diff options
Diffstat (limited to 'recipes-core/coreutils/coreutils-6.9')
12 files changed, 5026 insertions, 0 deletions
diff --git a/recipes-core/coreutils/coreutils-6.9/coreutils-6.9-cp-i-u.patch b/recipes-core/coreutils/coreutils-6.9/coreutils-6.9-cp-i-u.patch new file mode 100644 index 0000000..5452b46 --- /dev/null +++ b/recipes-core/coreutils/coreutils-6.9/coreutils-6.9-cp-i-u.patch | |||
@@ -0,0 +1,120 @@ | |||
1 | Upstream-Status: Inappropriate [legacy version] | ||
2 | |||
3 | This patch was imported from the Fedora Core 8 coreutils-6.9-9 package. | ||
4 | |||
5 | The package is stated as being Licensed as GPLv2+. | ||
6 | |||
7 | Signed-off-by: Mark Hatle <mark.hatle@windriver.com> | ||
8 | |||
9 | ---- | ||
10 | |||
11 | When "cp -i --update old new" would do nothing because "new" is | ||
12 | newer than "old", cp would nonetheless prompt for whether it is | ||
13 | ok to overwrite "new". Then, regardless of the response (because | ||
14 | of the --update option), cp would do nothing. | ||
15 | |||
16 | The following patch eliminates the unnecessary prompt in that case. | ||
17 | |||
18 | diff --git a/src/copy.c b/src/copy.c | ||
19 | index b7bf73b..0e549d2 100644 | ||
20 | --- a/src/copy.c | ||
21 | +++ b/src/copy.c | ||
22 | @@ -1210,6 +1210,30 @@ copy_internal (char const *src_name, char const *dst_name, | ||
23 | return false; | ||
24 | } | ||
25 | |||
26 | + if (!S_ISDIR (src_mode) && x->update) | ||
27 | + { | ||
28 | + /* When preserving time stamps (but not moving within a file | ||
29 | + system), don't worry if the destination time stamp is | ||
30 | + less than the source merely because of time stamp | ||
31 | + truncation. */ | ||
32 | + int options = ((x->preserve_timestamps | ||
33 | + && ! (x->move_mode | ||
34 | + && dst_sb.st_dev == src_sb.st_dev)) | ||
35 | + ? UTIMECMP_TRUNCATE_SOURCE | ||
36 | + : 0); | ||
37 | + | ||
38 | + if (0 <= utimecmp (dst_name, &dst_sb, &src_sb, options)) | ||
39 | + { | ||
40 | + /* We're using --update and the destination is not older | ||
41 | + than the source, so do not copy or move. Pretend the | ||
42 | + rename succeeded, so the caller (if it's mv) doesn't | ||
43 | + end up removing the source file. */ | ||
44 | + if (rename_succeeded) | ||
45 | + *rename_succeeded = true; | ||
46 | + return true; | ||
47 | + } | ||
48 | + } | ||
49 | + | ||
50 | /* When there is an existing destination file, we may end up | ||
51 | returning early, and hence not copying/moving the file. | ||
52 | This may be due to an interactive `negative' reply to the | ||
53 | @@ -1302,30 +1326,6 @@ copy_internal (char const *src_name, char const *dst_name, | ||
54 | return false; | ||
55 | } | ||
56 | } | ||
57 | - | ||
58 | - if (x->update) | ||
59 | - { | ||
60 | - /* When preserving time stamps (but not moving within a file | ||
61 | - system), don't worry if the destination time stamp is | ||
62 | - less than the source merely because of time stamp | ||
63 | - truncation. */ | ||
64 | - int options = ((x->preserve_timestamps | ||
65 | - && ! (x->move_mode | ||
66 | - && dst_sb.st_dev == src_sb.st_dev)) | ||
67 | - ? UTIMECMP_TRUNCATE_SOURCE | ||
68 | - : 0); | ||
69 | - | ||
70 | - if (0 <= utimecmp (dst_name, &dst_sb, &src_sb, options)) | ||
71 | - { | ||
72 | - /* We're using --update and the destination is not older | ||
73 | - than the source, so do not copy or move. Pretend the | ||
74 | - rename succeeded, so the caller (if it's mv) doesn't | ||
75 | - end up removing the source file. */ | ||
76 | - if (rename_succeeded) | ||
77 | - *rename_succeeded = true; | ||
78 | - return true; | ||
79 | - } | ||
80 | - } | ||
81 | } | ||
82 | |||
83 | if (x->move_mode) | ||
84 | diff --git a/tests/mv/update b/tests/mv/update | ||
85 | index 0c06024..6c3d149 100755 | ||
86 | --- a/tests/mv/update | ||
87 | +++ b/tests/mv/update | ||
88 | @@ -1,7 +1,7 @@ | ||
89 | #!/bin/sh | ||
90 | # make sure --update works as advertised | ||
91 | |||
92 | -# Copyright (C) 2001, 2004, 2006 Free Software Foundation, Inc. | ||
93 | +# Copyright (C) 2001, 2004, 2006-2007 Free Software Foundation, Inc. | ||
94 | |||
95 | # This program is free software; you can redistribute it and/or modify | ||
96 | # it under the terms of the GNU General Public License as published by | ||
97 | @@ -46,11 +46,16 @@ fi | ||
98 | |||
99 | fail=0 | ||
100 | |||
101 | -for cp_or_mv in cp mv; do | ||
102 | - # This is a no-op. | ||
103 | - $cp_or_mv --update old new || fail=1 | ||
104 | - case "`cat new`" in new) ;; *) fail=1 ;; esac | ||
105 | - case "`cat old`" in old) ;; *) fail=1 ;; esac | ||
106 | +for interactive in '' -i; do | ||
107 | + for cp_or_mv in cp mv; do | ||
108 | + # This is a no-op, with no prompt. | ||
109 | + # With coreutils-6.9 and earlier, using --update with -i would | ||
110 | + # mistakenly elicit a prompt. | ||
111 | + $cp_or_mv $interactive --update old new < /dev/null > out 2>&1 || fail=1 | ||
112 | + test -s out && fail=1 | ||
113 | + case "`cat new`" in new) ;; *) fail=1 ;; esac | ||
114 | + case "`cat old`" in old) ;; *) fail=1 ;; esac | ||
115 | + done | ||
116 | done | ||
117 | |||
118 | # This will actually perform the rename. | ||
119 | -- | ||
120 | 1.5.3.rc1.16.g9d6f | ||
diff --git a/recipes-core/coreutils/coreutils-6.9/coreutils-fix-install.patch b/recipes-core/coreutils/coreutils-6.9/coreutils-fix-install.patch new file mode 100644 index 0000000..88f61fa --- /dev/null +++ b/recipes-core/coreutils/coreutils-6.9/coreutils-fix-install.patch | |||
@@ -0,0 +1,101 @@ | |||
1 | Upstream-Status: Inappropriate [legacy version] | ||
2 | |||
3 | The install command doesn't over write the dangling symlink, for | ||
4 | example: | ||
5 | |||
6 | $ install fileA /tmp/fileA | ||
7 | |||
8 | If /tmp/fileA is a dangling symlink, there would be an error: | ||
9 | |||
10 | install: cannot create regular file '/tmp/fileA': File exists | ||
11 | |||
12 | This is because of the following code in copy.c: | ||
13 | |||
14 | if (!new_dst) | ||
15 | { | ||
16 | if (XSTAT (x, dst_name, &dst_sb) != 0) | ||
17 | { | ||
18 | if (errno != ENOENT) | ||
19 | { | ||
20 | error (0, errno, _("cannot stat %s"), quote (dst_name)); | ||
21 | return false; | ||
22 | } | ||
23 | else | ||
24 | { | ||
25 | new_dst = true; | ||
26 | } | ||
27 | } | ||
28 | |||
29 | XSTAT() use stat() for dst_name(the dangling symlink /tmp/fileA) when | ||
30 | install.c invokes it, and stat will set errno to ENOENT, and then | ||
31 | new_dst will be set to true which means that /tmp/fileA doesn't exist, | ||
32 | then we will create /tmp/fileA without remove it first, so the error | ||
33 | comes. | ||
34 | |||
35 | This is fixed in a way which adds the member cmd_install in | ||
36 | struct cp_options to make sure my change only affected to the install | ||
37 | command and use lstat to fix the problem. | ||
38 | |||
39 | Signed-off-by: Robert Yang <liezhi.yang@windriver.com> | ||
40 | Signed-off-by: Mark Hatle <mark.hatle@windriver.com> | ||
41 | |||
42 | --- | ||
43 | src/copy.c | 10 +++++++++- | ||
44 | src/copy.h | 3 +++ | ||
45 | src/install.c | 1 + | ||
46 | 3 files changed, 13 insertions(+), 1 deletions(-) | ||
47 | |||
48 | diff --git a/src/copy.c b/src/copy.c | ||
49 | --- a/src/copy.c | ||
50 | +++ b/src/copy.c | ||
51 | @@ -1029,6 +1029,7 @@ copy_internal (char const *src_name, char const *dst_name, | ||
52 | bool delayed_ok; | ||
53 | bool copied_as_regular = false; | ||
54 | bool preserve_metadata; | ||
55 | + int dst_stat_result; | ||
56 | |||
57 | if (x->move_mode && rename_succeeded) | ||
58 | *rename_succeeded = false; | ||
59 | @@ -1069,7 +1070,14 @@ copy_internal (char const *src_name, char const *dst_name, | ||
60 | |||
61 | if (!new_dst) | ||
62 | { | ||
63 | - if (XSTAT (x, dst_name, &dst_sb) != 0) | ||
64 | + if ( x->cmd_install && ( x->backup_type == no_backups)) | ||
65 | + dst_stat_result = lstat (dst_name, &dst_sb); | ||
66 | + else | ||
67 | + { | ||
68 | + dst_stat_result = XSTAT (x, dst_name, &dst_sb); | ||
69 | + } | ||
70 | + | ||
71 | + if (dst_stat_result != 0) | ||
72 | { | ||
73 | if (errno != ENOENT) | ||
74 | { | ||
75 | diff --git a/src/copy.h b/src/copy.h | ||
76 | --- a/src/copy.h | ||
77 | +++ b/src/copy.h | ||
78 | @@ -114,6 +114,9 @@ struct cp_options | ||
79 | If that fails, then resort to copying. */ | ||
80 | bool move_mode; | ||
81 | |||
82 | + /* For the install command */ | ||
83 | + bool cmd_install; | ||
84 | + | ||
85 | /* Whether this process has appropriate privileges to chown a file | ||
86 | whose owner is not the effective user ID. */ | ||
87 | bool chown_privileges; | ||
88 | diff --git a/src/install.c b/src/install.c | ||
89 | --- a/src/install.c | ||
90 | +++ b/src/install.c | ||
91 | @@ -149,6 +149,7 @@ cp_option_init (struct cp_options *x) | ||
92 | x->hard_link = false; | ||
93 | x->interactive = I_UNSPECIFIED; | ||
94 | x->move_mode = false; | ||
95 | + x->cmd_install = true; | ||
96 | x->chown_privileges = chown_privileges (); | ||
97 | x->one_file_system = false; | ||
98 | x->preserve_ownership = false; | ||
99 | -- | ||
100 | 1.7.0.1 | ||
101 | |||
diff --git a/recipes-core/coreutils/coreutils-6.9/coreutils-fix-texinfo.patch b/recipes-core/coreutils/coreutils-6.9/coreutils-fix-texinfo.patch new file mode 100644 index 0000000..3ae5a2f --- /dev/null +++ b/recipes-core/coreutils/coreutils-6.9/coreutils-fix-texinfo.patch | |||
@@ -0,0 +1,375 @@ | |||
1 | From 170be4023bbf9e9698a709e03265945588ac8e01 Mon Sep 17 00:00:00 2001 | ||
2 | From: Robert Yang <liezhi.yang@windriver.com> | ||
3 | Date: Tue, 26 Nov 2013 00:21:50 +0800 | ||
4 | Subject: [PATCH] doc/coreutils.texi: Use '@item' instead of '@itemx' | ||
5 | |||
6 | Use '@item' instead of '@itemx' in several places, as Texinfo 5 refuses | ||
7 | to process an '@itemx' that is not preceded by an '@item'. Ensure that | ||
8 | node extended names in menus and sectioning are consistent, and that | ||
9 | ordering and presence of nodes in menus and in the actual text are | ||
10 | consistent as well. | ||
11 | |||
12 | Upstream-Status: Backport [From: coreutils.7620.n7.nabble.com, bug#11828] | ||
13 | |||
14 | Signed-off-by: Robert Yang <liezhi.yang@windriver.com> | ||
15 | --- | ||
16 | doc/coreutils.texi | 82 +++++++++++++++++++++++++++--------------------------- | ||
17 | 1 file changed, 41 insertions(+), 41 deletions(-) | ||
18 | |||
19 | diff --git a/doc/coreutils.texi b/doc/coreutils.texi | ||
20 | index 588147f..2dae3fe 100644 | ||
21 | --- a/doc/coreutils.texi | ||
22 | +++ b/doc/coreutils.texi | ||
23 | @@ -555,7 +555,7 @@ symbolic link to a directory. @xref{Target directory}. | ||
24 | @end macro | ||
25 | |||
26 | @macro optSi | ||
27 | -@itemx --si | ||
28 | +@item --si | ||
29 | @opindex --si | ||
30 | @cindex SI output | ||
31 | Append an SI-style abbreviation to each size, such as @samp{M} for | ||
32 | @@ -578,7 +578,7 @@ Use the @option{--si} option if you prefer powers of 1000. | ||
33 | @end macro | ||
34 | |||
35 | @macro optStripTrailingSlashes | ||
36 | -@itemx @w{@kbd{--strip-trailing-slashes}} | ||
37 | +@item @w{@kbd{--strip-trailing-slashes}} | ||
38 | @opindex --strip-trailing-slashes | ||
39 | @cindex stripping trailing slashes | ||
40 | Remove any trailing slashes from each @var{source} argument. | ||
41 | @@ -2496,7 +2496,7 @@ by 1048576. | ||
42 | However, if @var{n} starts with a @samp{-}, | ||
43 | print all but the last @var{n} bytes of each file. | ||
44 | |||
45 | -@itemx -n @var{n} | ||
46 | +@item -n @var{n} | ||
47 | @itemx --lines=@var{n} | ||
48 | @opindex -n | ||
49 | @opindex --lines | ||
50 | @@ -2633,7 +2633,7 @@ This option is the same as @option{--follow=name --retry}. That is, tail | ||
51 | will attempt to reopen a file when it is removed. Should this fail, tail | ||
52 | will keep trying until it becomes accessible again. | ||
53 | |||
54 | -@itemx --retry | ||
55 | +@item --retry | ||
56 | @opindex --retry | ||
57 | This option is useful mainly when following by name (i.e., with | ||
58 | @option{--follow=name}). | ||
59 | @@ -2641,7 +2641,7 @@ Without this option, when tail encounters a file that doesn't | ||
60 | exist or is otherwise inaccessible, it reports that fact and | ||
61 | never checks it again. | ||
62 | |||
63 | -@itemx --sleep-interval=@var{number} | ||
64 | +@item --sleep-interval=@var{number} | ||
65 | @opindex --sleep-interval | ||
66 | Change the number of seconds to wait between iterations (the default is 1.0). | ||
67 | During one iteration, every specified file is checked to see if it has | ||
68 | @@ -2651,7 +2651,7 @@ Historical implementations of @command{tail} have required that | ||
69 | an arbitrary floating point number (using a period before any | ||
70 | fractional digits). | ||
71 | |||
72 | -@itemx --pid=@var{pid} | ||
73 | +@item --pid=@var{pid} | ||
74 | @opindex --pid | ||
75 | When following by name or by descriptor, you may specify the process ID, | ||
76 | @var{pid}, of the sole writer of all @var{file} arguments. Then, shortly | ||
77 | @@ -2674,7 +2674,7 @@ terminate until long after the real writer has terminated. | ||
78 | Note that @option{--pid} cannot be supported on some systems; @command{tail} | ||
79 | will print a warning if this is the case. | ||
80 | |||
81 | -@itemx --max-unchanged-stats=@var{n} | ||
82 | +@item --max-unchanged-stats=@var{n} | ||
83 | @opindex --max-unchanged-stats | ||
84 | When tailing a file by name, if there have been @var{n} (default | ||
85 | n=@value{DEFAULT_MAX_N_UNCHANGED_STATS_BETWEEN_OPENS}) consecutive | ||
86 | @@ -2686,7 +2686,7 @@ number of seconds between when tail prints the last pre-rotation lines | ||
87 | and when it prints the lines that have accumulated in the new log file. | ||
88 | This option is meaningful only when following by name. | ||
89 | |||
90 | -@itemx -n @var{n} | ||
91 | +@item -n @var{n} | ||
92 | @itemx --lines=@var{n} | ||
93 | @opindex -n | ||
94 | @opindex --lines | ||
95 | @@ -2817,7 +2817,7 @@ option. | ||
96 | @opindex --numeric-suffixes | ||
97 | Use digits in suffixes rather than lower-case letters. | ||
98 | |||
99 | -@itemx --verbose | ||
100 | +@item --verbose | ||
101 | @opindex --verbose | ||
102 | Write a diagnostic to standard error just before each output file is opened. | ||
103 | |||
104 | @@ -3055,7 +3055,7 @@ Print only the newline counts. | ||
105 | @opindex --max-line-length | ||
106 | Print only the maximum line lengths. | ||
107 | |||
108 | -@itemx --files0-from=@var{FILE} | ||
109 | +@item --files0-from=@var{FILE} | ||
110 | @opindex --files0-from=@var{FILE} | ||
111 | @cindex including files from @command{du} | ||
112 | Rather than processing files named on the command line, process those | ||
113 | @@ -3250,7 +3250,7 @@ an MD5 checksum inconsistent with the associated file, or if no valid | ||
114 | line is found, @command{md5sum} exits with nonzero status. Otherwise, | ||
115 | it exits successfully. | ||
116 | |||
117 | -@itemx --status | ||
118 | +@item --status | ||
119 | @opindex --status | ||
120 | @cindex verifying MD5 checksums | ||
121 | This option is useful only when verifying checksums. | ||
122 | @@ -5837,7 +5837,7 @@ command line unless the @option{--dereference-command-line} (@option{-H}), | ||
123 | If a command line argument specifies a symbolic link, show information | ||
124 | for the file the link references rather than for the link itself. | ||
125 | |||
126 | -@itemx --dereference-command-line-symlink-to-dir | ||
127 | +@item --dereference-command-line-symlink-to-dir | ||
128 | @opindex --dereference-command-line-symlink-to-dir | ||
129 | @cindex symbolic links, dereferencing | ||
130 | Do not dereference symbolic links, with one exception: | ||
131 | @@ -7015,15 +7015,15 @@ If specified, the @var{attribute_list} must be a comma-separated list | ||
132 | of one or more of the following strings: | ||
133 | |||
134 | @table @samp | ||
135 | -@itemx mode | ||
136 | +@item mode | ||
137 | Preserve the file mode bits and access control lists. | ||
138 | -@itemx ownership | ||
139 | +@item ownership | ||
140 | Preserve the owner and group. On most modern systems, | ||
141 | only users with appropriate privileges may change the owner of a file, | ||
142 | and ordinary users | ||
143 | may preserve the group ownership of a file only if they happen to be | ||
144 | a member of the desired group. | ||
145 | -@itemx timestamps | ||
146 | +@item timestamps | ||
147 | Preserve the times of last access and last modification, when possible. | ||
148 | In general, it is not possible to preserve these attributes | ||
149 | when the affected file is a symbolic link. | ||
150 | @@ -7031,12 +7031,12 @@ However, FreeBSD now provides the @code{lutimes} function, which makes | ||
151 | it possibile even for symbolic links. However, this implementation does | ||
152 | not yet take advantage of that. | ||
153 | @c FIXME: once we provide lutimes support, update the above. | ||
154 | -@itemx links | ||
155 | +@item links | ||
156 | Preserve in the destination files | ||
157 | any links between corresponding source files. | ||
158 | @c Give examples illustrating how hard links are preserved. | ||
159 | @c Also, show how soft links map to hard links with -L and -H. | ||
160 | -@itemx all | ||
161 | +@item all | ||
162 | Preserve all file attributes. | ||
163 | Equivalent to specifying all of the above. | ||
164 | @end table | ||
165 | @@ -7049,12 +7049,12 @@ mode bits of the corresponding source file, minus the bits set in the | ||
166 | umask and minus the set-user-ID and set-group-ID bits. | ||
167 | @xref{File permissions}. | ||
168 | |||
169 | -@itemx @w{@kbd{--no-preserve}=@var{attribute_list}} | ||
170 | +@item @w{@kbd{--no-preserve}=@var{attribute_list}} | ||
171 | @cindex file information, preserving | ||
172 | Do not preserve the specified attributes. The @var{attribute_list} | ||
173 | has the same form as for @option{--preserve}. | ||
174 | |||
175 | -@itemx --parents | ||
176 | +@item --parents | ||
177 | @opindex --parents | ||
178 | @cindex parent directories and @command{cp} | ||
179 | Form the name of each destination file by appending to the target | ||
180 | @@ -7070,7 +7070,7 @@ cp --parents a/b/c existing_dir | ||
181 | copies the file @file{a/b/c} to @file{existing_dir/a/b/c}, creating | ||
182 | any missing intermediate directories. | ||
183 | |||
184 | -@itemx @w{@kbd{--reply}=@var{how}} | ||
185 | +@item @w{@kbd{--reply}=@var{how}} | ||
186 | @opindex --reply | ||
187 | @cindex interactivity | ||
188 | @c FIXME: remove in 2008 | ||
189 | @@ -7742,7 +7742,7 @@ Prompt whether to overwrite each existing destination file, regardless | ||
190 | of its permissions. | ||
191 | If the response is not affirmative, the file is skipped. | ||
192 | |||
193 | -@itemx @w{@kbd{--reply}=@var{how}} | ||
194 | +@item @w{@kbd{--reply}=@var{how}} | ||
195 | @opindex --reply | ||
196 | @cindex interactivity | ||
197 | @c FIXME: remove in 2008 | ||
198 | @@ -7847,7 +7847,7 @@ files are named or if a recursive removal is requested. Ignore any | ||
199 | previous @option{--force} (@option{-f}) option. Equivalent to | ||
200 | @option{--interactive=once}. | ||
201 | |||
202 | -@itemx --interactive [=@var{when}] | ||
203 | +@item --interactive [=@var{when}] | ||
204 | @opindex --interactive | ||
205 | Specify when to issue an interactive prompt. @var{when} may be | ||
206 | omitted, or one of: | ||
207 | @@ -7866,7 +7866,7 @@ removal is requested. Equivalent to @option{-I}. | ||
208 | Specifying @option{--interactive} and no @var{when} is equivalent to | ||
209 | @option{--interactive=always}. | ||
210 | |||
211 | -@itemx --one-file-system | ||
212 | +@item --one-file-system | ||
213 | @opindex --one-file-system | ||
214 | @cindex one file system, restricting @command{rm} to | ||
215 | When removing a hierarchy recursively, skip any directory that is on a | ||
216 | @@ -7884,7 +7884,7 @@ warn about and skip directories on other file systems. | ||
217 | Of course, this will not save your @file{/home} if it and your | ||
218 | chroot happen to be on the same file system. | ||
219 | |||
220 | -@itemx --preserve-root | ||
221 | +@item --preserve-root | ||
222 | @opindex --preserve-root | ||
223 | @cindex root directory, disallow recursive destruction | ||
224 | Fail upon any attempt to remove the root directory, @file{/}, | ||
225 | @@ -7892,7 +7892,7 @@ when used with the @option{--recursive} option. | ||
226 | This is the default behavior. | ||
227 | @xref{Treating / specially}. | ||
228 | |||
229 | -@itemx --no-preserve-root | ||
230 | +@item --no-preserve-root | ||
231 | @opindex --no-preserve-root | ||
232 | @cindex root directory, allow recursive destruction | ||
233 | Do not treat @file{/} specially when removing recursively. | ||
234 | @@ -8874,7 +8874,7 @@ actually changes. | ||
235 | Do not print error messages about files whose ownership cannot be | ||
236 | changed. | ||
237 | |||
238 | -@itemx @w{@kbd{--from}=@var{old-owner}} | ||
239 | +@item @w{@kbd{--from}=@var{old-owner}} | ||
240 | @opindex --from | ||
241 | @cindex symbolic links, changing owner | ||
242 | Change a @var{file}'s ownership only if it has current attributes specified | ||
243 | @@ -8928,14 +8928,14 @@ is a symbolic link. | ||
244 | By default, no diagnostic is issued for symbolic links encountered | ||
245 | during a recursive traversal, but see @option{--verbose}. | ||
246 | |||
247 | -@itemx --preserve-root | ||
248 | +@item --preserve-root | ||
249 | @opindex --preserve-root | ||
250 | @cindex root directory, disallow recursive modification | ||
251 | Fail upon any attempt to recursively change the root directory, @file{/}. | ||
252 | Without @option{--recursive}, this option has no effect. | ||
253 | @xref{Treating / specially}. | ||
254 | |||
255 | -@itemx --no-preserve-root | ||
256 | +@item --no-preserve-root | ||
257 | @opindex --no-preserve-root | ||
258 | @cindex root directory, allow recursive modification | ||
259 | Cancel the effect of any preceding @option{--preserve-root} option. | ||
260 | @@ -9054,14 +9054,14 @@ is a symbolic link. | ||
261 | By default, no diagnostic is issued for symbolic links encountered | ||
262 | during a recursive traversal, but see @option{--verbose}. | ||
263 | |||
264 | -@itemx --preserve-root | ||
265 | +@item --preserve-root | ||
266 | @opindex --preserve-root | ||
267 | @cindex root directory, disallow recursive modification | ||
268 | Fail upon any attempt to recursively change the root directory, @file{/}. | ||
269 | Without @option{--recursive}, this option has no effect. | ||
270 | @xref{Treating / specially}. | ||
271 | |||
272 | -@itemx --no-preserve-root | ||
273 | +@item --no-preserve-root | ||
274 | @opindex --no-preserve-root | ||
275 | @cindex root directory, allow recursive modification | ||
276 | Cancel the effect of any preceding @option{--preserve-root} option. | ||
277 | @@ -9175,14 +9175,14 @@ actually changes. | ||
278 | Do not print error messages about files whose permissions cannot be | ||
279 | changed. | ||
280 | |||
281 | -@itemx --preserve-root | ||
282 | +@item --preserve-root | ||
283 | @opindex --preserve-root | ||
284 | @cindex root directory, disallow recursive modification | ||
285 | Fail upon any attempt to recursively change the root directory, @file{/}. | ||
286 | Without @option{--recursive}, this option has no effect. | ||
287 | @xref{Treating / specially}. | ||
288 | |||
289 | -@itemx --no-preserve-root | ||
290 | +@item --no-preserve-root | ||
291 | @opindex --no-preserve-root | ||
292 | @cindex root directory, allow recursive modification | ||
293 | Cancel the effect of any preceding @option{--preserve-root} option. | ||
294 | @@ -9603,7 +9603,7 @@ The program accepts the following options. Also see @ref{Common options}. | ||
295 | @opindex --all | ||
296 | Show counts for all files, not just directories. | ||
297 | |||
298 | -@itemx --apparent-size | ||
299 | +@item --apparent-size | ||
300 | @opindex --apparent-size | ||
301 | Print apparent sizes, rather than disk usage. The apparent size of a | ||
302 | file is the number of bytes reported by @code{wc -c} on regular files, | ||
303 | @@ -9654,7 +9654,7 @@ Does not affect other symbolic links. This is helpful for finding | ||
304 | out the disk usage of directories, such as @file{/usr/tmp}, which | ||
305 | are often symbolic links. | ||
306 | |||
307 | -@itemx --files0-from=@var{FILE} | ||
308 | +@item --files0-from=@var{FILE} | ||
309 | @opindex --files0-from=@var{FILE} | ||
310 | @cindex including files from @command{du} | ||
311 | Rather than processing files named on the command line, process those | ||
312 | @@ -9733,7 +9733,7 @@ Output a null byte at the end of each line, rather than a newline. | ||
313 | This option enables other programs to parse the output of @command{du} | ||
314 | even when that output would contain file names with embedded newlines. | ||
315 | |||
316 | -@itemx --si | ||
317 | +@item --si | ||
318 | @opindex --si | ||
319 | @cindex SI output | ||
320 | Append an SI-style abbreviation to each size, such as @samp{MB} for | ||
321 | @@ -9754,13 +9754,13 @@ Display only a total for each argument. | ||
322 | Report the size of each directory separately, not including the sizes | ||
323 | of subdirectories. | ||
324 | |||
325 | -@itemx --time | ||
326 | +@item --time | ||
327 | @opindex --time | ||
328 | @cindex last modified dates, displaying in @command{du} | ||
329 | Show time of the most recent modification of any file in the directory, | ||
330 | or any of its subdirectories. | ||
331 | |||
332 | -@itemx --time=ctime | ||
333 | +@item --time=ctime | ||
334 | @itemx --time=status | ||
335 | @itemx --time=use | ||
336 | @opindex --time | ||
337 | @@ -9770,7 +9770,7 @@ or any of its subdirectories. | ||
338 | Show the most recent status change time (the @samp{ctime} in the inode) of | ||
339 | any file in the directory, instead of the modification time. | ||
340 | |||
341 | -@itemx --time=atime | ||
342 | +@item --time=atime | ||
343 | @itemx --time=access | ||
344 | @opindex --time | ||
345 | @opindex atime@r{, show the most recent} | ||
346 | @@ -9911,7 +9911,7 @@ $ stat --format=%d:%i / /usr | ||
347 | 2057:2 | ||
348 | @end example | ||
349 | |||
350 | -@itemx --printf=@var{format} | ||
351 | +@item --printf=@var{format} | ||
352 | @opindex --printf=@var{format} | ||
353 | @cindex output format | ||
354 | Use @var{format} rather than the default format. | ||
355 | @@ -12240,7 +12240,7 @@ Overrides all other options. | ||
356 | @opindex -s | ||
357 | Ignored; for compatibility with other versions of @command{who}. | ||
358 | |||
359 | -@itemx -u | ||
360 | +@item -u | ||
361 | @opindex -u | ||
362 | @cindex idle time | ||
363 | After the login time, print the number of hours and minutes that the | ||
364 | @@ -12254,7 +12254,7 @@ user has been idle. @samp{.} means the user was active in the last minute. | ||
365 | List only the entries that correspond to processes via which the | ||
366 | system is waiting for a user to login. The user name is always @samp{LOGIN}. | ||
367 | |||
368 | -@itemx --lookup | ||
369 | +@item --lookup | ||
370 | @opindex --lookup | ||
371 | Attempt to canonicalize hostnames found in utmp through a DNS lookup. This | ||
372 | is not the default because it can cause significant delays on systems with | ||
373 | -- | ||
374 | 1.8.3.1 | ||
375 | |||
diff --git a/recipes-core/coreutils/coreutils-6.9/coreutils-i18n.patch b/recipes-core/coreutils/coreutils-6.9/coreutils-i18n.patch new file mode 100644 index 0000000..6537223 --- /dev/null +++ b/recipes-core/coreutils/coreutils-6.9/coreutils-i18n.patch | |||
@@ -0,0 +1,4051 @@ | |||
1 | Upstream-Status: Inappropriate [legacy version] | ||
2 | |||
3 | This patch was imported from the Fedora Core 8 coreutils-6.9-9 package. | ||
4 | |||
5 | The package is stated as being Licensed as GPLv2+. | ||
6 | |||
7 | The comment indicates that the purpose is lin18nux/lsb compliance. | ||
8 | |||
9 | Signed-off-by: Mark Hatle <mark.hatle@windriver.com> | ||
10 | |||
11 | --- /dev/null 2007-03-01 09:16:39.219409909 +0000 | ||
12 | +++ coreutils-6.8+/tests/sort/sort-mb-tests 2007-03-01 15:08:24.000000000 +0000 | ||
13 | @@ -0,0 +1,58 @@ | ||
14 | +#! /bin/sh | ||
15 | +case $# in | ||
16 | + 0) xx='../../src/sort';; | ||
17 | + *) xx="$1";; | ||
18 | +esac | ||
19 | +test "$VERBOSE" && echo=echo || echo=: | ||
20 | +$echo testing program: $xx | ||
21 | +errors=0 | ||
22 | +test "$srcdir" || srcdir=. | ||
23 | +test "$VERBOSE" && $xx --version 2> /dev/null | ||
24 | + | ||
25 | +export LC_ALL=en_US.UTF-8 | ||
26 | +locale -k LC_CTYPE 2>&1 | grep -q charmap.*UTF-8 || exit 77 | ||
27 | +errors=0 | ||
28 | + | ||
29 | +$xx -t ï¼ -k2 -n mb1.I > mb1.O | ||
30 | +code=$? | ||
31 | +if test $code != 0; then | ||
32 | + $echo "Test mb1 failed: $xx return code $code differs from expected value 0" 1>&2 | ||
33 | + errors=`expr $errors + 1` | ||
34 | +else | ||
35 | + cmp mb1.O $srcdir/mb1.X > /dev/null 2>&1 | ||
36 | + case $? in | ||
37 | + 0) if test "$VERBOSE"; then $echo "passed mb1"; fi;; | ||
38 | + 1) $echo "Test mb1 failed: files mb1.O and $srcdir/mb1.X differ" 1>&2 | ||
39 | + (diff -c mb1.O $srcdir/mb1.X) 2> /dev/null | ||
40 | + errors=`expr $errors + 1`;; | ||
41 | + 2) $echo "Test mb1 may have failed." 1>&2 | ||
42 | + $echo The command "cmp mb1.O $srcdir/mb1.X" failed. 1>&2 | ||
43 | + errors=`expr $errors + 1`;; | ||
44 | + esac | ||
45 | +fi | ||
46 | + | ||
47 | +$xx -t ï¼ -k4 -n mb2.I > mb2.O | ||
48 | +code=$? | ||
49 | +if test $code != 0; then | ||
50 | + $echo "Test mb2 failed: $xx return code $code differs from expected value 0" 1>&2 | ||
51 | + errors=`expr $errors + 1` | ||
52 | +else | ||
53 | + cmp mb2.O $srcdir/mb2.X > /dev/null 2>&1 | ||
54 | + case $? in | ||
55 | + 0) if test "$VERBOSE"; then $echo "passed mb2"; fi;; | ||
56 | + 1) $echo "Test mb2 failed: files mb2.O and $srcdir/mb2.X differ" 1>&2 | ||
57 | + (diff -c mb2.O $srcdir/mb2.X) 2> /dev/null | ||
58 | + errors=`expr $errors + 1`;; | ||
59 | + 2) $echo "Test mb2 may have failed." 1>&2 | ||
60 | + $echo The command "cmp mb2.O $srcdir/mb2.X" failed. 1>&2 | ||
61 | + errors=`expr $errors + 1`;; | ||
62 | + esac | ||
63 | +fi | ||
64 | + | ||
65 | +if test $errors = 0; then | ||
66 | + $echo Passed all 113 tests. 1>&2 | ||
67 | +else | ||
68 | + $echo Failed $errors tests. 1>&2 | ||
69 | +fi | ||
70 | +test $errors = 0 || errors=1 | ||
71 | +exit $errors | ||
72 | --- /dev/null 2007-03-01 09:16:39.219409909 +0000 | ||
73 | +++ coreutils-6.8+/tests/sort/mb2.I 2007-03-01 15:08:24.000000000 +0000 | ||
74 | @@ -0,0 +1,4 @@ | ||
75 | +Appleï¼ ï¼¡ï¼¡10ï¼ ï¼ 20 | ||
76 | +Bananaï¼ ï¼¡ï¼¡5ï¼ ï¼ 30 | ||
77 | +Citrusï¼ ï¼¡ï¼¡20ï¼ ï¼ 5 | ||
78 | +Cherryï¼ ï¼¡ï¼¡30ï¼ ï¼ 10 | ||
79 | --- /dev/null 2007-03-01 09:16:39.219409909 +0000 | ||
80 | +++ coreutils-6.8+/tests/sort/mb2.X 2007-03-01 15:08:24.000000000 +0000 | ||
81 | @@ -0,0 +1,4 @@ | ||
82 | +Citrusï¼ ï¼¡ï¼¡20ï¼ ï¼ 5 | ||
83 | +Cherryï¼ ï¼¡ï¼¡30ï¼ ï¼ 10 | ||
84 | +Appleï¼ ï¼¡ï¼¡10ï¼ ï¼ 20 | ||
85 | +Bananaï¼ ï¼¡ï¼¡5ï¼ ï¼ 30 | ||
86 | --- /dev/null 2007-03-01 09:16:39.219409909 +0000 | ||
87 | +++ coreutils-6.8+/tests/sort/mb1.I 2007-03-01 15:08:24.000000000 +0000 | ||
88 | @@ -0,0 +1,4 @@ | ||
89 | +Appleï¼ 10 | ||
90 | +Bananaï¼ 5 | ||
91 | +Citrusï¼ 20 | ||
92 | +Cherryï¼ 30 | ||
93 | --- /dev/null 2007-03-01 09:16:39.219409909 +0000 | ||
94 | +++ coreutils-6.8+/tests/sort/mb1.X 2007-03-01 15:08:24.000000000 +0000 | ||
95 | @@ -0,0 +1,4 @@ | ||
96 | +Bananaï¼ 5 | ||
97 | +Appleï¼ 10 | ||
98 | +Citrusï¼ 20 | ||
99 | +Cherryï¼ 30 | ||
100 | --- coreutils-6.8+/tests/sort/Makefile.am.i18n 2007-01-24 07:47:37.000000000 +0000 | ||
101 | +++ coreutils-6.8+/tests/sort/Makefile.am 2007-03-01 15:09:59.000000000 +0000 | ||
102 | @@ -66,15 +66,17 @@ | ||
103 | bigfield.O bigfield.E | ||
104 | ##test-files-end | ||
105 | |||
106 | -EXTRA_DIST = Test.pm $x-tests $(explicit) $(maint_gen) | ||
107 | -noinst_SCRIPTS = $x-tests | ||
108 | +run_gen += mb1.0 mb2.0 | ||
109 | + | ||
110 | +EXTRA_DIST = Test.pm $x-tests $(explicit) $(maint_gen) mb1.I mb1.X mb2.I mb2.X | ||
111 | +noinst_SCRIPTS = $x-tests # $x-mb-tests | ||
112 | TESTS_ENVIRONMENT = \ | ||
113 | CU_TEST_NAME=`basename $(abs_srcdir)`,$$tst \ | ||
114 | PATH="$(VG_PATH_PREFIX)`pwd`/../../src$(PATH_SEPARATOR)$$PATH" | ||
115 | |||
116 | editpl = sed -e 's,@''PERL''@,$(PERL),g' -e 's,@''srcdir''@,$(srcdir),g' | ||
117 | |||
118 | -TESTS = $x-tests | ||
119 | +TESTS = $x-tests $x-mb-tests | ||
120 | |||
121 | mk_script = $(srcdir)/../mk-script | ||
122 | $(srcdir)/$x-tests: $(mk_script) Test.pm Makefile.am | ||
123 | --- coreutils-6.8+/lib/linebuffer.h.i18n 2005-05-14 07:44:24.000000000 +0100 | ||
124 | +++ coreutils-6.8+/lib/linebuffer.h 2007-03-01 15:08:24.000000000 +0000 | ||
125 | @@ -22,6 +22,11 @@ | ||
126 | |||
127 | # include <stdio.h> | ||
128 | |||
129 | +/* Get mbstate_t. */ | ||
130 | +# if HAVE_WCHAR_H | ||
131 | +# include <wchar.h> | ||
132 | +# endif | ||
133 | + | ||
134 | /* A `struct linebuffer' holds a line of text. */ | ||
135 | |||
136 | struct linebuffer | ||
137 | @@ -29,6 +34,9 @@ | ||
138 | size_t size; /* Allocated. */ | ||
139 | size_t length; /* Used. */ | ||
140 | char *buffer; | ||
141 | +# if HAVE_WCHAR_H | ||
142 | + mbstate_t state; | ||
143 | +# endif | ||
144 | }; | ||
145 | |||
146 | /* Initialize linebuffer LINEBUFFER for use. */ | ||
147 | --- coreutils-6.8+/src/expand.c.i18n 2007-01-14 15:41:28.000000000 +0000 | ||
148 | +++ coreutils-6.8+/src/expand.c 2007-03-01 15:08:24.000000000 +0000 | ||
149 | @@ -38,11 +38,28 @@ | ||
150 | #include <stdio.h> | ||
151 | #include <getopt.h> | ||
152 | #include <sys/types.h> | ||
153 | + | ||
154 | +/* Get mbstate_t, mbrtowc(), wcwidth(). */ | ||
155 | +#if HAVE_WCHAR_H | ||
156 | +# include <wchar.h> | ||
157 | +#endif | ||
158 | + | ||
159 | #include "system.h" | ||
160 | #include "error.h" | ||
161 | #include "quote.h" | ||
162 | #include "xstrndup.h" | ||
163 | |||
164 | +/* MB_LEN_MAX is incorrectly defined to be 1 in at least one GCC | ||
165 | + installation; work around this configuration error. */ | ||
166 | +#if !defined MB_LEN_MAX || MB_LEN_MAX < 2 | ||
167 | +# define MB_LEN_MAX 16 | ||
168 | +#endif | ||
169 | + | ||
170 | +/* Some systems, like BeOS, have multibyte encodings but lack mbstate_t. */ | ||
171 | +#if HAVE_MBRTOWC && defined mbstate_t | ||
172 | +# define mbrtowc(pwc, s, n, ps) (mbrtowc) (pwc, s, n, 0) | ||
173 | +#endif | ||
174 | + | ||
175 | /* The official name of this program (e.g., no `g' prefix). */ | ||
176 | #define PROGRAM_NAME "expand" | ||
177 | |||
178 | @@ -183,6 +200,7 @@ | ||
179 | stops = num_start + len - 1; | ||
180 | } | ||
181 | } | ||
182 | + | ||
183 | else | ||
184 | { | ||
185 | error (0, 0, _("tab size contains invalid character(s): %s"), | ||
186 | @@ -365,6 +383,142 @@ | ||
187 | } | ||
188 | } | ||
189 | |||
190 | +#if HAVE_MBRTOWC | ||
191 | +static void | ||
192 | +expand_multibyte (void) | ||
193 | +{ | ||
194 | + FILE *fp; /* Input strem. */ | ||
195 | + mbstate_t i_state; /* Current shift state of the input stream. */ | ||
196 | + mbstate_t i_state_bak; /* Back up the I_STATE. */ | ||
197 | + mbstate_t o_state; /* Current shift state of the output stream. */ | ||
198 | + char buf[MB_LEN_MAX + BUFSIZ]; /* For spooling a read byte sequence. */ | ||
199 | + char *bufpos; /* Next read position of BUF. */ | ||
200 | + size_t buflen = 0; /* The length of the byte sequence in buf. */ | ||
201 | + wchar_t wc; /* A gotten wide character. */ | ||
202 | + size_t mblength; /* The byte size of a multibyte character | ||
203 | + which shows as same character as WC. */ | ||
204 | + int tab_index = 0; /* Index in `tab_list' of next tabstop. */ | ||
205 | + int column = 0; /* Column on screen of the next char. */ | ||
206 | + int next_tab_column; /* Column the next tab stop is on. */ | ||
207 | + int convert = 1; /* If nonzero, perform translations. */ | ||
208 | + | ||
209 | + fp = next_file ((FILE *) NULL); | ||
210 | + if (fp == NULL) | ||
211 | + return; | ||
212 | + | ||
213 | + memset (&o_state, '\0', sizeof(mbstate_t)); | ||
214 | + memset (&i_state, '\0', sizeof(mbstate_t)); | ||
215 | + | ||
216 | + for (;;) | ||
217 | + { | ||
218 | + /* Refill the buffer BUF. */ | ||
219 | + if (buflen < MB_LEN_MAX && !feof(fp) && !ferror(fp)) | ||
220 | + { | ||
221 | + memmove (buf, bufpos, buflen); | ||
222 | + buflen += fread (buf + buflen, sizeof(char), BUFSIZ, fp); | ||
223 | + bufpos = buf; | ||
224 | + } | ||
225 | + | ||
226 | + /* No character is left in BUF. */ | ||
227 | + if (buflen < 1) | ||
228 | + { | ||
229 | + fp = next_file (fp); | ||
230 | + | ||
231 | + if (fp == NULL) | ||
232 | + break; /* No more files. */ | ||
233 | + else | ||
234 | + { | ||
235 | + memset (&i_state, '\0', sizeof(mbstate_t)); | ||
236 | + continue; | ||
237 | + } | ||
238 | + } | ||
239 | + | ||
240 | + /* Get a wide character. */ | ||
241 | + i_state_bak = i_state; | ||
242 | + mblength = mbrtowc (&wc, bufpos, buflen, &i_state); | ||
243 | + | ||
244 | + switch (mblength) | ||
245 | + { | ||
246 | + case (size_t)-1: /* illegal byte sequence. */ | ||
247 | + case (size_t)-2: | ||
248 | + mblength = 1; | ||
249 | + i_state = i_state_bak; | ||
250 | + if (convert) | ||
251 | + { | ||
252 | + ++column; | ||
253 | + if (convert_entire_line == 0) | ||
254 | + convert = 0; | ||
255 | + } | ||
256 | + putchar (*bufpos); | ||
257 | + break; | ||
258 | + | ||
259 | + case 0: /* null. */ | ||
260 | + mblength = 1; | ||
261 | + if (convert && convert_entire_line == 0) | ||
262 | + convert = 0; | ||
263 | + putchar ('\0'); | ||
264 | + break; | ||
265 | + | ||
266 | + default: | ||
267 | + if (wc == L'\n') /* LF. */ | ||
268 | + { | ||
269 | + tab_index = 0; | ||
270 | + column = 0; | ||
271 | + convert = 1; | ||
272 | + putchar ('\n'); | ||
273 | + } | ||
274 | + else if (wc == L'\t' && convert) /* Tab. */ | ||
275 | + { | ||
276 | + if (tab_size == 0) | ||
277 | + { | ||
278 | + /* Do not let tab_index == first_free_tab; | ||
279 | + stop when it is 1 less. */ | ||
280 | + while (tab_index < first_free_tab - 1 | ||
281 | + && column >= tab_list[tab_index]) | ||
282 | + tab_index++; | ||
283 | + next_tab_column = tab_list[tab_index]; | ||
284 | + if (tab_index < first_free_tab - 1) | ||
285 | + tab_index++; | ||
286 | + if (column >= next_tab_column) | ||
287 | + next_tab_column = column + 1; | ||
288 | + } | ||
289 | + else | ||
290 | + next_tab_column = column + tab_size - column % tab_size; | ||
291 | + | ||
292 | + while (column < next_tab_column) | ||
293 | + { | ||
294 | + putchar (' '); | ||
295 | + ++column; | ||
296 | + } | ||
297 | + } | ||
298 | + else /* Others. */ | ||
299 | + { | ||
300 | + if (convert) | ||
301 | + { | ||
302 | + if (wc == L'\b') | ||
303 | + { | ||
304 | + if (column > 0) | ||
305 | + --column; | ||
306 | + } | ||
307 | + else | ||
308 | + { | ||
309 | + int width; /* The width of WC. */ | ||
310 | + | ||
311 | + width = wcwidth (wc); | ||
312 | + column += (width > 0) ? width : 0; | ||
313 | + if (convert_entire_line == 0) | ||
314 | + convert = 0; | ||
315 | + } | ||
316 | + } | ||
317 | + fwrite (bufpos, sizeof(char), mblength, stdout); | ||
318 | + } | ||
319 | + } | ||
320 | + buflen -= mblength; | ||
321 | + bufpos += mblength; | ||
322 | + } | ||
323 | +} | ||
324 | +#endif | ||
325 | + | ||
326 | int | ||
327 | main (int argc, char **argv) | ||
328 | { | ||
329 | @@ -429,7 +583,12 @@ | ||
330 | |||
331 | file_list = (optind < argc ? &argv[optind] : stdin_argv); | ||
332 | |||
333 | - expand (); | ||
334 | +#if HAVE_MBRTOWC | ||
335 | + if (MB_CUR_MAX > 1) | ||
336 | + expand_multibyte (); | ||
337 | + else | ||
338 | +#endif | ||
339 | + expand (); | ||
340 | |||
341 | if (have_read_stdin && fclose (stdin) != 0) | ||
342 | error (EXIT_FAILURE, errno, "-"); | ||
343 | --- coreutils-6.8+/src/join.c.i18n 2007-01-14 15:41:28.000000000 +0000 | ||
344 | +++ coreutils-6.8+/src/join.c 2007-03-01 15:08:24.000000000 +0000 | ||
345 | @@ -23,16 +23,30 @@ | ||
346 | #include <sys/types.h> | ||
347 | #include <getopt.h> | ||
348 | |||
349 | +/* Get mbstate_t, mbrtowc(), mbrtowc(), wcwidth(). */ | ||
350 | +#if HAVE_WCHAR_H | ||
351 | +# include <wchar.h> | ||
352 | +#endif | ||
353 | + | ||
354 | +/* Get iswblank(), towupper. */ | ||
355 | +#if HAVE_WCTYPE_H | ||
356 | +# include <wctype.h> | ||
357 | +#endif | ||
358 | + | ||
359 | #include "system.h" | ||
360 | #include "error.h" | ||
361 | #include "hard-locale.h" | ||
362 | #include "linebuffer.h" | ||
363 | -#include "memcasecmp.h" | ||
364 | #include "quote.h" | ||
365 | #include "stdio--.h" | ||
366 | #include "xmemcoll.h" | ||
367 | #include "xstrtol.h" | ||
368 | |||
369 | +/* Some systems, like BeOS, have multibyte encodings but lack mbstate_t. */ | ||
370 | +#if HAVE_MBRTOWC && defined mbstate_t | ||
371 | +# define mbrtowc(pwc, s, n, ps) (mbrtowc) (pwc, s, n, 0) | ||
372 | +#endif | ||
373 | + | ||
374 | /* The official name of this program (e.g., no `g' prefix). */ | ||
375 | #define PROGRAM_NAME "join" | ||
376 | |||
377 | @@ -104,10 +118,12 @@ | ||
378 | /* Last element in `outlist', where a new element can be added. */ | ||
379 | static struct outlist *outlist_end = &outlist_head; | ||
380 | |||
381 | -/* Tab character separating fields. If negative, fields are separated | ||
382 | - by any nonempty string of blanks, otherwise by exactly one | ||
383 | - tab character whose value (when cast to unsigned char) equals TAB. */ | ||
384 | -static int tab = -1; | ||
385 | +/* Tab character separating fields. If NULL, fields are separated | ||
386 | + by any nonempty string of blanks. */ | ||
387 | +static char *tab = NULL; | ||
388 | + | ||
389 | +/* The number of bytes used for tab. */ | ||
390 | +static size_t tablen = 0; | ||
391 | |||
392 | static struct option const longopts[] = | ||
393 | { | ||
394 | @@ -190,6 +206,8 @@ | ||
395 | |||
396 | /* Fill in the `fields' structure in LINE. */ | ||
397 | |||
398 | +/* Fill in the `fields' structure in LINE. */ | ||
399 | + | ||
400 | static void | ||
401 | xfields (struct line *line) | ||
402 | { | ||
403 | @@ -199,10 +217,11 @@ | ||
404 | if (ptr == lim) | ||
405 | return; | ||
406 | |||
407 | - if (0 <= tab) | ||
408 | + if (tab != NULL) | ||
409 | { | ||
410 | + unsigned char t = tab[0]; | ||
411 | char *sep; | ||
412 | - for (; (sep = memchr (ptr, tab, lim - ptr)) != NULL; ptr = sep + 1) | ||
413 | + for (; (sep = memchr (ptr, t, lim - ptr)) != NULL; ptr = sep + 1) | ||
414 | extract_field (line, ptr, sep - ptr); | ||
415 | } | ||
416 | else | ||
417 | @@ -229,6 +248,148 @@ | ||
418 | extract_field (line, ptr, lim - ptr); | ||
419 | } | ||
420 | |||
421 | +#if HAVE_MBRTOWC | ||
422 | +static void | ||
423 | +xfields_multibyte (struct line *line) | ||
424 | +{ | ||
425 | + char *ptr = line->buf.buffer; | ||
426 | + char const *lim = ptr + line->buf.length - 1; | ||
427 | + wchar_t wc = 0; | ||
428 | + size_t mblength = 1; | ||
429 | + mbstate_t state, state_bak; | ||
430 | + | ||
431 | + memset (&state, 0, sizeof (mbstate_t)); | ||
432 | + | ||
433 | + if (ptr == lim) | ||
434 | + return; | ||
435 | + | ||
436 | + if (tab != NULL) | ||
437 | + { | ||
438 | + unsigned char t = tab[0]; | ||
439 | + char *sep = ptr; | ||
440 | + for (; ptr < lim; ptr = sep + mblength) | ||
441 | + { | ||
442 | + sep = ptr; | ||
443 | + while (sep < lim) | ||
444 | + { | ||
445 | + state_bak = state; | ||
446 | + mblength = mbrtowc (&wc, sep, lim - sep + 1, &state); | ||
447 | + | ||
448 | + if (mblength == (size_t)-1 || mblength == (size_t)-2) | ||
449 | + { | ||
450 | + mblength = 1; | ||
451 | + state = state_bak; | ||
452 | + } | ||
453 | + mblength = (mblength < 1) ? 1 : mblength; | ||
454 | + | ||
455 | + if (mblength == tablen && !memcmp (sep, tab, mblength)) | ||
456 | + break; | ||
457 | + else | ||
458 | + { | ||
459 | + sep += mblength; | ||
460 | + continue; | ||
461 | + } | ||
462 | + } | ||
463 | + | ||
464 | + if (sep == lim) | ||
465 | + break; | ||
466 | + | ||
467 | + extract_field (line, ptr, sep - ptr); | ||
468 | + } | ||
469 | + } | ||
470 | + else | ||
471 | + { | ||
472 | + /* Skip leading blanks before the first field. */ | ||
473 | + while(ptr < lim) | ||
474 | + { | ||
475 | + state_bak = state; | ||
476 | + mblength = mbrtowc (&wc, ptr, lim - ptr + 1, &state); | ||
477 | + | ||
478 | + if (mblength == (size_t)-1 || mblength == (size_t)-2) | ||
479 | + { | ||
480 | + mblength = 1; | ||
481 | + state = state_bak; | ||
482 | + break; | ||
483 | + } | ||
484 | + mblength = (mblength < 1) ? 1 : mblength; | ||
485 | + | ||
486 | + if (!iswblank(wc)) | ||
487 | + break; | ||
488 | + ptr += mblength; | ||
489 | + } | ||
490 | + | ||
491 | + do | ||
492 | + { | ||
493 | + char *sep; | ||
494 | + state_bak = state; | ||
495 | + mblength = mbrtowc (&wc, ptr, lim - ptr + 1, &state); | ||
496 | + if (mblength == (size_t)-1 || mblength == (size_t)-2) | ||
497 | + { | ||
498 | + mblength = 1; | ||
499 | + state = state_bak; | ||
500 | + break; | ||
501 | + } | ||
502 | + mblength = (mblength < 1) ? 1 : mblength; | ||
503 | + | ||
504 | + sep = ptr + mblength; | ||
505 | + while (sep != lim) | ||
506 | + { | ||
507 | + state_bak = state; | ||
508 | + mblength = mbrtowc (&wc, sep, lim - sep + 1, &state); | ||
509 | + if (mblength == (size_t)-1 || mblength == (size_t)-2) | ||
510 | + { | ||
511 | + mblength = 1; | ||
512 | + state = state_bak; | ||
513 | + break; | ||
514 | + } | ||
515 | + mblength = (mblength < 1) ? 1 : mblength; | ||
516 | + | ||
517 | + if (iswblank (wc)) | ||
518 | + break; | ||
519 | + | ||
520 | + sep += mblength; | ||
521 | + } | ||
522 | + | ||
523 | + extract_field (line, ptr, sep - ptr); | ||
524 | + if (sep == lim) | ||
525 | + return; | ||
526 | + | ||
527 | + state_bak = state; | ||
528 | + mblength = mbrtowc (&wc, sep, lim - sep + 1, &state); | ||
529 | + if (mblength == (size_t)-1 || mblength == (size_t)-2) | ||
530 | + { | ||
531 | + mblength = 1; | ||
532 | + state = state_bak; | ||
533 | + break; | ||
534 | + } | ||
535 | + mblength = (mblength < 1) ? 1 : mblength; | ||
536 | + | ||
537 | + ptr = sep + mblength; | ||
538 | + while (ptr != lim) | ||
539 | + { | ||
540 | + state_bak = state; | ||
541 | + mblength = mbrtowc (&wc, ptr, lim - ptr + 1, &state); | ||
542 | + if (mblength == (size_t)-1 || mblength == (size_t)-2) | ||
543 | + { | ||
544 | + mblength = 1; | ||
545 | + state = state_bak; | ||
546 | + break; | ||
547 | + } | ||
548 | + mblength = (mblength < 1) ? 1 : mblength; | ||
549 | + | ||
550 | + if (!iswblank (wc)) | ||
551 | + break; | ||
552 | + | ||
553 | + ptr += mblength; | ||
554 | + } | ||
555 | + } | ||
556 | + while (ptr != lim); | ||
557 | + } | ||
558 | + | ||
559 | + extract_field (line, ptr, lim - ptr); | ||
560 | +} | ||
561 | +#endif | ||
562 | + | ||
563 | /* Read a line from FP into LINE and split it into fields. | ||
564 | Return true if successful. */ | ||
565 | |||
566 | @@ -249,6 +410,11 @@ | ||
567 | line->nfields_allocated = 0; | ||
568 | line->nfields = 0; | ||
569 | line->fields = NULL; | ||
570 | +#if HAVE_MBRTOWC | ||
571 | + if (MB_CUR_MAX > 1) | ||
572 | + xfields_multibyte (line); | ||
573 | + else | ||
574 | +#endif | ||
575 | xfields (line); | ||
576 | return true; | ||
577 | } | ||
578 | @@ -303,56 +469,114 @@ | ||
579 | keycmp (struct line const *line1, struct line const *line2) | ||
580 | { | ||
581 | /* Start of field to compare in each file. */ | ||
582 | - char *beg1; | ||
583 | - char *beg2; | ||
584 | - | ||
585 | - size_t len1; | ||
586 | - size_t len2; /* Length of fields to compare. */ | ||
587 | + char *beg[2]; | ||
588 | + char *copy[2]; | ||
589 | + size_t len[2]; /* Length of fields to compare. */ | ||
590 | int diff; | ||
591 | + int i, j; | ||
592 | |||
593 | if (join_field_1 < line1->nfields) | ||
594 | { | ||
595 | - beg1 = line1->fields[join_field_1].beg; | ||
596 | - len1 = line1->fields[join_field_1].len; | ||
597 | + beg[0] = line1->fields[join_field_1].beg; | ||
598 | + len[0] = line1->fields[join_field_1].len; | ||
599 | } | ||
600 | else | ||
601 | { | ||
602 | - beg1 = NULL; | ||
603 | - len1 = 0; | ||
604 | + beg[0] = NULL; | ||
605 | + len[0] = 0; | ||
606 | } | ||
607 | |||
608 | if (join_field_2 < line2->nfields) | ||
609 | { | ||
610 | - beg2 = line2->fields[join_field_2].beg; | ||
611 | - len2 = line2->fields[join_field_2].len; | ||
612 | + beg[1] = line2->fields[join_field_2].beg; | ||
613 | + len[1] = line2->fields[join_field_2].len; | ||
614 | } | ||
615 | else | ||
616 | { | ||
617 | - beg2 = NULL; | ||
618 | - len2 = 0; | ||
619 | + beg[1] = NULL; | ||
620 | + len[1] = 0; | ||
621 | } | ||
622 | |||
623 | - if (len1 == 0) | ||
624 | - return len2 == 0 ? 0 : -1; | ||
625 | - if (len2 == 0) | ||
626 | + if (len[0] == 0) | ||
627 | + return len[1] == 0 ? 0 : -1; | ||
628 | + if (len[1] == 0) | ||
629 | return 1; | ||
630 | |||
631 | if (ignore_case) | ||
632 | { | ||
633 | - /* FIXME: ignore_case does not work with NLS (in particular, | ||
634 | - with multibyte chars). */ | ||
635 | - diff = memcasecmp (beg1, beg2, MIN (len1, len2)); | ||
636 | +#ifdef HAVE_MBRTOWC | ||
637 | + if (MB_CUR_MAX > 1) | ||
638 | + { | ||
639 | + size_t mblength; | ||
640 | + wchar_t wc, uwc; | ||
641 | + mbstate_t state, state_bak; | ||
642 | + | ||
643 | + memset (&state, '\0', sizeof (mbstate_t)); | ||
644 | + | ||
645 | + for (i = 0; i < 2; i++) | ||
646 | + { | ||
647 | + copy[i] = alloca (len[i] + 1); | ||
648 | + | ||
649 | + for (j = 0; j < MIN (len[0], len[1]);) | ||
650 | + { | ||
651 | + state_bak = state; | ||
652 | + mblength = mbrtowc (&wc, beg[i] + j, len[i] - j, &state); | ||
653 | + | ||
654 | + switch (mblength) | ||
655 | + { | ||
656 | + case (size_t) -1: | ||
657 | + case (size_t) -2: | ||
658 | + state = state_bak; | ||
659 | + /* Fall through */ | ||
660 | + case 0: | ||
661 | + mblength = 1; | ||
662 | + break; | ||
663 | + | ||
664 | + default: | ||
665 | + uwc = towupper (wc); | ||
666 | + | ||
667 | + if (uwc != wc) | ||
668 | + { | ||
669 | + mbstate_t state_wc; | ||
670 | + | ||
671 | + memset (&state_wc, '\0', sizeof (mbstate_t)); | ||
672 | + wcrtomb (copy[i] + j, uwc, &state_wc); | ||
673 | + } | ||
674 | + else | ||
675 | + memcpy (copy[i] + j, beg[i] + j, mblength); | ||
676 | + } | ||
677 | + j += mblength; | ||
678 | + } | ||
679 | + copy[i][j] = '\0'; | ||
680 | + } | ||
681 | + } | ||
682 | + else | ||
683 | +#endif | ||
684 | + { | ||
685 | + for (i = 0; i < 2; i++) | ||
686 | + { | ||
687 | + copy[i] = alloca (len[i] + 1); | ||
688 | + | ||
689 | + for (j = 0; j < MIN (len[0], len[1]); j++) | ||
690 | + copy[i][j] = toupper (beg[i][j]); | ||
691 | + | ||
692 | + copy[i][j] = '\0'; | ||
693 | + } | ||
694 | + } | ||
695 | } | ||
696 | else | ||
697 | { | ||
698 | - if (hard_LC_COLLATE) | ||
699 | - return xmemcoll (beg1, len1, beg2, len2); | ||
700 | - diff = memcmp (beg1, beg2, MIN (len1, len2)); | ||
701 | + copy[0] = (unsigned char *) beg[0]; | ||
702 | + copy[1] = (unsigned char *) beg[1]; | ||
703 | } | ||
704 | |||
705 | + if (hard_LC_COLLATE) | ||
706 | + return xmemcoll ((char *) copy[0], len[0], (char *) copy[1], len[1]); | ||
707 | + diff = memcmp (copy[0], copy[1], MIN (len[0], len[1])); | ||
708 | + | ||
709 | if (diff) | ||
710 | return diff; | ||
711 | - return len1 < len2 ? -1 : len1 != len2; | ||
712 | + return len[0] - len[1]; | ||
713 | } | ||
714 | |||
715 | /* Print field N of LINE if it exists and is nonempty, otherwise | ||
716 | @@ -377,11 +601,18 @@ | ||
717 | |||
718 | /* Print the join of LINE1 and LINE2. */ | ||
719 | |||
720 | +#define PUT_TAB_CHAR \ | ||
721 | + do \ | ||
722 | + { \ | ||
723 | + (tab != NULL) ? \ | ||
724 | + fwrite(tab, sizeof(char), tablen, stdout) : putchar (' '); \ | ||
725 | + } \ | ||
726 | + while (0) | ||
727 | + | ||
728 | static void | ||
729 | prjoin (struct line const *line1, struct line const *line2) | ||
730 | { | ||
731 | const struct outlist *outlist; | ||
732 | - char output_separator = tab < 0 ? ' ' : tab; | ||
733 | |||
734 | outlist = outlist_head.next; | ||
735 | if (outlist) | ||
736 | @@ -397,12 +628,12 @@ | ||
737 | if (o->file == 0) | ||
738 | { | ||
739 | if (line1 == &uni_blank) | ||
740 | - { | ||
741 | + { | ||
742 | line = line2; | ||
743 | field = join_field_2; | ||
744 | } | ||
745 | else | ||
746 | - { | ||
747 | + { | ||
748 | line = line1; | ||
749 | field = join_field_1; | ||
750 | } | ||
751 | @@ -416,7 +647,7 @@ | ||
752 | o = o->next; | ||
753 | if (o == NULL) | ||
754 | break; | ||
755 | - putchar (output_separator); | ||
756 | + PUT_TAB_CHAR; | ||
757 | } | ||
758 | putchar ('\n'); | ||
759 | } | ||
760 | @@ -434,23 +665,23 @@ | ||
761 | prfield (join_field_1, line1); | ||
762 | for (i = 0; i < join_field_1 && i < line1->nfields; ++i) | ||
763 | { | ||
764 | - putchar (output_separator); | ||
765 | + PUT_TAB_CHAR; | ||
766 | prfield (i, line1); | ||
767 | } | ||
768 | for (i = join_field_1 + 1; i < line1->nfields; ++i) | ||
769 | { | ||
770 | - putchar (output_separator); | ||
771 | + PUT_TAB_CHAR; | ||
772 | prfield (i, line1); | ||
773 | } | ||
774 | |||
775 | for (i = 0; i < join_field_2 && i < line2->nfields; ++i) | ||
776 | { | ||
777 | - putchar (output_separator); | ||
778 | + PUT_TAB_CHAR; | ||
779 | prfield (i, line2); | ||
780 | } | ||
781 | for (i = join_field_2 + 1; i < line2->nfields; ++i) | ||
782 | { | ||
783 | - putchar (output_separator); | ||
784 | + PUT_TAB_CHAR; | ||
785 | prfield (i, line2); | ||
786 | } | ||
787 | putchar ('\n'); | ||
788 | @@ -859,20 +1090,41 @@ | ||
789 | |||
790 | case 't': | ||
791 | { | ||
792 | - unsigned char newtab = optarg[0]; | ||
793 | - if (! newtab) | ||
794 | + char *newtab; | ||
795 | + size_t newtablen; | ||
796 | + if (! optarg[0]) | ||
797 | error (EXIT_FAILURE, 0, _("empty tab")); | ||
798 | - if (optarg[1]) | ||
799 | + newtab = xstrdup (optarg); | ||
800 | +#if HAVE_MBRTOWC | ||
801 | + if (MB_CUR_MAX > 1) | ||
802 | + { | ||
803 | + mbstate_t state; | ||
804 | + | ||
805 | + memset (&state, 0, sizeof (mbstate_t)); | ||
806 | + newtablen = mbrtowc (NULL, newtab, | ||
807 | + strnlen (newtab, MB_LEN_MAX), | ||
808 | + &state); | ||
809 | + if (newtablen == (size_t) 0 | ||
810 | + || newtablen == (size_t) -1 | ||
811 | + || newtablen == (size_t) -2) | ||
812 | + newtablen = 1; | ||
813 | + } | ||
814 | + else | ||
815 | +#endif | ||
816 | + newtablen = 1; | ||
817 | + | ||
818 | + if (newtablen == 1 && newtab[1]) | ||
819 | + { | ||
820 | + if (STREQ (newtab, "\\0")) | ||
821 | + newtab[0] = '\0'; | ||
822 | + } | ||
823 | + if (tab != NULL && strcmp (tab, newtab)) | ||
824 | { | ||
825 | - if (STREQ (optarg, "\\0")) | ||
826 | - newtab = '\0'; | ||
827 | - else | ||
828 | - error (EXIT_FAILURE, 0, _("multi-character tab %s"), | ||
829 | - quote (optarg)); | ||
830 | + free (newtab); | ||
831 | + error (EXIT_FAILURE, 0, _("incompatible tabs")); | ||
832 | } | ||
833 | - if (0 <= tab && tab != newtab) | ||
834 | - error (EXIT_FAILURE, 0, _("incompatible tabs")); | ||
835 | tab = newtab; | ||
836 | + tablen = newtablen; | ||
837 | } | ||
838 | break; | ||
839 | |||
840 | --- coreutils-6.8+/src/uniq.c.i18n 2007-01-14 15:41:28.000000000 +0000 | ||
841 | +++ coreutils-6.8+/src/uniq.c 2007-03-01 15:08:24.000000000 +0000 | ||
842 | @@ -23,6 +23,16 @@ | ||
843 | #include <getopt.h> | ||
844 | #include <sys/types.h> | ||
845 | |||
846 | +/* Get mbstate_t, mbrtowc(). */ | ||
847 | +#if HAVE_WCHAR_H | ||
848 | +# include <wchar.h> | ||
849 | +#endif | ||
850 | + | ||
851 | +/* Get isw* functions. */ | ||
852 | +#if HAVE_WCTYPE_H | ||
853 | +# include <wctype.h> | ||
854 | +#endif | ||
855 | + | ||
856 | #include "system.h" | ||
857 | #include "argmatch.h" | ||
858 | #include "linebuffer.h" | ||
859 | @@ -32,7 +42,19 @@ | ||
860 | #include "quote.h" | ||
861 | #include "xmemcoll.h" | ||
862 | #include "xstrtol.h" | ||
863 | -#include "memcasecmp.h" | ||
864 | +#include "xmemcoll.h" | ||
865 | + | ||
866 | +/* MB_LEN_MAX is incorrectly defined to be 1 in at least one GCC | ||
867 | + installation; work around this configuration error. */ | ||
868 | +#if !defined MB_LEN_MAX || MB_LEN_MAX < 2 | ||
869 | +# define MB_LEN_MAX 16 | ||
870 | +#endif | ||
871 | + | ||
872 | +/* Some systems, like BeOS, have multibyte encodings but lack mbstate_t. */ | ||
873 | +#if HAVE_MBRTOWC && defined mbstate_t | ||
874 | +# define mbrtowc(pwc, s, n, ps) (mbrtowc) (pwc, s, n, 0) | ||
875 | +#endif | ||
876 | + | ||
877 | |||
878 | /* The official name of this program (e.g., no `g' prefix). */ | ||
879 | #define PROGRAM_NAME "uniq" | ||
880 | @@ -109,6 +131,10 @@ | ||
881 | /* Select whether/how to delimit groups of duplicate lines. */ | ||
882 | static enum delimit_method delimit_groups; | ||
883 | |||
884 | +/* Function pointers. */ | ||
885 | +static char * | ||
886 | +(*find_field) (struct linebuffer *line); | ||
887 | + | ||
888 | static struct option const longopts[] = | ||
889 | { | ||
890 | {"count", no_argument, NULL, 'c'}, | ||
891 | @@ -198,7 +224,7 @@ | ||
892 | return a pointer to the beginning of the line's field to be compared. */ | ||
893 | |||
894 | static char * | ||
895 | -find_field (const struct linebuffer *line) | ||
896 | +find_field_uni (struct linebuffer *line) | ||
897 | { | ||
898 | size_t count; | ||
899 | char *lp = line->buffer; | ||
900 | @@ -219,6 +245,83 @@ | ||
901 | return lp + i; | ||
902 | } | ||
903 | |||
904 | +#if HAVE_MBRTOWC | ||
905 | + | ||
906 | +# define MBCHAR_TO_WCHAR(WC, MBLENGTH, LP, POS, SIZE, STATEP, CONVFAIL) \ | ||
907 | + do \ | ||
908 | + { \ | ||
909 | + mbstate_t state_bak; \ | ||
910 | + \ | ||
911 | + CONVFAIL = 0; \ | ||
912 | + state_bak = *STATEP; \ | ||
913 | + \ | ||
914 | + MBLENGTH = mbrtowc (&WC, LP + POS, SIZE - POS, STATEP); \ | ||
915 | + \ | ||
916 | + switch (MBLENGTH) \ | ||
917 | + { \ | ||
918 | + case (size_t)-2: \ | ||
919 | + case (size_t)-1: \ | ||
920 | + *STATEP = state_bak; \ | ||
921 | + CONVFAIL++; \ | ||
922 | + /* Fall through */ \ | ||
923 | + case 0: \ | ||
924 | + MBLENGTH = 1; \ | ||
925 | + } \ | ||
926 | + } \ | ||
927 | + while (0) | ||
928 | + | ||
929 | +static char * | ||
930 | +find_field_multi (struct linebuffer *line) | ||
931 | +{ | ||
932 | + size_t count; | ||
933 | + char *lp = line->buffer; | ||
934 | + size_t size = line->length - 1; | ||
935 | + size_t pos; | ||
936 | + size_t mblength; | ||
937 | + wchar_t wc; | ||
938 | + mbstate_t *statep; | ||
939 | + int convfail; | ||
940 | + | ||
941 | + pos = 0; | ||
942 | + statep = &(line->state); | ||
943 | + | ||
944 | + /* skip fields. */ | ||
945 | + for (count = 0; count < skip_fields && pos < size; count++) | ||
946 | + { | ||
947 | + while (pos < size) | ||
948 | + { | ||
949 | + MBCHAR_TO_WCHAR (wc, mblength, lp, pos, size, statep, convfail); | ||
950 | + | ||
951 | + if (convfail || !iswblank (wc)) | ||
952 | + { | ||
953 | + pos += mblength; | ||
954 | + break; | ||
955 | + } | ||
956 | + pos += mblength; | ||
957 | + } | ||
958 | + | ||
959 | + while (pos < size) | ||
960 | + { | ||
961 | + MBCHAR_TO_WCHAR (wc, mblength, lp, pos, size, statep, convfail); | ||
962 | + | ||
963 | + if (!convfail && iswblank (wc)) | ||
964 | + break; | ||
965 | + | ||
966 | + pos += mblength; | ||
967 | + } | ||
968 | + } | ||
969 | + | ||
970 | + /* skip fields. */ | ||
971 | + for (count = 0; count < skip_chars && pos < size; count++) | ||
972 | + { | ||
973 | + MBCHAR_TO_WCHAR (wc, mblength, lp, pos, size, statep, convfail); | ||
974 | + pos += mblength; | ||
975 | + } | ||
976 | + | ||
977 | + return lp + pos; | ||
978 | +} | ||
979 | +#endif | ||
980 | + | ||
981 | /* Return false if two strings OLD and NEW match, true if not. | ||
982 | OLD and NEW point not to the beginnings of the lines | ||
983 | but rather to the beginnings of the fields to compare. | ||
984 | @@ -227,6 +330,8 @@ | ||
985 | static bool | ||
986 | different (char *old, char *new, size_t oldlen, size_t newlen) | ||
987 | { | ||
988 | + char *copy_old, *copy_new; | ||
989 | + | ||
990 | if (check_chars < oldlen) | ||
991 | oldlen = check_chars; | ||
992 | if (check_chars < newlen) | ||
993 | @@ -234,14 +339,92 @@ | ||
994 | |||
995 | if (ignore_case) | ||
996 | { | ||
997 | - /* FIXME: This should invoke strcoll somehow. */ | ||
998 | - return oldlen != newlen || memcasecmp (old, new, oldlen); | ||
999 | + size_t i; | ||
1000 | + | ||
1001 | + copy_old = alloca (oldlen + 1); | ||
1002 | + copy_new = alloca (oldlen + 1); | ||
1003 | + | ||
1004 | + for (i = 0; i < oldlen; i++) | ||
1005 | + { | ||
1006 | + copy_old[i] = toupper (old[i]); | ||
1007 | + copy_new[i] = toupper (new[i]); | ||
1008 | + } | ||
1009 | } | ||
1010 | - else if (hard_LC_COLLATE) | ||
1011 | - return xmemcoll (old, oldlen, new, newlen) != 0; | ||
1012 | else | ||
1013 | - return oldlen != newlen || memcmp (old, new, oldlen); | ||
1014 | + { | ||
1015 | + copy_old = (char *)old; | ||
1016 | + copy_new = (char *)new; | ||
1017 | + } | ||
1018 | + | ||
1019 | + return xmemcoll (copy_old, oldlen, copy_new, newlen); | ||
1020 | +} | ||
1021 | + | ||
1022 | +#if HAVE_MBRTOWC | ||
1023 | +static int | ||
1024 | +different_multi (const char *old, const char *new, size_t oldlen, size_t newlen, mbstate_t oldstate, mbstate_t newstate) | ||
1025 | +{ | ||
1026 | + size_t i, j, chars; | ||
1027 | + const char *str[2]; | ||
1028 | + char *copy[2]; | ||
1029 | + size_t len[2]; | ||
1030 | + mbstate_t state[2]; | ||
1031 | + size_t mblength; | ||
1032 | + wchar_t wc, uwc; | ||
1033 | + mbstate_t state_bak; | ||
1034 | + | ||
1035 | + str[0] = old; | ||
1036 | + str[1] = new; | ||
1037 | + len[0] = oldlen; | ||
1038 | + len[1] = newlen; | ||
1039 | + state[0] = oldstate; | ||
1040 | + state[1] = newstate; | ||
1041 | + | ||
1042 | + for (i = 0; i < 2; i++) | ||
1043 | + { | ||
1044 | + copy[i] = alloca (len[i] + 1); | ||
1045 | + | ||
1046 | + for (j = 0, chars = 0; j < len[i] && chars < check_chars; chars++) | ||
1047 | + { | ||
1048 | + state_bak = state[i]; | ||
1049 | + mblength = mbrtowc (&wc, str[i] + j, len[i] - j, &(state[i])); | ||
1050 | + | ||
1051 | + switch (mblength) | ||
1052 | + { | ||
1053 | + case (size_t)-1: | ||
1054 | + case (size_t)-2: | ||
1055 | + state[i] = state_bak; | ||
1056 | + /* Fall through */ | ||
1057 | + case 0: | ||
1058 | + mblength = 1; | ||
1059 | + break; | ||
1060 | + | ||
1061 | + default: | ||
1062 | + if (ignore_case) | ||
1063 | + { | ||
1064 | + uwc = towupper (wc); | ||
1065 | + | ||
1066 | + if (uwc != wc) | ||
1067 | + { | ||
1068 | + mbstate_t state_wc; | ||
1069 | + | ||
1070 | + memset (&state_wc, '\0', sizeof(mbstate_t)); | ||
1071 | + wcrtomb (copy[i] + j, uwc, &state_wc); | ||
1072 | + } | ||
1073 | + else | ||
1074 | + memcpy (copy[i] + j, str[i] + j, mblength); | ||
1075 | + } | ||
1076 | + else | ||
1077 | + memcpy (copy[i] + j, str[i] + j, mblength); | ||
1078 | + } | ||
1079 | + j += mblength; | ||
1080 | + } | ||
1081 | + copy[i][j] = '\0'; | ||
1082 | + len[i] = j; | ||
1083 | + } | ||
1084 | + | ||
1085 | + return xmemcoll (copy[0], len[0], copy[1], len[1]); | ||
1086 | } | ||
1087 | +#endif | ||
1088 | |||
1089 | /* Output the line in linebuffer LINE to standard output | ||
1090 | provided that the switches say it should be output. | ||
1091 | @@ -295,15 +478,43 @@ | ||
1092 | { | ||
1093 | char *prevfield IF_LINT (= NULL); | ||
1094 | size_t prevlen IF_LINT (= 0); | ||
1095 | +#if HAVE_MBRTOWC | ||
1096 | + mbstate_t prevstate; | ||
1097 | + | ||
1098 | + memset (&prevstate, '\0', sizeof (mbstate_t)); | ||
1099 | +#endif | ||
1100 | |||
1101 | while (!feof (stdin)) | ||
1102 | { | ||
1103 | char *thisfield; | ||
1104 | size_t thislen; | ||
1105 | +#if HAVE_MBRTOWC | ||
1106 | + mbstate_t thisstate; | ||
1107 | +#endif | ||
1108 | + | ||
1109 | if (readlinebuffer (thisline, stdin) == 0) | ||
1110 | break; | ||
1111 | thisfield = find_field (thisline); | ||
1112 | thislen = thisline->length - 1 - (thisfield - thisline->buffer); | ||
1113 | +#if HAVE_MBRTOWC | ||
1114 | + if (MB_CUR_MAX > 1) | ||
1115 | + { | ||
1116 | + thisstate = thisline->state; | ||
1117 | + | ||
1118 | + if (prevline->length == 0 || different_multi | ||
1119 | + (thisfield, prevfield, thislen, prevlen, thisstate, prevstate)) | ||
1120 | + { | ||
1121 | + fwrite (thisline->buffer, sizeof (char), | ||
1122 | + thisline->length, stdout); | ||
1123 | + | ||
1124 | + SWAP_LINES (prevline, thisline); | ||
1125 | + prevfield = thisfield; | ||
1126 | + prevlen = thislen; | ||
1127 | + prevstate = thisstate; | ||
1128 | + } | ||
1129 | + } | ||
1130 | + else | ||
1131 | +#endif | ||
1132 | if (prevline->length == 0 | ||
1133 | || different (thisfield, prevfield, thislen, prevlen)) | ||
1134 | { | ||
1135 | @@ -322,17 +533,26 @@ | ||
1136 | size_t prevlen; | ||
1137 | uintmax_t match_count = 0; | ||
1138 | bool first_delimiter = true; | ||
1139 | +#if HAVE_MBRTOWC | ||
1140 | + mbstate_t prevstate; | ||
1141 | +#endif | ||
1142 | |||
1143 | if (readlinebuffer (prevline, stdin) == 0) | ||
1144 | goto closefiles; | ||
1145 | prevfield = find_field (prevline); | ||
1146 | prevlen = prevline->length - 1 - (prevfield - prevline->buffer); | ||
1147 | +#if HAVE_MBRTOWC | ||
1148 | + prevstate = prevline->state; | ||
1149 | +#endif | ||
1150 | |||
1151 | while (!feof (stdin)) | ||
1152 | { | ||
1153 | bool match; | ||
1154 | char *thisfield; | ||
1155 | size_t thislen; | ||
1156 | +#if HAVE_MBRTOWC | ||
1157 | + mbstate_t thisstate; | ||
1158 | +#endif | ||
1159 | if (readlinebuffer (thisline, stdin) == 0) | ||
1160 | { | ||
1161 | if (ferror (stdin)) | ||
1162 | @@ -341,6 +561,15 @@ | ||
1163 | } | ||
1164 | thisfield = find_field (thisline); | ||
1165 | thislen = thisline->length - 1 - (thisfield - thisline->buffer); | ||
1166 | +#if HAVE_MBRTOWC | ||
1167 | + if (MB_CUR_MAX > 1) | ||
1168 | + { | ||
1169 | + thisstate = thisline->state; | ||
1170 | + match = !different_multi (thisfield, prevfield, | ||
1171 | + thislen, prevlen, thisstate, prevstate); | ||
1172 | + } | ||
1173 | + else | ||
1174 | +#endif | ||
1175 | match = !different (thisfield, prevfield, thislen, prevlen); | ||
1176 | match_count += match; | ||
1177 | |||
1178 | @@ -373,6 +602,9 @@ | ||
1179 | SWAP_LINES (prevline, thisline); | ||
1180 | prevfield = thisfield; | ||
1181 | prevlen = thislen; | ||
1182 | +#if HAVE_MBRTOWC | ||
1183 | + prevstate = thisstate; | ||
1184 | +#endif | ||
1185 | if (!match) | ||
1186 | match_count = 0; | ||
1187 | } | ||
1188 | @@ -417,6 +649,19 @@ | ||
1189 | |||
1190 | atexit (close_stdout); | ||
1191 | |||
1192 | +#if HAVE_MBRTOWC | ||
1193 | + if (MB_CUR_MAX > 1) | ||
1194 | + { | ||
1195 | + find_field = find_field_multi; | ||
1196 | + } | ||
1197 | + else | ||
1198 | +#endif | ||
1199 | + { | ||
1200 | + find_field = find_field_uni; | ||
1201 | + } | ||
1202 | + | ||
1203 | + | ||
1204 | + | ||
1205 | skip_chars = 0; | ||
1206 | skip_fields = 0; | ||
1207 | check_chars = SIZE_MAX; | ||
1208 | --- coreutils-6.8+/src/fold.c.i18n 2007-02-23 12:01:47.000000000 +0000 | ||
1209 | +++ coreutils-6.8+/src/fold.c 2007-03-01 15:08:24.000000000 +0000 | ||
1210 | @@ -23,11 +23,33 @@ | ||
1211 | #include <getopt.h> | ||
1212 | #include <sys/types.h> | ||
1213 | |||
1214 | +/* Get mbstate_t, mbrtowc(), wcwidth(). */ | ||
1215 | +#if HAVE_WCHAR_H | ||
1216 | +# include <wchar.h> | ||
1217 | +#endif | ||
1218 | + | ||
1219 | +/* Get iswprint(), iswblank(), wcwidth(). */ | ||
1220 | +#if HAVE_WCTYPE_H | ||
1221 | +# include <wctype.h> | ||
1222 | +#endif | ||
1223 | + | ||
1224 | #include "system.h" | ||
1225 | #include "error.h" | ||
1226 | #include "quote.h" | ||
1227 | #include "xstrtol.h" | ||
1228 | |||
1229 | +/* MB_LEN_MAX is incorrectly defined to be 1 in at least one GCC | ||
1230 | + installation; work around this configuration error. */ | ||
1231 | +#if !defined MB_LEN_MAX || MB_LEN_MAX < 2 | ||
1232 | +# undef MB_LEN_MAX | ||
1233 | +# define MB_LEN_MAX 16 | ||
1234 | +#endif | ||
1235 | + | ||
1236 | +/* Some systems, like BeOS, have multibyte encodings but lack mbstate_t. */ | ||
1237 | +#if HAVE_MBRTOWC && defined mbstate_t | ||
1238 | +# define mbrtowc(pwc, s, n, ps) (mbrtowc) (pwc, s, n, 0) | ||
1239 | +#endif | ||
1240 | + | ||
1241 | #define TAB_WIDTH 8 | ||
1242 | |||
1243 | /* The official name of this program (e.g., no `g' prefix). */ | ||
1244 | @@ -35,23 +57,44 @@ | ||
1245 | |||
1246 | #define AUTHORS "David MacKenzie" | ||
1247 | |||
1248 | +#define FATAL_ERROR(Message) \ | ||
1249 | + do \ | ||
1250 | + { \ | ||
1251 | + error (0, 0, (Message)); \ | ||
1252 | + usage (2); \ | ||
1253 | + } \ | ||
1254 | + while (0) | ||
1255 | + | ||
1256 | +enum operating_mode | ||
1257 | +{ | ||
1258 | + /* Fold texts by columns that are at the given positions. */ | ||
1259 | + column_mode, | ||
1260 | + | ||
1261 | + /* Fold texts by bytes that are at the given positions. */ | ||
1262 | + byte_mode, | ||
1263 | + | ||
1264 | + /* Fold texts by characters that are at the given positions. */ | ||
1265 | + character_mode, | ||
1266 | +}; | ||
1267 | + | ||
1268 | /* The name this program was run with. */ | ||
1269 | char *program_name; | ||
1270 | |||
1271 | +/* The argument shows current mode. (Default: column_mode) */ | ||
1272 | +static enum operating_mode operating_mode; | ||
1273 | + | ||
1274 | /* If nonzero, try to break on whitespace. */ | ||
1275 | static bool break_spaces; | ||
1276 | |||
1277 | -/* If nonzero, count bytes, not column positions. */ | ||
1278 | -static bool count_bytes; | ||
1279 | - | ||
1280 | /* If nonzero, at least one of the files we read was standard input. */ | ||
1281 | static bool have_read_stdin; | ||
1282 | |||
1283 | -static char const shortopts[] = "bsw:0::1::2::3::4::5::6::7::8::9::"; | ||
1284 | +static char const shortopts[] = "bcsw:0::1::2::3::4::5::6::7::8::9::"; | ||
1285 | |||
1286 | static struct option const longopts[] = | ||
1287 | { | ||
1288 | {"bytes", no_argument, NULL, 'b'}, | ||
1289 | + {"characters", no_argument, NULL, 'c'}, | ||
1290 | {"spaces", no_argument, NULL, 's'}, | ||
1291 | {"width", required_argument, NULL, 'w'}, | ||
1292 | {GETOPT_HELP_OPTION_DECL}, | ||
1293 | @@ -81,6 +124,7 @@ | ||
1294 | "), stdout); | ||
1295 | fputs (_("\ | ||
1296 | -b, --bytes count bytes rather than columns\n\ | ||
1297 | + -c, --characters count characters rather than columns\n\ | ||
1298 | -s, --spaces break at spaces\n\ | ||
1299 | -w, --width=WIDTH use WIDTH columns instead of 80\n\ | ||
1300 | "), stdout); | ||
1301 | @@ -98,7 +142,7 @@ | ||
1302 | static size_t | ||
1303 | adjust_column (size_t column, char c) | ||
1304 | { | ||
1305 | - if (!count_bytes) | ||
1306 | + if (operating_mode != byte_mode) | ||
1307 | { | ||
1308 | if (c == '\b') | ||
1309 | { | ||
1310 | @@ -121,30 +165,14 @@ | ||
1311 | to stdout, with maximum line length WIDTH. | ||
1312 | Return true if successful. */ | ||
1313 | |||
1314 | -static bool | ||
1315 | -fold_file (char const *filename, size_t width) | ||
1316 | +static void | ||
1317 | +fold_text (FILE *istream, size_t width, int *saved_errno) | ||
1318 | { | ||
1319 | - FILE *istream; | ||
1320 | int c; | ||
1321 | size_t column = 0; /* Screen column where next char will go. */ | ||
1322 | size_t offset_out = 0; /* Index in `line_out' for next char. */ | ||
1323 | static char *line_out = NULL; | ||
1324 | static size_t allocated_out = 0; | ||
1325 | - int saved_errno; | ||
1326 | - | ||
1327 | - if (STREQ (filename, "-")) | ||
1328 | - { | ||
1329 | - istream = stdin; | ||
1330 | - have_read_stdin = true; | ||
1331 | - } | ||
1332 | - else | ||
1333 | - istream = fopen (filename, "r"); | ||
1334 | - | ||
1335 | - if (istream == NULL) | ||
1336 | - { | ||
1337 | - error (0, errno, "%s", filename); | ||
1338 | - return false; | ||
1339 | - } | ||
1340 | |||
1341 | while ((c = getc (istream)) != EOF) | ||
1342 | { | ||
1343 | @@ -172,6 +200,15 @@ | ||
1344 | bool found_blank = false; | ||
1345 | size_t logical_end = offset_out; | ||
1346 | |||
1347 | + /* If LINE_OUT has no wide character, | ||
1348 | + put a new wide character in LINE_OUT | ||
1349 | + if column is bigger than width. */ | ||
1350 | + if (offset_out == 0) | ||
1351 | + { | ||
1352 | + line_out[offset_out++] = c; | ||
1353 | + continue; | ||
1354 | + } | ||
1355 | + | ||
1356 | /* Look for the last blank. */ | ||
1357 | while (logical_end) | ||
1358 | { | ||
1359 | @@ -218,11 +255,225 @@ | ||
1360 | line_out[offset_out++] = c; | ||
1361 | } | ||
1362 | |||
1363 | - saved_errno = errno; | ||
1364 | + *saved_errno = errno; | ||
1365 | + | ||
1366 | + if (offset_out) | ||
1367 | + fwrite (line_out, sizeof (char), (size_t) offset_out, stdout); | ||
1368 | + | ||
1369 | + free(line_out); | ||
1370 | +} | ||
1371 | + | ||
1372 | +#if HAVE_MBRTOWC | ||
1373 | +static void | ||
1374 | +fold_multibyte_text (FILE *istream, size_t width, int *saved_errno) | ||
1375 | +{ | ||
1376 | + char buf[MB_LEN_MAX + BUFSIZ]; /* For spooling a read byte sequence. */ | ||
1377 | + size_t buflen = 0; /* The length of the byte sequence in buf. */ | ||
1378 | + char *bufpos; /* Next read position of BUF. */ | ||
1379 | + wint_t wc; /* A gotten wide character. */ | ||
1380 | + size_t mblength; /* The byte size of a multibyte character which shows | ||
1381 | + as same character as WC. */ | ||
1382 | + mbstate_t state, state_bak; /* State of the stream. */ | ||
1383 | + int convfail; /* 1, when conversion is failed. Otherwise 0. */ | ||
1384 | + | ||
1385 | + char *line_out = NULL; | ||
1386 | + size_t offset_out = 0; /* Index in `line_out' for next char. */ | ||
1387 | + size_t allocated_out = 0; | ||
1388 | + | ||
1389 | + int increment; | ||
1390 | + size_t column = 0; | ||
1391 | + | ||
1392 | + size_t last_blank_pos; | ||
1393 | + size_t last_blank_column; | ||
1394 | + int is_blank_seen; | ||
1395 | + int last_blank_increment; | ||
1396 | + int is_bs_following_last_blank; | ||
1397 | + size_t bs_following_last_blank_num; | ||
1398 | + int is_cr_after_last_blank; | ||
1399 | + | ||
1400 | +#define CLEAR_FLAGS \ | ||
1401 | + do \ | ||
1402 | + { \ | ||
1403 | + last_blank_pos = 0; \ | ||
1404 | + last_blank_column = 0; \ | ||
1405 | + is_blank_seen = 0; \ | ||
1406 | + is_bs_following_last_blank = 0; \ | ||
1407 | + bs_following_last_blank_num = 0; \ | ||
1408 | + is_cr_after_last_blank = 0; \ | ||
1409 | + } \ | ||
1410 | + while (0) | ||
1411 | + | ||
1412 | +#define START_NEW_LINE \ | ||
1413 | + do \ | ||
1414 | + { \ | ||
1415 | + putchar ('\n'); \ | ||
1416 | + column = 0; \ | ||
1417 | + offset_out = 0; \ | ||
1418 | + CLEAR_FLAGS; \ | ||
1419 | + } \ | ||
1420 | + while (0) | ||
1421 | + | ||
1422 | + CLEAR_FLAGS; | ||
1423 | + memset (&state, '\0', sizeof(mbstate_t)); | ||
1424 | + | ||
1425 | + for (;; bufpos += mblength, buflen -= mblength) | ||
1426 | + { | ||
1427 | + if (buflen < MB_LEN_MAX && !feof (istream) && !ferror (istream)) | ||
1428 | + { | ||
1429 | + memmove (buf, bufpos, buflen); | ||
1430 | + buflen += fread (buf + buflen, sizeof(char), BUFSIZ, istream); | ||
1431 | + bufpos = buf; | ||
1432 | + } | ||
1433 | + | ||
1434 | + if (buflen < 1) | ||
1435 | + break; | ||
1436 | + | ||
1437 | + /* Get a wide character. */ | ||
1438 | + convfail = 0; | ||
1439 | + state_bak = state; | ||
1440 | + mblength = mbrtowc ((wchar_t *)&wc, bufpos, buflen, &state); | ||
1441 | + | ||
1442 | + switch (mblength) | ||
1443 | + { | ||
1444 | + case (size_t)-1: | ||
1445 | + case (size_t)-2: | ||
1446 | + convfail++; | ||
1447 | + state = state_bak; | ||
1448 | + /* Fall through. */ | ||
1449 | + | ||
1450 | + case 0: | ||
1451 | + mblength = 1; | ||
1452 | + break; | ||
1453 | + } | ||
1454 | + | ||
1455 | +rescan: | ||
1456 | + if (operating_mode == byte_mode) /* byte mode */ | ||
1457 | + increment = mblength; | ||
1458 | + else if (operating_mode == character_mode) /* character mode */ | ||
1459 | + increment = 1; | ||
1460 | + else /* column mode */ | ||
1461 | + { | ||
1462 | + if (convfail) | ||
1463 | + increment = 1; | ||
1464 | + else | ||
1465 | + { | ||
1466 | + switch (wc) | ||
1467 | + { | ||
1468 | + case L'\n': | ||
1469 | + fwrite (line_out, sizeof(char), offset_out, stdout); | ||
1470 | + START_NEW_LINE; | ||
1471 | + continue; | ||
1472 | + | ||
1473 | + case L'\b': | ||
1474 | + increment = (column > 0) ? -1 : 0; | ||
1475 | + break; | ||
1476 | + | ||
1477 | + case L'\r': | ||
1478 | + increment = -1 * column; | ||
1479 | + break; | ||
1480 | + | ||
1481 | + case L'\t': | ||
1482 | + increment = 8 - column % 8; | ||
1483 | + break; | ||
1484 | + | ||
1485 | + default: | ||
1486 | + increment = wcwidth (wc); | ||
1487 | + increment = (increment < 0) ? 0 : increment; | ||
1488 | + } | ||
1489 | + } | ||
1490 | + } | ||
1491 | + | ||
1492 | + if (column + increment > width && break_spaces && last_blank_pos) | ||
1493 | + { | ||
1494 | + fwrite (line_out, sizeof(char), last_blank_pos, stdout); | ||
1495 | + putchar ('\n'); | ||
1496 | + | ||
1497 | + offset_out = offset_out - last_blank_pos; | ||
1498 | + column = column - last_blank_column + ((is_cr_after_last_blank) | ||
1499 | + ? last_blank_increment : bs_following_last_blank_num); | ||
1500 | + memmove (line_out, line_out + last_blank_pos, offset_out); | ||
1501 | + CLEAR_FLAGS; | ||
1502 | + goto rescan; | ||
1503 | + } | ||
1504 | + | ||
1505 | + if (column + increment > width && column != 0) | ||
1506 | + { | ||
1507 | + fwrite (line_out, sizeof(char), offset_out, stdout); | ||
1508 | + START_NEW_LINE; | ||
1509 | + goto rescan; | ||
1510 | + } | ||
1511 | + | ||
1512 | + if (allocated_out < offset_out + mblength) | ||
1513 | + { | ||
1514 | + allocated_out += 1024; | ||
1515 | + line_out = xrealloc (line_out, allocated_out); | ||
1516 | + } | ||
1517 | + | ||
1518 | + memcpy (line_out + offset_out, bufpos, mblength); | ||
1519 | + offset_out += mblength; | ||
1520 | + column += increment; | ||
1521 | + | ||
1522 | + if (is_blank_seen && !convfail && wc == L'\r') | ||
1523 | + is_cr_after_last_blank = 1; | ||
1524 | + | ||
1525 | + if (is_bs_following_last_blank && !convfail && wc == L'\b') | ||
1526 | + ++bs_following_last_blank_num; | ||
1527 | + else | ||
1528 | + is_bs_following_last_blank = 0; | ||
1529 | + | ||
1530 | + if (break_spaces && !convfail && iswblank (wc)) | ||
1531 | + { | ||
1532 | + last_blank_pos = offset_out; | ||
1533 | + last_blank_column = column; | ||
1534 | + is_blank_seen = 1; | ||
1535 | + last_blank_increment = increment; | ||
1536 | + is_bs_following_last_blank = 1; | ||
1537 | + bs_following_last_blank_num = 0; | ||
1538 | + is_cr_after_last_blank = 0; | ||
1539 | + } | ||
1540 | + } | ||
1541 | + | ||
1542 | + *saved_errno = errno; | ||
1543 | |||
1544 | if (offset_out) | ||
1545 | fwrite (line_out, sizeof (char), (size_t) offset_out, stdout); | ||
1546 | |||
1547 | + free(line_out); | ||
1548 | +} | ||
1549 | +#endif | ||
1550 | + | ||
1551 | +/* Fold file FILENAME, or standard input if FILENAME is "-", | ||
1552 | + to stdout, with maximum line length WIDTH. | ||
1553 | + Return 0 if successful, 1 if an error occurs. */ | ||
1554 | + | ||
1555 | +static bool | ||
1556 | +fold_file (char *filename, size_t width) | ||
1557 | +{ | ||
1558 | + FILE *istream; | ||
1559 | + int saved_errno; | ||
1560 | + | ||
1561 | + if (STREQ (filename, "-")) | ||
1562 | + { | ||
1563 | + istream = stdin; | ||
1564 | + have_read_stdin = 1; | ||
1565 | + } | ||
1566 | + else | ||
1567 | + istream = fopen (filename, "r"); | ||
1568 | + | ||
1569 | + if (istream == NULL) | ||
1570 | + { | ||
1571 | + error (0, errno, "%s", filename); | ||
1572 | + return 1; | ||
1573 | + } | ||
1574 | + | ||
1575 | + /* Define how ISTREAM is being folded. */ | ||
1576 | +#if HAVE_MBRTOWC | ||
1577 | + if (MB_CUR_MAX > 1) | ||
1578 | + fold_multibyte_text (istream, width, &saved_errno); | ||
1579 | + else | ||
1580 | +#endif | ||
1581 | + fold_text (istream, width, &saved_errno); | ||
1582 | + | ||
1583 | if (ferror (istream)) | ||
1584 | { | ||
1585 | error (0, saved_errno, "%s", filename); | ||
1586 | @@ -255,7 +506,8 @@ | ||
1587 | |||
1588 | atexit (close_stdout); | ||
1589 | |||
1590 | - break_spaces = count_bytes = have_read_stdin = false; | ||
1591 | + operating_mode = column_mode; | ||
1592 | + break_spaces = have_read_stdin = false; | ||
1593 | |||
1594 | while ((optc = getopt_long (argc, argv, shortopts, longopts, NULL)) != -1) | ||
1595 | { | ||
1596 | @@ -264,7 +516,15 @@ | ||
1597 | switch (optc) | ||
1598 | { | ||
1599 | case 'b': /* Count bytes rather than columns. */ | ||
1600 | - count_bytes = true; | ||
1601 | + if (operating_mode != column_mode) | ||
1602 | + FATAL_ERROR (_("only one way of folding may be specified")); | ||
1603 | + operating_mode = byte_mode; | ||
1604 | + break; | ||
1605 | + | ||
1606 | + case 'c': | ||
1607 | + if (operating_mode != column_mode) | ||
1608 | + FATAL_ERROR (_("only one way of folding may be specified")); | ||
1609 | + operating_mode = character_mode; | ||
1610 | break; | ||
1611 | |||
1612 | case 's': /* Break at word boundaries. */ | ||
1613 | --- coreutils-6.8+/src/sort.c.i18n 2007-02-24 11:23:23.000000000 +0000 | ||
1614 | +++ coreutils-6.8+/src/sort.c 2007-03-01 15:10:57.000000000 +0000 | ||
1615 | @@ -23,10 +23,19 @@ | ||
1616 | |||
1617 | #include <config.h> | ||
1618 | |||
1619 | +#include <assert.h> | ||
1620 | #include <getopt.h> | ||
1621 | #include <sys/types.h> | ||
1622 | #include <sys/wait.h> | ||
1623 | #include <signal.h> | ||
1624 | +#if HAVE_WCHAR_H | ||
1625 | +# include <wchar.h> | ||
1626 | +#endif | ||
1627 | +/* Get isw* functions. */ | ||
1628 | +#if HAVE_WCTYPE_H | ||
1629 | +# include <wctype.h> | ||
1630 | +#endif | ||
1631 | + | ||
1632 | #include "system.h" | ||
1633 | #include "argmatch.h" | ||
1634 | #include "error.h" | ||
1635 | @@ -116,14 +125,38 @@ | ||
1636 | /* Thousands separator; if -1, then there isn't one. */ | ||
1637 | static int thousands_sep; | ||
1638 | |||
1639 | +static int force_general_numcompare = 0; | ||
1640 | + | ||
1641 | /* Nonzero if the corresponding locales are hard. */ | ||
1642 | static bool hard_LC_COLLATE; | ||
1643 | -#if HAVE_NL_LANGINFO | ||
1644 | +#if HAVE_LANGINFO_CODESET | ||
1645 | static bool hard_LC_TIME; | ||
1646 | #endif | ||
1647 | |||
1648 | #define NONZERO(x) ((x) != 0) | ||
1649 | |||
1650 | +/* get a multibyte character's byte length. */ | ||
1651 | +#define GET_BYTELEN_OF_CHAR(LIM, PTR, MBLENGTH, STATE) \ | ||
1652 | + do \ | ||
1653 | + { \ | ||
1654 | + wchar_t wc; \ | ||
1655 | + mbstate_t state_bak; \ | ||
1656 | + \ | ||
1657 | + state_bak = STATE; \ | ||
1658 | + mblength = mbrtowc (&wc, PTR, LIM - PTR, &STATE); \ | ||
1659 | + \ | ||
1660 | + switch (MBLENGTH) \ | ||
1661 | + { \ | ||
1662 | + case (size_t)-1: \ | ||
1663 | + case (size_t)-2: \ | ||
1664 | + STATE = state_bak; \ | ||
1665 | + /* Fall through. */ \ | ||
1666 | + case 0: \ | ||
1667 | + MBLENGTH = 1; \ | ||
1668 | + } \ | ||
1669 | + } \ | ||
1670 | + while (0) | ||
1671 | + | ||
1672 | /* The kind of blanks for '-b' to skip in various options. */ | ||
1673 | enum blanktype { bl_start, bl_end, bl_both }; | ||
1674 | |||
1675 | @@ -261,13 +294,11 @@ | ||
1676 | they were read if all keys compare equal. */ | ||
1677 | static bool stable; | ||
1678 | |||
1679 | -/* If TAB has this value, blanks separate fields. */ | ||
1680 | -enum { TAB_DEFAULT = CHAR_MAX + 1 }; | ||
1681 | - | ||
1682 | -/* Tab character separating fields. If TAB_DEFAULT, then fields are | ||
1683 | +/* Tab character separating fields. If tab_length is 0, then fields are | ||
1684 | separated by the empty string between a non-blank character and a blank | ||
1685 | character. */ | ||
1686 | -static int tab = TAB_DEFAULT; | ||
1687 | +static char tab[MB_LEN_MAX + 1]; | ||
1688 | +static size_t tab_length = 0; | ||
1689 | |||
1690 | /* Flag to remove consecutive duplicate lines from the output. | ||
1691 | Only the last of a sequence of equal lines will be output. */ | ||
1692 | @@ -639,6 +670,44 @@ | ||
1693 | update_proc (pid); | ||
1694 | } | ||
1695 | |||
1696 | +/* Function pointers. */ | ||
1697 | +static void | ||
1698 | +(*inittables) (void); | ||
1699 | +static char * | ||
1700 | +(*begfield) (const struct line*, const struct keyfield *); | ||
1701 | +static char * | ||
1702 | +(*limfield) (const struct line*, const struct keyfield *); | ||
1703 | +static int | ||
1704 | +(*getmonth) (char const *, size_t); | ||
1705 | +static int | ||
1706 | +(*keycompare) (const struct line *, const struct line *); | ||
1707 | +static int | ||
1708 | +(*numcompare) (const char *, const char *); | ||
1709 | + | ||
1710 | +/* Test for white space multibyte character. | ||
1711 | + Set LENGTH the byte length of investigated multibyte character. */ | ||
1712 | +#if HAVE_MBRTOWC | ||
1713 | +static int | ||
1714 | +ismbblank (const char *str, size_t len, size_t *length) | ||
1715 | +{ | ||
1716 | + size_t mblength; | ||
1717 | + wchar_t wc; | ||
1718 | + mbstate_t state; | ||
1719 | + | ||
1720 | + memset (&state, '\0', sizeof(mbstate_t)); | ||
1721 | + mblength = mbrtowc (&wc, str, len, &state); | ||
1722 | + | ||
1723 | + if (mblength == (size_t)-1 || mblength == (size_t)-2) | ||
1724 | + { | ||
1725 | + *length = 1; | ||
1726 | + return 0; | ||
1727 | + } | ||
1728 | + | ||
1729 | + *length = (mblength < 1) ? 1 : mblength; | ||
1730 | + return iswblank (wc); | ||
1731 | +} | ||
1732 | +#endif | ||
1733 | + | ||
1734 | /* Clean up any remaining temporary files. */ | ||
1735 | |||
1736 | static void | ||
1737 | @@ -978,7 +1047,7 @@ | ||
1738 | free (node); | ||
1739 | } | ||
1740 | |||
1741 | -#if HAVE_NL_LANGINFO | ||
1742 | +#if HAVE_LANGINFO_CODESET | ||
1743 | |||
1744 | static int | ||
1745 | struct_month_cmp (const void *m1, const void *m2) | ||
1746 | @@ -993,7 +1062,7 @@ | ||
1747 | /* Initialize the character class tables. */ | ||
1748 | |||
1749 | static void | ||
1750 | -inittables (void) | ||
1751 | +inittables_uni (void) | ||
1752 | { | ||
1753 | size_t i; | ||
1754 | |||
1755 | @@ -1005,7 +1074,7 @@ | ||
1756 | fold_toupper[i] = toupper (i); | ||
1757 | } | ||
1758 | |||
1759 | -#if HAVE_NL_LANGINFO | ||
1760 | +#if HAVE_LANGINFO_CODESET | ||
1761 | /* If we're not in the "C" locale, read different names for months. */ | ||
1762 | if (hard_LC_TIME) | ||
1763 | { | ||
1764 | @@ -1031,6 +1100,64 @@ | ||
1765 | #endif | ||
1766 | } | ||
1767 | |||
1768 | +#if HAVE_MBRTOWC | ||
1769 | +static void | ||
1770 | +inittables_mb (void) | ||
1771 | +{ | ||
1772 | + int i, j, k, l; | ||
1773 | + char *name, *s; | ||
1774 | + size_t s_len, mblength; | ||
1775 | + char mbc[MB_LEN_MAX]; | ||
1776 | + wchar_t wc, pwc; | ||
1777 | + mbstate_t state_mb, state_wc; | ||
1778 | + | ||
1779 | + for (i = 0; i < MONTHS_PER_YEAR; i++) | ||
1780 | + { | ||
1781 | + s = (char *) nl_langinfo (ABMON_1 + i); | ||
1782 | + s_len = strlen (s); | ||
1783 | + monthtab[i].name = name = (char *) xmalloc (s_len + 1); | ||
1784 | + monthtab[i].val = i + 1; | ||
1785 | + | ||
1786 | + memset (&state_mb, '\0', sizeof (mbstate_t)); | ||
1787 | + memset (&state_wc, '\0', sizeof (mbstate_t)); | ||
1788 | + | ||
1789 | + for (j = 0; j < s_len;) | ||
1790 | + { | ||
1791 | + if (!ismbblank (s + j, s_len - j, &mblength)) | ||
1792 | + break; | ||
1793 | + j += mblength; | ||
1794 | + } | ||
1795 | + | ||
1796 | + for (k = 0; j < s_len;) | ||
1797 | + { | ||
1798 | + mblength = mbrtowc (&wc, (s + j), (s_len - j), &state_mb); | ||
1799 | + assert (mblength != (size_t)-1 && mblength != (size_t)-2); | ||
1800 | + if (mblength == 0) | ||
1801 | + break; | ||
1802 | + | ||
1803 | + pwc = towupper (wc); | ||
1804 | + if (pwc == wc) | ||
1805 | + { | ||
1806 | + memcpy (mbc, s + j, mblength); | ||
1807 | + j += mblength; | ||
1808 | + } | ||
1809 | + else | ||
1810 | + { | ||
1811 | + j += mblength; | ||
1812 | + mblength = wcrtomb (mbc, pwc, &state_wc); | ||
1813 | + assert (mblength != (size_t)0 && mblength != (size_t)-1); | ||
1814 | + } | ||
1815 | + | ||
1816 | + for (l = 0; l < mblength; l++) | ||
1817 | + name[k++] = mbc[l]; | ||
1818 | + } | ||
1819 | + name[k] = '\0'; | ||
1820 | + } | ||
1821 | + qsort ((void *) monthtab, MONTHS_PER_YEAR, | ||
1822 | + sizeof (struct month), struct_month_cmp); | ||
1823 | +} | ||
1824 | +#endif | ||
1825 | + | ||
1826 | /* Specify the amount of main memory to use when sorting. */ | ||
1827 | static void | ||
1828 | specify_sort_size (char const *s) | ||
1829 | @@ -1241,7 +1368,7 @@ | ||
1830 | by KEY in LINE. */ | ||
1831 | |||
1832 | static char * | ||
1833 | -begfield (const struct line *line, const struct keyfield *key) | ||
1834 | +begfield_uni (const struct line *line, const struct keyfield *key) | ||
1835 | { | ||
1836 | char *ptr = line->text, *lim = ptr + line->length - 1; | ||
1837 | size_t sword = key->sword; | ||
1838 | @@ -1251,10 +1378,10 @@ | ||
1839 | /* The leading field separator itself is included in a field when -t | ||
1840 | is absent. */ | ||
1841 | |||
1842 | - if (tab != TAB_DEFAULT) | ||
1843 | + if (tab_length) | ||
1844 | while (ptr < lim && sword--) | ||
1845 | { | ||
1846 | - while (ptr < lim && *ptr != tab) | ||
1847 | + while (ptr < lim && *ptr != tab[0]) | ||
1848 | ++ptr; | ||
1849 | if (ptr < lim) | ||
1850 | ++ptr; | ||
1851 | @@ -1282,11 +1409,70 @@ | ||
1852 | return ptr; | ||
1853 | } | ||
1854 | |||
1855 | +#if HAVE_MBRTOWC | ||
1856 | +static char * | ||
1857 | +begfield_mb (const struct line *line, const struct keyfield *key) | ||
1858 | +{ | ||
1859 | + int i; | ||
1860 | + char *ptr = line->text, *lim = ptr + line->length - 1; | ||
1861 | + size_t sword = key->sword; | ||
1862 | + size_t schar = key->schar; | ||
1863 | + size_t mblength; | ||
1864 | + mbstate_t state; | ||
1865 | + | ||
1866 | + memset (&state, '\0', sizeof(mbstate_t)); | ||
1867 | + | ||
1868 | + if (tab_length) | ||
1869 | + while (ptr < lim && sword--) | ||
1870 | + { | ||
1871 | + while (ptr < lim && memcmp (ptr, tab, tab_length) != 0) | ||
1872 | + { | ||
1873 | + GET_BYTELEN_OF_CHAR (lim, ptr, mblength, state); | ||
1874 | + ptr += mblength; | ||
1875 | + } | ||
1876 | + if (ptr < lim) | ||
1877 | + { | ||
1878 | + GET_BYTELEN_OF_CHAR (lim, ptr, mblength, state); | ||
1879 | + ptr += mblength; | ||
1880 | + } | ||
1881 | + } | ||
1882 | + else | ||
1883 | + while (ptr < lim && sword--) | ||
1884 | + { | ||
1885 | + while (ptr < lim && ismbblank (ptr, lim - ptr, &mblength)) | ||
1886 | + ptr += mblength; | ||
1887 | + if (ptr < lim) | ||
1888 | + { | ||
1889 | + GET_BYTELEN_OF_CHAR (lim, ptr, mblength, state); | ||
1890 | + ptr += mblength; | ||
1891 | + } | ||
1892 | + while (ptr < lim && !ismbblank (ptr, lim - ptr, &mblength)) | ||
1893 | + ptr += mblength; | ||
1894 | + } | ||
1895 | + | ||
1896 | + if (key->skipsblanks) | ||
1897 | + while (ptr < lim && ismbblank (ptr, lim - ptr, &mblength)) | ||
1898 | + ptr += mblength; | ||
1899 | + | ||
1900 | + for (i = 0; i < schar; i++) | ||
1901 | + { | ||
1902 | + GET_BYTELEN_OF_CHAR (lim, ptr, mblength, state); | ||
1903 | + | ||
1904 | + if (ptr + mblength > lim) | ||
1905 | + break; | ||
1906 | + else | ||
1907 | + ptr += mblength; | ||
1908 | + } | ||
1909 | + | ||
1910 | + return ptr; | ||
1911 | +} | ||
1912 | +#endif | ||
1913 | + | ||
1914 | /* Return the limit of (a pointer to the first character after) the field | ||
1915 | in LINE specified by KEY. */ | ||
1916 | |||
1917 | static char * | ||
1918 | -limfield (const struct line *line, const struct keyfield *key) | ||
1919 | +limfield_uni (const struct line *line, const struct keyfield *key) | ||
1920 | { | ||
1921 | char *ptr = line->text, *lim = ptr + line->length - 1; | ||
1922 | size_t eword = key->eword, echar = key->echar; | ||
1923 | @@ -1299,10 +1485,10 @@ | ||
1924 | `beginning' is the first character following the delimiting TAB. | ||
1925 | Otherwise, leave PTR pointing at the first `blank' character after | ||
1926 | the preceding field. */ | ||
1927 | - if (tab != TAB_DEFAULT) | ||
1928 | + if (tab_length) | ||
1929 | while (ptr < lim && eword--) | ||
1930 | { | ||
1931 | - while (ptr < lim && *ptr != tab) | ||
1932 | + while (ptr < lim && *ptr != tab[0]) | ||
1933 | ++ptr; | ||
1934 | if (ptr < lim && (eword | echar)) | ||
1935 | ++ptr; | ||
1936 | @@ -1348,10 +1534,10 @@ | ||
1937 | */ | ||
1938 | |||
1939 | /* Make LIM point to the end of (one byte past) the current field. */ | ||
1940 | - if (tab != TAB_DEFAULT) | ||
1941 | + if (tab_length) | ||
1942 | { | ||
1943 | char *newlim; | ||
1944 | - newlim = memchr (ptr, tab, lim - ptr); | ||
1945 | + newlim = memchr (ptr, tab[0], lim - ptr); | ||
1946 | if (newlim) | ||
1947 | lim = newlim; | ||
1948 | } | ||
1949 | @@ -1384,6 +1570,107 @@ | ||
1950 | return ptr; | ||
1951 | } | ||
1952 | |||
1953 | +#if HAVE_MBRTOWC | ||
1954 | +static char * | ||
1955 | +limfield_mb (const struct line *line, const struct keyfield *key) | ||
1956 | +{ | ||
1957 | + char *ptr = line->text, *lim = ptr + line->length - 1; | ||
1958 | + size_t eword = key->eword, echar = key->echar; | ||
1959 | + int i; | ||
1960 | + size_t mblength; | ||
1961 | + mbstate_t state; | ||
1962 | + | ||
1963 | + memset (&state, '\0', sizeof(mbstate_t)); | ||
1964 | + | ||
1965 | + if (tab_length) | ||
1966 | + while (ptr < lim && eword--) | ||
1967 | + { | ||
1968 | + while (ptr < lim && memcmp (ptr, tab, tab_length) != 0) | ||
1969 | + { | ||
1970 | + GET_BYTELEN_OF_CHAR (lim, ptr, mblength, state); | ||
1971 | + ptr += mblength; | ||
1972 | + } | ||
1973 | + if (ptr < lim && (eword | echar)) | ||
1974 | + { | ||
1975 | + GET_BYTELEN_OF_CHAR (lim, ptr, mblength, state); | ||
1976 | + ptr += mblength; | ||
1977 | + } | ||
1978 | + } | ||
1979 | + else | ||
1980 | + while (ptr < lim && eword--) | ||
1981 | + { | ||
1982 | + while (ptr < lim && ismbblank (ptr, lim - ptr, &mblength)) | ||
1983 | + ptr += mblength; | ||
1984 | + if (ptr < lim) | ||
1985 | + { | ||
1986 | + GET_BYTELEN_OF_CHAR (lim, ptr, mblength, state); | ||
1987 | + ptr += mblength; | ||
1988 | + } | ||
1989 | + while (ptr < lim && !ismbblank (ptr, lim - ptr, &mblength)) | ||
1990 | + ptr += mblength; | ||
1991 | + } | ||
1992 | + | ||
1993 | + | ||
1994 | +# ifdef POSIX_UNSPECIFIED | ||
1995 | + /* Make LIM point to the end of (one byte past) the current field. */ | ||
1996 | + if (tab_length) | ||
1997 | + { | ||
1998 | + char *newlim, *p; | ||
1999 | + | ||
2000 | + newlim = NULL; | ||
2001 | + for (p = ptr; p < lim;) | ||
2002 | + { | ||
2003 | + if (memcmp (p, tab, tab_length) == 0) | ||
2004 | + { | ||
2005 | + newlim = p; | ||
2006 | + break; | ||
2007 | + } | ||
2008 | + | ||
2009 | + GET_BYTELEN_OF_CHAR (lim, ptr, mblength, state); | ||
2010 | + p += mblength; | ||
2011 | + } | ||
2012 | + } | ||
2013 | + else | ||
2014 | + { | ||
2015 | + char *newlim; | ||
2016 | + newlim = ptr; | ||
2017 | + | ||
2018 | + while (newlim < lim && ismbblank (newlim, lim - newlim, &mblength)) | ||
2019 | + newlim += mblength; | ||
2020 | + if (ptr < lim) | ||
2021 | + { | ||
2022 | + GET_BYTELEN_OF_CHAR (lim, ptr, mblength, state); | ||
2023 | + ptr += mblength; | ||
2024 | + } | ||
2025 | + while (newlim < lim && !ismbblank (newlim, lim - newlim, &mblength)) | ||
2026 | + newlim += mblength; | ||
2027 | + lim = newlim; | ||
2028 | + } | ||
2029 | +# endif | ||
2030 | + | ||
2031 | + /* If we're skipping leading blanks, don't start counting characters | ||
2032 | + * until after skipping past any leading blanks. */ | ||
2033 | + if (key->skipsblanks) | ||
2034 | + while (ptr < lim && ismbblank (ptr, lim - ptr, &mblength)) | ||
2035 | + ptr += mblength; | ||
2036 | + | ||
2037 | + memset (&state, '\0', sizeof(mbstate_t)); | ||
2038 | + | ||
2039 | + /* Advance PTR by ECHAR (if possible), but no further than LIM. */ | ||
2040 | + for (i = 0; i < echar; i++) | ||
2041 | + { | ||
2042 | + GET_BYTELEN_OF_CHAR (lim, ptr, mblength, state); | ||
2043 | + | ||
2044 | + if (ptr + mblength > lim) | ||
2045 | + break; | ||
2046 | + else | ||
2047 | + ptr += mblength; | ||
2048 | + } | ||
2049 | + | ||
2050 | + return ptr; | ||
2051 | +} | ||
2052 | +#endif | ||
2053 | + | ||
2054 | /* Fill BUF reading from FP, moving buf->left bytes from the end | ||
2055 | of buf->buf to the beginning first. If EOF is reached and the | ||
2056 | file wasn't terminated by a newline, supply one. Set up BUF's line | ||
2057 | @@ -1466,8 +1753,24 @@ | ||
2058 | else | ||
2059 | { | ||
2060 | if (key->skipsblanks) | ||
2061 | - while (blanks[to_uchar (*line_start)]) | ||
2062 | - line_start++; | ||
2063 | + { | ||
2064 | +#if HAVE_MBRTOWC | ||
2065 | + if (MB_CUR_MAX > 1) | ||
2066 | + { | ||
2067 | + size_t mblength; | ||
2068 | + mbstate_t state; | ||
2069 | + memset (&state, '\0', sizeof(mbstate_t)); | ||
2070 | + while (line_start < line->keylim && | ||
2071 | + ismbblank (line_start, | ||
2072 | + line->keylim - line_start, | ||
2073 | + &mblength)) | ||
2074 | + line_start += mblength; | ||
2075 | + } | ||
2076 | + else | ||
2077 | +#endif | ||
2078 | + while (blanks[to_uchar (*line_start)]) | ||
2079 | + line_start++; | ||
2080 | + } | ||
2081 | line->keybeg = line_start; | ||
2082 | } | ||
2083 | } | ||
2084 | @@ -1500,7 +1803,7 @@ | ||
2085 | hideously fast. */ | ||
2086 | |||
2087 | static int | ||
2088 | -numcompare (const char *a, const char *b) | ||
2089 | +numcompare_uni (const char *a, const char *b) | ||
2090 | { | ||
2091 | while (blanks[to_uchar (*a)]) | ||
2092 | a++; | ||
2093 | @@ -1510,6 +1813,25 @@ | ||
2094 | return strnumcmp (a, b, decimal_point, thousands_sep); | ||
2095 | } | ||
2096 | |||
2097 | +#if HAVE_MBRTOWC | ||
2098 | +static int | ||
2099 | +numcompare_mb (const char *a, const char *b) | ||
2100 | +{ | ||
2101 | + size_t mblength, len; | ||
2102 | + len = strlen (a); /* okay for UTF-8 */ | ||
2103 | + while (*a && ismbblank (a, len > MB_CUR_MAX ? MB_CUR_MAX : len, &mblength)) | ||
2104 | + { | ||
2105 | + a += mblength; | ||
2106 | + len -= mblength; | ||
2107 | + } | ||
2108 | + len = strlen (b); /* okay for UTF-8 */ | ||
2109 | + while (*b && ismbblank (b, len > MB_CUR_MAX ? MB_CUR_MAX : len, &mblength)) | ||
2110 | + b += mblength; | ||
2111 | + | ||
2112 | + return strnumcmp (a, b, decimal_point, thousands_sep); | ||
2113 | +} | ||
2114 | +#endif /* HAV_EMBRTOWC */ | ||
2115 | + | ||
2116 | static int | ||
2117 | general_numcompare (const char *sa, const char *sb) | ||
2118 | { | ||
2119 | @@ -1543,7 +1865,7 @@ | ||
2120 | Return 0 if the name in S is not recognized. */ | ||
2121 | |||
2122 | static int | ||
2123 | -getmonth (char const *month, size_t len) | ||
2124 | +getmonth_uni (char const *month, size_t len) | ||
2125 | { | ||
2126 | size_t lo = 0; | ||
2127 | size_t hi = MONTHS_PER_YEAR; | ||
2128 | @@ -1698,11 +2020,79 @@ | ||
2129 | return diff; | ||
2130 | } | ||
2131 | |||
2132 | +#if HAVE_MBRTOWC | ||
2133 | +static int | ||
2134 | +getmonth_mb (const char *s, size_t len) | ||
2135 | +{ | ||
2136 | + char *month; | ||
2137 | + register size_t i; | ||
2138 | + register int lo = 0, hi = MONTHS_PER_YEAR, result; | ||
2139 | + char *tmp; | ||
2140 | + size_t wclength, mblength; | ||
2141 | + const char **pp; | ||
2142 | + const wchar_t **wpp; | ||
2143 | + wchar_t *month_wcs; | ||
2144 | + mbstate_t state; | ||
2145 | + | ||
2146 | + while (len > 0 && ismbblank (s, len, &mblength)) | ||
2147 | + { | ||
2148 | + s += mblength; | ||
2149 | + len -= mblength; | ||
2150 | + } | ||
2151 | + | ||
2152 | + if (len == 0) | ||
2153 | + return 0; | ||
2154 | + | ||
2155 | + month = (char *) alloca (len + 1); | ||
2156 | + | ||
2157 | + tmp = (char *) alloca (len + 1); | ||
2158 | + memcpy (tmp, s, len); | ||
2159 | + tmp[len] = '\0'; | ||
2160 | + pp = (const char **)&tmp; | ||
2161 | + month_wcs = (wchar_t *) alloca ((len + 1) * sizeof (wchar_t)); | ||
2162 | + memset (&state, '\0', sizeof(mbstate_t)); | ||
2163 | + | ||
2164 | + wclength = mbsrtowcs (month_wcs, pp, len + 1, &state); | ||
2165 | + assert (wclength != (size_t)-1 && *pp == NULL); | ||
2166 | + | ||
2167 | + for (i = 0; i < wclength; i++) | ||
2168 | + { | ||
2169 | + month_wcs[i] = towupper(month_wcs[i]); | ||
2170 | + if (iswblank (month_wcs[i])) | ||
2171 | + { | ||
2172 | + month_wcs[i] = L'\0'; | ||
2173 | + break; | ||
2174 | + } | ||
2175 | + } | ||
2176 | + | ||
2177 | + wpp = (const wchar_t **)&month_wcs; | ||
2178 | + | ||
2179 | + mblength = wcsrtombs (month, wpp, len + 1, &state); | ||
2180 | + assert (mblength != (-1) && *wpp == NULL); | ||
2181 | + | ||
2182 | + do | ||
2183 | + { | ||
2184 | + int ix = (lo + hi) / 2; | ||
2185 | + | ||
2186 | + if (strncmp (month, monthtab[ix].name, strlen (monthtab[ix].name)) < 0) | ||
2187 | + hi = ix; | ||
2188 | + else | ||
2189 | + lo = ix; | ||
2190 | + } | ||
2191 | + while (hi - lo > 1); | ||
2192 | + | ||
2193 | + result = (!strncmp (month, monthtab[lo].name, strlen (monthtab[lo].name)) | ||
2194 | + ? monthtab[lo].val : 0); | ||
2195 | + | ||
2196 | + return result; | ||
2197 | +} | ||
2198 | +#endif | ||
2199 | + | ||
2200 | /* Compare two lines A and B trying every key in sequence until there | ||
2201 | are no more keys or a difference is found. */ | ||
2202 | |||
2203 | static int | ||
2204 | -keycompare (const struct line *a, const struct line *b) | ||
2205 | +keycompare_uni (const struct line *a, const struct line *b) | ||
2206 | { | ||
2207 | struct keyfield const *key = keylist; | ||
2208 | |||
2209 | @@ -1875,6 +2265,179 @@ | ||
2210 | return key->reverse ? -diff : diff; | ||
2211 | } | ||
2212 | |||
2213 | +#if HAVE_MBRTOWC | ||
2214 | +static int | ||
2215 | +keycompare_mb (const struct line *a, const struct line *b) | ||
2216 | +{ | ||
2217 | + struct keyfield *key = keylist; | ||
2218 | + | ||
2219 | + /* For the first iteration only, the key positions have been | ||
2220 | + precomputed for us. */ | ||
2221 | + char *texta = a->keybeg; | ||
2222 | + char *textb = b->keybeg; | ||
2223 | + char *lima = a->keylim; | ||
2224 | + char *limb = b->keylim; | ||
2225 | + | ||
2226 | + size_t mblength_a, mblength_b; | ||
2227 | + wchar_t wc_a, wc_b; | ||
2228 | + mbstate_t state_a, state_b; | ||
2229 | + | ||
2230 | + int diff; | ||
2231 | + | ||
2232 | + memset (&state_a, '\0', sizeof(mbstate_t)); | ||
2233 | + memset (&state_b, '\0', sizeof(mbstate_t)); | ||
2234 | + | ||
2235 | + for (;;) | ||
2236 | + { | ||
2237 | + unsigned char *translate = (unsigned char *) key->translate; | ||
2238 | + bool const *ignore = key->ignore; | ||
2239 | + | ||
2240 | + /* Find the lengths. */ | ||
2241 | + size_t lena = lima <= texta ? 0 : lima - texta; | ||
2242 | + size_t lenb = limb <= textb ? 0 : limb - textb; | ||
2243 | + | ||
2244 | + /* Actually compare the fields. */ | ||
2245 | + if (key->random) | ||
2246 | + diff = compare_random (texta, lena, textb, lenb); | ||
2247 | + else if (key->numeric | key->general_numeric) | ||
2248 | + { | ||
2249 | + char savea = *lima, saveb = *limb; | ||
2250 | + | ||
2251 | + *lima = *limb = '\0'; | ||
2252 | + if (force_general_numcompare) | ||
2253 | + diff = general_numcompare (texta, textb); | ||
2254 | + else | ||
2255 | + diff = ((key->numeric ? numcompare : general_numcompare) | ||
2256 | + (texta, textb)); | ||
2257 | + *lima = savea, *limb = saveb; | ||
2258 | + } | ||
2259 | + else if (key->month) | ||
2260 | + diff = getmonth (texta, lena) - getmonth (textb, lenb); | ||
2261 | + else | ||
2262 | + { | ||
2263 | + if (ignore || translate) | ||
2264 | + { | ||
2265 | + char *copy_a = (char *) alloca (lena + 1 + lenb + 1); | ||
2266 | + char *copy_b = copy_a + lena + 1; | ||
2267 | + size_t new_len_a, new_len_b; | ||
2268 | + size_t i, j; | ||
2269 | + | ||
2270 | + /* Ignore and/or translate chars before comparing. */ | ||
2271 | +# define IGNORE_CHARS(NEW_LEN, LEN, TEXT, COPY, WC, MBLENGTH, STATE) \ | ||
2272 | + do \ | ||
2273 | + { \ | ||
2274 | + wchar_t uwc; \ | ||
2275 | + char mbc[MB_LEN_MAX]; \ | ||
2276 | + mbstate_t state_wc; \ | ||
2277 | + \ | ||
2278 | + for (NEW_LEN = i = 0; i < LEN;) \ | ||
2279 | + { \ | ||
2280 | + mbstate_t state_bak; \ | ||
2281 | + \ | ||
2282 | + state_bak = STATE; \ | ||
2283 | + MBLENGTH = mbrtowc (&WC, TEXT + i, LEN - i, &STATE); \ | ||
2284 | + \ | ||
2285 | + if (MBLENGTH == (size_t)-2 || MBLENGTH == (size_t)-1 \ | ||
2286 | + || MBLENGTH == 0) \ | ||
2287 | + { \ | ||
2288 | + if (MBLENGTH == (size_t)-2 || MBLENGTH == (size_t)-1) \ | ||
2289 | + STATE = state_bak; \ | ||
2290 | + if (!ignore) \ | ||
2291 | + COPY[NEW_LEN++] = TEXT[i++]; \ | ||
2292 | + continue; \ | ||
2293 | + } \ | ||
2294 | + \ | ||
2295 | + if (ignore) \ | ||
2296 | + { \ | ||
2297 | + if ((ignore == nonprinting && !iswprint (WC)) \ | ||
2298 | + || (ignore == nondictionary \ | ||
2299 | + && !iswalnum (WC) && !iswblank (WC))) \ | ||
2300 | + { \ | ||
2301 | + i += MBLENGTH; \ | ||
2302 | + continue; \ | ||
2303 | + } \ | ||
2304 | + } \ | ||
2305 | + \ | ||
2306 | + if (translate) \ | ||
2307 | + { \ | ||
2308 | + \ | ||
2309 | + uwc = towupper(WC); \ | ||
2310 | + if (WC == uwc) \ | ||
2311 | + { \ | ||
2312 | + memcpy (mbc, TEXT + i, MBLENGTH); \ | ||
2313 | + i += MBLENGTH; \ | ||
2314 | + } \ | ||
2315 | + else \ | ||
2316 | + { \ | ||
2317 | + i += MBLENGTH; \ | ||
2318 | + WC = uwc; \ | ||
2319 | + memset (&state_wc, '\0', sizeof (mbstate_t)); \ | ||
2320 | + \ | ||
2321 | + MBLENGTH = wcrtomb (mbc, WC, &state_wc); \ | ||
2322 | + assert (MBLENGTH != (size_t)-1 && MBLENGTH != 0); \ | ||
2323 | + } \ | ||
2324 | + \ | ||
2325 | + for (j = 0; j < MBLENGTH; j++) \ | ||
2326 | + COPY[NEW_LEN++] = mbc[j]; \ | ||
2327 | + } \ | ||
2328 | + else \ | ||
2329 | + for (j = 0; j < MBLENGTH; j++) \ | ||
2330 | + COPY[NEW_LEN++] = TEXT[i++]; \ | ||
2331 | + } \ | ||
2332 | + COPY[NEW_LEN] = '\0'; \ | ||
2333 | + } \ | ||
2334 | + while (0) | ||
2335 | + IGNORE_CHARS (new_len_a, lena, texta, copy_a, | ||
2336 | + wc_a, mblength_a, state_a); | ||
2337 | + IGNORE_CHARS (new_len_b, lenb, textb, copy_b, | ||
2338 | + wc_b, mblength_b, state_b); | ||
2339 | + diff = xmemcoll (copy_a, new_len_a, copy_b, new_len_b); | ||
2340 | + } | ||
2341 | + else if (lena == 0) | ||
2342 | + diff = - NONZERO (lenb); | ||
2343 | + else if (lenb == 0) | ||
2344 | + goto greater; | ||
2345 | + else | ||
2346 | + diff = xmemcoll (texta, lena, textb, lenb); | ||
2347 | + } | ||
2348 | + | ||
2349 | + if (diff) | ||
2350 | + goto not_equal; | ||
2351 | + | ||
2352 | + key = key->next; | ||
2353 | + if (! key) | ||
2354 | + break; | ||
2355 | + | ||
2356 | + /* Find the beginning and limit of the next field. */ | ||
2357 | + if (key->eword != -1) | ||
2358 | + lima = limfield (a, key), limb = limfield (b, key); | ||
2359 | + else | ||
2360 | + lima = a->text + a->length - 1, limb = b->text + b->length - 1; | ||
2361 | + | ||
2362 | + if (key->sword != -1) | ||
2363 | + texta = begfield (a, key), textb = begfield (b, key); | ||
2364 | + else | ||
2365 | + { | ||
2366 | + texta = a->text, textb = b->text; | ||
2367 | + if (key->skipsblanks) | ||
2368 | + { | ||
2369 | + while (texta < lima && ismbblank (texta, lima - texta, &mblength_a)) | ||
2370 | + texta += mblength_a; | ||
2371 | + while (textb < limb && ismbblank (textb, limb - textb, &mblength_b)) | ||
2372 | + textb += mblength_b; | ||
2373 | + } | ||
2374 | + } | ||
2375 | + } | ||
2376 | + | ||
2377 | + return 0; | ||
2378 | + | ||
2379 | +greater: | ||
2380 | + diff = 1; | ||
2381 | +not_equal: | ||
2382 | + return key->reverse ? -diff : diff; | ||
2383 | +} | ||
2384 | +#endif | ||
2385 | + | ||
2386 | /* Compare two lines A and B, returning negative, zero, or positive | ||
2387 | depending on whether A compares less than, equal to, or greater than B. */ | ||
2388 | |||
2389 | @@ -2744,7 +3305,7 @@ | ||
2390 | initialize_exit_failure (SORT_FAILURE); | ||
2391 | |||
2392 | hard_LC_COLLATE = hard_locale (LC_COLLATE); | ||
2393 | -#if HAVE_NL_LANGINFO | ||
2394 | +#if HAVE_LANGINFO_CODESET | ||
2395 | hard_LC_TIME = hard_locale (LC_TIME); | ||
2396 | #endif | ||
2397 | |||
2398 | @@ -2765,6 +3326,27 @@ | ||
2399 | thousands_sep = -1; | ||
2400 | } | ||
2401 | |||
2402 | +#if HAVE_MBRTOWC | ||
2403 | + if (MB_CUR_MAX > 1) | ||
2404 | + { | ||
2405 | + inittables = inittables_mb; | ||
2406 | + begfield = begfield_mb; | ||
2407 | + limfield = limfield_mb; | ||
2408 | + getmonth = getmonth_mb; | ||
2409 | + keycompare = keycompare_mb; | ||
2410 | + numcompare = numcompare_mb; | ||
2411 | + } | ||
2412 | + else | ||
2413 | +#endif | ||
2414 | + { | ||
2415 | + inittables = inittables_uni; | ||
2416 | + begfield = begfield_uni; | ||
2417 | + limfield = limfield_uni; | ||
2418 | + getmonth = getmonth_uni; | ||
2419 | + keycompare = keycompare_uni; | ||
2420 | + numcompare = numcompare_uni; | ||
2421 | + } | ||
2422 | + | ||
2423 | have_read_stdin = false; | ||
2424 | inittables (); | ||
2425 | |||
2426 | @@ -3015,13 +3597,35 @@ | ||
2427 | |||
2428 | case 't': | ||
2429 | { | ||
2430 | - char newtab = optarg[0]; | ||
2431 | - if (! newtab) | ||
2432 | + char newtab[MB_LEN_MAX + 1]; | ||
2433 | + size_t newtab_length = 1; | ||
2434 | + strncpy (newtab, optarg, MB_LEN_MAX); | ||
2435 | + if (! newtab[0]) | ||
2436 | error (SORT_FAILURE, 0, _("empty tab")); | ||
2437 | - if (optarg[1]) | ||
2438 | +#if HAVE_MBRTOWC | ||
2439 | + if (MB_CUR_MAX > 1) | ||
2440 | + { | ||
2441 | + wchar_t wc; | ||
2442 | + mbstate_t state; | ||
2443 | + size_t i; | ||
2444 | + | ||
2445 | + memset (&state, '\0', sizeof (mbstate_t)); | ||
2446 | + newtab_length = mbrtowc (&wc, newtab, strnlen (newtab, | ||
2447 | + MB_LEN_MAX), | ||
2448 | + &state); | ||
2449 | + switch (newtab_length) | ||
2450 | + { | ||
2451 | + case (size_t) -1: | ||
2452 | + case (size_t) -2: | ||
2453 | + case 0: | ||
2454 | + newtab_length = 1; | ||
2455 | + } | ||
2456 | + } | ||
2457 | +#endif | ||
2458 | + if (newtab_length == 1 && optarg[1]) | ||
2459 | { | ||
2460 | if (STREQ (optarg, "\\0")) | ||
2461 | - newtab = '\0'; | ||
2462 | + newtab[0] = '\0'; | ||
2463 | else | ||
2464 | { | ||
2465 | /* Provoke with `sort -txx'. Complain about | ||
2466 | @@ -3032,9 +3636,12 @@ | ||
2467 | quote (optarg)); | ||
2468 | } | ||
2469 | } | ||
2470 | - if (tab != TAB_DEFAULT && tab != newtab) | ||
2471 | + if (tab_length | ||
2472 | + && (tab_length != newtab_length | ||
2473 | + || memcmp (tab, newtab, tab_length) != 0)) | ||
2474 | error (SORT_FAILURE, 0, _("incompatible tabs")); | ||
2475 | - tab = newtab; | ||
2476 | + memcpy (tab, newtab, newtab_length); | ||
2477 | + tab_length = newtab_length; | ||
2478 | } | ||
2479 | break; | ||
2480 | |||
2481 | --- coreutils-6.8+/src/unexpand.c.i18n 2007-01-14 15:41:28.000000000 +0000 | ||
2482 | +++ coreutils-6.8+/src/unexpand.c 2007-03-01 15:08:24.000000000 +0000 | ||
2483 | @@ -39,11 +39,28 @@ | ||
2484 | #include <stdio.h> | ||
2485 | #include <getopt.h> | ||
2486 | #include <sys/types.h> | ||
2487 | + | ||
2488 | +/* Get mbstate_t, mbrtowc(), wcwidth(). */ | ||
2489 | +#if HAVE_WCHAR_H | ||
2490 | +# include <wchar.h> | ||
2491 | +#endif | ||
2492 | + | ||
2493 | #include "system.h" | ||
2494 | #include "error.h" | ||
2495 | #include "quote.h" | ||
2496 | #include "xstrndup.h" | ||
2497 | |||
2498 | +/* MB_LEN_MAX is incorrectly defined to be 1 in at least one GCC | ||
2499 | + installation; work around this configuration error. */ | ||
2500 | +#if !defined MB_LEN_MAX || MB_LEN_MAX < 2 | ||
2501 | +# define MB_LEN_MAX 16 | ||
2502 | +#endif | ||
2503 | + | ||
2504 | +/* Some systems, like BeOS, have multibyte encodings but lack mbstate_t. */ | ||
2505 | +#if HAVE_MBRTOWC && defined mbstate_t | ||
2506 | +# define mbrtowc(pwc, s, n, ps) (mbrtowc) (pwc, s, n, 0) | ||
2507 | +#endif | ||
2508 | + | ||
2509 | /* The official name of this program (e.g., no `g' prefix). */ | ||
2510 | #define PROGRAM_NAME "unexpand" | ||
2511 | |||
2512 | @@ -110,6 +127,208 @@ | ||
2513 | {NULL, 0, NULL, 0} | ||
2514 | }; | ||
2515 | |||
2516 | +static FILE *next_file (FILE *fp); | ||
2517 | + | ||
2518 | +#if HAVE_MBRTOWC | ||
2519 | +static void | ||
2520 | +unexpand_multibyte (void) | ||
2521 | +{ | ||
2522 | + FILE *fp; /* Input stream. */ | ||
2523 | + mbstate_t i_state; /* Current shift state of the input stream. */ | ||
2524 | + mbstate_t i_state_bak; /* Back up the I_STATE. */ | ||
2525 | + mbstate_t o_state; /* Current shift state of the output stream. */ | ||
2526 | + char buf[MB_LEN_MAX + BUFSIZ]; /* For spooling a read byte sequence. */ | ||
2527 | + char *bufpos; /* Next read position of BUF. */ | ||
2528 | + size_t buflen = 0; /* The length of the byte sequence in buf. */ | ||
2529 | + wint_t wc; /* A gotten wide character. */ | ||
2530 | + size_t mblength; /* The byte size of a multibyte character | ||
2531 | + which shows as same character as WC. */ | ||
2532 | + | ||
2533 | + /* Index in `tab_list' of next tabstop: */ | ||
2534 | + int tab_index = 0; /* For calculating width of pending tabs. */ | ||
2535 | + int print_tab_index = 0; /* For printing as many tabs as possible. */ | ||
2536 | + unsigned int column = 0; /* Column on screen of next char. */ | ||
2537 | + int next_tab_column; /* Column the next tab stop is on. */ | ||
2538 | + int convert = 1; /* If nonzero, perform translations. */ | ||
2539 | + unsigned int pending = 0; /* Pending columns of blanks. */ | ||
2540 | + | ||
2541 | + fp = next_file ((FILE *) NULL); | ||
2542 | + if (fp == NULL) | ||
2543 | + return; | ||
2544 | + | ||
2545 | + memset (&o_state, '\0', sizeof(mbstate_t)); | ||
2546 | + memset (&i_state, '\0', sizeof(mbstate_t)); | ||
2547 | + | ||
2548 | + for (;;) | ||
2549 | + { | ||
2550 | + if (buflen < MB_LEN_MAX && !feof(fp) && !ferror(fp)) | ||
2551 | + { | ||
2552 | + memmove (buf, bufpos, buflen); | ||
2553 | + buflen += fread (buf + buflen, sizeof(char), BUFSIZ, fp); | ||
2554 | + bufpos = buf; | ||
2555 | + } | ||
2556 | + | ||
2557 | + /* Get a wide character. */ | ||
2558 | + if (buflen < 1) | ||
2559 | + { | ||
2560 | + mblength = 1; | ||
2561 | + wc = WEOF; | ||
2562 | + } | ||
2563 | + else | ||
2564 | + { | ||
2565 | + i_state_bak = i_state; | ||
2566 | + mblength = mbrtowc ((wchar_t *)&wc, bufpos, buflen, &i_state); | ||
2567 | + } | ||
2568 | + | ||
2569 | + if (mblength == (size_t)-1 || mblength == (size_t)-2) | ||
2570 | + { | ||
2571 | + i_state = i_state_bak; | ||
2572 | + wc = L'\0'; | ||
2573 | + } | ||
2574 | + | ||
2575 | + if (wc == L' ' && convert && column < INT_MAX) | ||
2576 | + { | ||
2577 | + ++pending; | ||
2578 | + ++column; | ||
2579 | + } | ||
2580 | + else if (wc == L'\t' && convert) | ||
2581 | + { | ||
2582 | + if (tab_size == 0) | ||
2583 | + { | ||
2584 | + /* Do not let tab_index == first_free_tab; | ||
2585 | + stop when it is 1 less. */ | ||
2586 | + while (tab_index < first_free_tab - 1 | ||
2587 | + && column >= tab_list[tab_index]) | ||
2588 | + tab_index++; | ||
2589 | + next_tab_column = tab_list[tab_index]; | ||
2590 | + if (tab_index < first_free_tab - 1) | ||
2591 | + tab_index++; | ||
2592 | + if (column >= next_tab_column) | ||
2593 | + { | ||
2594 | + convert = 0; /* Ran out of tab stops. */ | ||
2595 | + goto flush_pend_mb; | ||
2596 | + } | ||
2597 | + } | ||
2598 | + else | ||
2599 | + { | ||
2600 | + next_tab_column = column + tab_size - column % tab_size; | ||
2601 | + } | ||
2602 | + pending += next_tab_column - column; | ||
2603 | + column = next_tab_column; | ||
2604 | + } | ||
2605 | + else | ||
2606 | + { | ||
2607 | +flush_pend_mb: | ||
2608 | + /* Flush pending spaces. Print as many tabs as possible, | ||
2609 | + then print the rest as spaces. */ | ||
2610 | + if (pending == 1) | ||
2611 | + { | ||
2612 | + putchar (' '); | ||
2613 | + pending = 0; | ||
2614 | + } | ||
2615 | + column -= pending; | ||
2616 | + while (pending > 0) | ||
2617 | + { | ||
2618 | + if (tab_size == 0) | ||
2619 | + { | ||
2620 | + /* Do not let print_tab_index == first_free_tab; | ||
2621 | + stop when it is 1 less. */ | ||
2622 | + while (print_tab_index < first_free_tab - 1 | ||
2623 | + && column >= tab_list[print_tab_index]) | ||
2624 | + print_tab_index++; | ||
2625 | + next_tab_column = tab_list[print_tab_index]; | ||
2626 | + if (print_tab_index < first_free_tab - 1) | ||
2627 | + print_tab_index++; | ||
2628 | + } | ||
2629 | + else | ||
2630 | + { | ||
2631 | + next_tab_column = | ||
2632 | + column + tab_size - column % tab_size; | ||
2633 | + } | ||
2634 | + if (next_tab_column - column <= pending) | ||
2635 | + { | ||
2636 | + putchar ('\t'); | ||
2637 | + pending -= next_tab_column - column; | ||
2638 | + column = next_tab_column; | ||
2639 | + } | ||
2640 | + else | ||
2641 | + { | ||
2642 | + --print_tab_index; | ||
2643 | + column += pending; | ||
2644 | + while (pending != 0) | ||
2645 | + { | ||
2646 | + putchar (' '); | ||
2647 | + pending--; | ||
2648 | + } | ||
2649 | + } | ||
2650 | + } | ||
2651 | + | ||
2652 | + if (wc == WEOF) | ||
2653 | + { | ||
2654 | + fp = next_file (fp); | ||
2655 | + if (fp == NULL) | ||
2656 | + break; /* No more files. */ | ||
2657 | + else | ||
2658 | + { | ||
2659 | + memset (&i_state, '\0', sizeof(mbstate_t)); | ||
2660 | + continue; | ||
2661 | + } | ||
2662 | + } | ||
2663 | + | ||
2664 | + if (mblength == (size_t)-1 || mblength == (size_t)-2) | ||
2665 | + { | ||
2666 | + if (convert) | ||
2667 | + { | ||
2668 | + ++column; | ||
2669 | + if (convert_entire_line == 0) | ||
2670 | + convert = 0; | ||
2671 | + } | ||
2672 | + mblength = 1; | ||
2673 | + putchar (buf[0]); | ||
2674 | + } | ||
2675 | + else if (mblength == 0) | ||
2676 | + { | ||
2677 | + if (convert && convert_entire_line == 0) | ||
2678 | + convert = 0; | ||
2679 | + mblength = 1; | ||
2680 | + putchar ('\0'); | ||
2681 | + } | ||
2682 | + else | ||
2683 | + { | ||
2684 | + if (convert) | ||
2685 | + { | ||
2686 | + if (wc == L'\b') | ||
2687 | + { | ||
2688 | + if (column > 0) | ||
2689 | + --column; | ||
2690 | + } | ||
2691 | + else | ||
2692 | + { | ||
2693 | + int width; /* The width of WC. */ | ||
2694 | + | ||
2695 | + width = wcwidth (wc); | ||
2696 | + column += (width > 0) ? width : 0; | ||
2697 | + if (convert_entire_line == 0) | ||
2698 | + convert = 0; | ||
2699 | + } | ||
2700 | + } | ||
2701 | + | ||
2702 | + if (wc == L'\n') | ||
2703 | + { | ||
2704 | + tab_index = print_tab_index = 0; | ||
2705 | + column = pending = 0; | ||
2706 | + convert = 1; | ||
2707 | + } | ||
2708 | + fwrite (bufpos, sizeof(char), mblength, stdout); | ||
2709 | + } | ||
2710 | + } | ||
2711 | + buflen -= mblength; | ||
2712 | + bufpos += mblength; | ||
2713 | + } | ||
2714 | +} | ||
2715 | +#endif | ||
2716 | + | ||
2717 | + | ||
2718 | void | ||
2719 | usage (int status) | ||
2720 | { | ||
2721 | @@ -531,7 +750,12 @@ | ||
2722 | |||
2723 | file_list = (optind < argc ? &argv[optind] : stdin_argv); | ||
2724 | |||
2725 | - unexpand (); | ||
2726 | +#if HAVE_MBRTOWC | ||
2727 | + if (MB_CUR_MAX > 1) | ||
2728 | + unexpand_multibyte (); | ||
2729 | + else | ||
2730 | +#endif | ||
2731 | + unexpand (); | ||
2732 | |||
2733 | if (have_read_stdin && fclose (stdin) != 0) | ||
2734 | error (EXIT_FAILURE, errno, "-"); | ||
2735 | --- coreutils-6.8+/src/pr.c.i18n 2007-01-14 15:41:28.000000000 +0000 | ||
2736 | +++ coreutils-6.8+/src/pr.c 2007-03-01 15:08:24.000000000 +0000 | ||
2737 | @@ -313,6 +313,32 @@ | ||
2738 | |||
2739 | #include <getopt.h> | ||
2740 | #include <sys/types.h> | ||
2741 | + | ||
2742 | +/* Get MB_LEN_MAX. */ | ||
2743 | +#include <limits.h> | ||
2744 | +/* MB_LEN_MAX is incorrectly defined to be 1 in at least one GCC | ||
2745 | + installation; work around this configuration error. */ | ||
2746 | +#if !defined MB_LEN_MAX || MB_LEN_MAX == 1 | ||
2747 | +# define MB_LEN_MAX 16 | ||
2748 | +#endif | ||
2749 | + | ||
2750 | +/* Get MB_CUR_MAX. */ | ||
2751 | +#include <stdlib.h> | ||
2752 | + | ||
2753 | +/* Solaris 2.5 has a bug: <wchar.h> must be included before <wctype.h>. */ | ||
2754 | +/* Get mbstate_t, mbrtowc(), wcwidth(). */ | ||
2755 | +#if HAVE_WCHAR_H | ||
2756 | +# include <wchar.h> | ||
2757 | +#endif | ||
2758 | + | ||
2759 | +/* Get iswprint(). -- for wcwidth(). */ | ||
2760 | +#if HAVE_WCTYPE_H | ||
2761 | +# include <wctype.h> | ||
2762 | +#endif | ||
2763 | +#if !defined iswprint && !HAVE_ISWPRINT | ||
2764 | +# define iswprint(wc) 1 | ||
2765 | +#endif | ||
2766 | + | ||
2767 | #include "system.h" | ||
2768 | #include "error.h" | ||
2769 | #include "hard-locale.h" | ||
2770 | @@ -324,6 +350,18 @@ | ||
2771 | #include "strftime.h" | ||
2772 | #include "xstrtol.h" | ||
2773 | |||
2774 | +/* Some systems, like BeOS, have multibyte encodings but lack mbstate_t. */ | ||
2775 | +#if HAVE_MBRTOWC && defined mbstate_t | ||
2776 | +# define mbrtowc(pwc, s, n, ps) (mbrtowc) (pwc, s, n, 0) | ||
2777 | +#endif | ||
2778 | + | ||
2779 | +#ifndef HAVE_DECL_WCWIDTH | ||
2780 | +"this configure-time declaration test was not run" | ||
2781 | +#endif | ||
2782 | +#if !HAVE_DECL_WCWIDTH | ||
2783 | +extern int wcwidth (); | ||
2784 | +#endif | ||
2785 | + | ||
2786 | /* The official name of this program (e.g., no `g' prefix). */ | ||
2787 | #define PROGRAM_NAME "pr" | ||
2788 | |||
2789 | @@ -416,7 +454,20 @@ | ||
2790 | |||
2791 | #define NULLCOL (COLUMN *)0 | ||
2792 | |||
2793 | -static int char_to_clump (char c); | ||
2794 | +/* Funtion pointers to switch functions for single byte locale or for | ||
2795 | + multibyte locale. If multibyte functions do not exist in your sysytem, | ||
2796 | + these pointers always point the function for single byte locale. */ | ||
2797 | +static void (*print_char) (char c); | ||
2798 | +static int (*char_to_clump) (char c); | ||
2799 | + | ||
2800 | +/* Functions for single byte locale. */ | ||
2801 | +static void print_char_single (char c); | ||
2802 | +static int char_to_clump_single (char c); | ||
2803 | + | ||
2804 | +/* Functions for multibyte locale. */ | ||
2805 | +static void print_char_multi (char c); | ||
2806 | +static int char_to_clump_multi (char c); | ||
2807 | + | ||
2808 | static bool read_line (COLUMN *p); | ||
2809 | static bool print_page (void); | ||
2810 | static bool print_stored (COLUMN *p); | ||
2811 | @@ -426,6 +477,7 @@ | ||
2812 | static void pad_across_to (int position); | ||
2813 | static void add_line_number (COLUMN *p); | ||
2814 | static void getoptarg (char *arg, char switch_char, char *character, | ||
2815 | + int *character_length, int *character_width, | ||
2816 | int *number); | ||
2817 | void usage (int status); | ||
2818 | static void print_files (int number_of_files, char **av); | ||
2819 | @@ -440,7 +492,6 @@ | ||
2820 | static void pad_down (int lines); | ||
2821 | static void read_rest_of_line (COLUMN *p); | ||
2822 | static void skip_read (COLUMN *p, int column_number); | ||
2823 | -static void print_char (char c); | ||
2824 | static void cleanup (void); | ||
2825 | static void print_sep_string (void); | ||
2826 | static void separator_string (const char *optarg_S); | ||
2827 | @@ -455,7 +506,7 @@ | ||
2828 | we store the leftmost columns contiguously in buff. | ||
2829 | To print a line from buff, get the index of the first character | ||
2830 | from line_vector[i], and print up to line_vector[i + 1]. */ | ||
2831 | -static char *buff; | ||
2832 | +static unsigned char *buff; | ||
2833 | |||
2834 | /* Index of the position in buff where the next character | ||
2835 | will be stored. */ | ||
2836 | @@ -559,7 +610,7 @@ | ||
2837 | static bool untabify_input = false; | ||
2838 | |||
2839 | /* (-e) The input tab character. */ | ||
2840 | -static char input_tab_char = '\t'; | ||
2841 | +static char input_tab_char[MB_LEN_MAX] = "\t"; | ||
2842 | |||
2843 | /* (-e) Tabstops are at chars_per_tab, 2*chars_per_tab, 3*chars_per_tab, ... | ||
2844 | where the leftmost column is 1. */ | ||
2845 | @@ -569,7 +620,10 @@ | ||
2846 | static bool tabify_output = false; | ||
2847 | |||
2848 | /* (-i) The output tab character. */ | ||
2849 | -static char output_tab_char = '\t'; | ||
2850 | +static char output_tab_char[MB_LEN_MAX] = "\t"; | ||
2851 | + | ||
2852 | +/* (-i) The byte length of output tab character. */ | ||
2853 | +static int output_tab_char_length = 1; | ||
2854 | |||
2855 | /* (-i) The width of the output tab. */ | ||
2856 | static int chars_per_output_tab = 8; | ||
2857 | @@ -643,7 +697,13 @@ | ||
2858 | static bool numbered_lines = false; | ||
2859 | |||
2860 | /* (-n) Character which follows each line number. */ | ||
2861 | -static char number_separator = '\t'; | ||
2862 | +static char number_separator[MB_LEN_MAX] = "\t"; | ||
2863 | + | ||
2864 | +/* (-n) The byte length of the character which follows each line number. */ | ||
2865 | +static int number_separator_length = 1; | ||
2866 | + | ||
2867 | +/* (-n) The character width of the character which follows each line number. */ | ||
2868 | +static int number_separator_width = 0; | ||
2869 | |||
2870 | /* (-n) line counting starts with 1st line of input file (not with 1st | ||
2871 | line of 1st page printed). */ | ||
2872 | @@ -696,6 +756,7 @@ | ||
2873 | -a|COLUMN|-m is a `space' and with the -J option a `tab'. */ | ||
2874 | static char *col_sep_string = ""; | ||
2875 | static int col_sep_length = 0; | ||
2876 | +static int col_sep_width = 0; | ||
2877 | static char *column_separator = " "; | ||
2878 | static char *line_separator = "\t"; | ||
2879 | |||
2880 | @@ -852,6 +913,13 @@ | ||
2881 | col_sep_length = (int) strlen (optarg_S); | ||
2882 | col_sep_string = xmalloc (col_sep_length + 1); | ||
2883 | strcpy (col_sep_string, optarg_S); | ||
2884 | + | ||
2885 | +#if HAVE_MBRTOWC | ||
2886 | + if (MB_CUR_MAX > 1) | ||
2887 | + col_sep_width = mbswidth (col_sep_string, 0); | ||
2888 | + else | ||
2889 | +#endif | ||
2890 | + col_sep_width = col_sep_length; | ||
2891 | } | ||
2892 | |||
2893 | int | ||
2894 | @@ -877,6 +945,21 @@ | ||
2895 | |||
2896 | atexit (close_stdout); | ||
2897 | |||
2898 | +/* Define which functions are used, the ones for single byte locale or the ones | ||
2899 | + for multibyte locale. */ | ||
2900 | +#if HAVE_MBRTOWC | ||
2901 | + if (MB_CUR_MAX > 1) | ||
2902 | + { | ||
2903 | + print_char = print_char_multi; | ||
2904 | + char_to_clump = char_to_clump_multi; | ||
2905 | + } | ||
2906 | + else | ||
2907 | +#endif | ||
2908 | + { | ||
2909 | + print_char = print_char_single; | ||
2910 | + char_to_clump = char_to_clump_single; | ||
2911 | + } | ||
2912 | + | ||
2913 | n_files = 0; | ||
2914 | file_names = (argc > 1 | ||
2915 | ? xmalloc ((argc - 1) * sizeof (char *)) | ||
2916 | @@ -949,8 +1032,12 @@ | ||
2917 | break; | ||
2918 | case 'e': | ||
2919 | if (optarg) | ||
2920 | - getoptarg (optarg, 'e', &input_tab_char, | ||
2921 | - &chars_per_input_tab); | ||
2922 | + { | ||
2923 | + int dummy_length, dummy_width; | ||
2924 | + | ||
2925 | + getoptarg (optarg, 'e', input_tab_char, &dummy_length, | ||
2926 | + &dummy_width, &chars_per_input_tab); | ||
2927 | + } | ||
2928 | /* Could check tab width > 0. */ | ||
2929 | untabify_input = true; | ||
2930 | break; | ||
2931 | @@ -963,8 +1050,12 @@ | ||
2932 | break; | ||
2933 | case 'i': | ||
2934 | if (optarg) | ||
2935 | - getoptarg (optarg, 'i', &output_tab_char, | ||
2936 | - &chars_per_output_tab); | ||
2937 | + { | ||
2938 | + int dummy_width; | ||
2939 | + | ||
2940 | + getoptarg (optarg, 'i', output_tab_char, &output_tab_char_length, | ||
2941 | + &dummy_width, &chars_per_output_tab); | ||
2942 | + } | ||
2943 | /* Could check tab width > 0. */ | ||
2944 | tabify_output = true; | ||
2945 | break; | ||
2946 | @@ -991,8 +1082,8 @@ | ||
2947 | case 'n': | ||
2948 | numbered_lines = true; | ||
2949 | if (optarg) | ||
2950 | - getoptarg (optarg, 'n', &number_separator, | ||
2951 | - &chars_per_number); | ||
2952 | + getoptarg (optarg, 'n', number_separator, &number_separator_length, | ||
2953 | + &number_separator_width, &chars_per_number); | ||
2954 | break; | ||
2955 | case 'N': | ||
2956 | skip_count = false; | ||
2957 | @@ -1031,7 +1122,7 @@ | ||
2958 | old_s = false; | ||
2959 | /* Reset an additional input of -s, -S dominates -s */ | ||
2960 | col_sep_string = ""; | ||
2961 | - col_sep_length = 0; | ||
2962 | + col_sep_length = col_sep_width = 0; | ||
2963 | use_col_separator = true; | ||
2964 | if (optarg) | ||
2965 | separator_string (optarg); | ||
2966 | @@ -1188,10 +1279,45 @@ | ||
2967 | a number. */ | ||
2968 | |||
2969 | static void | ||
2970 | -getoptarg (char *arg, char switch_char, char *character, int *number) | ||
2971 | +getoptarg (char *arg, char switch_char, char *character, int *character_length, | ||
2972 | + int *character_width, int *number) | ||
2973 | { | ||
2974 | if (!ISDIGIT (*arg)) | ||
2975 | - *character = *arg++; | ||
2976 | + { | ||
2977 | +#ifdef HAVE_MBRTOWC | ||
2978 | + if (MB_CUR_MAX > 1) /* for multibyte locale. */ | ||
2979 | + { | ||
2980 | + wchar_t wc; | ||
2981 | + size_t mblength; | ||
2982 | + int width; | ||
2983 | + mbstate_t state = {'\0'}; | ||
2984 | + | ||
2985 | + mblength = mbrtowc (&wc, arg, strnlen(arg, MB_LEN_MAX), &state); | ||
2986 | + | ||
2987 | + if (mblength == (size_t)-1 || mblength == (size_t)-2) | ||
2988 | + { | ||
2989 | + *character_length = 1; | ||
2990 | + *character_width = 1; | ||
2991 | + } | ||
2992 | + else | ||
2993 | + { | ||
2994 | + *character_length = (mblength < 1) ? 1 : mblength; | ||
2995 | + width = wcwidth (wc); | ||
2996 | + *character_width = (width < 0) ? 0 : width; | ||
2997 | + } | ||
2998 | + | ||
2999 | + strncpy (character, arg, *character_length); | ||
3000 | + arg += *character_length; | ||
3001 | + } | ||
3002 | + else /* for single byte locale. */ | ||
3003 | +#endif | ||
3004 | + { | ||
3005 | + *character = *arg++; | ||
3006 | + *character_length = 1; | ||
3007 | + *character_width = 1; | ||
3008 | + } | ||
3009 | + } | ||
3010 | + | ||
3011 | if (*arg) | ||
3012 | { | ||
3013 | long int tmp_long; | ||
3014 | @@ -1256,7 +1382,7 @@ | ||
3015 | else | ||
3016 | col_sep_string = column_separator; | ||
3017 | |||
3018 | - col_sep_length = 1; | ||
3019 | + col_sep_length = col_sep_width = 1; | ||
3020 | use_col_separator = true; | ||
3021 | } | ||
3022 | /* It's rather pointless to define a TAB separator with column | ||
3023 | @@ -1288,11 +1414,11 @@ | ||
3024 | TAB_WIDTH (chars_per_input_tab, chars_per_number); */ | ||
3025 | |||
3026 | /* Estimate chars_per_text without any margin and keep it constant. */ | ||
3027 | - if (number_separator == '\t') | ||
3028 | + if (number_separator[0] == '\t') | ||
3029 | number_width = chars_per_number + | ||
3030 | TAB_WIDTH (chars_per_default_tab, chars_per_number); | ||
3031 | else | ||
3032 | - number_width = chars_per_number + 1; | ||
3033 | + number_width = chars_per_number + number_separator_width; | ||
3034 | |||
3035 | /* The number is part of the column width unless we are | ||
3036 | printing files in parallel. */ | ||
3037 | @@ -1307,7 +1433,7 @@ | ||
3038 | } | ||
3039 | |||
3040 | chars_per_column = (chars_per_line - chars_used_by_number - | ||
3041 | - (columns - 1) * col_sep_length) / columns; | ||
3042 | + (columns - 1) * col_sep_width) / columns; | ||
3043 | |||
3044 | if (chars_per_column < 1) | ||
3045 | error (EXIT_FAILURE, 0, _("page width too narrow")); | ||
3046 | @@ -1432,7 +1558,7 @@ | ||
3047 | |||
3048 | /* Enlarge p->start_position of first column to use the same form of | ||
3049 | padding_not_printed with all columns. */ | ||
3050 | - h = h + col_sep_length; | ||
3051 | + h = h + col_sep_width; | ||
3052 | |||
3053 | /* This loop takes care of all but the rightmost column. */ | ||
3054 | |||
3055 | @@ -1466,7 +1592,7 @@ | ||
3056 | } | ||
3057 | else | ||
3058 | { | ||
3059 | - h = h_next + col_sep_length; | ||
3060 | + h = h_next + col_sep_width; | ||
3061 | h_next = h + chars_per_column; | ||
3062 | } | ||
3063 | } | ||
3064 | @@ -1756,9 +1882,9 @@ | ||
3065 | align_column (COLUMN *p) | ||
3066 | { | ||
3067 | padding_not_printed = p->start_position; | ||
3068 | - if (padding_not_printed - col_sep_length > 0) | ||
3069 | + if (padding_not_printed - col_sep_width > 0) | ||
3070 | { | ||
3071 | - pad_across_to (padding_not_printed - col_sep_length); | ||
3072 | + pad_across_to (padding_not_printed - col_sep_width); | ||
3073 | padding_not_printed = ANYWHERE; | ||
3074 | } | ||
3075 | |||
3076 | @@ -2029,13 +2155,13 @@ | ||
3077 | /* May be too generous. */ | ||
3078 | buff = X2REALLOC (buff, &buff_allocated); | ||
3079 | } | ||
3080 | - buff[buff_current++] = c; | ||
3081 | + buff[buff_current++] = (unsigned char) c; | ||
3082 | } | ||
3083 | |||
3084 | static void | ||
3085 | add_line_number (COLUMN *p) | ||
3086 | { | ||
3087 | - int i; | ||
3088 | + int i, j; | ||
3089 | char *s; | ||
3090 | int left_cut; | ||
3091 | |||
3092 | @@ -2058,22 +2184,24 @@ | ||
3093 | /* Tabification is assumed for multiple columns, also for n-separators, | ||
3094 | but `default n-separator = TAB' hasn't been given priority over | ||
3095 | equal column_width also specified by POSIX. */ | ||
3096 | - if (number_separator == '\t') | ||
3097 | + if (number_separator[0] == '\t') | ||
3098 | { | ||
3099 | i = number_width - chars_per_number; | ||
3100 | while (i-- > 0) | ||
3101 | (p->char_func) (' '); | ||
3102 | } | ||
3103 | else | ||
3104 | - (p->char_func) (number_separator); | ||
3105 | + for (j = 0; j < number_separator_length; j++) | ||
3106 | + (p->char_func) (number_separator[j]); | ||
3107 | } | ||
3108 | else | ||
3109 | /* To comply with POSIX, we avoid any expansion of default TAB | ||
3110 | separator with a single column output. No column_width requirement | ||
3111 | has to be considered. */ | ||
3112 | { | ||
3113 | - (p->char_func) (number_separator); | ||
3114 | - if (number_separator == '\t') | ||
3115 | + for (j = 0; j < number_separator_length; j++) | ||
3116 | + (p->char_func) (number_separator[j]); | ||
3117 | + if (number_separator[0] == '\t') | ||
3118 | output_position = POS_AFTER_TAB (chars_per_output_tab, | ||
3119 | output_position); | ||
3120 | } | ||
3121 | @@ -2234,7 +2362,7 @@ | ||
3122 | while (goal - h_old > 1 | ||
3123 | && (h_new = POS_AFTER_TAB (chars_per_output_tab, h_old)) <= goal) | ||
3124 | { | ||
3125 | - putchar (output_tab_char); | ||
3126 | + fwrite (output_tab_char, sizeof(char), output_tab_char_length, stdout); | ||
3127 | h_old = h_new; | ||
3128 | } | ||
3129 | while (++h_old <= goal) | ||
3130 | @@ -2254,6 +2382,7 @@ | ||
3131 | { | ||
3132 | char *s; | ||
3133 | int l = col_sep_length; | ||
3134 | + int not_space_flag; | ||
3135 | |||
3136 | s = col_sep_string; | ||
3137 | |||
3138 | @@ -2267,6 +2396,7 @@ | ||
3139 | { | ||
3140 | for (; separators_not_printed > 0; --separators_not_printed) | ||
3141 | { | ||
3142 | + not_space_flag = 0; | ||
3143 | while (l-- > 0) | ||
3144 | { | ||
3145 | /* 3 types of sep_strings: spaces only, spaces and chars, | ||
3146 | @@ -2280,12 +2410,15 @@ | ||
3147 | } | ||
3148 | else | ||
3149 | { | ||
3150 | + not_space_flag = 1; | ||
3151 | if (spaces_not_printed > 0) | ||
3152 | print_white_space (); | ||
3153 | putchar (*s++); | ||
3154 | - ++output_position; | ||
3155 | } | ||
3156 | } | ||
3157 | + if (not_space_flag) | ||
3158 | + output_position += col_sep_width; | ||
3159 | + | ||
3160 | /* sep_string ends with some spaces */ | ||
3161 | if (spaces_not_printed > 0) | ||
3162 | print_white_space (); | ||
3163 | @@ -2313,7 +2446,7 @@ | ||
3164 | required number of tabs and spaces. */ | ||
3165 | |||
3166 | static void | ||
3167 | -print_char (char c) | ||
3168 | +print_char_single (char c) | ||
3169 | { | ||
3170 | if (tabify_output) | ||
3171 | { | ||
3172 | @@ -2337,6 +2470,74 @@ | ||
3173 | putchar (c); | ||
3174 | } | ||
3175 | |||
3176 | +#ifdef HAVE_MBRTOWC | ||
3177 | +static void | ||
3178 | +print_char_multi (char c) | ||
3179 | +{ | ||
3180 | + static size_t mbc_pos = 0; | ||
3181 | + static char mbc[MB_LEN_MAX] = {'\0'}; | ||
3182 | + static mbstate_t state = {'\0'}; | ||
3183 | + mbstate_t state_bak; | ||
3184 | + wchar_t wc; | ||
3185 | + size_t mblength; | ||
3186 | + int width; | ||
3187 | + | ||
3188 | + if (tabify_output) | ||
3189 | + { | ||
3190 | + state_bak = state; | ||
3191 | + mbc[mbc_pos++] = c; | ||
3192 | + mblength = mbrtowc (&wc, mbc, mbc_pos, &state); | ||
3193 | + | ||
3194 | + while (mbc_pos > 0) | ||
3195 | + { | ||
3196 | + switch (mblength) | ||
3197 | + { | ||
3198 | + case (size_t)-2: | ||
3199 | + state = state_bak; | ||
3200 | + return; | ||
3201 | + | ||
3202 | + case (size_t)-1: | ||
3203 | + state = state_bak; | ||
3204 | + ++output_position; | ||
3205 | + putchar (mbc[0]); | ||
3206 | + memmove (mbc, mbc + 1, MB_CUR_MAX - 1); | ||
3207 | + --mbc_pos; | ||
3208 | + break; | ||
3209 | + | ||
3210 | + case 0: | ||
3211 | + mblength = 1; | ||
3212 | + | ||
3213 | + default: | ||
3214 | + if (wc == L' ') | ||
3215 | + { | ||
3216 | + memmove (mbc, mbc + mblength, MB_CUR_MAX - mblength); | ||
3217 | + --mbc_pos; | ||
3218 | + ++spaces_not_printed; | ||
3219 | + return; | ||
3220 | + } | ||
3221 | + else if (spaces_not_printed > 0) | ||
3222 | + print_white_space (); | ||
3223 | + | ||
3224 | + /* Nonprintables are assumed to have width 0, except L'\b'. */ | ||
3225 | + if ((width = wcwidth (wc)) < 1) | ||
3226 | + { | ||
3227 | + if (wc == L'\b') | ||
3228 | + --output_position; | ||
3229 | + } | ||
3230 | + else | ||
3231 | + output_position += width; | ||
3232 | + | ||
3233 | + fwrite (mbc, sizeof(char), mblength, stdout); | ||
3234 | + memmove (mbc, mbc + mblength, MB_CUR_MAX - mblength); | ||
3235 | + mbc_pos -= mblength; | ||
3236 | + } | ||
3237 | + } | ||
3238 | + return; | ||
3239 | + } | ||
3240 | + putchar (c); | ||
3241 | +} | ||
3242 | +#endif | ||
3243 | + | ||
3244 | /* Skip to page PAGE before printing. | ||
3245 | PAGE may be larger than total number of pages. */ | ||
3246 | |||
3247 | @@ -2517,9 +2718,9 @@ | ||
3248 | align_empty_cols = false; | ||
3249 | } | ||
3250 | |||
3251 | - if (padding_not_printed - col_sep_length > 0) | ||
3252 | + if (padding_not_printed - col_sep_width > 0) | ||
3253 | { | ||
3254 | - pad_across_to (padding_not_printed - col_sep_length); | ||
3255 | + pad_across_to (padding_not_printed - col_sep_width); | ||
3256 | padding_not_printed = ANYWHERE; | ||
3257 | } | ||
3258 | |||
3259 | @@ -2620,9 +2821,9 @@ | ||
3260 | } | ||
3261 | } | ||
3262 | |||
3263 | - if (padding_not_printed - col_sep_length > 0) | ||
3264 | + if (padding_not_printed - col_sep_width > 0) | ||
3265 | { | ||
3266 | - pad_across_to (padding_not_printed - col_sep_length); | ||
3267 | + pad_across_to (padding_not_printed - col_sep_width); | ||
3268 | padding_not_printed = ANYWHERE; | ||
3269 | } | ||
3270 | |||
3271 | @@ -2635,8 +2836,8 @@ | ||
3272 | if (spaces_not_printed == 0) | ||
3273 | { | ||
3274 | output_position = p->start_position + end_vector[line]; | ||
3275 | - if (p->start_position - col_sep_length == chars_per_margin) | ||
3276 | - output_position -= col_sep_length; | ||
3277 | + if (p->start_position - col_sep_width == chars_per_margin) | ||
3278 | + output_position -= col_sep_width; | ||
3279 | } | ||
3280 | |||
3281 | return true; | ||
3282 | @@ -2655,7 +2856,7 @@ | ||
3283 | number of characters is 1.) */ | ||
3284 | |||
3285 | static int | ||
3286 | -char_to_clump (char c) | ||
3287 | +char_to_clump_single (char c) | ||
3288 | { | ||
3289 | unsigned char uc = c; | ||
3290 | char *s = clump_buff; | ||
3291 | @@ -2665,10 +2866,10 @@ | ||
3292 | int chars; | ||
3293 | int chars_per_c = 8; | ||
3294 | |||
3295 | - if (c == input_tab_char) | ||
3296 | + if (c == input_tab_char[0]) | ||
3297 | chars_per_c = chars_per_input_tab; | ||
3298 | |||
3299 | - if (c == input_tab_char || c == '\t') | ||
3300 | + if (c == input_tab_char[0] || c == '\t') | ||
3301 | { | ||
3302 | width = TAB_WIDTH (chars_per_c, input_position); | ||
3303 | |||
3304 | @@ -2739,6 +2940,154 @@ | ||
3305 | return chars; | ||
3306 | } | ||
3307 | |||
3308 | +#ifdef HAVE_MBRTOWC | ||
3309 | +static int | ||
3310 | +char_to_clump_multi (char c) | ||
3311 | +{ | ||
3312 | + static size_t mbc_pos = 0; | ||
3313 | + static char mbc[MB_LEN_MAX] = {'\0'}; | ||
3314 | + static mbstate_t state = {'\0'}; | ||
3315 | + mbstate_t state_bak; | ||
3316 | + wchar_t wc; | ||
3317 | + size_t mblength; | ||
3318 | + int wc_width; | ||
3319 | + register char *s = clump_buff; | ||
3320 | + register int i, j; | ||
3321 | + char esc_buff[4]; | ||
3322 | + int width; | ||
3323 | + int chars; | ||
3324 | + int chars_per_c = 8; | ||
3325 | + | ||
3326 | + state_bak = state; | ||
3327 | + mbc[mbc_pos++] = c; | ||
3328 | + mblength = mbrtowc (&wc, mbc, mbc_pos, &state); | ||
3329 | + | ||
3330 | + width = 0; | ||
3331 | + chars = 0; | ||
3332 | + while (mbc_pos > 0) | ||
3333 | + { | ||
3334 | + switch (mblength) | ||
3335 | + { | ||
3336 | + case (size_t)-2: | ||
3337 | + state = state_bak; | ||
3338 | + return 0; | ||
3339 | + | ||
3340 | + case (size_t)-1: | ||
3341 | + state = state_bak; | ||
3342 | + mblength = 1; | ||
3343 | + | ||
3344 | + if (use_esc_sequence || use_cntrl_prefix) | ||
3345 | + { | ||
3346 | + width = +4; | ||
3347 | + chars = +4; | ||
3348 | + *s++ = '\\'; | ||
3349 | + sprintf (esc_buff, "%03o", mbc[0]); | ||
3350 | + for (i = 0; i <= 2; ++i) | ||
3351 | + *s++ = (int) esc_buff[i]; | ||
3352 | + } | ||
3353 | + else | ||
3354 | + { | ||
3355 | + width += 1; | ||
3356 | + chars += 1; | ||
3357 | + *s++ = mbc[0]; | ||
3358 | + } | ||
3359 | + break; | ||
3360 | + | ||
3361 | + case 0: | ||
3362 | + mblength = 1; | ||
3363 | + /* Fall through */ | ||
3364 | + | ||
3365 | + default: | ||
3366 | + if (memcmp (mbc, input_tab_char, mblength) == 0) | ||
3367 | + chars_per_c = chars_per_input_tab; | ||
3368 | + | ||
3369 | + if (memcmp (mbc, input_tab_char, mblength) == 0 || c == '\t') | ||
3370 | + { | ||
3371 | + int width_inc; | ||
3372 | + | ||
3373 | + width_inc = TAB_WIDTH (chars_per_c, input_position); | ||
3374 | + width += width_inc; | ||
3375 | + | ||
3376 | + if (untabify_input) | ||
3377 | + { | ||
3378 | + for (i = width_inc; i; --i) | ||
3379 | + *s++ = ' '; | ||
3380 | + chars += width_inc; | ||
3381 | + } | ||
3382 | + else | ||
3383 | + { | ||
3384 | + for (i = 0; i < mblength; i++) | ||
3385 | + *s++ = mbc[i]; | ||
3386 | + chars += mblength; | ||
3387 | + } | ||
3388 | + } | ||
3389 | + else if ((wc_width = wcwidth (wc)) < 1) | ||
3390 | + { | ||
3391 | + if (use_esc_sequence) | ||
3392 | + { | ||
3393 | + for (i = 0; i < mblength; i++) | ||
3394 | + { | ||
3395 | + width += 4; | ||
3396 | + chars += 4; | ||
3397 | + *s++ = '\\'; | ||
3398 | + sprintf (esc_buff, "%03o", c); | ||
3399 | + for (j = 0; j <= 2; ++j) | ||
3400 | + *s++ = (int) esc_buff[j]; | ||
3401 | + } | ||
3402 | + } | ||
3403 | + else if (use_cntrl_prefix) | ||
3404 | + { | ||
3405 | + if (wc < 0200) | ||
3406 | + { | ||
3407 | + width += 2; | ||
3408 | + chars += 2; | ||
3409 | + *s++ = '^'; | ||
3410 | + *s++ = wc ^ 0100; | ||
3411 | + } | ||
3412 | + else | ||
3413 | + { | ||
3414 | + for (i = 0; i < mblength; i++) | ||
3415 | + { | ||
3416 | + width += 4; | ||
3417 | + chars += 4; | ||
3418 | + *s++ = '\\'; | ||
3419 | + sprintf (esc_buff, "%03o", c); | ||
3420 | + for (j = 0; j <= 2; ++j) | ||
3421 | + *s++ = (int) esc_buff[j]; | ||
3422 | + } | ||
3423 | + } | ||
3424 | + } | ||
3425 | + else if (wc == L'\b') | ||
3426 | + { | ||
3427 | + width += -1; | ||
3428 | + chars += 1; | ||
3429 | + *s++ = c; | ||
3430 | + } | ||
3431 | + else | ||
3432 | + { | ||
3433 | + width += 0; | ||
3434 | + chars += mblength; | ||
3435 | + for (i = 0; i < mblength; i++) | ||
3436 | + *s++ = mbc[i]; | ||
3437 | + } | ||
3438 | + } | ||
3439 | + else | ||
3440 | + { | ||
3441 | + width += wc_width; | ||
3442 | + chars += mblength; | ||
3443 | + for (i = 0; i < mblength; i++) | ||
3444 | + *s++ = mbc[i]; | ||
3445 | + } | ||
3446 | + } | ||
3447 | + memmove (mbc, mbc + mblength, MB_CUR_MAX - mblength); | ||
3448 | + mbc_pos -= mblength; | ||
3449 | + } | ||
3450 | + | ||
3451 | + input_position += width; | ||
3452 | + return chars; | ||
3453 | +} | ||
3454 | +#endif | ||
3455 | + | ||
3456 | /* We've just printed some files and need to clean up things before | ||
3457 | looking for more options and printing the next batch of files. | ||
3458 | |||
3459 | --- coreutils-6.8+/src/cut.c.i18n 2007-01-14 15:41:28.000000000 +0000 | ||
3460 | +++ coreutils-6.8+/src/cut.c 2007-03-01 15:08:24.000000000 +0000 | ||
3461 | @@ -29,6 +29,11 @@ | ||
3462 | #include <assert.h> | ||
3463 | #include <getopt.h> | ||
3464 | #include <sys/types.h> | ||
3465 | + | ||
3466 | +/* Get mbstate_t, mbrtowc(). */ | ||
3467 | +#if HAVE_WCHAR_H | ||
3468 | +# include <wchar.h> | ||
3469 | +#endif | ||
3470 | #include "system.h" | ||
3471 | |||
3472 | #include "error.h" | ||
3473 | @@ -37,6 +42,18 @@ | ||
3474 | #include "quote.h" | ||
3475 | #include "xstrndup.h" | ||
3476 | |||
3477 | +/* MB_LEN_MAX is incorrectly defined to be 1 in at least one GCC | ||
3478 | + installation; work around this configuration error. */ | ||
3479 | +#if !defined MB_LEN_MAX || MB_LEN_MAX < 2 | ||
3480 | +# undef MB_LEN_MAX | ||
3481 | +# define MB_LEN_MAX 16 | ||
3482 | +#endif | ||
3483 | + | ||
3484 | +/* Some systems, like BeOS, have multibyte encodings but lack mbstate_t. */ | ||
3485 | +#if HAVE_MBRTOWC && defined mbstate_t | ||
3486 | +# define mbrtowc(pwc, s, n, ps) (mbrtowc) (pwc, s, n, 0) | ||
3487 | +#endif | ||
3488 | + | ||
3489 | /* The official name of this program (e.g., no `g' prefix). */ | ||
3490 | #define PROGRAM_NAME "cut" | ||
3491 | |||
3492 | @@ -67,6 +84,52 @@ | ||
3493 | } \ | ||
3494 | while (0) | ||
3495 | |||
3496 | +/* Refill the buffer BUF to get a multibyte character. */ | ||
3497 | +#define REFILL_BUFFER(BUF, BUFPOS, BUFLEN, STREAM) \ | ||
3498 | + do \ | ||
3499 | + { \ | ||
3500 | + if (BUFLEN < MB_LEN_MAX && !feof (STREAM) && !ferror (STREAM)) \ | ||
3501 | + { \ | ||
3502 | + memmove (BUF, BUFPOS, BUFLEN); \ | ||
3503 | + BUFLEN += fread (BUF + BUFLEN, sizeof(char), BUFSIZ, STREAM); \ | ||
3504 | + BUFPOS = BUF; \ | ||
3505 | + } \ | ||
3506 | + } \ | ||
3507 | + while (0) | ||
3508 | + | ||
3509 | +/* Get wide character on BUFPOS. BUFPOS is not included after that. | ||
3510 | + If byte sequence is not valid as a character, CONVFAIL is 1. Otherwise 0. */ | ||
3511 | +#define GET_NEXT_WC_FROM_BUFFER(WC, BUFPOS, BUFLEN, MBLENGTH, STATE, CONVFAIL) \ | ||
3512 | + do \ | ||
3513 | + { \ | ||
3514 | + mbstate_t state_bak; \ | ||
3515 | + \ | ||
3516 | + if (BUFLEN < 1) \ | ||
3517 | + { \ | ||
3518 | + WC = WEOF; \ | ||
3519 | + break; \ | ||
3520 | + } \ | ||
3521 | + \ | ||
3522 | + /* Get a wide character. */ \ | ||
3523 | + CONVFAIL = 0; \ | ||
3524 | + state_bak = STATE; \ | ||
3525 | + MBLENGTH = mbrtowc ((wchar_t *)&WC, BUFPOS, BUFLEN, &STATE); \ | ||
3526 | + \ | ||
3527 | + switch (MBLENGTH) \ | ||
3528 | + { \ | ||
3529 | + case (size_t)-1: \ | ||
3530 | + case (size_t)-2: \ | ||
3531 | + CONVFAIL++; \ | ||
3532 | + STATE = state_bak; \ | ||
3533 | + /* Fall througn. */ \ | ||
3534 | + \ | ||
3535 | + case 0: \ | ||
3536 | + MBLENGTH = 1; \ | ||
3537 | + break; \ | ||
3538 | + } \ | ||
3539 | + } \ | ||
3540 | + while (0) | ||
3541 | + | ||
3542 | struct range_pair | ||
3543 | { | ||
3544 | size_t lo; | ||
3545 | @@ -85,7 +148,7 @@ | ||
3546 | /* The number of bytes allocated for FIELD_1_BUFFER. */ | ||
3547 | static size_t field_1_bufsize; | ||
3548 | |||
3549 | -/* The largest field or byte index used as an endpoint of a closed | ||
3550 | +/* The largest byte, character or field index used as an endpoint of a closed | ||
3551 | or degenerate range specification; this doesn't include the starting | ||
3552 | index of right-open-ended ranges. For example, with either range spec | ||
3553 | `2-5,9-', `2-3,5,9-' this variable would be set to 5. */ | ||
3554 | @@ -97,10 +160,11 @@ | ||
3555 | |||
3556 | /* This is a bit vector. | ||
3557 | In byte mode, which bytes to output. | ||
3558 | + In character mode, which characters to output. | ||
3559 | In field mode, which DELIM-separated fields to output. | ||
3560 | - Both bytes and fields are numbered starting with 1, | ||
3561 | + Bytes, characters and fields are numbered starting with 1, | ||
3562 | so the zeroth bit of this array is unused. | ||
3563 | - A field or byte K has been selected if | ||
3564 | + A byte, character or field K has been selected if | ||
3565 | (K <= MAX_RANGE_ENDPOINT and is_printable_field(K)) | ||
3566 | || (EOL_RANGE_START > 0 && K >= EOL_RANGE_START). */ | ||
3567 | static unsigned char *printable_field; | ||
3568 | @@ -109,9 +173,12 @@ | ||
3569 | { | ||
3570 | undefined_mode, | ||
3571 | |||
3572 | - /* Output characters that are in the given bytes. */ | ||
3573 | + /* Output bytes that are at the given positions. */ | ||
3574 | byte_mode, | ||
3575 | |||
3576 | + /* Output characters that are at the given positions. */ | ||
3577 | + character_mode, | ||
3578 | + | ||
3579 | /* Output the given delimeter-separated fields. */ | ||
3580 | field_mode | ||
3581 | }; | ||
3582 | @@ -121,6 +188,13 @@ | ||
3583 | |||
3584 | static enum operating_mode operating_mode; | ||
3585 | |||
3586 | +/* If nonzero, when in byte mode, don't split multibyte characters. */ | ||
3587 | +static int byte_mode_character_aware; | ||
3588 | + | ||
3589 | +/* If nonzero, the function for single byte locale is work | ||
3590 | + if this program runs on multibyte locale. */ | ||
3591 | +static int force_singlebyte_mode; | ||
3592 | + | ||
3593 | /* If true do not output lines containing no delimeter characters. | ||
3594 | Otherwise, all such lines are printed. This option is valid only | ||
3595 | with field mode. */ | ||
3596 | @@ -132,6 +206,9 @@ | ||
3597 | |||
3598 | /* The delimeter character for field mode. */ | ||
3599 | static unsigned char delim; | ||
3600 | +#if HAVE_WCHAR_H | ||
3601 | +static wchar_t wcdelim; | ||
3602 | +#endif | ||
3603 | |||
3604 | /* True if the --output-delimiter=STRING option was specified. */ | ||
3605 | static bool output_delimiter_specified; | ||
3606 | @@ -205,7 +282,7 @@ | ||
3607 | -f, --fields=LIST select only these fields; also print any line\n\ | ||
3608 | that contains no delimiter character, unless\n\ | ||
3609 | the -s option is specified\n\ | ||
3610 | - -n (ignored)\n\ | ||
3611 | + -n with -b: don't split multibyte characters\n\ | ||
3612 | "), stdout); | ||
3613 | fputs (_("\ | ||
3614 | --complement complement the set of selected bytes, characters\n\ | ||
3615 | @@ -362,7 +439,7 @@ | ||
3616 | in_digits = false; | ||
3617 | /* Starting a range. */ | ||
3618 | if (dash_found) | ||
3619 | - FATAL_ERROR (_("invalid byte or field list")); | ||
3620 | + FATAL_ERROR (_("invalid byte, character or field list")); | ||
3621 | dash_found = true; | ||
3622 | fieldstr++; | ||
3623 | |||
3624 | @@ -387,14 +464,16 @@ | ||
3625 | if (value == 0) | ||
3626 | { | ||
3627 | /* `n-'. From `initial' to end of line. */ | ||
3628 | - eol_range_start = initial; | ||
3629 | + if (eol_range_start == 0 || | ||
3630 | + (eol_range_start != 0 && eol_range_start > initial)) | ||
3631 | + eol_range_start = initial; | ||
3632 | field_found = true; | ||
3633 | } | ||
3634 | else | ||
3635 | { | ||
3636 | /* `m-n' or `-n' (1-n). */ | ||
3637 | if (value < initial) | ||
3638 | - FATAL_ERROR (_("invalid byte or field list")); | ||
3639 | + FATAL_ERROR (_("invalid byte, character or field list")); | ||
3640 | |||
3641 | /* Is there already a range going to end of line? */ | ||
3642 | if (eol_range_start != 0) | ||
3643 | @@ -467,6 +546,9 @@ | ||
3644 | if (operating_mode == byte_mode) | ||
3645 | error (0, 0, | ||
3646 | _("byte offset %s is too large"), quote (bad_num)); | ||
3647 | + else if (operating_mode == character_mode) | ||
3648 | + error (0, 0, | ||
3649 | + _("character offset %s is too large"), quote (bad_num)); | ||
3650 | else | ||
3651 | error (0, 0, | ||
3652 | _("field number %s is too large"), quote (bad_num)); | ||
3653 | @@ -477,7 +559,7 @@ | ||
3654 | fieldstr++; | ||
3655 | } | ||
3656 | else | ||
3657 | - FATAL_ERROR (_("invalid byte or field list")); | ||
3658 | + FATAL_ERROR (_("invalid byte, character or field list")); | ||
3659 | } | ||
3660 | |||
3661 | max_range_endpoint = 0; | ||
3662 | @@ -570,6 +652,63 @@ | ||
3663 | } | ||
3664 | } | ||
3665 | |||
3666 | +#if HAVE_MBRTOWC | ||
3667 | +/* This function is in use for the following case. | ||
3668 | + | ||
3669 | + 1. Read from the stream STREAM, printing to standard output any selected | ||
3670 | + characters. | ||
3671 | + | ||
3672 | + 2. Read from stream STREAM, printing to standard output any selected bytes, | ||
3673 | + without splitting multibyte characters. */ | ||
3674 | + | ||
3675 | +static void | ||
3676 | +cut_characters_or_cut_bytes_no_split (FILE *stream) | ||
3677 | +{ | ||
3678 | + int idx; /* number of bytes or characters in the line so far. */ | ||
3679 | + char buf[MB_LEN_MAX + BUFSIZ]; /* For spooling a read byte sequence. */ | ||
3680 | + char *bufpos; /* Next read position of BUF. */ | ||
3681 | + size_t buflen; /* The length of the byte sequence in buf. */ | ||
3682 | + wint_t wc; /* A gotten wide character. */ | ||
3683 | + size_t mblength; /* The byte size of a multibyte character which shows | ||
3684 | + as same character as WC. */ | ||
3685 | + mbstate_t state; /* State of the stream. */ | ||
3686 | + int convfail; /* 1, when conversion is failed. Otherwise 0. */ | ||
3687 | + | ||
3688 | + idx = 0; | ||
3689 | + buflen = 0; | ||
3690 | + bufpos = buf; | ||
3691 | + memset (&state, '\0', sizeof(mbstate_t)); | ||
3692 | + | ||
3693 | + while (1) | ||
3694 | + { | ||
3695 | + REFILL_BUFFER (buf, bufpos, buflen, stream); | ||
3696 | + | ||
3697 | + GET_NEXT_WC_FROM_BUFFER (wc, bufpos, buflen, mblength, state, convfail); | ||
3698 | + | ||
3699 | + if (wc == WEOF) | ||
3700 | + { | ||
3701 | + if (idx > 0) | ||
3702 | + putchar ('\n'); | ||
3703 | + break; | ||
3704 | + } | ||
3705 | + else if (wc == L'\n') | ||
3706 | + { | ||
3707 | + putchar ('\n'); | ||
3708 | + idx = 0; | ||
3709 | + } | ||
3710 | + else | ||
3711 | + { | ||
3712 | + idx += (operating_mode == byte_mode) ? mblength : 1; | ||
3713 | + if (print_kth (idx, NULL)) | ||
3714 | + fwrite (bufpos, mblength, sizeof(char), stdout); | ||
3715 | + } | ||
3716 | + | ||
3717 | + buflen -= mblength; | ||
3718 | + bufpos += mblength; | ||
3719 | + } | ||
3720 | +} | ||
3721 | +#endif | ||
3722 | + | ||
3723 | /* Read from stream STREAM, printing to standard output any selected fields. */ | ||
3724 | |||
3725 | static void | ||
3726 | @@ -692,13 +831,192 @@ | ||
3727 | } | ||
3728 | } | ||
3729 | |||
3730 | +#if HAVE_MBRTOWC | ||
3731 | +static void | ||
3732 | +cut_fields_mb (FILE *stream) | ||
3733 | +{ | ||
3734 | + int c; | ||
3735 | + unsigned int field_idx; | ||
3736 | + int found_any_selected_field; | ||
3737 | + int buffer_first_field; | ||
3738 | + int empty_input; | ||
3739 | + char buf[MB_LEN_MAX + BUFSIZ]; /* For spooling a read byte sequence. */ | ||
3740 | + char *bufpos; /* Next read position of BUF. */ | ||
3741 | + size_t buflen; /* The length of the byte sequence in buf. */ | ||
3742 | + wint_t wc = 0; /* A gotten wide character. */ | ||
3743 | + size_t mblength; /* The byte size of a multibyte character which shows | ||
3744 | + as same character as WC. */ | ||
3745 | + mbstate_t state; /* State of the stream. */ | ||
3746 | + int convfail; /* 1, when conversion is failed. Otherwise 0. */ | ||
3747 | + | ||
3748 | + found_any_selected_field = 0; | ||
3749 | + field_idx = 1; | ||
3750 | + bufpos = buf; | ||
3751 | + buflen = 0; | ||
3752 | + memset (&state, '\0', sizeof(mbstate_t)); | ||
3753 | + | ||
3754 | + c = getc (stream); | ||
3755 | + empty_input = (c == EOF); | ||
3756 | + if (c != EOF) | ||
3757 | + ungetc (c, stream); | ||
3758 | + else | ||
3759 | + wc = WEOF; | ||
3760 | + | ||
3761 | + /* To support the semantics of the -s flag, we may have to buffer | ||
3762 | + all of the first field to determine whether it is `delimited.' | ||
3763 | + But that is unnecessary if all non-delimited lines must be printed | ||
3764 | + and the first field has been selected, or if non-delimited lines | ||
3765 | + must be suppressed and the first field has *not* been selected. | ||
3766 | + That is because a non-delimited line has exactly one field. */ | ||
3767 | + buffer_first_field = (suppress_non_delimited ^ !print_kth (1, NULL)); | ||
3768 | + | ||
3769 | + while (1) | ||
3770 | + { | ||
3771 | + if (field_idx == 1 && buffer_first_field) | ||
3772 | + { | ||
3773 | + int len = 0; | ||
3774 | + | ||
3775 | + while (1) | ||
3776 | + { | ||
3777 | + REFILL_BUFFER (buf, bufpos, buflen, stream); | ||
3778 | + | ||
3779 | + GET_NEXT_WC_FROM_BUFFER | ||
3780 | + (wc, bufpos, buflen, mblength, state, convfail); | ||
3781 | + | ||
3782 | + if (wc == WEOF) | ||
3783 | + break; | ||
3784 | + | ||
3785 | + field_1_buffer = xrealloc (field_1_buffer, len + mblength); | ||
3786 | + memcpy (field_1_buffer + len, bufpos, mblength); | ||
3787 | + len += mblength; | ||
3788 | + buflen -= mblength; | ||
3789 | + bufpos += mblength; | ||
3790 | + | ||
3791 | + if (!convfail && (wc == L'\n' || wc == wcdelim)) | ||
3792 | + break; | ||
3793 | + } | ||
3794 | + | ||
3795 | + if (wc == WEOF) | ||
3796 | + break; | ||
3797 | + | ||
3798 | + /* If the first field extends to the end of line (it is not | ||
3799 | + delimited) and we are printing all non-delimited lines, | ||
3800 | + print this one. */ | ||
3801 | + if (convfail || (!convfail && wc != wcdelim)) | ||
3802 | + { | ||
3803 | + if (suppress_non_delimited) | ||
3804 | + { | ||
3805 | + /* Empty. */ | ||
3806 | + } | ||
3807 | + else | ||
3808 | + { | ||
3809 | + fwrite (field_1_buffer, sizeof (char), len, stdout); | ||
3810 | + /* Make sure the output line is newline terminated. */ | ||
3811 | + if (convfail || (!convfail && wc != L'\n')) | ||
3812 | + putchar ('\n'); | ||
3813 | + } | ||
3814 | + continue; | ||
3815 | + } | ||
3816 | + | ||
3817 | + if (print_kth (1, NULL)) | ||
3818 | + { | ||
3819 | + /* Print the field, but not the trailing delimiter. */ | ||
3820 | + fwrite (field_1_buffer, sizeof (char), len - 1, stdout); | ||
3821 | + found_any_selected_field = 1; | ||
3822 | + } | ||
3823 | + ++field_idx; | ||
3824 | + } | ||
3825 | + | ||
3826 | + if (wc != WEOF) | ||
3827 | + { | ||
3828 | + if (print_kth (field_idx, NULL)) | ||
3829 | + { | ||
3830 | + if (found_any_selected_field) | ||
3831 | + { | ||
3832 | + fwrite (output_delimiter_string, sizeof (char), | ||
3833 | + output_delimiter_length, stdout); | ||
3834 | + } | ||
3835 | + found_any_selected_field = 1; | ||
3836 | + } | ||
3837 | + | ||
3838 | + while (1) | ||
3839 | + { | ||
3840 | + REFILL_BUFFER (buf, bufpos, buflen, stream); | ||
3841 | + | ||
3842 | + GET_NEXT_WC_FROM_BUFFER | ||
3843 | + (wc, bufpos, buflen, mblength, state, convfail); | ||
3844 | + | ||
3845 | + if (wc == WEOF) | ||
3846 | + break; | ||
3847 | + else if (!convfail && (wc == wcdelim || wc == L'\n')) | ||
3848 | + { | ||
3849 | + buflen -= mblength; | ||
3850 | + bufpos += mblength; | ||
3851 | + break; | ||
3852 | + } | ||
3853 | + | ||
3854 | + if (print_kth (field_idx, NULL)) | ||
3855 | + fwrite (bufpos, mblength, sizeof(char), stdout); | ||
3856 | + | ||
3857 | + buflen -= mblength; | ||
3858 | + bufpos += mblength; | ||
3859 | + } | ||
3860 | + } | ||
3861 | + | ||
3862 | + if ((!convfail || wc == L'\n') && buflen < 1) | ||
3863 | + wc = WEOF; | ||
3864 | + | ||
3865 | + if (!convfail && wc == wcdelim) | ||
3866 | + ++field_idx; | ||
3867 | + else if (wc == WEOF || (!convfail && wc == L'\n')) | ||
3868 | + { | ||
3869 | + if (found_any_selected_field | ||
3870 | + || (!empty_input && !(suppress_non_delimited && field_idx == 1))) | ||
3871 | + putchar ('\n'); | ||
3872 | + if (wc == WEOF) | ||
3873 | + break; | ||
3874 | + field_idx = 1; | ||
3875 | + found_any_selected_field = 0; | ||
3876 | + } | ||
3877 | + } | ||
3878 | +} | ||
3879 | +#endif | ||
3880 | + | ||
3881 | static void | ||
3882 | cut_stream (FILE *stream) | ||
3883 | { | ||
3884 | - if (operating_mode == byte_mode) | ||
3885 | - cut_bytes (stream); | ||
3886 | +#if HAVE_MBRTOWC | ||
3887 | + if (MB_CUR_MAX > 1 && !force_singlebyte_mode) | ||
3888 | + { | ||
3889 | + switch (operating_mode) | ||
3890 | + { | ||
3891 | + case byte_mode: | ||
3892 | + if (byte_mode_character_aware) | ||
3893 | + cut_characters_or_cut_bytes_no_split (stream); | ||
3894 | + else | ||
3895 | + cut_bytes (stream); | ||
3896 | + break; | ||
3897 | + | ||
3898 | + case character_mode: | ||
3899 | + cut_characters_or_cut_bytes_no_split (stream); | ||
3900 | + break; | ||
3901 | + | ||
3902 | + case field_mode: | ||
3903 | + cut_fields_mb (stream); | ||
3904 | + break; | ||
3905 | + | ||
3906 | + default: | ||
3907 | + abort (); | ||
3908 | + } | ||
3909 | + } | ||
3910 | else | ||
3911 | - cut_fields (stream); | ||
3912 | +#endif | ||
3913 | + { | ||
3914 | + if (operating_mode == field_mode) | ||
3915 | + cut_fields (stream); | ||
3916 | + else | ||
3917 | + cut_bytes (stream); | ||
3918 | + } | ||
3919 | } | ||
3920 | |||
3921 | /* Process file FILE to standard output. | ||
3922 | @@ -748,6 +1066,8 @@ | ||
3923 | bool ok; | ||
3924 | bool delim_specified = false; | ||
3925 | char *spec_list_string IF_LINT(= NULL); | ||
3926 | + char mbdelim[MB_LEN_MAX + 1]; | ||
3927 | + size_t delimlen = 0; | ||
3928 | |||
3929 | initialize_main (&argc, &argv); | ||
3930 | program_name = argv[0]; | ||
3931 | @@ -770,7 +1090,6 @@ | ||
3932 | switch (optc) | ||
3933 | { | ||
3934 | case 'b': | ||
3935 | - case 'c': | ||
3936 | /* Build the byte list. */ | ||
3937 | if (operating_mode != undefined_mode) | ||
3938 | FATAL_ERROR (_("only one type of list may be specified")); | ||
3939 | @@ -778,6 +1097,14 @@ | ||
3940 | spec_list_string = optarg; | ||
3941 | break; | ||
3942 | |||
3943 | + case 'c': | ||
3944 | + /* Build the character list. */ | ||
3945 | + if (operating_mode != undefined_mode) | ||
3946 | + FATAL_ERROR (_("only one type of list may be specified")); | ||
3947 | + operating_mode = character_mode; | ||
3948 | + spec_list_string = optarg; | ||
3949 | + break; | ||
3950 | + | ||
3951 | case 'f': | ||
3952 | /* Build the field list. */ | ||
3953 | if (operating_mode != undefined_mode) | ||
3954 | @@ -789,10 +1116,35 @@ | ||
3955 | case 'd': | ||
3956 | /* New delimiter. */ | ||
3957 | /* Interpret -d '' to mean `use the NUL byte as the delimiter.' */ | ||
3958 | - if (optarg[0] != '\0' && optarg[1] != '\0') | ||
3959 | - FATAL_ERROR (_("the delimiter must be a single character")); | ||
3960 | - delim = optarg[0]; | ||
3961 | - delim_specified = true; | ||
3962 | +#if HAVE_MBRTOWC | ||
3963 | + { | ||
3964 | + if(MB_CUR_MAX > 1) | ||
3965 | + { | ||
3966 | + mbstate_t state; | ||
3967 | + | ||
3968 | + memset (&state, '\0', sizeof(mbstate_t)); | ||
3969 | + delimlen = mbrtowc (&wcdelim, optarg, strnlen(optarg, MB_LEN_MAX), &state); | ||
3970 | + | ||
3971 | + if (delimlen == (size_t)-1 || delimlen == (size_t)-2) | ||
3972 | + ++force_singlebyte_mode; | ||
3973 | + else | ||
3974 | + { | ||
3975 | + delimlen = (delimlen < 1) ? 1 : delimlen; | ||
3976 | + if (wcdelim != L'\0' && *(optarg + delimlen) != '\0') | ||
3977 | + FATAL_ERROR (_("the delimiter must be a single character")); | ||
3978 | + memcpy (mbdelim, optarg, delimlen); | ||
3979 | + } | ||
3980 | + } | ||
3981 | + | ||
3982 | + if (MB_CUR_MAX <= 1 || force_singlebyte_mode) | ||
3983 | +#endif | ||
3984 | + { | ||
3985 | + if (optarg[0] != '\0' && optarg[1] != '\0') | ||
3986 | + FATAL_ERROR (_("the delimiter must be a single character")); | ||
3987 | + delim = (unsigned char) optarg[0]; | ||
3988 | + } | ||
3989 | + delim_specified = true; | ||
3990 | + } | ||
3991 | break; | ||
3992 | |||
3993 | case OUTPUT_DELIMITER_OPTION: | ||
3994 | @@ -805,6 +1157,7 @@ | ||
3995 | break; | ||
3996 | |||
3997 | case 'n': | ||
3998 | + byte_mode_character_aware = 1; | ||
3999 | break; | ||
4000 | |||
4001 | case 's': | ||
4002 | @@ -827,7 +1180,7 @@ | ||
4003 | if (operating_mode == undefined_mode) | ||
4004 | FATAL_ERROR (_("you must specify a list of bytes, characters, or fields")); | ||
4005 | |||
4006 | - if (delim != '\0' && operating_mode != field_mode) | ||
4007 | + if (delim_specified && operating_mode != field_mode) | ||
4008 | FATAL_ERROR (_("an input delimiter may be specified only\ | ||
4009 | when operating on fields")); | ||
4010 | |||
4011 | @@ -854,15 +1207,34 @@ | ||
4012 | } | ||
4013 | |||
4014 | if (!delim_specified) | ||
4015 | - delim = '\t'; | ||
4016 | + { | ||
4017 | + delim = '\t'; | ||
4018 | +#ifdef HAVE_MBRTOWC | ||
4019 | + wcdelim = L'\t'; | ||
4020 | + mbdelim[0] = '\t'; | ||
4021 | + mbdelim[1] = '\0'; | ||
4022 | + delimlen = 1; | ||
4023 | +#endif | ||
4024 | + } | ||
4025 | |||
4026 | if (output_delimiter_string == NULL) | ||
4027 | { | ||
4028 | - static char dummy[2]; | ||
4029 | - dummy[0] = delim; | ||
4030 | - dummy[1] = '\0'; | ||
4031 | - output_delimiter_string = dummy; | ||
4032 | - output_delimiter_length = 1; | ||
4033 | +#ifdef HAVE_MBRTOWC | ||
4034 | + if (MB_CUR_MAX > 1 && !force_singlebyte_mode) | ||
4035 | + { | ||
4036 | + output_delimiter_string = xstrdup(mbdelim); | ||
4037 | + output_delimiter_length = delimlen; | ||
4038 | + } | ||
4039 | + | ||
4040 | + if (MB_CUR_MAX <= 1 || force_singlebyte_mode) | ||
4041 | +#endif | ||
4042 | + { | ||
4043 | + static char dummy[2]; | ||
4044 | + dummy[0] = delim; | ||
4045 | + dummy[1] = '\0'; | ||
4046 | + output_delimiter_string = dummy; | ||
4047 | + output_delimiter_length = 1; | ||
4048 | + } | ||
4049 | } | ||
4050 | |||
4051 | if (optind == argc) | ||
diff --git a/recipes-core/coreutils/coreutils-6.9/coreutils-ls-x.patch b/recipes-core/coreutils/coreutils-6.9/coreutils-ls-x.patch new file mode 100644 index 0000000..aba8742 --- /dev/null +++ b/recipes-core/coreutils/coreutils-6.9/coreutils-ls-x.patch | |||
@@ -0,0 +1,117 @@ | |||
1 | Upstream-Status: Inappropriate [legacy version] | ||
2 | |||
3 | This patch was imported from the Fedora Core 8 coreutils-6.9-9 package. | ||
4 | |||
5 | The package is stated as being Licensed as GPLv2+. | ||
6 | |||
7 | Signed-off-by: Mark Hatle <mark.hatle@windriver.com> | ||
8 | |||
9 | --- coreutils-6.9/src/ls.c.ls-x 2007-06-13 14:27:36.000000000 +0100 | ||
10 | +++ coreutils-6.9/src/ls.c 2007-06-13 14:28:42.000000000 +0100 | ||
11 | @@ -4151,16 +4151,16 @@ | ||
12 | size_t pos = 0; | ||
13 | size_t cols = calculate_columns (false); | ||
14 | struct column_info const *line_fmt = &column_info[cols - 1]; | ||
15 | - size_t name_length = length_of_file_name_and_frills (cwd_file); | ||
16 | + struct fileinfo const *f = sorted_file[0]; | ||
17 | + size_t name_length = length_of_file_name_and_frills (f); | ||
18 | size_t max_name_length = line_fmt->col_arr[0]; | ||
19 | |||
20 | /* Print first entry. */ | ||
21 | - print_file_name_and_frills (cwd_file); | ||
22 | + print_file_name_and_frills (f); | ||
23 | |||
24 | /* Now the rest. */ | ||
25 | for (filesno = 1; filesno < cwd_n_used; ++filesno) | ||
26 | { | ||
27 | - struct fileinfo const *f; | ||
28 | size_t col = filesno % cols; | ||
29 | |||
30 | if (col == 0) | ||
31 | --- coreutils-6.9/tests/ls/Makefile.am.ls-x 2007-03-18 21:36:43.000000000 +0000 | ||
32 | +++ coreutils-6.9/tests/ls/Makefile.am 2007-06-13 14:28:42.000000000 +0100 | ||
33 | @@ -24,7 +24,7 @@ | ||
34 | stat-dtype \ | ||
35 | inode dangle file-type recursive dired infloop \ | ||
36 | rt-1 time-1 symlink-slash follow-slink no-arg m-option \ | ||
37 | - stat-vs-dirent | ||
38 | + stat-vs-dirent x-option | ||
39 | |||
40 | EXTRA_DIST = $(TESTS) | ||
41 | TESTS_ENVIRONMENT = \ | ||
42 | --- /dev/null 2007-06-13 08:43:51.993263382 +0100 | ||
43 | +++ coreutils-6.9/tests/ls/x-option 2007-06-13 14:28:42.000000000 +0100 | ||
44 | @@ -0,0 +1,59 @@ | ||
45 | +#!/bin/sh | ||
46 | +# Exercise the -x option. | ||
47 | + | ||
48 | +# Copyright (C) 2007 Free Software Foundation, Inc. | ||
49 | + | ||
50 | +# This program is free software; you can redistribute it and/or modify | ||
51 | +# it under the terms of the GNU General Public License as published by | ||
52 | +# the Free Software Foundation; either version 2 of the License, or | ||
53 | +# (at your option) any later version. | ||
54 | + | ||
55 | +# This program is distributed in the hope that it will be useful, | ||
56 | +# but WITHOUT ANY WARRANTY; without even the implied warranty of | ||
57 | +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | ||
58 | +# GNU General Public License for more details. | ||
59 | + | ||
60 | +# You should have received a copy of the GNU General Public License | ||
61 | +# along with this program; if not, write to the Free Software | ||
62 | +# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA | ||
63 | +# 02110-1301, USA. | ||
64 | + | ||
65 | +if test "$VERBOSE" = yes; then | ||
66 | + set -x | ||
67 | + ls --version | ||
68 | +fi | ||
69 | + | ||
70 | +. $srcdir/../envvar-check | ||
71 | +. $srcdir/../lang-default | ||
72 | + | ||
73 | +pwd=`pwd` | ||
74 | +t0=`echo "$0"|sed 's,.*/,,'`.tmp; tmp=$t0/$$ | ||
75 | +trap 'status=$?; cd "$pwd" && chmod -R u+rwx $t0 && rm -rf $t0 && exit $status' 0 | ||
76 | +trap '(exit $?); exit $?' 1 2 13 15 | ||
77 | + | ||
78 | +framework_failure=0 | ||
79 | +mkdir -p $tmp || framework_failure=1 | ||
80 | +cd $tmp || framework_failure=1 | ||
81 | +mkdir subdir || framework_failure=1 | ||
82 | +touch subdir/b || framework_failure=1 | ||
83 | +touch subdir/a || framework_failure=1 | ||
84 | + | ||
85 | +if test $framework_failure = 1; then | ||
86 | + echo "$0: failure in testing framework" 1>&2 | ||
87 | + (exit 1); exit 1 | ||
88 | +fi | ||
89 | + | ||
90 | +fail=0 | ||
91 | + | ||
92 | +# Coreutils 6.8 and 6.9 would output this in the wrong order. | ||
93 | +ls -x subdir > out || fail=1 | ||
94 | +ls -rx subdir >> out || fail=1 | ||
95 | +cat <<\EOF > exp || fail=1 | ||
96 | +a b | ||
97 | +b a | ||
98 | +EOF | ||
99 | + | ||
100 | +cmp out exp || fail=1 | ||
101 | +test $fail = 1 && diff out exp 2> /dev/null | ||
102 | + | ||
103 | +(exit $fail); exit $fail | ||
104 | --- coreutils-6.9/NEWS.ls-x 2007-03-22 21:19:45.000000000 +0000 | ||
105 | +++ coreutils-6.9/NEWS 2007-06-13 14:28:42.000000000 +0100 | ||
106 | @@ -13,6 +13,11 @@ | ||
107 | Using pr -m -s (i.e. merging files, with TAB as the output separator) | ||
108 | no longer inserts extraneous spaces between output columns. | ||
109 | |||
110 | +** Bug fixes | ||
111 | + | ||
112 | + ls -x DIR would sometimes output the wrong string in place of the | ||
113 | + first entry. [introduced in coreutils-6.8] | ||
114 | + | ||
115 | |||
116 | * Noteworthy changes in release 6.8 (2007-02-24) [not-unstable] | ||
117 | |||
diff --git a/recipes-core/coreutils/coreutils-6.9/coreutils-overflow.patch b/recipes-core/coreutils/coreutils-6.9/coreutils-overflow.patch new file mode 100644 index 0000000..58074c0 --- /dev/null +++ b/recipes-core/coreutils/coreutils-6.9/coreutils-overflow.patch | |||
@@ -0,0 +1,19 @@ | |||
1 | Upstream-Status: Inappropriate [legacy version] | ||
2 | |||
3 | This patch was imported from the Fedora Core 8 coreutils-6.9-9 package. | ||
4 | |||
5 | The package is stated as being Licensed as GPLv2+. | ||
6 | |||
7 | Signed-off-by: Mark Hatle <mark.hatle@windriver.com> | ||
8 | |||
9 | --- coreutils-5.2.1/src/who.c.overflow 2005-05-25 09:59:06.000000000 +0100 | ||
10 | +++ coreutils-5.2.1/src/who.c 2005-05-25 10:00:31.000000000 +0100 | ||
11 | @@ -75,7 +75,7 @@ | ||
12 | # define NEW_TIME 0 | ||
13 | #endif | ||
14 | |||
15 | -#define IDLESTR_LEN 6 | ||
16 | +#define IDLESTR_LEN 10 | ||
17 | |||
18 | #if HAVE_STRUCT_XTMP_UT_PID | ||
19 | # define PIDSTR_DECL_AND_INIT(Var, Utmp_ent) \ | ||
diff --git a/recipes-core/coreutils/coreutils-6.9/coreutils_fix_for_automake-1.12.patch b/recipes-core/coreutils/coreutils-6.9/coreutils_fix_for_automake-1.12.patch new file mode 100644 index 0000000..64e5f12 --- /dev/null +++ b/recipes-core/coreutils/coreutils-6.9/coreutils_fix_for_automake-1.12.patch | |||
@@ -0,0 +1,32 @@ | |||
1 | Upstream-Status: Pending | ||
2 | |||
3 | automake 1.12 has depricated automatic de-ANSI-fication support | ||
4 | |||
5 | this patch avoids these kinds of errors: | ||
6 | |||
7 | | configure.ac:40: error: automatic de-ANSI-fication support has been removed | ||
8 | | /srv/home/nitin/builds/build-gcc47/tmp/sysroots/x86_64-linux/usr/share/aclocal-1.12/protos.m4:12: AM_C_PROTOTYPES is expanded from... | ||
9 | | /srv/home/nitin/builds/build-gcc47/tmp/work/x86_64-linux/coreutils-native-6.9-r2/coreutils-6.9/m4/jm-macros.m4:138: gl_CHECK_ALL_TYPES is expanded from... | ||
10 | | /srv/home/nitin/builds/build-gcc47/tmp/work/x86_64-linux/coreutils-native-6.9-r2/coreutils-6.9/m4/jm-macros.m4:24: coreutils_MACROS is expanded from... | ||
11 | | configure.ac:40: the top level | ||
12 | | autom4te: m4 failed with exit status: 1 | ||
13 | |||
14 | Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com> | ||
15 | 2012/05/04 | ||
16 | |||
17 | Index: coreutils-6.9/m4/jm-macros.m4 | ||
18 | =================================================================== | ||
19 | --- coreutils-6.9.orig/m4/jm-macros.m4 | ||
20 | +++ coreutils-6.9/m4/jm-macros.m4 | ||
21 | @@ -142,11 +142,6 @@ AC_DEFUN([gl_CHECK_ALL_TYPES], | ||
22 | dnl whether functions and headers are available, whether they work, etc. | ||
23 | AC_REQUIRE([AC_SYS_LARGEFILE]) | ||
24 | |||
25 | - dnl This test must precede tests of compiler characteristics like | ||
26 | - dnl that for the inline keyword, since it may change the degree to | ||
27 | - dnl which the compiler supports such features. | ||
28 | - AC_REQUIRE([AM_C_PROTOTYPES]) | ||
29 | - | ||
30 | dnl Checks for typedefs, structures, and compiler characteristics. | ||
31 | AC_REQUIRE([AC_C_BIGENDIAN]) | ||
32 | AC_REQUIRE([AC_C_VOLATILE]) | ||
diff --git a/recipes-core/coreutils/coreutils-6.9/fix_for_manpage_building.patch b/recipes-core/coreutils/coreutils-6.9/fix_for_manpage_building.patch new file mode 100644 index 0000000..e0d600a --- /dev/null +++ b/recipes-core/coreutils/coreutils-6.9/fix_for_manpage_building.patch | |||
@@ -0,0 +1,85 @@ | |||
1 | Upstream-Status: Inappropriate [legacy version] | ||
2 | |||
3 | Use native coreutils binaries to build manpages in cross environment. | ||
4 | This avoids man page build issues like this: | ||
5 | |||
6 | | Making all in man | ||
7 | | make[1]: Entering directory `/build_disk/poky_build/build1/tmp/work/i586-poky-linux/coreutils-6.9-r0/coreutils-6.9/man' | ||
8 | | Updating man page cut.1 | ||
9 | | Updating man page dir.1 | ||
10 | | Updating man page expand.1 | ||
11 | | Updating man page fold.1 | ||
12 | | Updating man page install.1 | ||
13 | | Updating man page join.1 | ||
14 | | Updating man page pr.1 | ||
15 | | Updating man page ls.1 | ||
16 | | Updating man page sort.1 | ||
17 | | Updating man page unexpand.1 | ||
18 | | Updating man page uniq.1 | ||
19 | | Updating man page who.1 | ||
20 | | Updating man page vdir.1 | ||
21 | | help2man: can't get `--help' info from dir.td/dir | ||
22 | | help2man: can't get `--help' info from cut.td/cut | ||
23 | | make[1]: *** [dir.1] Error 126 | ||
24 | | make[1]: *** Waiting for unfinished jobs.... | ||
25 | | help2man: can't get `--help' info from fold.td/fold | ||
26 | | help2man: can't get `--help' info from install.td/install | ||
27 | | help2man: can't get `--help' info from expand.td/expand | ||
28 | | help2man: can't get `--help' info from join.td/join | ||
29 | | make[1]: *** [cut.1] Error 126 | ||
30 | | make[1]: *** [fold.1] Error 126 | ||
31 | | make[1]: *** [install.1] Error 126 | ||
32 | | help2man: can't get `--help' info from sort.td/sort | ||
33 | | make[1]: *** [expand.1] Error 126 | ||
34 | | help2man: can't get `--help' info from pr.td/pr | ||
35 | | make[1]: *** [join.1] Error 126 | ||
36 | | help2man: can't get `--help' info from ls.td/ls | ||
37 | | help2man: can't get `--help' info from unexpand.td/unexpand | ||
38 | | help2man: can't get `--help' info from uniq.td/uniq | ||
39 | | help2man: can't get `--help' info from who.td/who | ||
40 | | make[1]: *** [sort.1] Error 126 | ||
41 | | make[1]: *** [pr.1] Error 126 | ||
42 | | help2man: can't get `--help' info from vdir.td/vdir | ||
43 | | make[1]: *** [ls.1] Error 126 | ||
44 | | make[1]: *** [uniq.1] Error 126 | ||
45 | | make[1]: *** [unexpand.1] Error 126 | ||
46 | | make[1]: *** [who.1] Error 126 | ||
47 | | make[1]: *** [vdir.1] Error 126 | ||
48 | | make[1]: Leaving directory `/build_disk/poky_build/build1/tmp/work/i586-poky-linux/coreutils-6.9-r0/coreutils-6.9/man' | ||
49 | | make: *** [all-recursive] Error 1 | ||
50 | | FATAL: oe_runmake failed | ||
51 | | ERROR: Function 'do_compile' failed (see /build_disk/poky_build/build1/tmp/work/i586-poky-linux/coreutils-6.9-r0/temp/log.do_compile.12780 for further information) | ||
52 | NOTE: package coreutils-6.9-r0: task do_compile: Failed | ||
53 | ERROR: Task 8 (/home/nitin/prj/poky.git/meta/recipes-core/coreutils/coreutils_6.9.bb, do_compile) failed with exit code '1' | ||
54 | |||
55 | |||
56 | This patch is made for gplv2 coreutils the recipe | ||
57 | Nitin A Kamble <nitin.a.kamble@intel.com> | ||
58 | 2011/03/17 | ||
59 | |||
60 | Index: man/Makefile.am | ||
61 | =================================================================== | ||
62 | --- a/man.orig/Makefile.am | ||
63 | +++ b/man/Makefile.am | ||
64 | @@ -167,7 +167,7 @@ mapped_name = `echo $*|sed 's/install/gi | ||
65 | $(PERL) -- $(srcdir)/help2man \ | ||
66 | --source='$(PACKAGE_STRING)' \ | ||
67 | --include=$(srcdir)/$*.x \ | ||
68 | - --output=$t/$@ $t/$*; \ | ||
69 | + --output=$t/$@ $*; \ | ||
70 | } \ | ||
71 | && sed 's|$*\.td/||g' $t/$@ > $@ \ | ||
72 | && chmod a-w $@ \ | ||
73 | Index: man/Makefile.in | ||
74 | =================================================================== | ||
75 | --- a/man.orig/Makefile.in | ||
76 | +++ b/man/Makefile.in | ||
77 | @@ -865,7 +865,7 @@ yes.1: $(common_dep) $(srcdir)/yes.x . | ||
78 | $(PERL) -- $(srcdir)/help2man \ | ||
79 | --source='$(PACKAGE_STRING)' \ | ||
80 | --include=$(srcdir)/$*.x \ | ||
81 | - --output=$t/$@ $t/$*; \ | ||
82 | + --output=$t/$@ $*; \ | ||
83 | } \ | ||
84 | && sed 's|$*\.td/||g' $t/$@ > $@ \ | ||
85 | && chmod a-w $@ \ | ||
diff --git a/recipes-core/coreutils/coreutils-6.9/futimens.patch b/recipes-core/coreutils/coreutils-6.9/futimens.patch new file mode 100644 index 0000000..5088106 --- /dev/null +++ b/recipes-core/coreutils/coreutils-6.9/futimens.patch | |||
@@ -0,0 +1,63 @@ | |||
1 | Upstream-Status: Inappropriate [legacy version] | ||
2 | |||
3 | # coreutils uses gnulib which conflicts with newer libc header on futimens | ||
4 | # this patch simply renames coreutils futimes to avoid confliction | ||
5 | # | ||
6 | # Signed-off-by: Kevin Tian <kevin.tian@intel.com>, 2010-08-18 | ||
7 | # (this patch is licensed under GPLv2) | ||
8 | |||
9 | diff --git a/lib/utimens.c b/lib/utimens.c | ||
10 | index 71bc510..ae870b8 100644 | ||
11 | --- a/lib/utimens.c | ||
12 | +++ b/lib/utimens.c | ||
13 | @@ -75,7 +75,7 @@ struct utimbuf | ||
14 | Return 0 on success, -1 (setting errno) on failure. */ | ||
15 | |||
16 | int | ||
17 | -futimens (int fd ATTRIBUTE_UNUSED, | ||
18 | +futimens_coreutils (int fd ATTRIBUTE_UNUSED, | ||
19 | char const *file, struct timespec const timespec[2]) | ||
20 | { | ||
21 | /* Some Linux-based NFS clients are buggy, and mishandle time stamps | ||
22 | @@ -185,5 +185,5 @@ futimens (int fd ATTRIBUTE_UNUSED, | ||
23 | int | ||
24 | utimens (char const *file, struct timespec const timespec[2]) | ||
25 | { | ||
26 | - return futimens (-1, file, timespec); | ||
27 | + return futimens_coreutils (-1, file, timespec); | ||
28 | } | ||
29 | diff --git a/lib/utimens.h b/lib/utimens.h | ||
30 | index 0097aaa..13fc45a 100644 | ||
31 | --- a/lib/utimens.h | ||
32 | +++ b/lib/utimens.h | ||
33 | @@ -1,3 +1,3 @@ | ||
34 | #include <time.h> | ||
35 | -int futimens (int, char const *, struct timespec const [2]); | ||
36 | +int futimens_coreutils (int, char const *, struct timespec const [2]); | ||
37 | int utimens (char const *, struct timespec const [2]); | ||
38 | diff --git a/src/copy.c b/src/copy.c | ||
39 | index 4bdb75c..04634f1 100644 | ||
40 | --- a/src/copy.c | ||
41 | +++ b/src/copy.c | ||
42 | @@ -518,7 +518,7 @@ copy_reg (char const *src_name, char const *dst_name, | ||
43 | timespec[0] = get_stat_atime (src_sb); | ||
44 | timespec[1] = get_stat_mtime (src_sb); | ||
45 | |||
46 | - if (futimens (dest_desc, dst_name, timespec) != 0) | ||
47 | + if (futimens_coreutils (dest_desc, dst_name, timespec) != 0) | ||
48 | { | ||
49 | error (0, errno, _("preserving times for %s"), quote (dst_name)); | ||
50 | if (x->require_preserve) | ||
51 | diff --git a/src/touch.c b/src/touch.c | ||
52 | index a79c26d..6ef317d 100644 | ||
53 | --- a/src/touch.c | ||
54 | +++ b/src/touch.c | ||
55 | @@ -182,7 +182,7 @@ touch (const char *file) | ||
56 | t = timespec; | ||
57 | } | ||
58 | |||
59 | - ok = (futimens (fd, (fd == STDOUT_FILENO ? NULL : file), t) == 0); | ||
60 | + ok = (futimens_coreutils (fd, (fd == STDOUT_FILENO ? NULL : file), t) == 0); | ||
61 | |||
62 | if (fd == STDIN_FILENO) | ||
63 | { | ||
diff --git a/recipes-core/coreutils/coreutils-6.9/gnulib_m4.patch b/recipes-core/coreutils/coreutils-6.9/gnulib_m4.patch new file mode 100644 index 0000000..8a8ffa7 --- /dev/null +++ b/recipes-core/coreutils/coreutils-6.9/gnulib_m4.patch | |||
@@ -0,0 +1,21 @@ | |||
1 | Upstream-Status: Inappropriate [legacy version] | ||
2 | |||
3 | # remove the line to cause recursive inclusion error from autoreconf, sicne | ||
4 | # newer autoconf has included this definition. Simply rename it here. | ||
5 | # | ||
6 | # Signed-off-by: Kevin Tian <kevin.tian@intel.com>, 2010-08-18 | ||
7 | # (this patch is licensed under GPLv2) | ||
8 | |||
9 | diff --git a/extensions.m4 b/extensions.m4 | ||
10 | index 143a9e5..f6558f1 100644 | ||
11 | --- a/m4/extensions.m4 | ||
12 | +++ b/m4/extensions.m4 | ||
13 | @@ -16,7 +16,7 @@ | ||
14 | # ------------------------ | ||
15 | # Enable extensions on systems that normally disable them, | ||
16 | # typically due to standards-conformance issues. | ||
17 | -AC_DEFUN([AC_USE_SYSTEM_EXTENSIONS], | ||
18 | +AC_DEFUN([AC_USE_SYSTEM_EXTENSIONS_DUMMY], | ||
19 | [ | ||
20 | AC_BEFORE([$0], [AC_COMPILE_IFELSE]) | ||
21 | AC_BEFORE([$0], [AC_RUN_IFELSE]) | ||
diff --git a/recipes-core/coreutils/coreutils-6.9/loadavg.patch b/recipes-core/coreutils/coreutils-6.9/loadavg.patch new file mode 100644 index 0000000..c72efd4 --- /dev/null +++ b/recipes-core/coreutils/coreutils-6.9/loadavg.patch | |||
@@ -0,0 +1,18 @@ | |||
1 | Remove hardcoded paths so OE's configure QA does not detect it and fail the builds | ||
2 | For cross compilation is less interesting to look into host paths for target libraries anyway | ||
3 | |||
4 | Upstream-Status: Inappropriate [OE Specific] | ||
5 | Signed-off-by: Khem Raj <raj.khem@gmail.com> | ||
6 | |||
7 | Index: coreutils-6.9/m4/getloadavg.m4 | ||
8 | =================================================================== | ||
9 | --- coreutils-6.9.orig/m4/getloadavg.m4 | ||
10 | +++ coreutils-6.9/m4/getloadavg.m4 | ||
11 | @@ -49,7 +49,6 @@ if test $gl_have_func = no; then | ||
12 | # There is a commonly available library for RS/6000 AIX. | ||
13 | # Since it is not a standard part of AIX, it might be installed locally. | ||
14 | gl_getloadavg_LIBS=$LIBS | ||
15 | - LIBS="-L/usr/local/lib $LIBS" | ||
16 | AC_CHECK_LIB(getloadavg, getloadavg, | ||
17 | [LIBS="-lgetloadavg $LIBS"], [LIBS=$gl_getloadavg_LIBS]) | ||
18 | fi | ||
diff --git a/recipes-core/coreutils/coreutils-6.9/man-touch.patch b/recipes-core/coreutils/coreutils-6.9/man-touch.patch new file mode 100644 index 0000000..59a4fe0 --- /dev/null +++ b/recipes-core/coreutils/coreutils-6.9/man-touch.patch | |||
@@ -0,0 +1,24 @@ | |||
1 | Upstream-Status: Inappropriate [legacy version] | ||
2 | |||
3 | # man page for 'touch' is generated differently from others. All other utilities | ||
4 | # are provided static man source files, while for 'touch' it requires help2man | ||
5 | # to invoke "touch --help" and then convert the output into the manual. Since touch | ||
6 | # is with target format which can't be invoked on build system, disable building | ||
7 | # 'touch' man page here. | ||
8 | # | ||
9 | # Signed-off-by: Kevin Tian <kevin.tian@intel.com>, 2010-08-18 | ||
10 | # (this patch is licensed under GPLv2) | ||
11 | |||
12 | diff --git a/man/Makefile.am b/man/Makefile.am | ||
13 | index 32df9d1..37b09e3 100644 | ||
14 | --- a/man/Makefile.am | ||
15 | +++ b/man/Makefile.am | ||
16 | @@ -27,7 +27,7 @@ dist_man_MANS = \ | ||
17 | paste.1 pathchk.1 pr.1 printenv.1 printf.1 ptx.1 pwd.1 readlink.1 \ | ||
18 | rm.1 rmdir.1 seq.1 sha1sum.1 sha224sum.1 sha256sum.1 sha384sum.1 sha512sum.1 \ | ||
19 | shred.1 shuf.1 sleep.1 sort.1 split.1 stat.1 \ | ||
20 | - su.1 sum.1 sync.1 tac.1 tail.1 tee.1 test.1 touch.1 tr.1 true.1 tsort.1 \ | ||
21 | + su.1 sum.1 sync.1 tac.1 tail.1 tee.1 test.1 tr.1 true.1 tsort.1 \ | ||
22 | tty.1 unexpand.1 uniq.1 unlink.1 vdir.1 wc.1 \ | ||
23 | whoami.1 yes.1 $(MAN) | ||
24 | optional_mans = \ | ||