.mpc 38;5;45 ++.ogg 38;5;45 ++.ra 38;5;45 ++.wav 38;5;45 + + # https://wiki.xiph.org/MIME_Types_and_File_Extensions +-.oga 00;36 +-.opus 00;36 +-.spx 00;36 +-.xspf 00;36 ++.oga 38;5;45 ++.opus 38;5;45 ++.spx 38;5;45 ++.xspf 38;5;45 +diff --git a/DIR_COLORS.lightbgcolor b/DIR_COLORS.lightbgcolor +index d2ea453..95d6879 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 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. + +@@ -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. ++ + # 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 +@@ -46,17 +55,17 @@ TERM xterm* + #NORMAL 00 # no color code at all + #FILE 00 # regular file: use no color at all + RESET 0 # reset to "normal" color +-DIR 01;34 # directory +-LINK 01;36 # symbolic link. (If you set this to 'target' instead of a ++DIR 00;34 # directory ++LINK 00;36 # symbolic link. (If you set this to 'target' instead of a + # numerical value, the color is as for the file pointed to.) + MULTIHARDLINK 00 # regular file with more than one link + FIFO 40;33 # pipe +-SOCK 01;35 # socket +-DOOR 01;35 # door ++SOCK 00;35 # socket ++DOOR 00;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;05;37;41 # ... and the files they point to + SETUID 37;41 # file that is setuid (u+s) + SETGID 30;43 # file that is setgid (g+s) + CAPABILITY 30;41 # file with capability +@@ -65,7 +74,7 @@ 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 files with execute permission: +-EXEC 01;32 ++EXEC 00;32 + + # 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. +@@ -83,105 +92,105 @@ EXEC 01;32 + #.csh 01;32 + + # archives or compressed (bright red) +-.tar 01;31 +-.tgz 01;31 +-.arc 01;31 +-.arj 01;31 +-.taz 01;31 +-.lha 01;31 +-.lz4 01;31 +-.lzh 01;31 +-.lzma 01;31 +-.tlz 01;31 +-.txz 01;31 +-.tzo 01;31 +-.t7z 01;31 +-.zip 01;31 +-.z 01;31 +-.dz 01;31 +-.gz 01;31 +-.lrz 01;31 +-.lz 01;31 +-.lzo 01;31 +-.xz 01;31 +-.zst 01;31 +-.tzst 01;31 +-.bz2 01;31 +-.bz 01;31 +-.tbz 01;31 +-.tbz2 01;31 +-.tz 01;31 +-.deb 01;31 +-.rpm 01;31 +-.jar 01;31 +-.war 01;31 +-.ear 01;31 +-.sar 01;31 +-.rar 01;31 +-.alz 01;31 +-.ace 01;31 +-.zoo 01;31 +-.cpio 01;31 +-.7z 01;31 +-.rz 01;31 +-.cab 01;31 +-.wim 01;31 +-.swm 01;31 +-.dwm 01;31 +-.esd 01;31 ++.tar 00;31 ++.tgz 00;31 ++.arc 00;31 ++.arj 00;31 ++.taz 00;31 ++.lha 00;31 ++.lz4 00;31 ++.lzh 00;31 ++.lzma 00;31 ++.tlz 00;31 ++.txz 00;31 ++.tzo 00;31 ++.t7z 00;31 ++.zip 00;31 ++.z 00;31 ++.dz 00;31 ++.gz 00;31 ++.lrz 00;31 ++.lz 00;31 ++.lzo 00;31 ++.xz 00;31 ++.zst 00;31 ++.tzst 00;31 ++.bz2 00;31 ++.bz 00;31 ++.tbz 00;31 ++.tbz2 00;31 ++.tz 00;31 ++.deb 00;31 ++.rpm 00;31 ++.jar 00;31 ++.war 00;31 ++.ear 00;31 ++.sar 00;31 ++.rar 00;31 ++.alz 00;31 ++.ace 00;31 ++.zoo 00;31 ++.cpio 00;31 ++.7z 00;31 ++.rz 00;31 ++.cab 00;31 ++.wim 00;31 ++.swm 00;31 ++.dwm 00;31 ++.esd 00;31 + + # image formats +-.jpg 01;35 +-.jpeg 01;35 +-.mjpg 01;35 +-.mjpeg 01;35 +-.gif 01;35 +-.bmp 01;35 +-.pbm 01;35 +-.pgm 01;35 +-.ppm 01;35 +-.tga 01;35 +-.xbm 01;35 +-.xpm 01;35 +-.tif 01;35 +-.tiff 01;35 +-.png 01;35 +-.svg 01;35 +-.svgz 01;35 +-.mng 01;35 +-.pcx 01;35 +-.mov 01;35 +-.mpg 01;35 +-.mpeg 01;35 +-.m2v 01;35 +-.mkv 01;35 +-.webm 01;35 +-.ogm 01;35 +-.mp4 01;35 +-.m4v 01;35 +-.mp4v 01;35 +-.vob 01;35 +-.qt 01;35 +-.nuv 01;35 +-.wmv 01;35 +-.asf 01;35 +-.rm 01;35 +-.rmvb 01;35 +-.flc 01;35 +-.avi 01;35 +-.fli 01;35 +-.flv 01;35 +-.gl 01;35 +-.dl 01;35 +-.xcf 01;35 +-.xwd 01;35 +-.yuv 01;35 +-.cgm 01;35 +-.emf 01;35 ++.jpg 00;35 ++.jpeg 00;35 ++.mjpg 00;35 ++.mjpeg 00;35 ++.gif 00;35 ++.bmp 00;35 ++.pbm 00;35 ++.pgm 00;35 ++.ppm 00;35 ++.tga 00;35 ++.xbm 00;35 ++.xpm 00;35 ++.tif 00;35 ++.tiff 00;35 ++.png 00;35 ++.svg 00;35 ++.svgz 00;35 ++.mng 00;35 ++.pcx 00;35 ++.mov 00;35 ++.mpg 00;35 ++.mpeg 00;35 ++.m2v 00;35 ++.mkv 00;35 ++.webm 00;35 ++.ogm 00;35 ++.mp4 00;35 ++.m4v 00;35 ++.mp4v 00;35 ++.vob 00;35 ++.qt 00;35 ++.nuv 00;35 ++.wmv 00;35 ++.asf 00;35 ++.rm 00;35 ++.rmvb 00;35 ++.flc 00;35 ++.avi 00;35 ++.fli 00;35 ++.flv 00;35 ++.gl 00;35 ++.dl 00;35 ++.xcf 00;35 ++.xwd 00;35 ++.yuv 00;35 ++.cgm 00;35 ++.emf 00;35 + + # https://wiki.xiph.org/MIME_Types_and_File_Extensions +-.ogv 01;35 +-.ogx 01;35 ++.ogv 00;35 ++.ogx 00;35 + + # audio formats + .aac 00;36 +-- +2.5.5 + diff --git a/coreutils-8.26-selinuxenable.patch b/coreutils-8.26-selinuxenable.patch new file mode 100644 index 0000000..d9b625a --- /dev/null +++ b/coreutils-8.26-selinuxenable.patch @@ -0,0 +1,26 @@ +From 6880c3dc9098b3337612850d1500b474aeb944ca Mon Sep 17 00:00:00 2001 +From: Kamil Dudka +Date: Tue, 29 Aug 2017 17:33:51 +0200 +Subject: [PATCH] require_selinux_(): use selinuxenabled(8) if available + +--- + init.cfg | 3 +++ + 1 file changed, 3 insertions(+) + +diff --git a/init.cfg b/init.cfg +index af6b581..f887b3a 100644 +--- a/init.cfg ++++ b/init.cfg +@@ -114,6 +114,9 @@ require_selinux_() + grep 'selinuxfs$' /proc/filesystems > /dev/null \ + || skip_ "this system lacks SELinux support" + ++ # use the 'selinuxenabled' utility if available ++ selinuxenabled; [ $? = 1 ] && skip_ "SELinux is disabled" ++ + # Independent of whether SELinux is enabled system-wide, + # the current file system may lack SELinux support. + # Also the current build may have SELinux support disabled. +-- +2.9.5 + diff --git a/coreutils-8.26-test-lock.patch b/coreutils-8.26-test-lock.patch new file mode 100644 index 0000000..d66928c --- /dev/null +++ b/coreutils-8.26-test-lock.patch @@ -0,0 +1,29 @@ +From 0d04ee8ddedb2bf33d64f148f246a3b7ec4fef21 Mon Sep 17 00:00:00 2001 +From: Kamil Dudka +Date: Mon, 23 Jan 2017 12:35:41 +0100 +Subject: [PATCH] test-lock: disable the rwlock test + +It hangs indefinitely if the system rwlock implementation does not +prevent writer starvation (and glibc does not implement it). + +Bug: http://www.mail-archive.com/bug-gnulib@gnu.org/msg33017.html +--- + gnulib-tests/test-lock.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/gnulib-tests/test-lock.c b/gnulib-tests/test-lock.c +index aa6de27..5af0a6c 100644 +--- a/gnulib-tests/test-lock.c ++++ b/gnulib-tests/test-lock.c +@@ -42,7 +42,7 @@ + Uncomment some of these, to verify that all tests crash if no locking + is enabled. */ + #define DO_TEST_LOCK 1 +-#define DO_TEST_RWLOCK 1 ++#define DO_TEST_RWLOCK 0 + #define DO_TEST_RECURSIVE_LOCK 1 + #define DO_TEST_ONCE 1 + +-- +2.7.4 + diff --git a/coreutils-8.30-renameatu.patch b/coreutils-8.30-renameatu.patch new file mode 100644 index 0000000..a5aa48f --- /dev/null +++ b/coreutils-8.30-renameatu.patch @@ -0,0 +1,451 @@ +From 57ee8db4fee8eb6772df1ff18d275594c0b034d4 Mon Sep 17 00:00:00 2001 +From: Paul Eggert +Date: Thu, 5 Jul 2018 09:22:09 -0700 +Subject: [PATCH 1/2] renameatu: rename from renameat2 + +It's looking like Glibc will add a renameat2 function +that is incompatible with Gnulib renameat2; see: +https://sourceware.org/ml/libc-alpha/2018-07/msg00064.html +To help avoid future confusion, rename renameat2 to something else. +Use the name 'renameatu', as the Gnulib function is close to the +Glibc function. Perhaps someday there will also be a renameat2 +Gnulib module, which mimicks the future glibc renameat2, but that +can wait as nobody seems to need such a module now. +* NEWS: Mention this. +* lib/renameatu.c: Rename from lib/renameat2.c. +* lib/renameatu.h: Rename from lib/renameat2.h. +* modules/renameatu: Rename from modules/renameat2. +* modules/renameatu-tests: Rename from modules/renameat2-tests. +All uses of "renameat2" in identifiers or file name +changed to "renameatu", except for two instances in +lib/renameatu.c that deal with the Linux kernel's +renameat2 syscall. + +Upstream-commit: 2522322e5304e7d86c63e607e2bc83c8d8b0a889 +Signed-off-by: Kamil Dudka +--- + gnulib-tests/gnulib.mk | 12 +++--- + .../{test-renameat2.c => test-renameatu.c} | 48 +++++++++++----------- + lib/backupfile.c | 4 +- + lib/gnulib.mk | 10 ++--- + lib/renameat.c | 4 +- + lib/{renameat2.c => renameatu.c} | 9 ++-- + lib/{renameat2.h => renameatu.h} | 8 ++-- + 7 files changed, 48 insertions(+), 47 deletions(-) + rename gnulib-tests/{test-renameat2.c => test-renameatu.c} (80%) + rename lib/{renameat2.c => renameatu.c} (94%) + rename lib/{renameat2.h => renameatu.h} (84%) + +diff --git a/gnulib-tests/gnulib.mk b/gnulib-tests/gnulib.mk +index be2b99e..891425b 100644 +--- a/gnulib-tests/gnulib.mk ++++ b/gnulib-tests/gnulib.mk +@@ -1750,14 +1750,14 @@ EXTRA_DIST += test-rename.h test-renameat.c signature.h macros.h + + ## end gnulib module renameat-tests + +-## begin gnulib module renameat2-tests ++## begin gnulib module renameatu-tests + +-TESTS += test-renameat2 +-check_PROGRAMS += test-renameat2 +-test_renameat2_LDADD = $(LDADD) @LIBINTL@ +-EXTRA_DIST += test-rename.h test-renameat2.c signature.h macros.h ++TESTS += test-renameatu ++check_PROGRAMS += test-renameatu ++test_renameatu_LDADD = $(LDADD) @LIBINTL@ ++EXTRA_DIST += test-rename.h test-renameatu.c signature.h macros.h + +-## end gnulib module renameat2-tests ++## end gnulib module renameatu-tests + + ## begin gnulib module rmdir-tests + +diff --git a/gnulib-tests/test-renameat2.c b/gnulib-tests/test-renameatu.c +similarity index 80% +rename from gnulib-tests/test-renameat2.c +rename to gnulib-tests/test-renameatu.c +index 0104890..988428b 100644 +--- a/gnulib-tests/test-renameat2.c ++++ b/gnulib-tests/test-renameatu.c +@@ -1,4 +1,4 @@ +-/* Test renameat2. ++/* Test renameatu. + Copyright (C) 2009-2018 Free Software Foundation, Inc. + + This program is free software: you can redistribute it and/or modify +@@ -18,12 +18,12 @@ + + #include + +-#include ++#include + + #include + + #include "signature.h" +-SIGNATURE_CHECK (renameat2, int, ++SIGNATURE_CHECK (renameatu, int, + (int, char const *, int, char const *, unsigned int)); + + #include +@@ -39,18 +39,18 @@ SIGNATURE_CHECK (renameat2, int, + #include "ignore-value.h" + #include "macros.h" + +-#define BASE "test-renameat2.t" ++#define BASE "test-renameatu.t" + + #include "test-rename.h" + + static int dfd1 = AT_FDCWD; + static int dfd2 = AT_FDCWD; + +-/* Wrapper to test renameat2 like rename. */ ++/* Wrapper to test renameatu like rename. */ + static int + do_rename (char const *name1, char const *name2) + { +- return renameat2 (dfd1, name1, dfd2, name2, 0); ++ return renameatu (dfd1, name1, dfd2, name2, 0); + } + + int +@@ -67,24 +67,24 @@ main (void) + /* Test behaviour for invalid file descriptors. */ + { + errno = 0; +- ASSERT (renameat2 (-1, "foo", AT_FDCWD, "bar", 0) == -1); ++ ASSERT (renameatu (-1, "foo", AT_FDCWD, "bar", 0) == -1); + ASSERT (errno == EBADF); + } + { + close (99); + errno = 0; +- ASSERT (renameat2 (99, "foo", AT_FDCWD, "bar", 0) == -1); ++ ASSERT (renameatu (99, "foo", AT_FDCWD, "bar", 0) == -1); + ASSERT (errno == EBADF); + } + ASSERT (close (creat (BASE "oo", 0600)) == 0); + { + errno = 0; +- ASSERT (renameat2 (AT_FDCWD, BASE "oo", -1, "bar", 0) == -1); ++ ASSERT (renameatu (AT_FDCWD, BASE "oo", -1, "bar", 0) == -1); + ASSERT (errno == EBADF); + } + { + errno = 0; +- ASSERT (renameat2 (AT_FDCWD, BASE "oo", 99, "bar", 0) == -1); ++ ASSERT (renameatu (AT_FDCWD, BASE "oo", 99, "bar", 0) == -1); + ASSERT (errno == EBADF); + } + ASSERT (unlink (BASE "oo") == 0); +@@ -133,13 +133,13 @@ main (void) + + ASSERT (sprintf (strchr (file1, '\0') - 2, "%02d", i) == 2); + ASSERT (sprintf (strchr (file2, '\0') - 2, "%02d", i + 1) == 2); +- ASSERT (renameat2 (fd1, file1, fd2, file2, 0) == 0); ++ ASSERT (renameatu (fd1, file1, fd2, file2, 0) == 0); + free (file1); + free (file2); + } + dfd2 = open ("..", O_RDONLY); + ASSERT (0 <= dfd2); +- ASSERT (renameat2 (dfd, "../" BASE "16", dfd2, BASE "17", 0) == 0); ++ ASSERT (renameatu (dfd, "../" BASE "16", dfd2, BASE "17", 0) == 0); + ASSERT (close (dfd2) == 0); + + /* Now we change back to the parent directory, and set dfd to "."; +@@ -152,47 +152,47 @@ main (void) + + ASSERT (close (creat (BASE "sub2/file", 0600)) == 0); + errno = 0; +- ASSERT (renameat2 (dfd, BASE "sub1", dfd, BASE "sub2", 0) == -1); ++ ASSERT (renameatu (dfd, BASE "sub1", dfd, BASE "sub2", 0) == -1); + ASSERT (errno == EEXIST || errno == ENOTEMPTY); + ASSERT (unlink (BASE "sub2/file") == 0); + errno = 0; +- ASSERT (renameat2 (dfd, BASE "sub2", dfd, BASE "sub1/.", 0) == -1); ++ ASSERT (renameatu (dfd, BASE "sub2", dfd, BASE "sub1/.", 0) == -1); + ASSERT (errno == EINVAL || errno == EISDIR || errno == EBUSY + || errno == ENOTEMPTY || errno == EEXIST + || errno == ENOENT /* WSL */); + errno = 0; +- ASSERT (renameat2 (dfd, BASE "sub2/.", dfd, BASE "sub1", 0) == -1); ++ ASSERT (renameatu (dfd, BASE "sub2/.", dfd, BASE "sub1", 0) == -1); + ASSERT (errno == EINVAL || errno == EBUSY || errno == EEXIST + || errno == ENOENT /* WSL */); + errno = 0; +- ASSERT (renameat2 (dfd, BASE "17", dfd, BASE "sub1", 0) == -1); ++ ASSERT (renameatu (dfd, BASE "17", dfd, BASE "sub1", 0) == -1); + ASSERT (errno == EISDIR); + errno = 0; +- ASSERT (renameat2 (dfd, BASE "nosuch", dfd, BASE "18", 0) == -1); ++ ASSERT (renameatu (dfd, BASE "nosuch", dfd, BASE "18", 0) == -1); + ASSERT (errno == ENOENT); + errno = 0; +- ASSERT (renameat2 (dfd, "", dfd, BASE "17", 0) == -1); ++ ASSERT (renameatu (dfd, "", dfd, BASE "17", 0) == -1); + ASSERT (errno == ENOENT); + errno = 0; +- ASSERT (renameat2 (dfd, BASE "17", dfd, "", 0) == -1); ++ ASSERT (renameatu (dfd, BASE "17", dfd, "", 0) == -1); + ASSERT (errno == ENOENT); + errno = 0; +- ASSERT (renameat2 (dfd, BASE "sub2", dfd, BASE "17", 0) == -1); ++ ASSERT (renameatu (dfd, BASE "sub2", dfd, BASE "17", 0) == -1); + ASSERT (errno == ENOTDIR); + errno = 0; +- ASSERT (renameat2 (dfd, BASE "17/", dfd, BASE "18", 0) == -1); ++ ASSERT (renameatu (dfd, BASE "17/", dfd, BASE "18", 0) == -1); + ASSERT (errno == ENOTDIR); + errno = 0; +- ASSERT (renameat2 (dfd, BASE "17", dfd, BASE "18/", 0) == -1); ++ ASSERT (renameatu (dfd, BASE "17", dfd, BASE "18/", 0) == -1); + ASSERT (errno == ENOTDIR || errno == ENOENT); + + /* Finally, make sure we cannot overwrite existing files. */ + ASSERT (close (creat (BASE "sub2/file", 0600)) == 0); + errno = 0; +- ASSERT ((renameat2 (dfd, BASE "sub2", dfd, BASE "sub1", RENAME_NOREPLACE) ++ ASSERT ((renameatu (dfd, BASE "sub2", dfd, BASE "sub1", RENAME_NOREPLACE) + == -1) + && errno == EEXIST); +- ASSERT ((renameat2 (dfd, BASE "sub2/file", dfd, BASE "17", RENAME_NOREPLACE) ++ ASSERT ((renameatu (dfd, BASE "sub2/file", dfd, BASE "17", RENAME_NOREPLACE) + == -1) + && errno == EEXIST); + +diff --git a/lib/backupfile.c b/lib/backupfile.c +index d438455..637be6c 100644 +--- a/lib/backupfile.c ++++ b/lib/backupfile.c +@@ -23,7 +23,7 @@ + #include "backup-internal.h" + + #include "dirname.h" +-#include "renameat2.h" ++#include "renameatu.h" + #include "xalloc-oversized.h" + + #include +@@ -353,7 +353,7 @@ backupfile_internal (char const *file, enum backup_type backup_type, bool rename + base_offset = 0; + } + unsigned flags = backup_type == simple_backups ? 0 : RENAME_NOREPLACE; +- if (renameat2 (AT_FDCWD, file, sdir, s + base_offset, flags) == 0) ++ if (renameatu (AT_FDCWD, file, sdir, s + base_offset, flags) == 0) + break; + int e = errno; + if (e != EEXIST) +diff --git a/lib/gnulib.mk b/lib/gnulib.mk +index 04473d5..0b747e3 100644 +--- a/lib/gnulib.mk ++++ b/lib/gnulib.mk +@@ -21,7 +21,7 @@ + # the same distribution terms as the rest of that program. + # + # Generated by gnulib-tool. +-# Reproduce by: gnulib-tool --import --local-dir=gl --lib=libcoreutils --source-base=lib --m4-base=m4 --doc-base=doc --tests-base=gnulib-tests --aux-dir=build-aux --with-tests --avoid=canonicalize-lgpl --avoid=dummy --makefile-name=gnulib.mk --no-conditional-dependencies --no-libtool --macro-prefix=gl acl alignof alloca announce-gen areadlink-with-size argmatch argv-iter assert autobuild backup-rename backupfile base32 base64 buffer-lcm c-strcase c-strtod c-strtold calloc-gnu canon-host canonicalize chown cloexec closein closeout config-h configmake crypto/md5 crypto/sha1 crypto/sha256 crypto/sha512 cycle-check d-ino d-type di-set diacrit dirfd dirname do-release-commit-and-tag dtoastr dup2 environ error euidaccess exclude exitfail explicit_bzero faccessat fadvise fchdir fchmodat fchownat fclose fcntl fcntl-safer fd-reopen fdatasync fdl fdopen fdutimensat file-has-acl file-type fileblocks filemode filenamecat filevercmp flexmember fnmatch-gnu fopen-safer fprintftime freopen freopen-safer fseeko fstatat fsusage fsync ftoastr ftruncate fts full-read full-write getgroups gethrxtime getline getloadavg getlogin getndelim2 getopt-gnu getpagesize getpass-gnu gettext-h gettime gettimeofday getugroups getusershell git-version-gen gitlog-to-changelog gnu-make gnu-web-doc-update gnumakefile gnupload group-member hard-locale hash hash-pjw heap host-os human idcache ignore-value inttostr inttypes isapipe isatty isblank largefile lchmod lchown ldtoastr lib-ignore linebuffer link link-follow linkat long-options lstat maintainer-makefile malloc-gnu manywarnings mbrlen mbrtowc mbsalign mbschr mbslen mbswidth memcasecmp memchr memcmp2 mempcpy memrchr mgetgroups mkancesdirs mkdir mkdir-p mkfifo mknod mkostemp mkstemp mktime modechange mountlist mpsort netinet_in non-recursive-gnulib-prefix-hack nproc nstrftime obstack open parse-datetime pathmax perl physmem pipe-posix pipe2 posix-shell posixtm posixver priv-set progname propername pthread putenv quote quotearg randint randperm read-file readlink readtokens readtokens0 readutmp realloc-gnu regex remove rename renameat renameat2 rmdir root-dev-ino rpmatch safe-read same save-cwd savedir savewd selinux-at setenv settime sig2str sigaction smack ssize_t stat-macros stat-size stat-time statat stdbool stdlib-safer stpcpy stpncpy strdup-posix strncat strnumcmp strsignal strtod strtoimax strtoumax symlinkat sys_ioctl sys_resource sys_stat sys_wait tempname termios time_rz timer-time timespec tzset uname unicodeio unistd-safer unlink-busy unlinkat unlocked-io unsetenv update-copyright uptime useless-if-before-free userspec utimecmp utimens vasprintf-posix vc-list-files verify verror version-etc-fsf wchar-single wcswidth wcwidth winsz-ioctl winsz-termios write-any-file xalloc xbinary-io xdectoint xfts xgetcwd xgetgroups xgethostname xmemcoll xnanosleep xprintf xprintf-posix xreadlink xstrtod xstrtoimax xstrtol xstrtold xstrtoumax year2038 yesno ++# Reproduce by: gnulib-tool --import --local-dir=gl --lib=libcoreutils --source-base=lib --m4-base=m4 --doc-base=doc --tests-base=gnulib-tests --aux-dir=build-aux --with-tests --avoid=canonicalize-lgpl --avoid=dummy --makefile-name=gnulib.mk --no-conditional-dependencies --no-libtool --macro-prefix=gl acl alignof alloca announce-gen areadlink-with-size argmatch argv-iter assert autobuild backup-rename backupfile base32 base64 buffer-lcm c-strcase c-strtod c-strtold calloc-gnu canon-host canonicalize chown cloexec closein closeout config-h configmake crypto/md5 crypto/sha1 crypto/sha256 crypto/sha512 cycle-check d-ino d-type di-set diacrit dirfd dirname do-release-commit-and-tag dtoastr dup2 environ error euidaccess exclude exitfail explicit_bzero faccessat fadvise fchdir fchmodat fchownat fclose fcntl fcntl-safer fd-reopen fdatasync fdl fdopen fdutimensat file-has-acl file-type fileblocks filemode filenamecat filevercmp flexmember fnmatch-gnu fopen-safer fprintftime freopen freopen-safer fseeko fstatat fsusage fsync ftoastr ftruncate fts full-read full-write getgroups gethrxtime getline getloadavg getlogin getndelim2 getopt-gnu getpagesize getpass-gnu gettext-h gettime gettimeofday getugroups getusershell git-version-gen gitlog-to-changelog gnu-make gnu-web-doc-update gnumakefile gnupload group-member hard-locale hash hash-pjw heap host-os human idcache ignore-value inttostr inttypes isapipe isatty isblank largefile lchmod lchown ldtoastr lib-ignore linebuffer link link-follow linkat long-options lstat maintainer-makefile malloc-gnu manywarnings mbrlen mbrtowc mbsalign mbschr mbslen mbswidth memcasecmp memchr memcmp2 mempcpy memrchr mgetgroups mkancesdirs mkdir mkdir-p mkfifo mknod mkostemp mkstemp mktime modechange mountlist mpsort netinet_in non-recursive-gnulib-prefix-hack nproc nstrftime obstack open parse-datetime pathmax perl physmem pipe-posix pipe2 posix-shell posixtm posixver priv-set progname propername pthread putenv quote quotearg randint randperm read-file readlink readtokens readtokens0 readutmp realloc-gnu regex remove rename renameat renameatu rmdir root-dev-ino rpmatch safe-read same save-cwd savedir savewd selinux-at setenv settime sig2str sigaction smack ssize_t stat-macros stat-size stat-time statat stdbool stdlib-safer stpcpy stpncpy strdup-posix strncat strnumcmp strsignal strtod strtoimax strtoumax symlinkat sys_ioctl sys_resource sys_stat sys_wait tempname termios time_rz timer-time timespec tzset uname unicodeio unistd-safer unlink-busy unlinkat unlocked-io unsetenv update-copyright uptime useless-if-before-free userspec utimecmp utimens vasprintf-posix vc-list-files verify verror version-etc-fsf wchar-single wcswidth wcwidth winsz-ioctl winsz-termios write-any-file xalloc xbinary-io xdectoint xfts xgetcwd xgetgroups xgethostname xmemcoll xnanosleep xprintf xprintf-posix xreadlink xstrtod xstrtoimax xstrtol xstrtold xstrtoumax year2038 yesno + + + MOSTLYCLEANFILES += lib/core lib/*.stackdump +@@ -3218,15 +3218,15 @@ EXTRA_lib_libcoreutils_a_SOURCES += lib/renameat.c + + ## end gnulib module renameat + +-## begin gnulib module renameat2 ++## begin gnulib module renameatu + +-lib_libcoreutils_a_SOURCES += lib/renameat2.c ++lib_libcoreutils_a_SOURCES += lib/renameatu.c + +-EXTRA_DIST += lib/at-func2.c lib/renameat2.h ++EXTRA_DIST += lib/at-func2.c lib/renameatu.h + + EXTRA_lib_libcoreutils_a_SOURCES += lib/at-func2.c + +-## end gnulib module renameat2 ++## end gnulib module renameatu + + ## begin gnulib module rewinddir + +diff --git a/lib/renameat.c b/lib/renameat.c +index 0cb7d33..67be22b 100644 +--- a/lib/renameat.c ++++ b/lib/renameat.c +@@ -16,10 +16,10 @@ + + #include + #include +-#include "renameat2.h" ++#include "renameatu.h" + + int + renameat (int fd1, char const *src, int fd2, char const *dst) + { +- return renameat2 (fd1, src, fd2, dst, 0); ++ return renameatu (fd1, src, fd2, dst, 0); + } +diff --git a/lib/renameat2.c b/lib/renameatu.c +similarity index 94% +rename from lib/renameat2.c +rename to lib/renameatu.c +index a295ec3..b013ad6 100644 +--- a/lib/renameat2.c ++++ b/lib/renameatu.c +@@ -18,7 +18,7 @@ + + #include + +-#include "renameat2.h" ++#include "renameatu.h" + + #include + #include +@@ -68,10 +68,13 @@ rename_noreplace (char const *src, char const *dst) + the restore_cwd fails, then give a diagnostic and exit nonzero. + + Obey FLAGS when doing the renaming. If FLAGS is zero, this +- function is equivalent to renameat (FD1, SRC, FD2, DST). */ ++ function is equivalent to renameat (FD1, SRC, FD2, DST). ++ Otherwise, attempt to implement FLAGS even if the implementation is ++ not atomic; this differs from the GNU/Linux native renameat2, ++ which fails if it cannot guarantee atomicity. */ + + int +-renameat2 (int fd1, char const *src, int fd2, char const *dst, ++renameatu (int fd1, char const *src, int fd2, char const *dst, + unsigned int flags) + { + int ret_val = -1; +diff --git a/lib/renameat2.h b/lib/renameatu.h +similarity index 84% +rename from lib/renameat2.h +rename to lib/renameatu.h +index aba7966..7d79775 100644 +--- a/lib/renameat2.h ++++ b/lib/renameatu.h +@@ -16,15 +16,13 @@ + + /* written by Paul Eggert */ + +-/* Get RENAME_* macros from linux/fs.h if present, otherwise supply ++/* Get RENAME_* macros from if present, otherwise supply + the traditional Linux values. */ +-#if HAVE_LINUX_FS_H +-# include +-#endif ++#include + #ifndef RENAME_NOREPLACE + # define RENAME_NOREPLACE (1 << 0) + # define RENAME_EXCHANGE (1 << 1) + # define RENAME_WHITEOUT (1 << 2) + #endif + +-extern int renameat2 (int, char const *, int, char const *, unsigned int); ++extern int renameatu (int, char const *, int, char const *, unsigned int); +-- +2.14.4 + + +From a6b7ff5ef538bbdff4550a56fed878e9cd951d6d Mon Sep 17 00:00:00 2001 +From: Paul Eggert +Date: Thu, 5 Jul 2018 09:33:48 -0700 +Subject: [PATCH 2/2] build: update gnulib submodule to latest + +* bootstrap.conf, src/copy.c, src/mv.c, src/shred.c: +Adjust to renaming of renameat2 to renameatu. + +Upstream-commit: 439741053256618eb651e6d43919df29625b8714 +Signed-off-by: Kamil Dudka +--- + bootstrap.conf | 2 +- + src/copy.c | 4 ++-- + src/mv.c | 4 ++-- + src/shred.c | 4 ++-- + 4 files changed, 7 insertions(+), 7 deletions(-) + +diff --git a/bootstrap.conf b/bootstrap.conf +index 4da4f94..fcf29dc 100644 +--- a/bootstrap.conf ++++ b/bootstrap.conf +@@ -210,7 +210,7 @@ gnulib_modules=" + remove + rename + renameat +- renameat2 ++ renameatu + rmdir + root-dev-ino + rpmatch +diff --git a/src/copy.c b/src/copy.c +index 58d2f6e..1a9cdd1 100644 +--- a/src/copy.c ++++ b/src/copy.c +@@ -53,7 +53,7 @@ + #include "ignore-value.h" + #include "ioblksize.h" + #include "quote.h" +-#include "renameat2.h" ++#include "renameatu.h" + #include "root-uid.h" + #include "same.h" + #include "savedir.h" +@@ -1873,7 +1873,7 @@ copy_internal (char const *src_name, char const *dst_name, + if (x->move_mode) + { + if (rename_errno < 0) +- rename_errno = (renameat2 (AT_FDCWD, src_name, AT_FDCWD, dst_name, ++ rename_errno = (renameatu (AT_FDCWD, src_name, AT_FDCWD, dst_name, + RENAME_NOREPLACE) + ? errno : 0); + new_dst = rename_errno == 0; +diff --git a/src/mv.c b/src/mv.c +index b6dd72d..36fd1af 100644 +--- a/src/mv.c ++++ b/src/mv.c +@@ -31,7 +31,7 @@ + #include "error.h" + #include "filenamecat.h" + #include "remove.h" +-#include "renameat2.h" ++#include "renameatu.h" + #include "root-dev-ino.h" + #include "priv-set.h" + +@@ -456,7 +456,7 @@ main (int argc, char **argv) + { + assert (2 <= n_files); + if (n_files == 2) +- x.rename_errno = (renameat2 (AT_FDCWD, file[0], AT_FDCWD, file[1], ++ x.rename_errno = (renameatu (AT_FDCWD, file[0], AT_FDCWD, file[1], + RENAME_NOREPLACE) + ? errno : 0); + if (x.rename_errno != 0 && target_directory_operand (file[n_files - 1])) +diff --git a/src/shred.c b/src/shred.c +index 2ddaadd..270b1e9 100644 +--- a/src/shred.c ++++ b/src/shred.c +@@ -93,7 +93,7 @@ + #include "human.h" + #include "randint.h" + #include "randread.h" +-#include "renameat2.h" ++#include "renameatu.h" + #include "stat-size.h" + + /* Default number of times to overwrite. */ +@@ -1096,7 +1096,7 @@ wipename (char *oldname, char const *qoldname, struct Options const *flags) + memset (base, nameset[0], len); + base[len] = 0; + bool rename_ok; +- while (! (rename_ok = (renameat2 (AT_FDCWD, oldname, AT_FDCWD, newname, ++ while (! (rename_ok = (renameatu (AT_FDCWD, oldname, AT_FDCWD, newname, + RENAME_NOREPLACE) + == 0)) + && errno == EEXIST && incname (base, len)) +-- +2.14.4 + diff --git a/coreutils-8.30.tar.xz b/coreutils-8.30.tar.xz new file mode 100644 index 0000000..0ea8f87 Binary files /dev/null and b/coreutils-8.30.tar.xz differ 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-colorls.csh b/coreutils-colorls.csh new file mode 100755 index 0000000..f631762 --- /dev/null +++ b/coreutils-colorls.csh @@ -0,0 +1,68 @@ +# skip everything for non-interactive shells +if (! $?prompt) exit + +# color-ls initialization +if ( $?USER_LS_COLORS ) then + if ( "$USER_LS_COLORS" != "" ) then + #when USER_LS_COLORS defined do not override user + #specified LS_COLORS and use them + goto finish + endif +endif + +alias ll 'ls -l' +alias l. 'ls -d .*' +set COLORS=/etc/DIR_COLORS + +if ($?TERM) then + if ( -e "/etc/DIR_COLORS.256color" ) then + if ( "`/usr/bin/tput colors`" == "256" ) then + set COLORS=/etc/DIR_COLORS.256color + endif + endif + if ( -e "/etc/DIR_COLORS.$TERM" ) then + set COLORS="/etc/DIR_COLORS.$TERM" + endif +endif +if ( -f ~/.dircolors ) set COLORS=~/.dircolors +if ( -f ~/.dir_colors ) set COLORS=~/.dir_colors +if ($?TERM) then + if ( -f ~/.dircolors."$TERM" ) set COLORS=~/.dircolors."$TERM" + if ( -f ~/.dir_colors."$TERM" ) set COLORS=~/.dir_colors."$TERM" +endif +set INCLUDE="`/usr/bin/cat "$COLORS" | /usr/bin/grep '^INCLUDE' | /usr/bin/cut -d ' ' -f2-`" + +if ( ! -e "$COLORS" ) exit + +set _tmp="`/usr/bin/mktemp .colorlsXXX -q --tmpdir=/tmp`" +#if mktemp fails, exit when include was active, otherwise use $COLORS file +if ( "$_tmp" == '' ) then + if ( "$INCLUDE" == '' ) then + eval "`/usr/bin/dircolors -c $COLORS`" + endif + goto cleanup +endif + +if ( "$INCLUDE" != '' ) /usr/bin/cat "$INCLUDE" >> $_tmp +/usr/bin/grep -v '^INCLUDE' "$COLORS" >> $_tmp + +eval "`/usr/bin/dircolors -c $_tmp`" + +/usr/bin/rm -f $_tmp + +if ( "$LS_COLORS" == '' ) exit +cleanup: +set color_none=`/usr/bin/sed -n '/^COLOR.*none/Ip' < $COLORS` +if ( "$color_none" != '' ) then + unset color_none + exit +endif +unset color_none +unset _tmp +unset INCLUDE +unset COLORS + +finish: +alias ll 'ls -l --color=auto' +alias l. 'ls -d .* --color=auto' +alias ls 'ls --color=auto' diff --git a/coreutils-colorls.sh b/coreutils-colorls.sh new file mode 100755 index 0000000..ac92268 --- /dev/null +++ b/coreutils-colorls.sh @@ -0,0 +1,57 @@ +# color-ls initialization + +# Skip all for noninteractive shells. +[ ! -t 0 ] && return + +#when USER_LS_COLORS defined do not override user LS_COLORS, but use them. +if [ -z "$USER_LS_COLORS" ]; then + + alias ll='ls -l' 2>/dev/null + alias l.='ls -d .*' 2>/dev/null + + INCLUDE= + COLORS= + + for colors in "$HOME/.dir_colors.$TERM" "$HOME/.dircolors.$TERM" \ + "$HOME/.dir_colors" "$HOME/.dircolors"; do + [ -e "$colors" ] && COLORS="$colors" && \ + INCLUDE="`/usr/bin/cat "$COLORS" | /usr/bin/grep '^INCLUDE' | /usr/bin/cut -d ' ' -f2-`" && \ + break + done + + [ -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" + + # Existence of $COLORS already checked above. + [ -n "$COLORS" ] || return + + if [ -e "$INCLUDE" ]; + then + TMP="`/usr/bin/mktemp .colorlsXXX -q --tmpdir=/tmp`" + [ -z "$TMP" ] && return + + /usr/bin/cat "$INCLUDE" >> $TMP + /usr/bin/grep -v '^INCLUDE' "$COLORS" >> $TMP + + eval "`/usr/bin/dircolors --sh $TMP 2>/dev/null`" + /usr/bin/rm -f $TMP + else + eval "`/usr/bin/dircolors --sh $COLORS 2>/dev/null`" + fi + + [ -z "$LS_COLORS" ] && return + /usr/bin/grep -qi "^COLOR.*none" $COLORS >/dev/null 2>/dev/null && return +fi + +unset TMP COLORS INCLUDE + +alias ll='ls -l --color=auto' 2>/dev/null +alias l.='ls -d .* --color=auto' 2>/dev/null +alias ls='ls --color=auto' 2>/dev/null diff --git a/coreutils-df-direct.patch b/coreutils-df-direct.patch new file mode 100644 index 0000000..248a6ae --- /dev/null +++ b/coreutils-df-direct.patch @@ -0,0 +1,174 @@ +diff --git a/doc/coreutils.texi b/doc/coreutils.texi +index a507280..400e135 100644 +--- a/doc/coreutils.texi ++++ b/doc/coreutils.texi +@@ -11303,6 +11303,13 @@ 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. + ++@item --direct ++@opindex --direct ++@cindex direct statfs for a file ++Do not resolve mount point and show statistics directly for a file. It can be ++especially useful for NFS mount points if there is a boundary between two ++storage policies behind the mount point. ++ + @item --total + @opindex --total + @cindex grand total of disk size, usage and available space +diff --git a/src/df.c b/src/df.c +index 8f760db..a7385fd 100644 +--- a/src/df.c ++++ b/src/df.c +@@ -120,6 +120,9 @@ static bool print_type; + /* If true, print a grand total at the end. */ + static bool print_grand_total; + ++/* If true, show statistics for a file instead of mount point. */ ++static bool direct_statfs; ++ + /* Grand total data. */ + static struct fs_usage grand_fsu; + +@@ -247,13 +250,15 @@ enum + NO_SYNC_OPTION = CHAR_MAX + 1, + SYNC_OPTION, + TOTAL_OPTION, +- OUTPUT_OPTION ++ OUTPUT_OPTION, ++ DIRECT_OPTION + }; + + static struct option const long_options[] = + { + {"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'}, +@@ -509,7 +514,10 @@ get_header (void) + for (col = 0; col < ncolumns; col++) + { + char *cell = NULL; +- char const *header = _(columns[col]->caption); ++ char const *header = (columns[col]->field == TARGET_FIELD ++ && direct_statfs)? ++ _("File") : ++ _(columns[col]->caption); + + if (columns[col]->field == SIZE_FIELD + && (header_mode == DEFAULT_MODE +@@ -1397,6 +1405,19 @@ get_point (const char *point, const struct stat *statp) + static void + get_entry (char const *name, struct stat const *statp) + { ++ if (direct_statfs) ++ { ++ char *resolved = canonicalize_file_name (name); ++ if (resolved) ++ { ++ char *mp = find_mount_point (name, statp); ++ get_dev (NULL, mp, resolved, NULL, NULL, false, false, NULL, false); ++ free(mp); ++ free (resolved); ++ return; ++ } ++ } ++ + if ((S_ISBLK (statp->st_mode) || S_ISCHR (statp->st_mode)) + && get_disk (name)) + return; +@@ -1467,6 +1488,7 @@ 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\ ++ --direct show statistics for a file instead of mount point\n\ + -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); +@@ -1557,6 +1579,9 @@ main (int argc, char **argv) + xstrtol_fatal (e, oi, c, long_options, optarg); + } + break; ++ case DIRECT_OPTION: ++ direct_statfs = true; ++ break; + case 'i': + if (header_mode == OUTPUT_MODE) + { +@@ -1653,6 +1678,13 @@ main (int argc, char **argv) + } + } + ++ if (direct_statfs && show_local_fs) ++ { ++ error (0, 0, _("options --direct and --local (-l) are mutually " ++ "exclusive")); ++ usage (EXIT_FAILURE); ++ } ++ + if (human_output_opts == -1) + { + if (posix_format) +diff --git a/tests/df/direct.sh b/tests/df/direct.sh +new file mode 100755 +index 0000000..8e4cfb8 +--- /dev/null ++++ b/tests/df/direct.sh +@@ -0,0 +1,55 @@ ++#!/bin/sh ++# Ensure "df --direct" works as documented ++ ++# Copyright (C) 2010 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=.}/init.sh"; path_prepend_ ../src ++print_ver_ df ++ ++df || skip_ "df fails" ++ ++DIR=`pwd` || framework_failure ++FILE="$DIR/file" ++touch "$FILE" || framework_failure ++echo "$FILE" > file_exp || framework_failure ++echo "Mounted on" > header_mounted_exp || framework_failure ++echo "File" > header_file_exp || framework_failure ++ ++fail=0 ++ ++df --portability "$FILE" > df_out || fail=1 ++df --portability --direct "$FILE" > df_direct_out || fail=1 ++df --portability --direct --local "$FILE" > /dev/null 2>&1 && fail=1 ++ ++# check df header ++$AWK '{ if (NR==1) print $6 " " $7; }' df_out > header_mounted_out \ ++ || framework_failure ++$AWK '{ if (NR==1) print $6; }' df_direct_out > header_file_out \ ++ || framework_failure ++compare header_mounted_out header_mounted_exp || fail=1 ++compare header_file_out header_file_exp || fail=1 ++ ++# check df output (without --direct) ++$AWK '{ if (NR==2) print $6; }' df_out > file_out \ ++ || framework_failure ++compare file_out file_exp && fail=1 ++ ++# check df output (with --direct) ++$AWK '{ if (NR==2) print $6; }' df_direct_out > file_out \ ++ || framework_failure ++compare file_out file_exp || fail=1 ++ ++Exit $fail diff --git a/coreutils-getgrouplist.patch b/coreutils-getgrouplist.patch new file mode 100644 index 0000000..5b35f42 --- /dev/null +++ b/coreutils-getgrouplist.patch @@ -0,0 +1,86 @@ +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 +@@ -121,9 +121,17 @@ 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 +153,25 @@ 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..b5f571f --- /dev/null +++ b/coreutils-i18n-expand-unexpand.patch @@ -0,0 +1,848 @@ +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/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/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 new file mode 100644 index 0000000..429675f --- /dev/null +++ b/coreutils-i18n.patch @@ -0,0 +1,3761 @@ +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 +--- + lib/linebuffer.h | 8 + + src/fold.c | 308 ++++++++++++++++-- + src/join.c | 359 ++++++++++++++++++--- + src/pr.c | 443 ++++++++++++++++++++++--- + src/sort.c | 764 +++++++++++++++++++++++++++++++++++++++++--- + src/uniq.c | 265 ++++++++++++++- + tests/i18n/sort.sh | 29 ++ + tests/local.mk | 2 + + tests/misc/expand.pl | 42 +++ + 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 +++ + 17 files changed, 2430 insertions(+), 160 deletions(-) + create mode 100755 tests/i18n/sort.sh + create mode 100755 tests/misc/sort-mb-tests.sh + +diff --git a/lib/linebuffer.h b/lib/linebuffer.h +index 64181af..9b8fe5a 100644 +--- a/lib/linebuffer.h ++++ b/lib/linebuffer.h +@@ -21,6 +21,11 @@ + + # include + ++/* Get mbstate_t. */ ++# if HAVE_WCHAR_H ++# include ++# endif ++ + /* A 'struct linebuffer' holds a line of text. */ + + struct linebuffer +@@ -28,6 +33,9 @@ struct linebuffer + size_t size; /* Allocated. */ + size_t length; /* Used. */ + char *buffer; ++# if HAVE_WCHAR_H ++ mbstate_t state; ++# endif + }; + + /* Initialize linebuffer LINEBUFFER for use. */ +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(), wcwidth(). */ ++#if HAVE_WCHAR_H ++# include ++#endif ++ ++/* Get iswprint(), iswblank(), wcwidth(). */ ++#if HAVE_WCTYPE_H ++# include ++#endif ++ + #include "system.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. */ ++#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 ++ + #define TAB_WIDTH 8 + + /* The official name of this program (e.g., no 'g' prefix). */ +@@ -35,20 +57,41 @@ + + #define AUTHORS proper_name ("David MacKenzie") + ++#define FATAL_ERROR(Message) \ ++ do \ ++ { \ ++ error (0, 0, (Message)); \ ++ usage (2); \ ++ } \ ++ while (0) ++ ++enum operating_mode ++{ ++ /* Fold texts by columns that are at the given positions. */ ++ column_mode, ++ ++ /* Fold texts by bytes that are at the given positions. */ ++ byte_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, try to break on whitespace. */ + static bool break_spaces; + +-/* If nonzero, count bytes, not column positions. */ +-static bool count_bytes; +- + /* If nonzero, at least one of the files we read was standard input. */ + static bool have_read_stdin; + +-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::"; + + 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 ++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) ++{ ++ char buf[MB_LEN_MAX + BUFSIZ]; /* For spooling a read byte sequence. */ ++ 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_bak; /* State of the stream. */ ++ int convfail = 0; /* 1, when conversion is failed. Otherwise 0. */ ++ ++ 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)); ++ ++ for (;; bufpos += mblength, buflen -= mblength) ++ { ++ if (buflen < MB_LEN_MAX && !feof (istream) && !ferror (istream)) ++ { ++ 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; ++ } ++ ++rescan: ++ if (operating_mode == byte_mode) /* byte mode */ ++ increment = mblength; ++ else if (operating_mode == character_mode) /* character mode */ ++ increment = 1; ++ else /* column mode */ ++ { ++ if (convfail) ++ increment = 1; ++ else ++ { ++ 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'\r': ++ increment = -1 * column; ++ break; ++ ++ case L'\t': ++ increment = 8 - column % 8; ++ break; ++ ++ default: ++ increment = wcwidth (wc); ++ increment = (increment < 0) ? 0 : increment; ++ } ++ } ++ } ++ ++ 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 ++ fold_text (istream, width, &saved_errno); ++ + if (ferror (istream)) + { + error (0, saved_errno, "%s", quotef (filename)); +@@ -252,7 +499,8 @@ main (int argc, char **argv) + + atexit (close_stdout); + +- break_spaces = count_bytes = have_read_stdin = false; ++ operating_mode = column_mode; ++ break_spaces = have_read_stdin = false; + + while ((optc = getopt_long (argc, argv, shortopts, longopts, NULL)) != -1) + { +@@ -261,7 +509,15 @@ main (int argc, char **argv) + switch (optc) + { + 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; ++ break; ++ ++ case 'c': ++ if (operating_mode != column_mode) ++ FATAL_ERROR (_("only one way of folding may be specified")); ++ operating_mode = character_mode; + break; + + 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 + ++/* 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 "die.h" + #include "error.h" + #include "fadvise.h" + #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); + } + ++#if HAVE_MBRTOWC ++static void ++xfields_multibyte (struct line *line) ++{ ++ 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; ++ ++ memset (&state, 0, sizeof (mbstate_t)); ++ ++ if (ptr >= 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; ++ } ++ 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 ++ + static void + freeline (struct line *line) + { +@@ -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 (jf_1 < line1->nfields) + { +- 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]; + } + + /* 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]; + ++#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); + } +@@ -1099,20 +1345,43 @@ 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; + +diff --git a/src/pr.c b/src/pr.c +index 26f221f..633f50e 100644 +--- a/src/pr.c ++++ b/src/pr.c +@@ -311,6 +311,24 @@ + + #include + #include ++ ++/* Get MB_LEN_MAX. */ ++#include ++/* 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 == 1 ++# define MB_LEN_MAX 16 ++#endif ++ ++/* Get MB_CUR_MAX. */ ++#include ++ ++/* Solaris 2.5 has a bug: must be included before . */ ++/* Get mbstate_t, mbrtowc(), wcwidth(). */ ++#if HAVE_WCHAR_H ++# include ++#endif ++ + #include "system.h" + #include "die.h" + #include "error.h" +@@ -324,6 +342,18 @@ + #include "xstrtol.h" + #include "xdectoint.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 ++ ++#ifndef HAVE_DECL_WCWIDTH ++"this configure-time declaration test was not run" ++#endif ++#if !HAVE_DECL_WCWIDTH ++extern int wcwidth (); ++#endif ++ + /* The official name of this program (e.g., no 'g' prefix). */ + #define PROGRAM_NAME "pr" + +@@ -416,7 +446,20 @@ struct COLUMN + + typedef struct COLUMN COLUMN; + +-static int char_to_clump (char c); ++/* Funtion pointers to switch functions for single byte locale or for ++ multibyte locale. If multibyte functions do not exist in your sysytem, ++ these pointers always point the function for single byte locale. */ ++static void (*print_char) (char c); ++static int (*char_to_clump) (char c); ++ ++/* Functions for single byte locale. */ ++static void print_char_single (char c); ++static int char_to_clump_single (char c); ++ ++/* Functions for multibyte locale. */ ++static void print_char_multi (char c); ++static int char_to_clump_multi (char c); ++ + static bool read_line (COLUMN *p); + static bool print_page (void); + static bool print_stored (COLUMN *p); +@@ -428,6 +471,7 @@ 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); + 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 (const char *optarg_S); +@@ -453,7 +496,7 @@ 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]. */ +-static char *buff; ++static unsigned char *buff; + + /* Index of the position in buff where the next character + will be stored. */ +@@ -557,7 +600,7 @@ static int chars_per_column; + static bool untabify_input = false; + + /* (-e) The input tab character. */ +-static char input_tab_char = '\t'; ++static char input_tab_char[MB_LEN_MAX] = "\t"; + + /* (-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; + static bool tabify_output = false; + + /* (-i) The output tab character. */ +-static char output_tab_char = '\t'; ++static char output_tab_char[MB_LEN_MAX] = "\t"; ++ ++/* (-i) The byte length of output tab character. */ ++static int output_tab_char_length = 1; + + /* (-i) The width of the output tab. */ + static int chars_per_output_tab = 8; +@@ -637,7 +683,13 @@ static int line_number; + static bool numbered_lines = false; + + /* (-n) Character which follows each line number. */ +-static char number_separator = '\t'; ++static char number_separator[MB_LEN_MAX] = "\t"; ++ ++/* (-n) The byte length of the character which follows each line number. */ ++static int number_separator_length = 1; ++ ++/* (-n) The character width of the character which follows each line number. */ ++static int number_separator_width = 0; + + /* (-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; + -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; ++static int col_sep_width = 0; + static char *column_separator = (char *) " "; + static char *line_separator = (char *) "\t"; + +@@ -851,6 +904,13 @@ separator_string (const char *optarg_S) + integer_overflow (); + col_sep_length = len; + col_sep_string = optarg_S; ++ ++#if HAVE_MBRTOWC ++ if (MB_CUR_MAX > 1) ++ col_sep_width = mbswidth (col_sep_string, 0); ++ else ++#endif ++ col_sep_width = col_sep_length; + } + + int +@@ -875,6 +935,21 @@ main (int argc, char **argv) + + atexit (close_stdout); + ++/* Define which functions are used, the ones for single byte locale or the ones ++ for multibyte locale. */ ++#if HAVE_MBRTOWC ++ if (MB_CUR_MAX > 1) ++ { ++ print_char = print_char_multi; ++ char_to_clump = char_to_clump_multi; ++ } ++ else ++#endif ++ { ++ print_char = print_char_single; ++ char_to_clump = char_to_clump_single; ++ } ++ + n_files = 0; + file_names = (argc > 1 + ? xnmalloc (argc - 1, sizeof (char *)) +@@ -951,8 +1026,12 @@ main (int argc, char **argv) + break; + case 'e': + if (optarg) +- getoptarg (optarg, 'e', &input_tab_char, +- &chars_per_input_tab); ++ { ++ int dummy_length, dummy_width; ++ ++ getoptarg (optarg, 'e', input_tab_char, &dummy_length, ++ &dummy_width, &chars_per_input_tab); ++ } + /* Could check tab width > 0. */ + untabify_input = true; + break; +@@ -965,8 +1044,12 @@ main (int argc, char **argv) + break; + case 'i': + if (optarg) +- getoptarg (optarg, 'i', &output_tab_char, +- &chars_per_output_tab); ++ { ++ int dummy_width; ++ ++ getoptarg (optarg, 'i', output_tab_char, &output_tab_char_length, ++ &dummy_width, &chars_per_output_tab); ++ } + /* Could check tab width > 0. */ + tabify_output = true; + break; +@@ -984,8 +1067,8 @@ main (int argc, char **argv) + case 'n': + numbered_lines = true; + if (optarg) +- getoptarg (optarg, 'n', &number_separator, +- &chars_per_number); ++ getoptarg (optarg, 'n', number_separator, &number_separator_length, ++ &number_separator_width, &chars_per_number); + break; + case 'N': + skip_count = false; +@@ -1010,6 +1093,7 @@ main (int argc, char **argv) + /* Reset an additional input of -s, -S dominates -s */ + col_sep_string = ""; + col_sep_length = 0; ++ col_sep_width = 0; + use_col_separator = true; + if (optarg) + separator_string (optarg); +@@ -1165,10 +1249,45 @@ getoptnum (const char *n_str, int min, int *num, const char *err) + a number. */ + + static void +-getoptarg (char *arg, char switch_char, char *character, int *number) ++getoptarg (char *arg, char switch_char, char *character, int *character_length, ++ int *character_width, int *number) + { + if (!ISDIGIT (*arg)) +- *character = *arg++; ++ { ++#ifdef HAVE_MBRTOWC ++ if (MB_CUR_MAX > 1) /* for multibyte locale. */ ++ { ++ wchar_t wc; ++ size_t mblength; ++ int width; ++ mbstate_t state = {'\0'}; ++ ++ mblength = mbrtowc (&wc, arg, strnlen(arg, MB_LEN_MAX), &state); ++ ++ if (mblength == (size_t)-1 || mblength == (size_t)-2) ++ { ++ *character_length = 1; ++ *character_width = 1; ++ } ++ else ++ { ++ *character_length = (mblength < 1) ? 1 : mblength; ++ width = wcwidth (wc); ++ *character_width = (width < 0) ? 0 : width; ++ } ++ ++ strncpy (character, arg, *character_length); ++ arg += *character_length; ++ } ++ else /* for single byte locale. */ ++#endif ++ { ++ *character = *arg++; ++ *character_length = 1; ++ *character_width = 1; ++ } ++ } ++ + if (*arg) + { + long int tmp_long; +@@ -1190,6 +1309,11 @@ static void + init_parameters (int number_of_files) + { + int chars_used_by_number = 0; ++ int mb_len = 1; ++#if HAVE_MBRTOWC ++ if (MB_CUR_MAX > 1) ++ mb_len = MB_LEN_MAX; ++#endif + + lines_per_body = lines_per_page - lines_per_header - lines_per_footer; + if (lines_per_body <= 0) +@@ -1227,7 +1351,7 @@ init_parameters (int number_of_files) + else + col_sep_string = column_separator; + +- col_sep_length = 1; ++ col_sep_length = col_sep_width = 1; + use_col_separator = true; + } + /* It's rather pointless to define a TAB separator with column +@@ -1257,11 +1381,11 @@ 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. */ +- if (number_separator == '\t') ++ if (number_separator[0] == '\t') + number_width = (chars_per_number + + TAB_WIDTH (chars_per_default_tab, chars_per_number)); + else +- number_width = chars_per_number + 1; ++ number_width = chars_per_number + number_separator_width; + + /* The number is part of the column width unless we are + printing files in parallel. */ +@@ -1270,7 +1394,7 @@ init_parameters (int number_of_files) + } + + int sep_chars, useful_chars; +- 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 (INT_SUBTRACT_WRAPV (chars_per_line - chars_used_by_number, sep_chars, + &useful_chars)) +@@ -1293,7 +1417,7 @@ 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); +- clump_buff = xmalloc (MAX (8, chars_per_input_tab)); ++ clump_buff = xmalloc (mb_len * MAX (8, chars_per_input_tab)); + } + + /* Open the necessary files, +@@ -1399,7 +1523,7 @@ init_funcs (void) + + /* Enlarge p->start_position of first column to use the same form of + padding_not_printed with all columns. */ +- h = h + col_sep_length; ++ h = h + col_sep_width; + + /* This loop takes care of all but the rightmost column. */ + +@@ -1433,7 +1557,7 @@ init_funcs (void) + } + else + { +- h = h_next + col_sep_length; ++ h = h_next + col_sep_width; + h_next = h + chars_per_column; + } + } +@@ -1724,9 +1848,9 @@ static void + align_column (COLUMN *p) + { + padding_not_printed = p->start_position; +- if (col_sep_length < padding_not_printed) ++ if (col_sep_width < padding_not_printed) + { +- pad_across_to (padding_not_printed - col_sep_length); ++ pad_across_to (padding_not_printed - col_sep_width); + padding_not_printed = ANYWHERE; + } + +@@ -2001,13 +2125,13 @@ store_char (char c) + /* May be too generous. */ + buff = X2REALLOC (buff, &buff_allocated); + } +- buff[buff_current++] = c; ++ buff[buff_current++] = (unsigned char) c; + } + + static void + add_line_number (COLUMN *p) + { +- int i; ++ int i, j; + char *s; + int num_width; + +@@ -2024,22 +2148,24 @@ 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. */ +- if (number_separator == '\t') ++ if (number_separator[0] == '\t') + { + i = number_width - chars_per_number; + while (i-- > 0) + (p->char_func) (' '); + } + else +- (p->char_func) (number_separator); ++ for (j = 0; j < number_separator_length; j++) ++ (p->char_func) (number_separator[j]); + } + else + /* To comply with POSIX, we avoid any expansion of default TAB + separator with a single column output. No column_width requirement + has to be considered. */ + { +- (p->char_func) (number_separator); +- if (number_separator == '\t') ++ for (j = 0; j < number_separator_length; j++) ++ (p->char_func) (number_separator[j]); ++ if (number_separator[0] == '\t') + output_position = POS_AFTER_TAB (chars_per_output_tab, + output_position); + } +@@ -2198,7 +2324,7 @@ print_white_space (void) + while (goal - h_old > 1 + && (h_new = POS_AFTER_TAB (chars_per_output_tab, h_old)) <= goal) + { +- putchar (output_tab_char); ++ fwrite (output_tab_char, sizeof(char), output_tab_char_length, stdout); + h_old = h_new; + } + while (++h_old <= goal) +@@ -2218,6 +2344,7 @@ print_sep_string (void) + { + char const *s = col_sep_string; + int l = col_sep_length; ++ int not_space_flag; + + if (separators_not_printed <= 0) + { +@@ -2229,6 +2356,7 @@ print_sep_string (void) + { + for (; separators_not_printed > 0; --separators_not_printed) + { ++ not_space_flag = 0; + while (l-- > 0) + { + /* 3 types of sep_strings: spaces only, spaces and chars, +@@ -2242,12 +2370,15 @@ print_sep_string (void) + } + else + { ++ not_space_flag = 1; + if (spaces_not_printed > 0) + print_white_space (); + putchar (*s++); +- ++output_position; + } + } ++ if (not_space_flag) ++ output_position += col_sep_width; ++ + /* sep_string ends with some spaces */ + if (spaces_not_printed > 0) + print_white_space (); +@@ -2275,7 +2406,7 @@ print_clump (COLUMN *p, int n, char *clump) + required number of tabs and spaces. */ + + static void +-print_char (char c) ++print_char_single (char c) + { + if (tabify_output) + { +@@ -2299,6 +2430,74 @@ print_char (char c) + putchar (c); + } + ++#ifdef HAVE_MBRTOWC ++static void ++print_char_multi (char c) ++{ ++ static size_t mbc_pos = 0; ++ static char mbc[MB_LEN_MAX] = {'\0'}; ++ static mbstate_t state = {'\0'}; ++ mbstate_t state_bak; ++ wchar_t wc; ++ size_t mblength; ++ int width; ++ ++ if (tabify_output) ++ { ++ state_bak = state; ++ mbc[mbc_pos++] = c; ++ mblength = mbrtowc (&wc, mbc, mbc_pos, &state); ++ ++ while (mbc_pos > 0) ++ { ++ switch (mblength) ++ { ++ case (size_t)-2: ++ state = state_bak; ++ return; ++ ++ case (size_t)-1: ++ state = state_bak; ++ ++output_position; ++ putchar (mbc[0]); ++ memmove (mbc, mbc + 1, MB_CUR_MAX - 1); ++ --mbc_pos; ++ break; ++ ++ case 0: ++ mblength = 1; ++ ++ default: ++ if (wc == L' ') ++ { ++ memmove (mbc, mbc + mblength, MB_CUR_MAX - mblength); ++ --mbc_pos; ++ ++spaces_not_printed; ++ return; ++ } ++ else if (spaces_not_printed > 0) ++ print_white_space (); ++ ++ /* Nonprintables are assumed to have width 0, except L'\b'. */ ++ if ((width = wcwidth (wc)) < 1) ++ { ++ if (wc == L'\b') ++ --output_position; ++ } ++ else ++ output_position += width; ++ ++ fwrite (mbc, sizeof(char), mblength, stdout); ++ memmove (mbc, mbc + mblength, MB_CUR_MAX - mblength); ++ mbc_pos -= mblength; ++ } ++ } ++ return; ++ } ++ putchar (c); ++} ++#endif ++ + /* Skip to page PAGE before printing. + PAGE may be larger than total number of pages. */ + +@@ -2476,9 +2675,9 @@ read_line (COLUMN *p) + align_empty_cols = false; + } + +- if (col_sep_length < padding_not_printed) ++ if (col_sep_width < padding_not_printed) + { +- pad_across_to (padding_not_printed - col_sep_length); ++ pad_across_to (padding_not_printed - col_sep_width); + padding_not_printed = ANYWHERE; + } + +@@ -2547,7 +2746,7 @@ print_stored (COLUMN *p) + COLUMN *q; + + int line = p->current_line++; +- char *first = &buff[line_vector[line]]; ++ unsigned char *first = &buff[line_vector[line]]; + /* FIXME + UMR: Uninitialized memory read: + * This is occurring while in: +@@ -2559,7 +2758,7 @@ print_stored (COLUMN *p) + xmalloc [xmalloc.c:94] + init_store_cols [pr.c:1648] + */ +- char *last = &buff[line_vector[line + 1]]; ++ unsigned char *last = &buff[line_vector[line + 1]]; + + pad_vertically = true; + +@@ -2579,9 +2778,9 @@ print_stored (COLUMN *p) + } + } + +- if (col_sep_length < padding_not_printed) ++ if (col_sep_width < padding_not_printed) + { +- pad_across_to (padding_not_printed - col_sep_length); ++ pad_across_to (padding_not_printed - col_sep_width); + padding_not_printed = ANYWHERE; + } + +@@ -2594,8 +2793,8 @@ print_stored (COLUMN *p) + if (spaces_not_printed == 0) + { + output_position = p->start_position + end_vector[line]; +- if (p->start_position - col_sep_length == chars_per_margin) +- output_position -= col_sep_length; ++ if (p->start_position - col_sep_width == chars_per_margin) ++ output_position -= col_sep_width; + } + + return true; +@@ -2614,7 +2813,7 @@ print_stored (COLUMN *p) + number of characters is 1.) */ + + static int +-char_to_clump (char c) ++char_to_clump_single (char c) + { + unsigned char uc = c; + char *s = clump_buff; +@@ -2624,10 +2823,10 @@ char_to_clump (char c) + int chars; + int chars_per_c = 8; + +- if (c == input_tab_char) ++ if (c == input_tab_char[0]) + chars_per_c = chars_per_input_tab; + +- if (c == input_tab_char || c == '\t') ++ if (c == input_tab_char[0] || c == '\t') + { + width = TAB_WIDTH (chars_per_c, input_position); + +@@ -2708,6 +2907,164 @@ char_to_clump (char c) + return chars; + } + ++#ifdef HAVE_MBRTOWC ++static int ++char_to_clump_multi (char c) ++{ ++ static size_t mbc_pos = 0; ++ static char mbc[MB_LEN_MAX] = {'\0'}; ++ static mbstate_t state = {'\0'}; ++ mbstate_t state_bak; ++ wchar_t wc; ++ size_t mblength; ++ int wc_width; ++ register char *s = clump_buff; ++ register int i, j; ++ char esc_buff[4]; ++ int width; ++ int chars; ++ int chars_per_c = 8; ++ ++ state_bak = state; ++ mbc[mbc_pos++] = c; ++ mblength = mbrtowc (&wc, mbc, mbc_pos, &state); ++ ++ width = 0; ++ chars = 0; ++ while (mbc_pos > 0) ++ { ++ switch (mblength) ++ { ++ case (size_t)-2: ++ state = state_bak; ++ return 0; ++ ++ case (size_t)-1: ++ state = state_bak; ++ mblength = 1; ++ ++ if (use_esc_sequence || use_cntrl_prefix) ++ { ++ width = +4; ++ chars = +4; ++ *s++ = '\\'; ++ sprintf (esc_buff, "%03o", (unsigned char) mbc[0]); ++ for (i = 0; i <= 2; ++i) ++ *s++ = (int) esc_buff[i]; ++ } ++ else ++ { ++ width += 1; ++ chars += 1; ++ *s++ = mbc[0]; ++ } ++ break; ++ ++ case 0: ++ mblength = 1; ++ /* Fall through */ ++ ++ default: ++ if (memcmp (mbc, input_tab_char, mblength) == 0) ++ chars_per_c = chars_per_input_tab; ++ ++ if (memcmp (mbc, input_tab_char, mblength) == 0 || c == '\t') ++ { ++ int width_inc; ++ ++ width_inc = TAB_WIDTH (chars_per_c, input_position); ++ width += width_inc; ++ ++ if (untabify_input) ++ { ++ for (i = width_inc; i; --i) ++ *s++ = ' '; ++ chars += width_inc; ++ } ++ else ++ { ++ for (i = 0; i < mblength; i++) ++ *s++ = mbc[i]; ++ chars += mblength; ++ } ++ } ++ else if ((wc_width = wcwidth (wc)) < 1) ++ { ++ if (use_esc_sequence) ++ { ++ for (i = 0; i < mblength; i++) ++ { ++ width += 4; ++ chars += 4; ++ *s++ = '\\'; ++ sprintf (esc_buff, "%03o", (unsigned char) mbc[i]); ++ for (j = 0; j <= 2; ++j) ++ *s++ = (int) esc_buff[j]; ++ } ++ } ++ else if (use_cntrl_prefix) ++ { ++ if (wc < 0200) ++ { ++ width += 2; ++ chars += 2; ++ *s++ = '^'; ++ *s++ = wc ^ 0100; ++ } ++ else ++ { ++ for (i = 0; i < mblength; i++) ++ { ++ width += 4; ++ chars += 4; ++ *s++ = '\\'; ++ sprintf (esc_buff, "%03o", (unsigned char) mbc[i]); ++ for (j = 0; j <= 2; ++j) ++ *s++ = (int) esc_buff[j]; ++ } ++ } ++ } ++ else if (wc == L'\b') ++ { ++ width += -1; ++ chars += 1; ++ *s++ = c; ++ } ++ else ++ { ++ width += 0; ++ chars += mblength; ++ for (i = 0; i < mblength; i++) ++ *s++ = mbc[i]; ++ } ++ } ++ else ++ { ++ width += wc_width; ++ chars += mblength; ++ for (i = 0; i < mblength; i++) ++ *s++ = mbc[i]; ++ } ++ } ++ memmove (mbc, mbc + mblength, MB_CUR_MAX - mblength); ++ mbc_pos -= mblength; ++ } ++ ++ /* Too many backspaces must put us in position 0 -- never negative. */ ++ if (width < 0 && input_position == 0) ++ { ++ chars = 0; ++ input_position = 0; ++ } ++ else if (width < 0 && input_position <= -width) ++ input_position = 0; ++ else ++ input_position += width; ++ ++ return chars; ++} ++#endif ++ + /* We've just printed some files and need to clean up things before + looking for more options and printing the next batch of files. + +diff --git a/src/sort.c b/src/sort.c +index 6d2eec5..f189a0d 100644 +--- a/src/sort.c ++++ b/src/sort.c +@@ -29,6 +29,14 @@ + #include + #include + #include ++#if HAVE_WCHAR_H ++# include ++#endif ++/* Get isw* functions. */ ++#if HAVE_WCTYPE_H ++# include ++#endif ++ + #include "system.h" + #include "argmatch.h" + #include "die.h" +@@ -169,14 +177,39 @@ 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; ++ + /* Nonzero if the corresponding locales are hard. */ + static bool hard_LC_COLLATE; +-#if HAVE_NL_LANGINFO ++#if HAVE_LANGINFO_CODESET + static bool hard_LC_TIME; + #endif + + #define NONZERO(x) ((x) != 0) + ++/* get a multibyte character's byte length. */ ++#define GET_BYTELEN_OF_CHAR(LIM, PTR, MBLENGTH, STATE) \ ++ do \ ++ { \ ++ wchar_t wc; \ ++ mbstate_t state_bak; \ ++ \ ++ state_bak = STATE; \ ++ mblength = mbrtowc (&wc, PTR, LIM - PTR, &STATE); \ ++ \ ++ switch (MBLENGTH) \ ++ { \ ++ case (size_t)-1: \ ++ case (size_t)-2: \ ++ STATE = state_bak; \ ++ /* Fall through. */ \ ++ case 0: \ ++ MBLENGTH = 1; \ ++ } \ ++ } \ ++ while (0) ++ + /* The kind of blanks for '-b' to skip in various options. */ + enum blanktype { bl_start, bl_end, bl_both }; + +@@ -350,13 +383,11 @@ 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 }; +- +-/* Tab character separating fields. If TAB_DEFAULT, then fields are ++/* Tab character separating fields. If tab_length is 0, then fields are + separated by the empty string between a non-blank character and a blank + character. */ +-static int tab = TAB_DEFAULT; ++static char tab[MB_LEN_MAX + 1]; ++static size_t tab_length = 0; + + /* Flag to remove consecutive duplicate lines from the output. + Only the last of a sequence of equal lines will be output. */ +@@ -814,6 +845,46 @@ reap_all (void) + reap (-1); + } + ++/* Function pointers. */ ++static void ++(*inittables) (void); ++static char * ++(*begfield) (const struct line*, const struct keyfield *); ++static char * ++(*limfield) (const struct line*, const struct keyfield *); ++static void ++(*skipblanks) (char **ptr, char *lim); ++static int ++(*getmonth) (char const *, size_t, char **); ++static int ++(*keycompare) (const struct line *, const struct line *); ++static int ++(*numcompare) (const char *, const char *); ++ ++/* Test for white space multibyte character. ++ Set LENGTH the byte length of investigated multibyte character. */ ++#if HAVE_MBRTOWC ++static int ++ismbblank (const char *str, size_t len, size_t *length) ++{ ++ size_t mblength; ++ wchar_t wc; ++ mbstate_t state; ++ ++ memset (&state, '\0', sizeof(mbstate_t)); ++ mblength = mbrtowc (&wc, str, len, &state); ++ ++ if (mblength == (size_t)-1 || mblength == (size_t)-2) ++ { ++ *length = 1; ++ return 0; ++ } ++ ++ *length = (mblength < 1) ? 1 : mblength; ++ return iswblank (wc) || wc == '\n'; ++} ++#endif ++ + /* Clean up any remaining temporary files. */ + + static void +@@ -1264,7 +1335,7 @@ zaptemp (char const *name) + free (node); + } + +-#if HAVE_NL_LANGINFO ++#if HAVE_LANGINFO_CODESET + + static int + struct_month_cmp (void const *m1, void const *m2) +@@ -1279,7 +1350,7 @@ struct_month_cmp (void const *m1, void const *m2) + /* Initialize the character class tables. */ + + static void +-inittables (void) ++inittables_uni (void) + { + size_t i; + +@@ -1291,7 +1362,7 @@ inittables (void) + fold_toupper[i] = toupper (i); + } + +-#if HAVE_NL_LANGINFO ++#if HAVE_LANGINFO_CODESET + /* If we're not in the "C" locale, read different names for months. */ + if (hard_LC_TIME) + { +@@ -1373,6 +1444,84 @@ specify_nmerge (int oi, char c, char const *s) + xstrtol_fatal (e, oi, c, long_options, s); + } + ++#if HAVE_MBRTOWC ++static void ++inittables_mb (void) ++{ ++ int i, j, k, l; ++ char *name, *s, *lc_time, *lc_ctype; ++ size_t s_len, mblength; ++ char mbc[MB_LEN_MAX]; ++ wchar_t wc, pwc; ++ mbstate_t state_mb, state_wc; ++ ++ lc_time = setlocale (LC_TIME, ""); ++ if (lc_time) ++ lc_time = xstrdup (lc_time); ++ ++ lc_ctype = setlocale (LC_CTYPE, ""); ++ if (lc_ctype) ++ lc_ctype = xstrdup (lc_ctype); ++ ++ if (lc_time && lc_ctype) ++ /* temporarily set LC_CTYPE to match LC_TIME, so that we can convert ++ * the names of months to upper case */ ++ setlocale (LC_CTYPE, lc_time); ++ ++ for (i = 0; i < MONTHS_PER_YEAR; i++) ++ { ++ s = (char *) nl_langinfo (ABMON_1 + i); ++ s_len = strlen (s); ++ monthtab[i].name = name = (char *) xmalloc (s_len + 1); ++ monthtab[i].val = i + 1; ++ ++ memset (&state_mb, '\0', sizeof (mbstate_t)); ++ memset (&state_wc, '\0', sizeof (mbstate_t)); ++ ++ for (j = 0; j < s_len;) ++ { ++ if (!ismbblank (s + j, s_len - j, &mblength)) ++ break; ++ j += mblength; ++ } ++ ++ for (k = 0; j < s_len;) ++ { ++ mblength = mbrtowc (&wc, (s + j), (s_len - j), &state_mb); ++ assert (mblength != (size_t)-1 && mblength != (size_t)-2); ++ if (mblength == 0) ++ break; ++ ++ pwc = towupper (wc); ++ if (pwc == wc) ++ { ++ memcpy (mbc, s + j, mblength); ++ j += mblength; ++ } ++ else ++ { ++ j += mblength; ++ mblength = wcrtomb (mbc, pwc, &state_wc); ++ assert (mblength != (size_t)0 && mblength != (size_t)-1); ++ } ++ ++ for (l = 0; l < mblength; l++) ++ name[k++] = mbc[l]; ++ } ++ name[k] = '\0'; ++ } ++ qsort ((void *) monthtab, MONTHS_PER_YEAR, ++ sizeof (struct month), struct_month_cmp); ++ ++ if (lc_time && lc_ctype) ++ /* restore the original locales */ ++ setlocale (LC_CTYPE, lc_ctype); ++ ++ free (lc_ctype); ++ free (lc_time); ++} ++#endif ++ + /* Specify the amount of main memory to use when sorting. */ + static void + specify_sort_size (int oi, char c, char const *s) +@@ -1604,7 +1753,7 @@ buffer_linelim (struct buffer const *buf) + by KEY in LINE. */ + + static char * +-begfield (struct line const *line, struct keyfield const *key) ++begfield_uni (const struct line *line, const struct keyfield *key) + { + char *ptr = line->text, *lim = ptr + line->length - 1; + size_t sword = key->sword; +@@ -1613,10 +1762,10 @@ begfield (struct line const *line, struct keyfield const *key) + /* The leading field separator itself is included in a field when -t + is absent. */ + +- if (tab != TAB_DEFAULT) ++ if (tab_length) + while (ptr < lim && sword--) + { +- while (ptr < lim && *ptr != tab) ++ while (ptr < lim && *ptr != tab[0]) + ++ptr; + if (ptr < lim) + ++ptr; +@@ -1642,11 +1791,70 @@ begfield (struct line const *line, struct keyfield const *key) + return ptr; + } + ++#if HAVE_MBRTOWC ++static char * ++begfield_mb (const struct line *line, const struct keyfield *key) ++{ ++ int i; ++ char *ptr = line->text, *lim = ptr + line->length - 1; ++ size_t sword = key->sword; ++ size_t schar = key->schar; ++ size_t mblength; ++ mbstate_t state; ++ ++ memset (&state, '\0', sizeof(mbstate_t)); ++ ++ if (tab_length) ++ while (ptr < lim && sword--) ++ { ++ while (ptr < lim && memcmp (ptr, tab, tab_length) != 0) ++ { ++ GET_BYTELEN_OF_CHAR (lim, ptr, mblength, state); ++ ptr += mblength; ++ } ++ if (ptr < lim) ++ { ++ GET_BYTELEN_OF_CHAR (lim, ptr, mblength, state); ++ ptr += mblength; ++ } ++ } ++ else ++ while (ptr < lim && sword--) ++ { ++ while (ptr < lim && ismbblank (ptr, lim - ptr, &mblength)) ++ ptr += mblength; ++ if (ptr < lim) ++ { ++ GET_BYTELEN_OF_CHAR (lim, ptr, mblength, state); ++ ptr += mblength; ++ } ++ while (ptr < lim && !ismbblank (ptr, lim - ptr, &mblength)) ++ ptr += mblength; ++ } ++ ++ if (key->skipsblanks) ++ while (ptr < lim && ismbblank (ptr, lim - ptr, &mblength)) ++ ptr += mblength; ++ ++ for (i = 0; i < schar; i++) ++ { ++ GET_BYTELEN_OF_CHAR (lim, ptr, mblength, state); ++ ++ if (ptr + mblength > lim) ++ break; ++ else ++ ptr += mblength; ++ } ++ ++ return ptr; ++} ++#endif ++ + /* Return the limit of (a pointer to the first character after) the field + in LINE specified by KEY. */ + + static char * +-limfield (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; +@@ -1661,10 +1869,10 @@ 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. */ +- if (tab != TAB_DEFAULT) ++ if (tab_length) + while (ptr < lim && eword--) + { +- while (ptr < lim && *ptr != tab) ++ while (ptr < lim && *ptr != tab[0]) + ++ptr; + if (ptr < lim && (eword || echar)) + ++ptr; +@@ -1710,10 +1918,10 @@ limfield (struct line const *line, struct keyfield const *key) + */ + + /* Make LIM point to the end of (one byte past) the current field. */ +- if (tab != TAB_DEFAULT) ++ if (tab_length) + { + char *newlim; +- newlim = memchr (ptr, tab, lim - ptr); ++ newlim = memchr (ptr, tab[0], lim - ptr); + if (newlim) + lim = newlim; + } +@@ -1744,6 +1952,130 @@ limfield (struct line const *line, struct keyfield const *key) + return ptr; + } + ++#if HAVE_MBRTOWC ++static char * ++limfield_mb (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; ++ int i; ++ size_t mblength; ++ mbstate_t state; ++ ++ if (echar == 0) ++ eword++; /* skip all of end field. */ ++ ++ memset (&state, '\0', sizeof(mbstate_t)); ++ ++ if (tab_length) ++ while (ptr < lim && eword--) ++ { ++ while (ptr < lim && memcmp (ptr, tab, tab_length) != 0) ++ { ++ GET_BYTELEN_OF_CHAR (lim, ptr, mblength, state); ++ ptr += mblength; ++ } ++ if (ptr < lim && (eword | echar)) ++ { ++ GET_BYTELEN_OF_CHAR (lim, ptr, mblength, state); ++ ptr += mblength; ++ } ++ } ++ else ++ while (ptr < lim && eword--) ++ { ++ while (ptr < lim && ismbblank (ptr, lim - ptr, &mblength)) ++ ptr += mblength; ++ if (ptr < lim) ++ { ++ GET_BYTELEN_OF_CHAR (lim, ptr, mblength, state); ++ ptr += mblength; ++ } ++ while (ptr < lim && !ismbblank (ptr, lim - ptr, &mblength)) ++ ptr += mblength; ++ } ++ ++ ++# ifdef POSIX_UNSPECIFIED ++ /* Make LIM point to the end of (one byte past) the current field. */ ++ if (tab_length) ++ { ++ char *newlim, *p; ++ ++ newlim = NULL; ++ for (p = ptr; p < lim;) ++ { ++ if (memcmp (p, tab, tab_length) == 0) ++ { ++ newlim = p; ++ break; ++ } ++ ++ GET_BYTELEN_OF_CHAR (lim, ptr, mblength, state); ++ p += mblength; ++ } ++ } ++ else ++ { ++ char *newlim; ++ newlim = ptr; ++ ++ while (newlim < lim && ismbblank (newlim, lim - newlim, &mblength)) ++ newlim += mblength; ++ if (ptr < lim) ++ { ++ GET_BYTELEN_OF_CHAR (lim, ptr, mblength, state); ++ ptr += mblength; ++ } ++ while (newlim < lim && !ismbblank (newlim, lim - newlim, &mblength)) ++ newlim += mblength; ++ lim = newlim; ++ } ++# endif ++ ++ if (echar != 0) ++ { ++ /* If we're skipping leading blanks, don't start counting characters ++ * until after skipping past any leading blanks. */ ++ if (key->skipeblanks) ++ while (ptr < lim && ismbblank (ptr, lim - ptr, &mblength)) ++ ptr += mblength; ++ ++ memset (&state, '\0', sizeof(mbstate_t)); ++ ++ /* Advance PTR by ECHAR (if possible), but no further than LIM. */ ++ for (i = 0; i < echar; i++) ++ { ++ GET_BYTELEN_OF_CHAR (lim, ptr, mblength, state); ++ ++ if (ptr + mblength > lim) ++ break; ++ else ++ ptr += mblength; ++ } ++ } ++ ++ return ptr; ++} ++#endif ++ ++static void ++skipblanks_uni (char **ptr, char *lim) ++{ ++ while (*ptr < lim && blanks[to_uchar (**ptr)]) ++ ++(*ptr); ++} ++ ++#if HAVE_MBRTOWC ++static void ++skipblanks_mb (char **ptr, char *lim) ++{ ++ size_t mblength; ++ while (*ptr < lim && ismbblank (*ptr, lim - *ptr, &mblength)) ++ (*ptr) += mblength; ++} ++#endif ++ + /* 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 +@@ -1830,8 +2162,22 @@ fillbuf (struct buffer *buf, FILE *fp, char const *file) + else + { + if (key->skipsblanks) +- while (blanks[to_uchar (*line_start)]) +- line_start++; ++ { ++#if HAVE_MBRTOWC ++ if (MB_CUR_MAX > 1) ++ { ++ size_t mblength; ++ while (line_start < line->keylim && ++ ismbblank (line_start, ++ line->keylim - line_start, ++ &mblength)) ++ line_start += mblength; ++ } ++ else ++#endif ++ while (blanks[to_uchar (*line_start)]) ++ line_start++; ++ } + line->keybeg = line_start; + } + } +@@ -1981,7 +2327,7 @@ human_numcompare (char const *a, char const *b) + hideously fast. */ + + static int +-numcompare (char const *a, char const *b) ++numcompare_uni (const char *a, const char *b) + { + while (blanks[to_uchar (*a)]) + a++; +@@ -1991,6 +2337,25 @@ numcompare (char const *a, char const *b) + return strnumcmp (a, b, decimal_point, thousands_sep); + } + ++#if HAVE_MBRTOWC ++static int ++numcompare_mb (const char *a, const char *b) ++{ ++ size_t mblength, len; ++ len = strlen (a); /* okay for UTF-8 */ ++ while (*a && ismbblank (a, len > MB_CUR_MAX ? MB_CUR_MAX : len, &mblength)) ++ { ++ a += mblength; ++ len -= mblength; ++ } ++ len = strlen (b); /* okay for UTF-8 */ ++ while (*b && ismbblank (b, len > MB_CUR_MAX ? MB_CUR_MAX : len, &mblength)) ++ b += mblength; ++ ++ return strnumcmp (a, b, decimal_point, thousands_sep); ++} ++#endif /* HAV_EMBRTOWC */ ++ + /* 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 once +@@ -2041,7 +2406,7 @@ general_numcompare (char const *sa, char const *sb) + Return 0 if the name in S is not recognized. */ + + static int +-getmonth (char const *month, char **ea) ++getmonth_uni (char const *month, size_t len, char **ea) + { + size_t lo = 0; + size_t hi = MONTHS_PER_YEAR; +@@ -2317,15 +2682,14 @@ debug_key (struct line const *line, struct keyfield const *key) + char saved = *lim; + *lim = '\0'; + +- while (blanks[to_uchar (*beg)]) +- beg++; ++ skipblanks (&beg, lim); + + char *tighter_lim = beg; + + if (lim < beg) + tighter_lim = lim; + else if (key->month) +- getmonth (beg, &tighter_lim); ++ getmonth (beg, lim-beg, &tighter_lim); + else if (key->general_numeric) + ignore_value (strtold (beg, &tighter_lim)); + else if (key->numeric || key->human_numeric) +@@ -2459,7 +2823,7 @@ 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 */ +- if (!zero_width && !gkey_only && tab == TAB_DEFAULT && !line_offset ++ if (!zero_width && !gkey_only && !tab_length && !line_offset + && ((!key->skipsblanks && !implicit_skip) + || (!key->skipsblanks && key->schar) + || (!key->skipeblanks && key->echar))) +@@ -2517,11 +2881,87 @@ key_warnings (struct keyfield const *gkey, bool gkey_only) + error (0, 0, _("option '-r' only applies to last-resort comparison")); + } + ++#if HAVE_MBRTOWC ++static int ++getmonth_mb (const char *s, size_t len, char **ea) ++{ ++ char *month; ++ register size_t i; ++ register int lo = 0, hi = MONTHS_PER_YEAR, result; ++ char *tmp; ++ size_t wclength, mblength; ++ const char *pp; ++ const wchar_t *wpp; ++ wchar_t *month_wcs; ++ mbstate_t state; ++ ++ while (len > 0 && ismbblank (s, len, &mblength)) ++ { ++ s += mblength; ++ len -= mblength; ++ } ++ ++ if (len == 0) ++ return 0; ++ ++ if (SIZE_MAX - len < 1) ++ xalloc_die (); ++ ++ month = (char *) xnmalloc (len + 1, MB_CUR_MAX); ++ ++ pp = tmp = (char *) xnmalloc (len + 1, MB_CUR_MAX); ++ memcpy (tmp, s, len); ++ tmp[len] = '\0'; ++ wpp = month_wcs = (wchar_t *) xnmalloc (len + 1, sizeof (wchar_t)); ++ memset (&state, '\0', sizeof (mbstate_t)); ++ ++ wclength = mbsrtowcs (month_wcs, &pp, len + 1, &state); ++ if (wclength == (size_t)-1 || pp != NULL) ++ error (SORT_FAILURE, 0, _("Invalid multibyte input %s."), quote(s)); ++ ++ for (i = 0; i < wclength; i++) ++ { ++ month_wcs[i] = towupper(month_wcs[i]); ++ if (iswblank (month_wcs[i])) ++ { ++ month_wcs[i] = L'\0'; ++ break; ++ } ++ } ++ ++ mblength = wcsrtombs (month, &wpp, (len + 1) * MB_CUR_MAX, &state); ++ assert (mblength != (-1) && wpp == NULL); ++ ++ do ++ { ++ int ix = (lo + hi) / 2; ++ ++ if (strncmp (month, monthtab[ix].name, strlen (monthtab[ix].name)) < 0) ++ hi = ix; ++ else ++ lo = ix; ++ } ++ while (hi - lo > 1); ++ ++ result = (!strncmp (month, monthtab[lo].name, strlen (monthtab[lo].name)) ++ ? monthtab[lo].val : 0); ++ ++ if (ea && result) ++ *ea = (char*) s + strlen (monthtab[lo].name); ++ ++ free (month); ++ free (tmp); ++ free (month_wcs); ++ ++ return result; ++} ++#endif ++ + /* Compare two lines A and B trying every key in sequence until there + are no more keys or a difference is found. */ + + static int +-keycompare (struct line const *a, struct line const *b) ++keycompare_uni (const struct line *a, const struct line *b) + { + struct keyfield *key = keylist; + +@@ -2606,7 +3046,7 @@ 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, 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) +@@ -2722,6 +3162,211 @@ keycompare (struct line const *a, struct line const *b) + return key->reverse ? -diff : diff; + } + ++#if HAVE_MBRTOWC ++static int ++keycompare_mb (const struct line *a, const struct line *b) ++{ ++ struct keyfield *key = keylist; ++ ++ /* For the first iteration only, the key positions have been ++ precomputed for us. */ ++ char *texta = a->keybeg; ++ char *textb = b->keybeg; ++ char *lima = a->keylim; ++ char *limb = b->keylim; ++ ++ size_t mblength_a, mblength_b; ++ wchar_t wc_a, wc_b; ++ mbstate_t state_a, state_b; ++ ++ int diff = 0; ++ ++ memset (&state_a, '\0', sizeof(mbstate_t)); ++ memset (&state_b, '\0', sizeof(mbstate_t)); ++ /* Ignore keys with start after end. */ ++ if (a->keybeg - a->keylim > 0) ++ return 0; ++ ++ ++ /* Ignore and/or translate chars before comparing. */ ++# define IGNORE_CHARS(NEW_LEN, LEN, TEXT, COPY, WC, MBLENGTH, STATE) \ ++ do \ ++ { \ ++ wchar_t uwc; \ ++ char mbc[MB_LEN_MAX]; \ ++ mbstate_t state_wc; \ ++ \ ++ for (NEW_LEN = i = 0; i < LEN;) \ ++ { \ ++ mbstate_t state_bak; \ ++ \ ++ state_bak = STATE; \ ++ MBLENGTH = mbrtowc (&WC, TEXT + i, LEN - i, &STATE); \ ++ \ ++ if (MBLENGTH == (size_t)-2 || MBLENGTH == (size_t)-1 \ ++ || MBLENGTH == 0) \ ++ { \ ++ if (MBLENGTH == (size_t)-2 || MBLENGTH == (size_t)-1) \ ++ STATE = state_bak; \ ++ if (!ignore) \ ++ COPY[NEW_LEN++] = TEXT[i]; \ ++ i++; \ ++ continue; \ ++ } \ ++ \ ++ if (ignore) \ ++ { \ ++ if ((ignore == nonprinting && !iswprint (WC)) \ ++ || (ignore == nondictionary \ ++ && !iswalnum (WC) && !iswblank (WC))) \ ++ { \ ++ i += MBLENGTH; \ ++ continue; \ ++ } \ ++ } \ ++ \ ++ if (translate) \ ++ { \ ++ \ ++ uwc = towupper(WC); \ ++ if (WC == uwc) \ ++ { \ ++ memcpy (mbc, TEXT + i, MBLENGTH); \ ++ i += MBLENGTH; \ ++ } \ ++ else \ ++ { \ ++ i += MBLENGTH; \ ++ WC = uwc; \ ++ memset (&state_wc, '\0', sizeof (mbstate_t)); \ ++ \ ++ MBLENGTH = wcrtomb (mbc, WC, &state_wc); \ ++ assert (MBLENGTH != (size_t)-1 && MBLENGTH != 0); \ ++ } \ ++ \ ++ for (j = 0; j < MBLENGTH; j++) \ ++ COPY[NEW_LEN++] = mbc[j]; \ ++ } \ ++ else \ ++ for (j = 0; j < MBLENGTH; j++) \ ++ COPY[NEW_LEN++] = TEXT[i++]; \ ++ } \ ++ COPY[NEW_LEN] = '\0'; \ ++ } \ ++ while (0) ++ ++ /* Actually compare the fields. */ ++ ++ for (;;) ++ { ++ /* Find the lengths. */ ++ size_t lena = lima <= texta ? 0 : lima - texta; ++ size_t lenb = limb <= textb ? 0 : limb - textb; ++ ++ char enda IF_LINT (= 0); ++ char endb IF_LINT (= 0); ++ ++ char const *translate = key->translate; ++ bool const *ignore = key->ignore; ++ ++ if (ignore || translate) ++ { ++ if (SIZE_MAX - lenb - 2 < lena) ++ xalloc_die (); ++ char *copy_a = (char *) xnmalloc (lena + lenb + 2, MB_CUR_MAX); ++ char *copy_b = copy_a + lena * MB_CUR_MAX + 1; ++ size_t new_len_a, new_len_b; ++ size_t i, j; ++ ++ IGNORE_CHARS (new_len_a, lena, texta, copy_a, ++ wc_a, mblength_a, state_a); ++ IGNORE_CHARS (new_len_b, lenb, textb, copy_b, ++ wc_b, mblength_b, state_b); ++ texta = copy_a; textb = copy_b; ++ lena = new_len_a; lenb = new_len_b; ++ } ++ else ++ { ++ /* Use the keys in-place, temporarily null-terminated. */ ++ enda = texta[lena]; texta[lena] = '\0'; ++ endb = textb[lenb]; textb[lenb] = '\0'; ++ } ++ ++ if (key->random) ++ diff = compare_random (texta, lena, textb, lenb); ++ else if (key->numeric | key->general_numeric | key->human_numeric) ++ { ++ char savea = *lima, saveb = *limb; ++ ++ *lima = *limb = '\0'; ++ diff = (key->numeric ? numcompare (texta, textb) ++ : key->general_numeric ? general_numcompare (texta, textb) ++ : human_numcompare (texta, textb)); ++ *lima = savea, *limb = saveb; ++ } ++ else if (key->version) ++ diff = filevercmp (texta, textb); ++ else if (key->month) ++ diff = getmonth (texta, lena, NULL) - getmonth (textb, lenb, NULL); ++ else if (lena == 0) ++ diff = - NONZERO (lenb); ++ else if (lenb == 0) ++ diff = 1; ++ else if (hard_LC_COLLATE && !folding) ++ { ++ diff = xmemcoll0 (texta, lena + 1, textb, lenb + 1); ++ } ++ else ++ { ++ diff = memcmp (texta, textb, MIN (lena, lenb)); ++ if (diff == 0) ++ diff = lena < lenb ? -1 : lena != lenb; ++ } ++ ++ if (ignore || translate) ++ free (texta); ++ else ++ { ++ texta[lena] = enda; ++ textb[lenb] = endb; ++ } ++ ++ if (diff) ++ goto not_equal; ++ ++ key = key->next; ++ if (! key) ++ break; ++ ++ /* Find the beginning and limit of the next field. */ ++ if (key->eword != -1) ++ lima = limfield (a, key), limb = limfield (b, key); ++ else ++ lima = a->text + a->length - 1, limb = b->text + b->length - 1; ++ ++ if (key->sword != -1) ++ texta = begfield (a, key), textb = begfield (b, key); ++ else ++ { ++ texta = a->text, textb = b->text; ++ if (key->skipsblanks) ++ { ++ while (texta < lima && ismbblank (texta, lima - texta, &mblength_a)) ++ texta += mblength_a; ++ while (textb < limb && ismbblank (textb, limb - textb, &mblength_b)) ++ textb += mblength_b; ++ } ++ } ++ } ++ ++not_equal: ++ if (key && key->reverse) ++ return -diff; ++ else ++ return diff; ++} ++#endif ++ + /* Compare two lines A and B, returning negative, zero, or positive + depending on whether A compares less than, equal to, or greater than B. */ + +@@ -2749,7 +3394,7 @@ compare (struct line const *a, struct line const *b) + diff = - NONZERO (blen); + else if (blen == 0) + diff = 1; +- else if (hard_LC_COLLATE) ++ else if (hard_LC_COLLATE && !folding) + { + /* xmemcoll0 is a performance enhancement as + it will not unconditionally write '\0' after the +@@ -4144,6 +4789,7 @@ set_ordering (char const *s, struct keyfield *key, enum blanktype blanktype) + break; + case 'f': + key->translate = fold_toupper; ++ folding = true; + break; + case 'g': + key->general_numeric = true; +@@ -4223,7 +4869,7 @@ main (int argc, char **argv) + initialize_exit_failure (SORT_FAILURE); + + hard_LC_COLLATE = hard_locale (LC_COLLATE); +-#if HAVE_NL_LANGINFO ++#if HAVE_LANGINFO_CODESET + hard_LC_TIME = hard_locale (LC_TIME); + #endif + +@@ -4244,6 +4890,29 @@ main (int argc, char **argv) + thousands_sep = -1; + } + ++#if HAVE_MBRTOWC ++ if (MB_CUR_MAX > 1) ++ { ++ inittables = inittables_mb; ++ begfield = begfield_mb; ++ limfield = limfield_mb; ++ skipblanks = skipblanks_mb; ++ getmonth = getmonth_mb; ++ keycompare = keycompare_mb; ++ numcompare = numcompare_mb; ++ } ++ else ++#endif ++ { ++ inittables = inittables_uni; ++ begfield = begfield_uni; ++ limfield = limfield_uni; ++ skipblanks = skipblanks_uni; ++ getmonth = getmonth_uni; ++ keycompare = keycompare_uni; ++ numcompare = numcompare_uni; ++ } ++ + have_read_stdin = false; + inittables (); + +@@ -4518,13 +5187,34 @@ main (int argc, char **argv) + + case 't': + { +- char newtab = optarg[0]; +- if (! newtab) ++ char newtab[MB_LEN_MAX + 1]; ++ size_t newtab_length = 1; ++ strncpy (newtab, optarg, MB_LEN_MAX); ++ if (! newtab[0]) + die (SORT_FAILURE, 0, _("empty tab")); +- if (optarg[1]) ++#if HAVE_MBRTOWC ++ if (MB_CUR_MAX > 1) ++ { ++ wchar_t wc; ++ mbstate_t state; ++ ++ memset (&state, '\0', sizeof (mbstate_t)); ++ newtab_length = mbrtowc (&wc, newtab, strnlen (newtab, ++ MB_LEN_MAX), ++ &state); ++ switch (newtab_length) ++ { ++ case (size_t) -1: ++ case (size_t) -2: ++ case 0: ++ newtab_length = 1; ++ } ++ } ++#endif ++ if (newtab_length == 1 && optarg[1]) + { + if (STREQ (optarg, "\\0")) +- newtab = '\0'; ++ newtab[0] = '\0'; + else + { + /* Provoke with 'sort -txx'. Complain about +@@ -4535,9 +5225,11 @@ 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)) + die (SORT_FAILURE, 0, _("incompatible tabs")); +- tab = newtab; ++ memcpy (tab, newtab, newtab_length); ++ tab_length = newtab_length; + } + break; + +@@ -4765,12 +5457,10 @@ 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 ++ ++/* Get isw* functions. */ ++#if HAVE_WCTYPE_H ++# include ++#endif ++#include ++ + #include "system.h" + #include "argmatch.h" + #include "linebuffer.h" +@@ -32,9 +43,21 @@ + #include "stdio--.h" + #include "xmemcoll.h" + #include "xstrtol.h" +-#include "memcasecmp.h" ++#include "xmemcoll.h" + #include "quote.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 ++# 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" + +@@ -144,6 +167,10 @@ enum + GROUP_OPTION = CHAR_MAX + 1 + }; + ++/* Function pointers. */ ++static char * ++(*find_field) (struct linebuffer *line); ++ + static struct option const longopts[] = + { + {"count", no_argument, NULL, 'c'}, +@@ -260,7 +287,7 @@ size_opt (char const *opt, char const *msgid) + return a pointer to the beginning of the line's field to be compared. */ + + 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; +@@ -280,6 +307,83 @@ find_field (struct linebuffer const *line) + return line->buffer + i; + } + ++#if HAVE_MBRTOWC ++ ++# 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) ++ ++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++) ++ { ++ while (pos < size) ++ { ++ 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; ++ } ++ } ++ ++ /* skip fields. */ ++ for (count = 0; count < skip_chars && pos < size; count++) ++ { ++ MBCHAR_TO_WCHAR (wc, mblength, lp, pos, size, statep, convfail); ++ pos += mblength; ++ } ++ ++ 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. +@@ -288,6 +392,8 @@ find_field (struct linebuffer const *line) + static bool + different (char *old, char *new, size_t oldlen, size_t newlen) + { ++ char *copy_old, *copy_new; ++ + if (check_chars < oldlen) + oldlen = check_chars; + if (check_chars < newlen) +@@ -295,15 +401,104 @@ different (char *old, char *new, size_t oldlen, size_t newlen) + + if (ignore_case) + { +- /* FIXME: This should invoke strcoll somehow. */ +- return oldlen != newlen || memcasecmp (old, new, oldlen); ++ size_t i; ++ ++ copy_old = xmalloc (oldlen + 1); ++ copy_new = xmalloc (oldlen + 1); ++ ++ for (i = 0; i < oldlen; i++) ++ { ++ copy_old[i] = toupper (old[i]); ++ copy_new[i] = toupper (new[i]); ++ } ++ bool rc = xmemcoll (copy_old, oldlen, copy_new, newlen); ++ free (copy_old); ++ free (copy_new); ++ return rc; + } +- else if (hard_LC_COLLATE) +- return xmemcoll (old, oldlen, new, newlen) != 0; + else +- return oldlen != newlen || memcmp (old, new, oldlen); ++ { ++ copy_old = (char *)old; ++ copy_new = (char *)new; ++ } ++ ++ return xmemcoll (copy_old, oldlen, copy_new, newlen); ++ + } + ++#if HAVE_MBRTOWC ++static int ++different_multi (const char *old, const char *new, size_t oldlen, size_t newlen, mbstate_t oldstate, mbstate_t newstate) ++{ ++ size_t i, j, chars; ++ const char *str[2]; ++ char *copy[2]; ++ size_t len[2]; ++ mbstate_t state[2]; ++ size_t mblength; ++ wchar_t wc, uwc; ++ mbstate_t state_bak; ++ ++ str[0] = old; ++ str[1] = new; ++ len[0] = oldlen; ++ len[1] = newlen; ++ state[0] = oldstate; ++ state[1] = newstate; ++ ++ for (i = 0; i < 2; i++) ++ { ++ copy[i] = xmalloc (len[i] + 1); ++ memset (copy[i], '\0', len[i] + 1); ++ ++ for (j = 0, chars = 0; j < len[i] && chars < check_chars; chars++) ++ { ++ state_bak = state[i]; ++ mblength = mbrtowc (&wc, str[i] + j, len[i] - j, &(state[i])); ++ ++ switch (mblength) ++ { ++ case (size_t)-1: ++ case (size_t)-2: ++ state[i] = state_bak; ++ /* Fall through */ ++ case 0: ++ mblength = 1; ++ break; ++ ++ default: ++ if (ignore_case) ++ { ++ 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, str[i] + j, mblength); ++ } ++ else ++ memcpy (copy[i] + j, str[i] + j, mblength); ++ } ++ j += mblength; ++ } ++ copy[i][j] = '\0'; ++ len[i] = j; ++ } ++ int rc = xmemcoll (copy[0], len[0], copy[1], len[1]); ++ free (copy[0]); ++ free (copy[1]); ++ return rc; ++ ++} ++#endif ++ + /* Output the line in linebuffer LINE to standard output + provided that the switches say it should be output. + MATCH is true if the line matches the previous line. +@@ -367,19 +562,38 @@ check_file (const char *infile, const char *outfile, char delimiter) + char *prevfield IF_LINT ( = NULL); + size_t prevlen IF_LINT ( = 0); + bool first_group_printed = false; ++#if HAVE_MBRTOWC ++ mbstate_t prevstate; ++ ++ memset (&prevstate, '\0', sizeof (mbstate_t)); ++#endif + + while (!feof (stdin)) + { + char *thisfield; + size_t thislen; + bool new_group; ++#if HAVE_MBRTOWC ++ mbstate_t thisstate; ++#endif + + if (readlinebuffer_delim (thisline, stdin, delimiter) == 0) + break; + + thisfield = find_field (thisline); + thislen = thisline->length - 1 - (thisfield - thisline->buffer); ++#if HAVE_MBRTOWC ++ if (MB_CUR_MAX > 1) ++ { ++ thisstate = thisline->state; + ++ new_group = (prevline->length == 0 ++ || different_multi (thisfield, prevfield, ++ thislen, prevlen, ++ thisstate, prevstate)); ++ } ++ else ++#endif + new_group = (prevline->length == 0 + || different (thisfield, prevfield, thislen, prevlen)); + +@@ -397,6 +611,10 @@ check_file (const char *infile, const char *outfile, char delimiter) + SWAP_LINES (prevline, thisline); + prevfield = thisfield; + prevlen = thislen; ++#if HAVE_MBRTOWC ++ if (MB_CUR_MAX > 1) ++ prevstate = thisstate; ++#endif + first_group_printed = true; + } + } +@@ -409,17 +627,26 @@ check_file (const char *infile, const char *outfile, char delimiter) + size_t prevlen; + uintmax_t match_count = 0; + bool first_delimiter = true; ++#if HAVE_MBRTOWC ++ mbstate_t prevstate; ++#endif + + if (readlinebuffer_delim (prevline, stdin, delimiter) == 0) + goto closefiles; + prevfield = find_field (prevline); + prevlen = prevline->length - 1 - (prevfield - prevline->buffer); ++#if HAVE_MBRTOWC ++ prevstate = prevline->state; ++#endif + + while (!feof (stdin)) + { + bool match; + char *thisfield; + size_t thislen; ++#if HAVE_MBRTOWC ++ mbstate_t thisstate = thisline->state; ++#endif + if (readlinebuffer_delim (thisline, stdin, delimiter) == 0) + { + if (ferror (stdin)) +@@ -428,6 +655,14 @@ check_file (const char *infile, const char *outfile, char delimiter) + } + thisfield = find_field (thisline); + thislen = thisline->length - 1 - (thisfield - thisline->buffer); ++#if HAVE_MBRTOWC ++ if (MB_CUR_MAX > 1) ++ { ++ match = !different_multi (thisfield, prevfield, ++ thislen, prevlen, thisstate, prevstate); ++ } ++ else ++#endif + match = !different (thisfield, prevfield, thislen, prevlen); + match_count += match; + +@@ -460,6 +695,9 @@ check_file (const char *infile, const char *outfile, char delimiter) + SWAP_LINES (prevline, thisline); + prevfield = thisfield; + prevlen = thislen; ++#if HAVE_MBRTOWC ++ prevstate = thisstate; ++#endif + if (!match) + match_count = 0; + } +@@ -506,6 +744,19 @@ main (int argc, char **argv) + + atexit (close_stdout); + ++#if HAVE_MBRTOWC ++ if (MB_CUR_MAX > 1) ++ { ++ find_field = find_field_multi; ++ } ++ else ++#endif ++ { ++ find_field = find_field_uni; ++ } ++ ++ ++ + 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 100755 +index 0000000..26c95de +--- /dev/null ++++ b/tests/i18n/sort.sh +@@ -0,0 +1,29 @@ ++#!/bin/sh ++# Verify sort's multi-byte support. ++ ++. "${srcdir=.}/tests/init.sh"; path_prepend_ ./src ++print_ver_ sort ++ ++export LC_ALL=en_US.UTF-8 ++locale -k LC_CTYPE | grep -q "charmap.*UTF-8" \ ++ || skip_ "No UTF-8 locale available" ++ ++# Enable heap consistency checkng on older systems ++export MALLOC_CHECK_=2 ++ ++ ++# check buffer overflow issue due to ++# expanding multi-byte representation due to case conversion ++# https://bugzilla.suse.com/show_bug.cgi?id=928749 ++cat < exp ++. ++ɑ ++EOF ++cat < out || fail=1 ++. ++ɑ ++EOF ++compare exp out || { fail=1; cat out; } ++ ++ ++Exit $fail +diff --git a/tests/local.mk b/tests/local.mk +index 568944e..192f776 100644 +--- a/tests/local.mk ++++ b/tests/local.mk +@@ -362,6 +362,8 @@ 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/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 8a9cad1..9293e39 100755 +--- a/tests/misc/expand.pl ++++ b/tests/misc/expand.pl +@@ -27,6 +27,15 @@ my $prog = 'expand'; + # Turn off localization of executable's output. + @ENV{qw(LANGUAGE LANG LC_ALL)} = ('C') x 3; + ++#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 $prog = 'expand'; ++my $try = "Try \`$prog --help' for more information.\n"; ++my $inval = "$prog: invalid byte, character or field list\n$try"; ++ + my @Tests = + ( + ['t1', '--tabs=3', {IN=>"a\tb"}, {OUT=>"a b"}], +@@ -168,6 +177,8 @@ my @Tests = + + + # Test errors ++ # FIXME: The following tests contain ‘quoting’ specific to LC_MESSAGES ++ # So we force LC_MESSAGES=C to make them pass. + ['e1', '--tabs="a"', {IN=>''}, {OUT=>''}, {EXIT=>1}, + {ERR => "$prog: tab size contains invalid character(s): 'a'\n"}], + ['e2', "-t $UINTMAX_OFLOW", {IN=>''}, {OUT=>''}, {EXIT=>1}, +@@ -184,6 +195,37 @@ my @Tests = + {ERR => "$prog: '/' specifier not at start of number: '/'\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 expand 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 => "LANG=$mb_locale LC_MESSAGES=C"}]; ++ } ++ push @Tests, @new; ++ } ++ ++ ++@Tests = triple_test \@Tests; ++ + 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 $delim = chr 0247; + sub t_subst ($) + { +@@ -329,8 +338,49 @@ 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 = $ENV{VERBOSE}; + +diff --git a/tests/misc/sort-mb-tests.sh b/tests/misc/sort-mb-tests.sh +new file mode 100755 +index 0000000..11836ba +--- /dev/null ++++ b/tests/misc/sort-mb-tests.sh +@@ -0,0 +1,45 @@ ++#!/bin/sh ++# Verify sort's multi-byte support. ++ ++. "${srcdir=.}/tests/init.sh"; path_prepend_ ./src ++print_ver_ sort ++ ++export LC_ALL=en_US.UTF-8 ++locale -k LC_CTYPE | grep -q "charmap.*UTF-8" \ ++ || skip_ "No UTF-8 locale available" ++ ++ ++cat < exp ++Banana＠5 ++Apple＠10 ++Citrus＠20 ++Cherry＠30 ++EOF ++ ++cat < out || fail=1 ++Apple＠10 ++Banana＠5 ++Citrus＠20 ++Cherry＠30 ++EOF ++ ++compare exp out || { fail=1; cat out; } ++ ++ ++cat < exp ++Citrus＠ＡＡ20＠＠5 ++Cherry＠ＡＡ30＠＠10 ++Apple＠ＡＡ10＠＠20 ++Banana＠ＡＡ5＠＠30 ++EOF ++ ++cat < out || fail=1 ++Apple＠ＡＡ10＠＠20 ++Banana＠ＡＡ5＠＠30 ++Citrus＠ＡＡ20＠＠5 ++Cherry＠ＡＡ30＠＠10 ++EOF ++ ++compare exp out || { fail=1; cat out; } ++ ++Exit $fail +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; + ++my $mb_locale; ++# uncommented according to upstream commit enabling multibyte paths ++$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$try"; ++ + # three empty files and one that says 'foo' + my @inputs = (+(map{{IN=> {"empty$_"=> ''}}}1..3), {IN=> {foo=> "foo\n"}}); + +@@ -77,6 +86,39 @@ my @Tests = + {OUT=>$big_input}], + ); + ++# Add _POSIX2_VERSION=199209 to the environment of each test ++# that uses an old-style option like +1. ++if ($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 sort 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 =~ "nmerge-."); ++ 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/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; + +-my $mb_locale = $ENV{LOCALE_FR_UTF8}; ++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 $try = "Try \`$prog --help' for more information.\n"; ++my $inval = "$prog: invalid byte, character or field list\n$try"; ++ + # 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 '-'. +@@ -423,6 +428,38 @@ foreach my $t (@Tests) + } + } + ++if ($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 sort 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; ++ } ++ #disable several failing tests until investigation, disable all tests with envvars set ++ next if (grep {ref $_ eq 'HASH' && exists $_->{ENV}} (@new_t)); ++ next if ($test_name =~ "18g" or $test_name =~ "sort-numeric" or $test_name =~ "08[ab]" or $test_name =~ "03[def]" or $test_name =~ "h4" or $test_name =~ "n1" or $test_name =~ "2[01]a"); ++ next if ($test_name =~ "11[ab]"); # avoid FP: expected result differs to MB result due to collation rules. ++ 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" +@@ -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; ++@Tests = grep {$_->[0] ne 'output-is-input-mb.p'} @Tests; + + my $save_temps = $ENV{DEBUG}; + my $verbose = $ENV{VERBOSE}; +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'; ++ ++my $try = "Try \`$prog --help' for more information.\n"; ++my $inval = "$prog: invalid byte, character or field list\n$try"; ++ + my @Tests = + ( + ['a1', {IN=> ' 'x 1 ."y\n"}, {OUT=> ' 'x 1 ."y\n"}], +@@ -128,6 +136,37 @@ my @Tests = + ['ts2', '-t5,8', {IN=>"x\t \t y\n"}, {OUT=>"x\t\t y\n"}], + ); + ++if ($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; ++ ++ # 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/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"; ++ + # 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 $mb_locale = 'C'; ++ + # When possible, create a "-z"-testing variant of each test. + sub add_z_variants($) + { +@@ -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; ++ ++ # 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=$mb_locale"}]; ++ } ++ push @Tests, @new; ++ } ++ ++# Remember that triple_test creates from each test with exactly one "IN" ++# file two more tests (.p and .r suffix on name) corresponding to reading ++# input from a file and from a pipe. The pipe-reading test would fail ++# due to a race condition about 1 in 20 times. ++# Remove the IN_PIPE version of the "output-is-input" test above. ++# The others aren't susceptible because they have three inputs each. ++ ++@Tests = grep {$_->[0] ne 'output-is-input.p'} @Tests; ++ + @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$try"; ++ + my @tv = ( + + # -b option is no longer an official option. But it's still working to +@@ -474,8 +483,48 @@ push @Tests, + {IN=>{2=>"a\n"}}, + {OUT=>"a\t\t\t\t \t\t\ta\n"} ]; + ++# Add _POSIX2_VERSION=199209 to the environment of each test ++# that uses an old-style option like +1. ++if ($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; ++ ++ # 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}; + +-- +2.7.4 + diff --git a/coreutils-provides.inc b/coreutils-provides.inc new file mode 100644 index 0000000..04f296c --- /dev/null +++ b/coreutils-provides.inc @@ -0,0 +1,38 @@ +Provides: bundled(gnulib) +Provides: /bin/basename +Provides: /bin/cat +Provides: /bin/chgrp +Provides: /bin/chmod +Provides: /bin/chown +Provides: /bin/cp +Provides: /bin/cut +Provides: /bin/date +Provides: /bin/dd +Provides: /bin/df +Provides: /bin/echo +Provides: /bin/env +Provides: /bin/false +Provides: /bin/ln +Provides: /bin/ls +Provides: /bin/mkdir +Provides: /bin/mknod +Provides: /bin/mktemp +Provides: /bin/mv +Provides: /bin/nice +Provides: /bin/pwd +Provides: /bin/readlink +Provides: /bin/rm +Provides: /bin/rmdir +Provides: /bin/sleep +Provides: /bin/sort +Provides: /bin/stty +Provides: /bin/sync +Provides: /bin/touch +Provides: /bin/true +Provides: /bin/uname +Provides: bundled(gnulib) +Provides: fileutils = %{version}-%{release} +Provides: mktemp = 4:%{version}-%{release} +Provides: sh-utils = %{version}-%{release} +Provides: stat = %{version}-%{release} +Provides: textutils = %{version}-%{release} 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 new file mode 100644 index 0000000..58dfb5e --- /dev/null +++ b/coreutils.spec @@ -0,0 +1,204 @@ +Name: coreutils +Version: 8.30 +Release: 5 +License: GPLv3+ +Summary: A set of basic GNU tools commonly used in shell scripts +Url: https://www.gnu.org/software/coreutils/ +Source0: https://ftp.gnu.org/gnu/%{name}/%{name}-%{version}.tar.xz +#Source 50 51 105 and 106 come from fedora +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$ + +#patch 1 - 950 from fedora +Patch1: coreutils-8.30-renameatu.patch +Patch100: coreutils-8.26-test-lock.patch +Patch105: coreutils-8.26-selinuxenable.patch +Patch101: coreutils-6.10-manpages.patch +Patch102: coreutils-8.25-DIR_COLORS.patch +Patch103: coreutils-8.2-uname-processortype.patch +Patch104: coreutils-df-direct.patch +Patch107: coreutils-8.4-mkdir-modenote.patch +Patch703: sh-utils-2.0.11-dateman.patch +Patch713: coreutils-4.5.3-langinfo.patch +Patch800: coreutils-i18n.patch +Patch801: coreutils-i18n-expand-unexpand.patch +Patch804: coreutils-i18n-cut-old.patch +Patch803: coreutils-i18n-fix-unexpand.patch +Patch805: coreutils-i18n-fix2-expand-unexpand.patch +Patch806: coreutils-i18n-un-expand-BOM.patch +Patch807: coreutils-i18n-sort-human.patch +Patch808: coreutils-i18n-fold-newline.patch +Patch908: coreutils-getgrouplist.patch +Patch950: coreutils-selinux.patch + +#openEuler +Patch6000: bugfix-remove-usr-local-lib-from-m4.patch +Patch6001: bugfix-dummy_help2man.patch +Patch6002: bugfix-selinux-flask.patch +Patch6003: echo-always-process-escapes-when-POSIXLY_CORRECT-is-.patch +Patch6004: sync-fix-open-fallback-bug.patch +Patch6005: tail-fix-handling-of-broken-pipes-with-SIGPIPE-ignor.patch +Patch6006: seq-output-decimal-points-consistently-with-invalid-.patch + +Conflicts: filesystem < 3 +# To avoid clobbering installs +Provides: /bin/sh + +Conflicts: %{name}-single +Obsoletes: %{name}-common +Provides: %{name}-common = %{version}-%{release} + +BuildRequires: attr, autoconf, automake, gcc, hostname, strace, texinfo +BuildRequires: gettext-devel, gmp-devel, libacl-devel, libattr-devel +BuildRequires: libcap-devel, libselinux-devel, libselinux-utils, openssl-devel + +Requires: ncurses, gmp +Requires(preun): /sbin/install-info +Requires(post): /sbin/install-info + +Provides: coreutils-full = %{version}-%{release} +%include %{SOURCE51} +Obsoletes: %{name} < 8.24 + +%description +These are the GNU core utilities. This package is the combination of +the old GNU fileutils, sh-utils, and textutils packages. + +%prep +%autosetup -N + +tee DIR_COLORS{,.256color,.lightbgcolor} /dev/null + +%autopatch -p1 + +(echo ">>> Fixing permissions on tests") 2>/dev/null +find tests -name '*.sh' -perm 0644 -print -exec chmod 0755 '{}' '+' +(echo "<<< done") 2>/dev/null + +autoreconf -fiv + +%build +export CFLAGS="$RPM_OPT_FLAGS -fno-strict-aliasing -fpic" +%{expand:%%global optflags %{optflags} -D_GNU_SOURCE=1} +mkdir separate && \ + (cd separate && ln -s ../configure || exit 1 +%configure --with-openssl \ + --cache-file=../config.cache \ + --enable-install-program=arch \ + --enable-no-install-program=kill,uptime \ + --with-tty-group \ + DEFAULT_POSIX2_VERSION=200112 alternative=199209 || : +make all %{?_smp_mflags} + +# make sure that parse-datetime.{c,y} ends up in debuginfo (#1555079) +ln -v ../lib/parse-datetime.{c,y} . + ) + +# Get the list of supported utilities +cp %SOURCE50 . + +%install +(cd separate && make DESTDIR=$RPM_BUILD_ROOT install) + +# chroot was in /usr/sbin : +mkdir -p $RPM_BUILD_ROOT/{%{_bindir},%{_sbindir}} +mv $RPM_BUILD_ROOT/{%_bindir,%_sbindir}/chroot + +mkdir -p $RPM_BUILD_ROOT%{_sysconfdir}/profile.d +install -p -c -m644 DIR_COLORS{,.256color,.lightbgcolor} $RPM_BUILD_ROOT%{_sysconfdir} +install -p -c -m644 %SOURCE105 $RPM_BUILD_ROOT%{_sysconfdir}/profile.d/colorls.sh +install -p -c -m644 %SOURCE106 $RPM_BUILD_ROOT%{_sysconfdir}/profile.d/colorls.csh + +%find_lang %name +# Add the %%lang(xyz) ownership for the LC_TIME dirs as well... +grep LC_TIME %name.lang | cut -d'/' -f1-6 | sed -e 's/) /) %%dir /g' >>%name.lang + +%preun +if [ $1 = 0 ]; then + if [ -f %{_infodir}/%{name}.info.gz ]; then + /sbin/install-info --delete %{_infodir}/%{name}.info.gz %{_infodir}/dir || : + fi +fi + +%post +if [ -f %{_infodir}/%{name}.info.gz ]; then + /sbin/install-info %{_infodir}/%{name}.info.gz %{_infodir}/dir || : +fi + +%files -f supported_utils -f %{name}.lang +%dir %{_libexecdir}/coreutils +%{_libexecdir}/coreutils/*.so +%config(noreplace) %{_sysconfdir}/DIR_COLORS* +%config(noreplace) %{_sysconfdir}/profile.d/* +%doc ABOUT-NLS NEWS README THANKS TODO +%license COPYING +%exclude %{_infodir}/dir + +%package_help +%files help +%{_infodir}/coreutils* +%{_mandir}/man*/* + +%changelog +* Thu Aug 29 2019 hexiaowen - 8.30-5 +- Package rebuild + +* Wed Aug 21 2019 gaoyi - 8.30-4.h8 +- Type: enhancement +- ID: NA +- SUG: NA +- DESC: remove patches' prefix starting with backport + +* Wed May 08 2019 gulining - 8.30-4.h7 +- Type:bugfix +- ID:NA +- SUG:NA +- DESC:revert patch for rtos + +* Wed May 8 2019 liusirui - 8.30-4.h6 +- Type:bugfix +- ID:NA +- SUG:NA +- DESC:rename some patches + +* Sat Apr 6 2019 luochunsheng - 8.30-4.h5 +- Type:enhancement +- ID:NA +- SUG:NA +- DESC:remove sensitive information + +* Fri Mar 15 2019 yanghua - 8.30-4.h4 +- Type:bugfix +- ID:NA +- SUG:restart +- DESC:echo always process escapes when POSIXLY_CORRECT is + sync fix open fallback bug + tail fix handling of broken pipes with SIGPIPE ignor + seq output decimal points consistently with invalid + +* Thu Feb 14 2019 zoujing - 8.30-4.h3 +- Type:bugfix +- ID:NA +- SUG:NA +- DESC:failed to exec scriptlet interpreter + +* Wed Jan 23 2019 sunguoshuai - 8.30-4.h2 +- Type:bugfix +- ID:NA +- SUG:NA +- DESC:sync + +* Fri Dec 28 2018 hushiyuan - 8.30-4.h1 +- Type:bugfix +- ID:NA +- SUG:NA +- DESC:add provides to coreutils-single to make it a drop-in replacement (#1572693) +- reintroduce very old Provides (mktemp, sh-utils, textwrap, fileutils, stat) + +* Thu Jul 12 2018 hexiaowen - 8.30-1 +- Pacakge init diff --git a/echo-always-process-escapes-when-POSIXLY_CORRECT-is-.patch b/echo-always-process-escapes-when-POSIXLY_CORRECT-is-.patch new file mode 100644 index 0000000..8741f98 --- /dev/null +++ b/echo-always-process-escapes-when-POSIXLY_CORRECT-is-.patch @@ -0,0 +1,165 @@ +From 8d328706c6ef0264dffb923f96431af1e26dcddd Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?P=C3=A1draig=20Brady?= +Date: Sun, 23 Sep 2018 19:35:24 -0700 +Subject: [PATCH 24/87] echo: always process escapes when POSIXLY_CORRECT is + set + +* src/echo.c (main): Always enable backslash processing if +POSIXLY_CORRECT is set. +* tests/misc/echo.sh: Add (the first) test for the echo command. +* tests/local.mk: Reference the new test. +* tests/misc/printf.sh: Update a stale comment. +* doc/coreutils.texi (echo invocation). Mention that POSIXLY_CORRECT +now always enables backslash processing. +* NEWS: Mention the change in behavior. +Fixes https://bugs.gnu.org/32703 +Issue identified by Eric Blake. +--- + doc/coreutils.texi | 2 +- + src/echo.c | 5 ++-- + tests/local.mk | 1 + + tests/misc/echo.sh | 66 ++++++++++++++++++++++++++++++++++++++++++++ + tests/misc/printf.sh | 5 ++-- + 5 files changed, 73 insertions(+), 6 deletions(-) + create mode 100755 tests/misc/echo.sh + +diff --git a/doc/coreutils.texi b/doc/coreutils.texi +index 30155bb86..f8339d73f 100644 +--- a/doc/coreutils.texi ++++ b/doc/coreutils.texi +@@ -12608,7 +12608,7 @@ If the @env{POSIXLY_CORRECT} environment variable is set, then when + @command{echo}'s first argument is not @option{-n} it outputs + option-like arguments instead of treating them as options. For + example, @code{echo -ne hello} outputs @samp{-ne hello} instead of +-plain @samp{hello}. ++plain @samp{hello}. Also backslash escapes are always enabled. + + POSIX does not require support for any options, and says + that the behavior of @command{echo} is implementation-defined if any +diff --git a/src/echo.c b/src/echo.c +index 9958a8ec6..2aee5acfb 100644 +--- a/src/echo.c ++++ b/src/echo.c +@@ -108,8 +108,9 @@ int + main (int argc, char **argv) + { + bool display_return = true; ++ bool posixly_correct = getenv ("POSIXLY_CORRECT"); + bool allow_options = +- (! getenv ("POSIXLY_CORRECT") ++ (! posixly_correct + || (! DEFAULT_ECHO_TO_XPG && 1 < argc && STREQ (argv[1], "-n"))); + + /* System V machines already have a /bin/sh with a v9 behavior. +@@ -189,7 +190,7 @@ main (int argc, char **argv) + + just_echo: + +- if (do_v9) ++ if (do_v9 || posixly_correct) + { + while (argc > 0) + { +diff --git a/tests/local.mk b/tests/local.mk +index 84d346979..5823886f8 100644 +--- a/tests/local.mk ++++ b/tests/local.mk +@@ -237,6 +237,7 @@ all_tests = \ + tests/fmt/base.pl \ + tests/fmt/long-line.sh \ + tests/fmt/goal-option.sh \ ++ tests/misc/echo.sh \ + tests/misc/env.sh \ + tests/misc/ptx.pl \ + tests/misc/test.pl \ +diff --git a/tests/misc/echo.sh b/tests/misc/echo.sh +new file mode 100755 +index 000000000..53e87fb71 +--- /dev/null ++++ b/tests/misc/echo.sh +@@ -0,0 +1,66 @@ ++#!/bin/sh ++# basic tests for echo ++ ++# Copyright (C) 2018 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 . ++ ++prog='env echo' ++ ++. "${srcdir=.}/tests/init.sh"; path_prepend_ ./src ++print_ver_ echo ++ ++ ++# Verify the methods of specifying "Escape": ++# Note 4 octal digits are allowed (unlike printf which uses up to 3) ++printf '%s\n' . . . . . | tr . '\033' > exp ++$prog -n -e '\x1b\n\e\n\33\n\033\n\0033\n' > out || fail=1 ++compare exp out || fail=1 ++ ++# Incomplete hex escapes are output as is (unlike printf) ++printf '\\x\n' > exp ++$prog -n -e '\x\n' > out || fail=1 ++compare exp out || fail=1 ++ ++# Always output -- (unlike printf) ++$prog -- 'foo' > out || fail=1 ++$prog -n -e -- 'foo\n' >> out || fail=1 ++cat <<\EOF > exp ++-- foo ++-- foo ++EOF ++compare exp out || fail=1 ++ ++# Ensure \c stops processing ++$prog -e 'foo\n\cbar' > out || fail=1 ++printf 'foo\n' > exp ++compare exp out || fail=1 ++ ++# With POSIXLY_CORRECT: ++# only -n as the first (separate) option enables option processing ++# -E is ignored ++# escapes are processed by default ++POSIXLY_CORRECT=1 $prog -n -E 'foo\n' > out || fail=1 ++POSIXLY_CORRECT=1 $prog -nE 'foo' >> out || fail=1 ++POSIXLY_CORRECT=1 $prog -E -n 'foo' >> out || fail=1 ++POSIXLY_CORRECT=1 $prog --version >> out || fail=1 ++cat <<\EOF > exp ++foo ++-nE foo ++-E -n foo ++--version ++EOF ++compare exp out || fail=1 ++ ++Exit $fail +diff --git a/tests/misc/printf.sh b/tests/misc/printf.sh +index 8716c56c8..dccabbe8c 100755 +--- a/tests/misc/printf.sh ++++ b/tests/misc/printf.sh +@@ -60,9 +60,8 @@ $prog '5 % +d\n' 234 >> out || fail=1 + # coreutils-5.0.1, it would print six bytes: "6 \41\n". + $prog '6 \41\n' | tr '\41' '!' >> out + +-# Note that as of coreutils-5.0.1, printf with a format of '\0002x' +-# prints a NUL byte followed by the digit '2' and an 'x'. +-# By contrast bash's printf outputs the same thing as $(printf '\2x') does. ++# Note that as of coreutils-5.0.1, printf with a format of '\0002y' ++# prints a NUL byte followed by the digit '2' and a 'y'. + $prog '7 \2y \02y \002y \0002y\n' |tr '\0\2' '*=' >> out + + $prog '8 %b %b %b %b\n' '\1y' '\01y' '\001y' '\0001y'|tr '\1' = >> out +-- +2.19.1 + diff --git a/seq-output-decimal-points-consistently-with-invalid-.patch b/seq-output-decimal-points-consistently-with-invalid-.patch new file mode 100644 index 0000000..f670bda --- /dev/null +++ b/seq-output-decimal-points-consistently-with-invalid-.patch @@ -0,0 +1,104 @@ +From 03fbb5ecec045d41b28c5ec9750ee94e4bea074d Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?P=C3=A1draig=20Brady?= +Date: Sat, 2 Feb 2019 19:16:02 -0800 +Subject: [PATCH 57/87] seq: output decimal points consistently with invalid + locales + +* src/seq.c (print_numbers): Only reset the locale if it +was successfully set originally. +* tests/misc/seq-locale.sh: Add a new test. +* tests/local.mk: Reference the new test. +* NEWS: Mention the fix. +--- + src/seq.c | 11 ++++++++--- + tests/local.mk | 1 + + tests/misc/seq-locale.sh | 28 ++++++++++++++++++++++++++++ + 3 files changed, 37 insertions(+), 3 deletions(-) + create mode 100755 tests/misc/seq-locale.sh + +diff --git a/src/seq.c b/src/seq.c +index 61d20fe62..b5913368a 100644 +--- a/src/seq.c ++++ b/src/seq.c +@@ -42,6 +42,9 @@ + + #define AUTHORS proper_name ("Ulrich Drepper") + ++/* True if the locale settings were honored. */ ++static bool locale_ok; ++ + /* If true print all number with equal width. */ + static bool equal_width; + +@@ -324,9 +327,11 @@ print_numbers (char const *fmt, struct layout layout, + long double x_val; + char *x_str; + int x_strlen; +- setlocale (LC_NUMERIC, "C"); ++ if (locale_ok) ++ setlocale (LC_NUMERIC, "C"); + x_strlen = asprintf (&x_str, fmt, x); +- setlocale (LC_NUMERIC, ""); ++ if (locale_ok) ++ setlocale (LC_NUMERIC, ""); + if (x_strlen < 0) + xalloc_die (); + x_str[x_strlen - layout.suffix_len] = '\0'; +@@ -559,7 +564,7 @@ main (int argc, char **argv) + + initialize_main (&argc, &argv); + set_program_name (argv[0]); +- setlocale (LC_ALL, ""); ++ locale_ok = !!setlocale (LC_ALL, ""); + bindtextdomain (PACKAGE, LOCALEDIR); + textdomain (PACKAGE); + +diff --git a/tests/local.mk b/tests/local.mk +index 290e30f62..4751886b2 100644 +--- a/tests/local.mk ++++ b/tests/local.mk +@@ -244,6 +244,7 @@ all_tests = \ + tests/misc/seq.pl \ + tests/misc/seq-epipe.sh \ + tests/misc/seq-io-errors.sh \ ++ tests/misc/seq-locale.sh \ + tests/misc/seq-long-double.sh \ + tests/misc/seq-precision.sh \ + tests/misc/head.pl \ +diff --git a/tests/misc/seq-locale.sh b/tests/misc/seq-locale.sh +new file mode 100755 +index 000000000..8a46ab7e4 +--- /dev/null ++++ b/tests/misc/seq-locale.sh +@@ -0,0 +1,28 @@ ++#!/bin/sh ++# Test for output with appropriate precision ++ ++# Copyright (C) 2019 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_ seq ++ ++# With coreutils-8.30 and earlier, the last decimal point would be ',' ++# when setlocale(LC_ALL, "") failed, but setlocale(LC_NUMERIC, "") succeeded. ++LC_ALL= LANG=invalid LC_NUMERIC=$LOCALE_FR_UTF8 seq 0.1 0.2 0.7 > out || fail=1 ++uniq -w2 out > out-merge || framework_failure_ ++test "$(wc -l < out-merge)" = 1 || { fail=1; cat out; } ++ ++Exit $fail +-- +2.19.1 + 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/supported_utils b/supported_utils new file mode 100644 index 0000000..8105ccf --- /dev/null +++ b/supported_utils @@ -0,0 +1,104 @@ +%{_bindir}/arch +%{_bindir}/b2sum +%{_bindir}/basename +%{_bindir}/cat +%{_bindir}/chgrp +%{_bindir}/chmod +%{_bindir}/chown +%{_bindir}/cp +%{_bindir}/cut +%{_bindir}/date +%{_bindir}/dd +%{_bindir}/df +%{_bindir}/echo +%{_bindir}/env +%{_bindir}/false +%{_bindir}/link +%{_bindir}/ln +%{_bindir}/ls +%{_bindir}/mkdir +%{_bindir}/mknod +%{_bindir}/mv +%{_bindir}/nice +%{_bindir}/pwd +%{_bindir}/readlink +%{_bindir}/rm +%{_bindir}/rmdir +%{_bindir}/sleep +%{_bindir}/sort +%{_bindir}/stty +%{_bindir}/sync +%{_bindir}/mktemp +%{_bindir}/touch +%{_bindir}/true +%{_bindir}/uname +%{_bindir}/unlink +%{_bindir}/[ +%{_bindir}/base32 +%{_bindir}/base64 +%{_bindir}/chcon +%{_bindir}/cksum +%{_bindir}/comm +%{_bindir}/csplit +%{_bindir}/dir +%{_bindir}/dircolors +%{_bindir}/dirname +%{_bindir}/du +%{_bindir}/expand +%{_bindir}/expr +%{_bindir}/factor +%{_bindir}/fmt +%{_bindir}/fold +%{_bindir}/groups +%{_bindir}/head +%{_bindir}/hostid +%{_bindir}/id +%{_bindir}/install +%{_bindir}/join +%{_bindir}/logname +%{_bindir}/md5sum +%{_bindir}/mkfifo +%{_bindir}/nl +%{_bindir}/nohup +%{_bindir}/nproc +%{_bindir}/numfmt +%{_bindir}/od +%{_bindir}/paste +%{_bindir}/pathchk +%{_bindir}/pinky +%{_bindir}/pr +%{_bindir}/printenv +%{_bindir}/printf +%{_bindir}/ptx +%{_bindir}/realpath +%{_bindir}/runcon +%{_bindir}/seq +%{_bindir}/sha1sum +%{_bindir}/sha224sum +%{_bindir}/sha256sum +%{_bindir}/sha384sum +%{_bindir}/sha512sum +%{_bindir}/shred +%{_bindir}/shuf +%{_bindir}/split +%{_bindir}/stat +%{_bindir}/stdbuf +%{_bindir}/sum +%{_bindir}/tac +%{_bindir}/tail +%{_bindir}/tee +%{_bindir}/test +%{_bindir}/timeout +%{_bindir}/tr +%{_bindir}/truncate +%{_bindir}/tsort +%{_bindir}/tty +%{_bindir}/unexpand +%{_bindir}/uniq +%{_bindir}/users +%{_bindir}/vdir +%{_bindir}/wc +%{_bindir}/who +%{_bindir}/whoami +%{_bindir}/yes +%{_sbindir}/chroot diff --git a/sync-fix-open-fallback-bug.patch b/sync-fix-open-fallback-bug.patch new file mode 100644 index 0000000..342a680 --- /dev/null +++ b/sync-fix-open-fallback-bug.patch @@ -0,0 +1,32 @@ +From 94d364f157f007f2b23c70863ac8eefe9b21229d Mon Sep 17 00:00:00 2001 +From: Paul Eggert +Date: Tue, 6 Nov 2018 10:35:16 -0800 +Subject: [PATCH 28/87] sync: fix open fallback bug + +Problem caught by Coverity Analysis +and reported by Kamil Dudka (Bug#33287). +* src/sync.c (sync_arg): Fix typo in fallback code. +--- + src/sync.c | 6 ++++-- + 1 file changed, 4 insertions(+), 2 deletions(-) + +diff --git a/src/sync.c b/src/sync.c +index bd3671a19..607fa8f7f 100644 +--- a/src/sync.c ++++ b/src/sync.c +@@ -111,8 +111,10 @@ sync_arg (enum sync_mode mode, char const *file) + if (open_flags != (O_WRONLY | O_NONBLOCK)) + fd = open (file, O_WRONLY | O_NONBLOCK); + if (fd < 0) +- error (0, rd_errno, _("error opening %s"), quoteaf (file)); +- return false; ++ { ++ error (0, rd_errno, _("error opening %s"), quoteaf (file)); ++ return false; ++ } + } + + /* We used O_NONBLOCK above to not hang with fifos, +-- +2.19.1 + diff --git a/tail-fix-handling-of-broken-pipes-with-SIGPIPE-ignor.patch b/tail-fix-handling-of-broken-pipes-with-SIGPIPE-ignor.patch new file mode 100644 index 0000000..8e3e046 --- /dev/null +++ b/tail-fix-handling-of-broken-pipes-with-SIGPIPE-ignor.patch @@ -0,0 +1,73 @@ +From fa50623394f491b975dbd7ad73193519dd721771 Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?P=C3=A1draig=20Brady?= +Date: Sun, 20 Jan 2019 00:13:15 -0800 +Subject: [PATCH 46/87] tail: fix handling of broken pipes with SIGPIPE ignored + +* init.cfg (trap_sigpipe_or_skip_): A new function refactored from... +* tests/misc/printf-surprise.sh: ...here. +* tests/misc/seq-epipe.sh. Likewise. +* src/tail.c (die_pipe): Ensure we exit upon sending SIGPIPE. +* tests/tail-2/pipe-f.sh: Ensure we exit even if SIGPIPE is ignored. +* NEWS: Mention the bug fix. +--- + init.cfg | 6 ++++++ + src/tail.c | 12 ++++++++++-- + 2 files changed, 16 insertions(+), 2 deletions(-) + +diff --git a/init.cfg b/init.cfg +index d99a0e3dd..739ba33df 100644 +--- a/init.cfg ++++ b/init.cfg +@@ -610,6 +610,12 @@ mkfifo_or_skip_() + fi + } + ++trap_sigpipe_or_skip_() ++{ ++ (trap '' PIPE && yes | :) 2>&1 | grep -qF 'Broken pipe' || ++ skip_ 'trapping SIGPIPE is not supported' ++} ++ + # Disable the current test if the working directory seems to have + # the setgid bit set. + skip_if_setgid_() +diff --git a/src/tail.c b/src/tail.c +index c63b61631..b8064853f 100644 +--- a/src/tail.c ++++ b/src/tail.c +@@ -330,6 +330,14 @@ named file in a way that accommodates renaming, removal and creation.\n\ + exit (status); + } + ++/* Ensure exit, either with SIGPIPE or EXIT_FAILURE status. */ ++static void ATTRIBUTE_NORETURN ++die_pipe (void) ++{ ++ raise (SIGPIPE); ++ exit (EXIT_FAILURE); ++} ++ + /* If the output has gone away, then terminate + as we would if we had written to this output. */ + static void +@@ -348,7 +356,7 @@ check_output_alive (void) + /* 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) +- raise (SIGPIPE); ++ die_pipe(); + } + + static bool +@@ -1645,7 +1653,7 @@ tail_forever_inotify (int wd, struct File_spec *f, size_t n_files, + { + /* readable event on STDOUT is equivalent to POLLERR, + and implies an error on output like broken pipe. */ +- raise (SIGPIPE); ++ die_pipe(); + } + else + break; +-- +2.19.1 +