Commit | Line | Data |
---|---|---|
a0d0e21e | 1 | package File::Find; |
3b825e41 | 2 | use 5.006; |
b75c8c73 | 3 | use strict; |
b395063c | 4 | use warnings; |
cd68ec93 | 5 | use warnings::register; |
7e8ec876 | 6 | our $VERSION = '1.21'; |
a0d0e21e | 7 | require Exporter; |
6280b799 | 8 | require Cwd; |
a0d0e21e | 9 | |
7bd31527 | 10 | # |
98dc9551 | 11 | # Modified to ensure sub-directory traversal order is not inverted by stack |
7bd31527 JH |
12 | # push and pops. That is remains in the same order as in the directory file, |
13 | # or user pre-processing (EG:sorted). | |
14 | # | |
15 | ||
f06db76b AD |
16 | =head1 NAME |
17 | ||
abfdd623 | 18 | File::Find - Traverse a directory tree. |
f06db76b AD |
19 | |
20 | =head1 SYNOPSIS | |
21 | ||
22 | use File::Find; | |
6d355c6e | 23 | find(\&wanted, @directories_to_search); |
f06db76b | 24 | sub wanted { ... } |
237437d0 | 25 | |
f06db76b | 26 | use File::Find; |
abfdd623 | 27 | finddepth(\&wanted, @directories_to_search); |
f06db76b | 28 | sub wanted { ... } |
3cb6de81 | 29 | |
81793b90 GS |
30 | use File::Find; |
31 | find({ wanted => \&process, follow => 1 }, '.'); | |
f06db76b AD |
32 | |
33 | =head1 DESCRIPTION | |
34 | ||
abfdd623 MS |
35 | These are functions for searching through directory trees doing work |
36 | on each file found similar to the Unix I<find> command. File::Find | |
37 | exports two functions, C<find> and C<finddepth>. They work similarly | |
38 | but have subtle differences. | |
39 | ||
40 | =over 4 | |
41 | ||
42 | =item B<find> | |
43 | ||
44 | find(\&wanted, @directories); | |
45 | find(\%options, @directories); | |
46 | ||
6eb87ff8 MJD |
47 | C<find()> does a depth-first search over the given C<@directories> in |
48 | the order they are given. For each file or directory found, it calls | |
49 | the C<&wanted> subroutine. (See below for details on how to use the | |
50 | C<&wanted> function). Additionally, for each directory found, it will | |
51 | C<chdir()> into that directory and continue the search, invoking the | |
52 | C<&wanted> function on each file or subdirectory in the directory. | |
abfdd623 MS |
53 | |
54 | =item B<finddepth> | |
55 | ||
56 | finddepth(\&wanted, @directories); | |
57 | finddepth(\%options, @directories); | |
58 | ||
b807ffce | 59 | C<finddepth()> works just like C<find()> except that it invokes the |
6eb87ff8 MJD |
60 | C<&wanted> function for a directory I<after> invoking it for the |
61 | directory's contents. It does a postorder traversal instead of a | |
62 | preorder traversal, working from the bottom of the directory tree up | |
63 | where C<find()> works from the top of the tree down. | |
abfdd623 MS |
64 | |
65 | =back | |
66 | ||
67 | =head2 %options | |
68 | ||
95e23d19 AL |
69 | The first argument to C<find()> is either a code reference to your |
70 | C<&wanted> function, or a hash reference describing the operations | |
71 | to be performed for each file. The | |
abfdd623 | 72 | code reference is described in L<The wanted function> below. |
20408e3c | 73 | |
81793b90 GS |
74 | Here are the possible keys for the hash: |
75 | ||
76 | =over 3 | |
77 | ||
78 | =item C<wanted> | |
79 | ||
abfdd623 | 80 | The value should be a code reference. This code reference is |
4c90698d RB |
81 | described in L<The wanted function> below. The C<&wanted> subroutine is |
82 | mandatory. | |
81793b90 GS |
83 | |
84 | =item C<bydepth> | |
85 | ||
86 | Reports the name of a directory only AFTER all its entries | |
95e23d19 | 87 | have been reported. Entry point C<finddepth()> is a shortcut for |
788c0c32 | 88 | specifying C<< { bydepth => 1 } >> in the first argument of C<find()>. |
81793b90 | 89 | |
719c805e JS |
90 | =item C<preprocess> |
91 | ||
3555aed3 SP |
92 | The value should be a code reference. This code reference is used to |
93 | preprocess the current directory. The name of the currently processed | |
95e23d19 AL |
94 | directory is in C<$File::Find::dir>. Your preprocessing function is |
95 | called after C<readdir()>, but before the loop that calls the C<wanted()> | |
3555aed3 SP |
96 | function. It is called with a list of strings (actually file/directory |
97 | names) and is expected to return a list of strings. The code can be | |
98 | used to sort the file/directory names alphabetically, numerically, | |
99 | or to filter out directory entries based on their name alone. When | |
7e47e6ff | 100 | I<follow> or I<follow_fast> are in effect, C<preprocess> is a no-op. |
719c805e JS |
101 | |
102 | =item C<postprocess> | |
103 | ||
3555aed3 SP |
104 | The value should be a code reference. It is invoked just before leaving |
105 | the currently processed directory. It is called in void context with no | |
106 | arguments. The name of the current directory is in C<$File::Find::dir>. This | |
107 | hook is handy for summarizing a directory, such as calculating its disk | |
108 | usage. When I<follow> or I<follow_fast> are in effect, C<postprocess> is a | |
7e47e6ff | 109 | no-op. |
719c805e | 110 | |
81793b90 GS |
111 | =item C<follow> |
112 | ||
113 | Causes symbolic links to be followed. Since directory trees with symbolic | |
114 | links (followed) may contain files more than once and may even have | |
115 | cycles, a hash has to be built up with an entry for each file. | |
116 | This might be expensive both in space and time for a large | |
f4d05d5f | 117 | directory tree. See L</follow_fast> and L</follow_skip> below. |
81793b90 GS |
118 | If either I<follow> or I<follow_fast> is in effect: |
119 | ||
120 | =over 6 | |
121 | ||
a45bd81d | 122 | =item * |
81793b90 | 123 | |
54bd407c RGS |
124 | It is guaranteed that an I<lstat> has been called before the user's |
125 | C<wanted()> function is called. This enables fast file checks involving S<_>. | |
126 | Note that this guarantee no longer holds if I<follow> or I<follow_fast> | |
127 | are not set. | |
81793b90 | 128 | |
a45bd81d | 129 | =item * |
81793b90 GS |
130 | |
131 | There is a variable C<$File::Find::fullname> which holds the absolute | |
615a2b9b SP |
132 | pathname of the file with all symbolic links resolved. If the link is |
133 | a dangling symbolic link, then fullname will be set to C<undef>. | |
81793b90 GS |
134 | |
135 | =back | |
136 | ||
204b4d7f SP |
137 | This is a no-op on Win32. |
138 | ||
81793b90 GS |
139 | =item C<follow_fast> |
140 | ||
f10e1564 RM |
141 | This is similar to I<follow> except that it may report some files more |
142 | than once. It does detect cycles, however. Since only symbolic links | |
143 | have to be hashed, this is much cheaper both in space and time. If | |
95e23d19 | 144 | processing a file more than once (by the user's C<wanted()> function) |
81793b90 GS |
145 | is worse than just taking time, the option I<follow> should be used. |
146 | ||
204b4d7f SP |
147 | This is also a no-op on Win32. |
148 | ||
81793b90 GS |
149 | =item C<follow_skip> |
150 | ||
151 | C<follow_skip==1>, which is the default, causes all files which are | |
152 | neither directories nor symbolic links to be ignored if they are about | |
3555aed3 | 153 | to be processed a second time. If a directory or a symbolic link |
81793b90 | 154 | are about to be processed a second time, File::Find dies. |
95e23d19 | 155 | |
81793b90 GS |
156 | C<follow_skip==0> causes File::Find to die if any file is about to be |
157 | processed a second time. | |
95e23d19 | 158 | |
81793b90 | 159 | C<follow_skip==2> causes File::Find to ignore any duplicate files and |
7e47e6ff | 160 | directories but to proceed normally otherwise. |
20408e3c | 161 | |
80e52b73 JH |
162 | =item C<dangling_symlinks> |
163 | ||
164 | If true and a code reference, will be called with the symbolic link | |
165 | name and the directory it lives in as arguments. Otherwise, if true | |
166 | and warnings are on, warning "symbolic_link_name is a dangling | |
167 | symbolic link\n" will be issued. If false, the dangling symbolic link | |
168 | will be silently ignored. | |
f06db76b | 169 | |
81793b90 GS |
170 | =item C<no_chdir> |
171 | ||
95e23d19 | 172 | Does not C<chdir()> to each directory as it recurses. The C<wanted()> |
81793b90 GS |
173 | function will need to be aware of this, of course. In this case, |
174 | C<$_> will be the same as C<$File::Find::name>. | |
175 | ||
176 | =item C<untaint> | |
177 | ||
178 | If find is used in taint-mode (-T command line switch or if EUID != UID | |
179 | or if EGID != GID) then internally directory names have to be untainted | |
7e47e6ff | 180 | before they can be chdir'ed to. Therefore they are checked against a regular |
3555aed3 SP |
181 | expression I<untaint_pattern>. Note that all names passed to the user's |
182 | I<wanted()> function are still tainted. If this option is used while | |
7e47e6ff | 183 | not in taint-mode, C<untaint> is a no-op. |
81793b90 GS |
184 | |
185 | =item C<untaint_pattern> | |
186 | ||
187 | See above. This should be set using the C<qr> quoting operator. | |
3555aed3 | 188 | The default is set to C<qr|^([-+@\w./]+)$|>. |
1cffc1dd | 189 | Note that the parentheses are vital. |
81793b90 GS |
190 | |
191 | =item C<untaint_skip> | |
192 | ||
3555aed3 | 193 | If set, a directory which fails the I<untaint_pattern> is skipped, |
7e47e6ff | 194 | including all its sub-directories. The default is to 'die' in such a case. |
81793b90 GS |
195 | |
196 | =back | |
197 | ||
abfdd623 MS |
198 | =head2 The wanted function |
199 | ||
95e23d19 AL |
200 | The C<wanted()> function does whatever verifications you want on |
201 | each file and directory. Note that despite its name, the C<wanted()> | |
202 | function is a generic callback function, and does B<not> tell | |
203 | File::Find if a file is "wanted" or not. In fact, its return value | |
204 | is ignored. | |
205 | ||
206 | The wanted function takes no arguments but rather does its work | |
abfdd623 MS |
207 | through a collection of variables. |
208 | ||
209 | =over 4 | |
210 | ||
f837ebe2 | 211 | =item C<$File::Find::dir> is the current directory name, |
abfdd623 MS |
212 | |
213 | =item C<$_> is the current filename within that directory | |
214 | ||
f837ebe2 | 215 | =item C<$File::Find::name> is the complete pathname to the file. |
abfdd623 MS |
216 | |
217 | =back | |
218 | ||
4c767a4e | 219 | The above variables have all been localized and may be changed without |
026810a8 | 220 | affecting data outside of the wanted function. |
4c767a4e | 221 | |
95e23d19 | 222 | For example, when examining the file F</some/path/foo.ext> you will have: |
abfdd623 MS |
223 | |
224 | $File::Find::dir = /some/path/ | |
225 | $_ = foo.ext | |
226 | $File::Find::name = /some/path/foo.ext | |
227 | ||
3555aed3 | 228 | You are chdir()'d to C<$File::Find::dir> when the function is called, |
abfdd623 MS |
229 | unless C<no_chdir> was specified. Note that when changing to |
230 | directories is in effect the root directory (F</>) is a somewhat | |
231 | special case inasmuch as the concatenation of C<$File::Find::dir>, | |
232 | C<'/'> and C<$_> is not literally equal to C<$File::Find::name>. The | |
233 | table below summarizes all variants: | |
5cf0a2f2 WL |
234 | |
235 | $File::Find::name $File::Find::dir $_ | |
236 | default / / . | |
237 | no_chdir=>0 /etc / etc | |
238 | /etc/x /etc x | |
abfdd623 | 239 | |
5cf0a2f2 WL |
240 | no_chdir=>1 / / / |
241 | /etc / /etc | |
242 | /etc/x /etc /etc/x | |
243 | ||
244 | ||
788c0c32 | 245 | When C<follow> or C<follow_fast> are in effect, there is |
f10e1564 RM |
246 | also a C<$File::Find::fullname>. The function may set |
247 | C<$File::Find::prune> to prune the tree unless C<bydepth> was | |
248 | specified. Unless C<follow> or C<follow_fast> is specified, for | |
249 | compatibility reasons (find.pl, find2perl) there are in addition the | |
250 | following globals available: C<$File::Find::topdir>, | |
251 | C<$File::Find::topdev>, C<$File::Find::topino>, | |
e7b91b67 | 252 | C<$File::Find::topmode> and C<$File::Find::topnlink>. |
47a735e8 | 253 | |
20408e3c | 254 | This library is useful for the C<find2perl> tool, which when fed, |
f06db76b AD |
255 | |
256 | find2perl / -name .nfs\* -mtime +7 \ | |
81793b90 | 257 | -exec rm -f {} \; -o -fstype nfs -prune |
f06db76b AD |
258 | |
259 | produces something like: | |
260 | ||
261 | sub wanted { | |
c7b9dd21 | 262 | /^\.nfs.*\z/s && |
81793b90 | 263 | (($dev, $ino, $mode, $nlink, $uid, $gid) = lstat($_)) && |
f06db76b AD |
264 | int(-M _) > 7 && |
265 | unlink($_) | |
266 | || | |
81793b90 | 267 | ($nlink || (($dev, $ino, $mode, $nlink, $uid, $gid) = lstat($_))) && |
f06db76b | 268 | $dev < 0 && |
6280b799 | 269 | ($File::Find::prune = 1); |
f06db76b AD |
270 | } |
271 | ||
43dece2a JH |
272 | Notice the C<_> in the above C<int(-M _)>: the C<_> is a magical |
273 | filehandle that caches the information from the preceding | |
95e23d19 | 274 | C<stat()>, C<lstat()>, or filetest. |
43dece2a | 275 | |
1cffc1dd JH |
276 | Here's another interesting wanted function. It will find all symbolic |
277 | links that don't resolve: | |
f06db76b AD |
278 | |
279 | sub wanted { | |
81793b90 | 280 | -l && !-e && print "bogus link: $File::Find::name\n"; |
237437d0 | 281 | } |
f06db76b | 282 | |
c82ae0bc | 283 | Note that you may mix directories and (non-directory) files in the list of |
04cc0b0d | 284 | directories to be searched by the C<wanted()> function. |
c82ae0bc JK |
285 | |
286 | find(\&wanted, "./foo", "./bar", "./baz/epsilon"); | |
287 | ||
288 | In the example above, no file in F<./baz/> other than F<./baz/epsilon> will be | |
289 | evaluated by C<wanted()>. | |
290 | ||
81793b90 GS |
291 | See also the script C<pfind> on CPAN for a nice application of this |
292 | module. | |
293 | ||
cd68ec93 RGS |
294 | =head1 WARNINGS |
295 | ||
296 | If you run your program with the C<-w> switch, or if you use the | |
297 | C<warnings> pragma, File::Find will report warnings for several weird | |
298 | situations. You can disable these warnings by putting the statement | |
299 | ||
300 | no warnings 'File::Find'; | |
301 | ||
302 | in the appropriate scope. See L<perllexwarn> for more info about lexical | |
303 | warnings. | |
304 | ||
81793b90 GS |
305 | =head1 CAVEAT |
306 | ||
5fa2bf2b DD |
307 | =over 2 |
308 | ||
309 | =item $dont_use_nlink | |
310 | ||
311 | You can set the variable C<$File::Find::dont_use_nlink> to 1, if you want to | |
6cf3b067 T |
312 | force File::Find to always stat directories. This was used for file systems |
313 | that do not have an C<nlink> count matching the number of sub-directories. | |
314 | Examples are ISO-9660 (CD-ROM), AFS, HPFS (OS/2 file system), FAT (DOS file | |
315 | system) and a couple of others. | |
5fa2bf2b | 316 | |
6cf3b067 T |
317 | You shouldn't need to set this variable, since File::Find should now detect |
318 | such file systems on-the-fly and switch itself to using stat. This works even | |
319 | for parts of your file system, like a mounted CD-ROM. | |
5fa2bf2b | 320 | |
6cf3b067 | 321 | If you do set C<$File::Find::dont_use_nlink> to 1, you will notice slow-downs. |
5fa2bf2b DD |
322 | |
323 | =item symlinks | |
324 | ||
f10e1564 | 325 | Be aware that the option to follow symbolic links can be dangerous. |
81793b90 GS |
326 | Depending on the structure of the directory tree (including symbolic |
327 | links to directories) you might traverse a given (physical) directory | |
3555aed3 | 328 | more than once (only if C<follow_fast> is in effect). |
81793b90 GS |
329 | Furthermore, deleting or changing files in a symbolically linked directory |
330 | might cause very unpleasant surprises, since you delete or change files | |
331 | in an unknown directory. | |
0530a6c4 | 332 | |
5fa2bf2b DD |
333 | =back |
334 | ||
6eb87ff8 MJD |
335 | =head1 BUGS AND CAVEATS |
336 | ||
337 | Despite the name of the C<finddepth()> function, both C<find()> and | |
338 | C<finddepth()> perform a depth-first search of the directory | |
339 | hierarchy. | |
340 | ||
a85af077 A |
341 | =head1 HISTORY |
342 | ||
343 | File::Find used to produce incorrect results if called recursively. | |
344 | During the development of perl 5.8 this bug was fixed. | |
345 | The first fixed version of File::Find was 1.01. | |
346 | ||
9b33fb8e DJ |
347 | =head1 SEE ALSO |
348 | ||
349 | find, find2perl. | |
350 | ||
f06db76b AD |
351 | =cut |
352 | ||
b75c8c73 MS |
353 | our @ISA = qw(Exporter); |
354 | our @EXPORT = qw(find finddepth); | |
6280b799 | 355 | |
a0d0e21e | 356 | |
81793b90 GS |
357 | use strict; |
358 | my $Is_VMS; | |
a0b245d5 | 359 | my $Is_Win32; |
81793b90 GS |
360 | |
361 | require File::Basename; | |
7e47e6ff | 362 | require File::Spec; |
81793b90 | 363 | |
9f826d6a BM |
364 | # Should ideally be my() not our() but local() currently |
365 | # refuses to operate on lexicals | |
366 | ||
367 | our %SLnkSeen; | |
368 | our ($wanted_callback, $avoid_nlink, $bydepth, $no_chdir, $follow, | |
719c805e | 369 | $follow_skip, $full_check, $untaint, $untaint_skip, $untaint_pat, |
80e52b73 | 370 | $pre_process, $post_process, $dangling_symlinks); |
81793b90 GS |
371 | |
372 | sub contract_name { | |
373 | my ($cdir,$fn) = @_; | |
374 | ||
7e47e6ff | 375 | return substr($cdir,0,rindex($cdir,'/')) if $fn eq $File::Find::current_dir; |
81793b90 GS |
376 | |
377 | $cdir = substr($cdir,0,rindex($cdir,'/')+1); | |
378 | ||
379 | $fn =~ s|^\./||; | |
380 | ||
381 | my $abs_name= $cdir . $fn; | |
382 | ||
383 | if (substr($fn,0,3) eq '../') { | |
51393fc0 | 384 | 1 while $abs_name =~ s!/[^/]*/\.\./+!/!; |
81793b90 GS |
385 | } |
386 | ||
387 | return $abs_name; | |
388 | } | |
389 | ||
81793b90 GS |
390 | sub PathCombine($$) { |
391 | my ($Base,$Name) = @_; | |
392 | my $AbsName; | |
393 | ||
862f843b NC |
394 | if (substr($Name,0,1) eq '/') { |
395 | $AbsName= $Name; | |
81793b90 GS |
396 | } |
397 | else { | |
862f843b NC |
398 | $AbsName= contract_name($Base,$Name); |
399 | } | |
81793b90 | 400 | |
862f843b NC |
401 | # (simple) check for recursion |
402 | my $newlen= length($AbsName); | |
403 | if ($newlen <= length($Base)) { | |
404 | if (($newlen == length($Base) || substr($Base,$newlen,1) eq '/') | |
405 | && $AbsName eq substr($Base,0,$newlen)) | |
406 | { | |
407 | return undef; | |
81793b90 GS |
408 | } |
409 | } | |
410 | return $AbsName; | |
411 | } | |
412 | ||
413 | sub Follow_SymLink($) { | |
414 | my ($AbsName) = @_; | |
415 | ||
416 | my ($NewName,$DEV, $INO); | |
417 | ($DEV, $INO)= lstat $AbsName; | |
418 | ||
419 | while (-l _) { | |
420 | if ($SLnkSeen{$DEV, $INO}++) { | |
421 | if ($follow_skip < 2) { | |
422 | die "$AbsName is encountered a second time"; | |
a0d0e21e LW |
423 | } |
424 | else { | |
81793b90 | 425 | return undef; |
a0d0e21e LW |
426 | } |
427 | } | |
81793b90 GS |
428 | $NewName= PathCombine($AbsName, readlink($AbsName)); |
429 | unless(defined $NewName) { | |
430 | if ($follow_skip < 2) { | |
431 | die "$AbsName is a recursive symbolic link"; | |
432 | } | |
433 | else { | |
434 | return undef; | |
a0d0e21e | 435 | } |
81793b90 GS |
436 | } |
437 | else { | |
438 | $AbsName= $NewName; | |
439 | } | |
440 | ($DEV, $INO) = lstat($AbsName); | |
441 | return undef unless defined $DEV; # dangling symbolic link | |
442 | } | |
443 | ||
cd68ec93 | 444 | if ($full_check && defined $DEV && $SLnkSeen{$DEV, $INO}++) { |
7e47e6ff | 445 | if ( ($follow_skip < 1) || ((-d _) && ($follow_skip < 2)) ) { |
81793b90 GS |
446 | die "$AbsName encountered a second time"; |
447 | } | |
448 | else { | |
449 | return undef; | |
450 | } | |
451 | } | |
452 | ||
453 | return $AbsName; | |
454 | } | |
455 | ||
17f410f9 | 456 | our($dir, $name, $fullname, $prune); |
81793b90 GS |
457 | sub _find_dir_symlnk($$$); |
458 | sub _find_dir($$$); | |
459 | ||
7e47e6ff JH |
460 | # check whether or not a scalar variable is tainted |
461 | # (code straight from the Camel, 3rd ed., page 561) | |
462 | sub is_tainted_pp { | |
463 | my $arg = shift; | |
464 | my $nada = substr($arg, 0, 0); # zero-length | |
465 | local $@; | |
466 | eval { eval "# $nada" }; | |
467 | return length($@) != 0; | |
3555aed3 | 468 | } |
7e47e6ff | 469 | |
81793b90 GS |
470 | sub _find_opt { |
471 | my $wanted = shift; | |
472 | die "invalid top directory" unless defined $_[0]; | |
473 | ||
9f826d6a BM |
474 | # This function must local()ize everything because callbacks may |
475 | # call find() or finddepth() | |
476 | ||
477 | local %SLnkSeen; | |
478 | local ($wanted_callback, $avoid_nlink, $bydepth, $no_chdir, $follow, | |
479 | $follow_skip, $full_check, $untaint, $untaint_skip, $untaint_pat, | |
80e52b73 | 480 | $pre_process, $post_process, $dangling_symlinks); |
4c621faf | 481 | local($dir, $name, $fullname, $prune); |
bc125c03 | 482 | local *_ = \my $a; |
9f826d6a | 483 | |
a0c9c202 | 484 | my $cwd = $wanted->{bydepth} ? Cwd::fastcwd() : Cwd::getcwd(); |
a1ccf0c4 JM |
485 | if ($Is_VMS) { |
486 | # VMS returns this by default in VMS format which just doesn't | |
487 | # work for the rest of this module. | |
488 | $cwd = VMS::Filespec::unixpath($cwd); | |
489 | ||
490 | # Apparently this is not expected to have a trailing space. | |
491 | # To attempt to make VMS/UNIX conversions mostly reversable, | |
492 | # a trailing slash is needed. The run-time functions ignore the | |
493 | # resulting double slash, but it causes the perl tests to fail. | |
494 | $cwd =~ s#/\z##; | |
495 | ||
496 | # This comes up in upper case now, but should be lower. | |
497 | # In the future this could be exact case, no need to change. | |
498 | } | |
80e52b73 JH |
499 | my $cwd_untainted = $cwd; |
500 | my $check_t_cwd = 1; | |
501 | $wanted_callback = $wanted->{wanted}; | |
502 | $bydepth = $wanted->{bydepth}; | |
503 | $pre_process = $wanted->{preprocess}; | |
504 | $post_process = $wanted->{postprocess}; | |
505 | $no_chdir = $wanted->{no_chdir}; | |
a0b245d5 AD |
506 | $full_check = $Is_Win32 ? 0 : $wanted->{follow}; |
507 | $follow = $Is_Win32 ? 0 : | |
1bb17459 | 508 | $full_check || $wanted->{follow_fast}; |
80e52b73 JH |
509 | $follow_skip = $wanted->{follow_skip}; |
510 | $untaint = $wanted->{untaint}; | |
511 | $untaint_pat = $wanted->{untaint_pattern}; | |
512 | $untaint_skip = $wanted->{untaint_skip}; | |
513 | $dangling_symlinks = $wanted->{dangling_symlinks}; | |
81793b90 | 514 | |
1cffc1dd | 515 | # for compatibility reasons (find.pl, find2perl) |
9f826d6a | 516 | local our ($topdir, $topdev, $topino, $topmode, $topnlink); |
81793b90 GS |
517 | |
518 | # a symbolic link to a directory doesn't increase the link count | |
519 | $avoid_nlink = $follow || $File::Find::dont_use_nlink; | |
3555aed3 | 520 | |
e7b91b67 | 521 | my ($abs_dir, $Is_Dir); |
81793b90 GS |
522 | |
523 | Proc_Top_Item: | |
4c621faf | 524 | foreach my $TOP (@_) { |
7e47e6ff JH |
525 | my $top_item = $TOP; |
526 | ||
3555aed3 SP |
527 | ($topdev,$topino,$topmode,$topnlink) = $follow ? stat $top_item : lstat $top_item; |
528 | ||
a0b245d5 AD |
529 | if ($Is_Win32) { |
530 | $top_item =~ s|[/\\]\z|| | |
531 | unless $top_item =~ m{^(?:\w:)?[/\\]$}; | |
7e47e6ff JH |
532 | } |
533 | else { | |
534 | $top_item =~ s|/\z|| unless $top_item eq '/'; | |
7e47e6ff JH |
535 | } |
536 | ||
537 | $Is_Dir= 0; | |
538 | ||
539 | if ($follow) { | |
540 | ||
862f843b NC |
541 | if (substr($top_item,0,1) eq '/') { |
542 | $abs_dir = $top_item; | |
7e47e6ff | 543 | } |
862f843b NC |
544 | elsif ($top_item eq $File::Find::current_dir) { |
545 | $abs_dir = $cwd; | |
546 | } | |
547 | else { # care about any ../ | |
548 | $top_item =~ s/\.dir\z//i if $Is_VMS; | |
549 | $abs_dir = contract_name("$cwd/",$top_item); | |
7e47e6ff JH |
550 | } |
551 | $abs_dir= Follow_SymLink($abs_dir); | |
552 | unless (defined $abs_dir) { | |
80e52b73 JH |
553 | if ($dangling_symlinks) { |
554 | if (ref $dangling_symlinks eq 'CODE') { | |
555 | $dangling_symlinks->($top_item, $cwd); | |
556 | } else { | |
cd68ec93 | 557 | warnings::warnif "$top_item is a dangling symbolic link\n"; |
80e52b73 JH |
558 | } |
559 | } | |
81793b90 | 560 | next Proc_Top_Item; |
7e47e6ff JH |
561 | } |
562 | ||
563 | if (-d _) { | |
a1ccf0c4 | 564 | $top_item =~ s/\.dir\z//i if $Is_VMS; |
81793b90 GS |
565 | _find_dir_symlnk($wanted, $abs_dir, $top_item); |
566 | $Is_Dir= 1; | |
7e47e6ff JH |
567 | } |
568 | } | |
81793b90 | 569 | else { # no follow |
7e47e6ff JH |
570 | $topdir = $top_item; |
571 | unless (defined $topnlink) { | |
cd68ec93 | 572 | warnings::warnif "Can't stat $top_item: $!\n"; |
7e47e6ff JH |
573 | next Proc_Top_Item; |
574 | } | |
575 | if (-d _) { | |
544ff7a7 | 576 | $top_item =~ s/\.dir\z//i if $Is_VMS; |
e7b91b67 | 577 | _find_dir($wanted, $top_item, $topnlink); |
81793b90 | 578 | $Is_Dir= 1; |
7e47e6ff | 579 | } |
237437d0 | 580 | else { |
81793b90 | 581 | $abs_dir= $top_item; |
7e47e6ff JH |
582 | } |
583 | } | |
81793b90 | 584 | |
7e47e6ff | 585 | unless ($Is_Dir) { |
81793b90 | 586 | unless (($_,$dir) = File::Basename::fileparse($abs_dir)) { |
862f843b | 587 | ($dir,$_) = ('./', $top_item); |
81793b90 GS |
588 | } |
589 | ||
7e47e6ff JH |
590 | $abs_dir = $dir; |
591 | if (( $untaint ) && (is_tainted($dir) )) { | |
592 | ( $abs_dir ) = $dir =~ m|$untaint_pat|; | |
81793b90 GS |
593 | unless (defined $abs_dir) { |
594 | if ($untaint_skip == 0) { | |
7e47e6ff | 595 | die "directory $dir is still tainted"; |
81793b90 GS |
596 | } |
597 | else { | |
598 | next Proc_Top_Item; | |
599 | } | |
600 | } | |
7e47e6ff | 601 | } |
81793b90 | 602 | |
7e47e6ff | 603 | unless ($no_chdir || chdir $abs_dir) { |
cd68ec93 | 604 | warnings::warnif "Couldn't chdir $abs_dir: $!\n"; |
7e47e6ff JH |
605 | next Proc_Top_Item; |
606 | } | |
719911cc | 607 | |
7e47e6ff | 608 | $name = $abs_dir . $_; # $File::Find::name |
3bb6d3e5 | 609 | $_ = $name if $no_chdir; |
719911cc | 610 | |
abfdd623 | 611 | { $wanted_callback->() }; # protect against wild "next" |
81793b90 | 612 | |
7e47e6ff | 613 | } |
81793b90 | 614 | |
7e47e6ff JH |
615 | unless ( $no_chdir ) { |
616 | if ( ($check_t_cwd) && (($untaint) && (is_tainted($cwd) )) ) { | |
617 | ( $cwd_untainted ) = $cwd =~ m|$untaint_pat|; | |
618 | unless (defined $cwd_untainted) { | |
619 | die "insecure cwd in find(depth)"; | |
620 | } | |
621 | $check_t_cwd = 0; | |
622 | } | |
623 | unless (chdir $cwd_untainted) { | |
624 | die "Can't cd to $cwd: $!\n"; | |
625 | } | |
626 | } | |
81793b90 GS |
627 | } |
628 | } | |
629 | ||
630 | # API: | |
631 | # $wanted | |
632 | # $p_dir : "parent directory" | |
633 | # $nlink : what came back from the stat | |
634 | # preconditions: | |
635 | # chdir (if not no_chdir) to dir | |
636 | ||
637 | sub _find_dir($$$) { | |
638 | my ($wanted, $p_dir, $nlink) = @_; | |
639 | my ($CdLvl,$Level) = (0,0); | |
640 | my @Stack; | |
641 | my @filenames; | |
642 | my ($subcount,$sub_nlink); | |
643 | my $SE= []; | |
644 | my $dir_name= $p_dir; | |
7e47e6ff | 645 | my $dir_pref; |
39e79f6b | 646 | my $dir_rel = $File::Find::current_dir; |
7e47e6ff | 647 | my $tainted = 0; |
5fa2bf2b | 648 | my $no_nlink; |
7e47e6ff | 649 | |
a0b245d5 AD |
650 | if ($Is_Win32) { |
651 | $dir_pref | |
652 | = ($p_dir =~ m{^(?:\w:[/\\]?|[/\\])$} ? $p_dir : "$p_dir/" ); | |
653 | } elsif ($Is_VMS) { | |
a1ccf0c4 JM |
654 | |
655 | # VMS is returning trailing .dir on directories | |
656 | # and trailing . on files and symbolic links | |
657 | # in UNIX syntax. | |
658 | # | |
659 | ||
660 | $p_dir =~ s/\.(dir)?$//i unless $p_dir eq '.'; | |
661 | ||
1e9c9d75 | 662 | $dir_pref = ($p_dir =~ m/[\]>]+$/ ? $p_dir : "$p_dir/" ); |
7e47e6ff JH |
663 | } |
664 | else { | |
665 | $dir_pref= ( $p_dir eq '/' ? '/' : "$p_dir/" ); | |
7e47e6ff | 666 | } |
81793b90 GS |
667 | |
668 | local ($dir, $name, $prune, *DIR); | |
7e47e6ff JH |
669 | |
670 | unless ( $no_chdir || ($p_dir eq $File::Find::current_dir)) { | |
81793b90 | 671 | my $udir = $p_dir; |
7e47e6ff JH |
672 | if (( $untaint ) && (is_tainted($p_dir) )) { |
673 | ( $udir ) = $p_dir =~ m|$untaint_pat|; | |
81793b90 GS |
674 | unless (defined $udir) { |
675 | if ($untaint_skip == 0) { | |
676 | die "directory $p_dir is still tainted"; | |
677 | } | |
678 | else { | |
679 | return; | |
680 | } | |
237437d0 | 681 | } |
a0d0e21e | 682 | } |
8d8eebbf | 683 | unless (chdir ($Is_VMS && $udir !~ /[\/\[<]+/ ? "./$udir" : $udir)) { |
cd68ec93 | 684 | warnings::warnif "Can't cd to $udir: $!\n"; |
81793b90 GS |
685 | return; |
686 | } | |
687 | } | |
7e47e6ff JH |
688 | |
689 | # push the starting directory | |
57e73c4b | 690 | push @Stack,[$CdLvl,$p_dir,$dir_rel,-1] if $bydepth; |
81793b90 GS |
691 | |
692 | while (defined $SE) { | |
693 | unless ($bydepth) { | |
3555aed3 SP |
694 | $dir= $p_dir; # $File::Find::dir |
695 | $name= $dir_name; # $File::Find::name | |
7e47e6ff | 696 | $_= ($no_chdir ? $dir_name : $dir_rel ); # $_ |
81793b90 | 697 | # prune may happen here |
7e47e6ff | 698 | $prune= 0; |
abfdd623 | 699 | { $wanted_callback->() }; # protect against wild "next" |
7e47e6ff | 700 | next if $prune; |
81793b90 | 701 | } |
7e47e6ff | 702 | |
81793b90 | 703 | # change to that directory |
7e47e6ff | 704 | unless ($no_chdir || ($dir_rel eq $File::Find::current_dir)) { |
81793b90 | 705 | my $udir= $dir_rel; |
7e47e6ff JH |
706 | if ( ($untaint) && (($tainted) || ($tainted = is_tainted($dir_rel) )) ) { |
707 | ( $udir ) = $dir_rel =~ m|$untaint_pat|; | |
81793b90 GS |
708 | unless (defined $udir) { |
709 | if ($untaint_skip == 0) { | |
862f843b | 710 | die "directory (" . ($p_dir ne '/' ? $p_dir : '') . "/) $dir_rel is still tainted"; |
7e47e6ff | 711 | } else { # $untaint_skip == 1 |
3555aed3 | 712 | next; |
81793b90 GS |
713 | } |
714 | } | |
715 | } | |
8d8eebbf | 716 | unless (chdir ($Is_VMS && $udir !~ /[\/\[<]+/ ? "./$udir" : $udir)) { |
862f843b NC |
717 | warnings::warnif "Can't cd to (" . |
718 | ($p_dir ne '/' ? $p_dir : '') . "/) $udir: $!\n"; | |
81793b90 GS |
719 | next; |
720 | } | |
721 | $CdLvl++; | |
722 | } | |
723 | ||
3555aed3 | 724 | $dir= $dir_name; # $File::Find::dir |
81793b90 GS |
725 | |
726 | # Get the list of files in the current directory. | |
7e47e6ff | 727 | unless (opendir DIR, ($no_chdir ? $dir_name : $File::Find::current_dir)) { |
cd68ec93 | 728 | warnings::warnif "Can't opendir($dir_name): $!\n"; |
81793b90 GS |
729 | next; |
730 | } | |
731 | @filenames = readdir DIR; | |
732 | closedir(DIR); | |
abfdd623 | 733 | @filenames = $pre_process->(@filenames) if $pre_process; |
719c805e | 734 | push @Stack,[$CdLvl,$dir_name,"",-2] if $post_process; |
81793b90 | 735 | |
98dc9551 | 736 | # default: use whatever was specified |
5fa2bf2b DD |
737 | # (if $nlink >= 2, and $avoid_nlink == 0, this will switch back) |
738 | $no_nlink = $avoid_nlink; | |
739 | # if dir has wrong nlink count, force switch to slower stat method | |
740 | $no_nlink = 1 if ($nlink < 2); | |
741 | ||
742 | if ($nlink == 2 && !$no_nlink) { | |
81793b90 GS |
743 | # This dir has no subdirectories. |
744 | for my $FN (@filenames) { | |
a1ccf0c4 JM |
745 | if ($Is_VMS) { |
746 | # Big hammer here - Compensate for VMS trailing . and .dir | |
747 | # No win situation until this is changed, but this | |
748 | # will handle the majority of the cases with breaking the fewest | |
749 | ||
750 | $FN =~ s/\.dir\z//i; | |
751 | $FN =~ s#\.$## if ($FN ne '.'); | |
752 | } | |
7e47e6ff | 753 | next if $FN =~ $File::Find::skip_pattern; |
81793b90 | 754 | |
7e47e6ff JH |
755 | $name = $dir_pref . $FN; # $File::Find::name |
756 | $_ = ($no_chdir ? $name : $FN); # $_ | |
abfdd623 | 757 | { $wanted_callback->() }; # protect against wild "next" |
81793b90 GS |
758 | } |
759 | ||
760 | } | |
761 | else { | |
762 | # This dir has subdirectories. | |
763 | $subcount = $nlink - 2; | |
764 | ||
7bd31527 JH |
765 | # HACK: insert directories at this position. so as to preserve |
766 | # the user pre-processed ordering of files. | |
767 | # EG: directory traversal is in user sorted order, not at random. | |
768 | my $stack_top = @Stack; | |
769 | ||
81793b90 | 770 | for my $FN (@filenames) { |
7e47e6ff | 771 | next if $FN =~ $File::Find::skip_pattern; |
5fa2bf2b | 772 | if ($subcount > 0 || $no_nlink) { |
81793b90 GS |
773 | # Seen all the subdirs? |
774 | # check for directoriness. | |
775 | # stat is faster for a file in the current directory | |
07867069 | 776 | $sub_nlink = (lstat ($no_chdir ? $dir_pref . $FN : $FN))[3]; |
81793b90 GS |
777 | |
778 | if (-d _) { | |
779 | --$subcount; | |
544ff7a7 | 780 | $FN =~ s/\.dir\z//i if $Is_VMS; |
7bd31527 JH |
781 | # HACK: replace push to preserve dir traversal order |
782 | #push @Stack,[$CdLvl,$dir_name,$FN,$sub_nlink]; | |
783 | splice @Stack, $stack_top, 0, | |
784 | [$CdLvl,$dir_name,$FN,$sub_nlink]; | |
81793b90 GS |
785 | } |
786 | else { | |
7e47e6ff JH |
787 | $name = $dir_pref . $FN; # $File::Find::name |
788 | $_= ($no_chdir ? $name : $FN); # $_ | |
abfdd623 | 789 | { $wanted_callback->() }; # protect against wild "next" |
81793b90 GS |
790 | } |
791 | } | |
07867069 | 792 | else { |
7e47e6ff JH |
793 | $name = $dir_pref . $FN; # $File::Find::name |
794 | $_= ($no_chdir ? $name : $FN); # $_ | |
abfdd623 | 795 | { $wanted_callback->() }; # protect against wild "next" |
81793b90 GS |
796 | } |
797 | } | |
798 | } | |
17b275ff RA |
799 | } |
800 | continue { | |
57e73c4b | 801 | while ( defined ($SE = pop @Stack) ) { |
81793b90 GS |
802 | ($Level, $p_dir, $dir_rel, $nlink) = @$SE; |
803 | if ($CdLvl > $Level && !$no_chdir) { | |
7e47e6ff | 804 | my $tmp; |
862f843b | 805 | if ($Is_VMS) { |
d8101854 CB |
806 | $tmp = '[' . ('-' x ($CdLvl-$Level)) . ']'; |
807 | } | |
7e47e6ff JH |
808 | else { |
809 | $tmp = join('/',('..') x ($CdLvl-$Level)); | |
810 | } | |
d8101854 | 811 | die "Can't cd to $tmp from $dir_name" |
7e47e6ff | 812 | unless chdir ($tmp); |
81793b90 GS |
813 | $CdLvl = $Level; |
814 | } | |
7e47e6ff | 815 | |
b242981a AD |
816 | if ($Is_Win32) { |
817 | $dir_name = ($p_dir =~ m{^(?:\w:[/\\]?|[/\\])$} | |
818 | ? "$p_dir$dir_rel" : "$p_dir/$dir_rel"); | |
3555aed3 SP |
819 | $dir_pref = "$dir_name/"; |
820 | } | |
1e9c9d75 CB |
821 | elsif ($^O eq 'VMS') { |
822 | if ($p_dir =~ m/[\]>]+$/) { | |
823 | $dir_name = $p_dir; | |
824 | $dir_name =~ s/([\]>]+)$/.$dir_rel$1/; | |
825 | $dir_pref = $dir_name; | |
826 | } | |
827 | else { | |
828 | $dir_name = "$p_dir/$dir_rel"; | |
829 | $dir_pref = "$dir_name/"; | |
830 | } | |
831 | } | |
7e47e6ff JH |
832 | else { |
833 | $dir_name = ($p_dir eq '/' ? "/$dir_rel" : "$p_dir/$dir_rel"); | |
834 | $dir_pref = "$dir_name/"; | |
835 | } | |
836 | ||
719c805e | 837 | if ( $nlink == -2 ) { |
7e47e6ff | 838 | $name = $dir = $p_dir; # $File::Find::name / dir |
39e79f6b | 839 | $_ = $File::Find::current_dir; |
abfdd623 | 840 | $post_process->(); # End-of-directory processing |
7e47e6ff JH |
841 | } |
842 | elsif ( $nlink < 0 ) { # must be finddepth, report dirname now | |
843 | $name = $dir_name; | |
862f843b NC |
844 | if ( substr($name,-2) eq '/.' ) { |
845 | substr($name, length($name) == 2 ? -1 : -2) = ''; | |
7e47e6ff | 846 | } |
862f843b NC |
847 | $dir = $p_dir; |
848 | $_ = ($no_chdir ? $dir_name : $dir_rel ); | |
849 | if ( substr($_,-2) eq '/.' ) { | |
850 | substr($_, length($_) == 2 ? -1 : -2) = ''; | |
7e47e6ff | 851 | } |
abfdd623 | 852 | { $wanted_callback->() }; # protect against wild "next" |
7e47e6ff JH |
853 | } |
854 | else { | |
855 | push @Stack,[$CdLvl,$p_dir,$dir_rel,-1] if $bydepth; | |
856 | last; | |
857 | } | |
81793b90 | 858 | } |
a0d0e21e LW |
859 | } |
860 | } | |
861 | ||
81793b90 GS |
862 | |
863 | # API: | |
864 | # $wanted | |
865 | # $dir_loc : absolute location of a dir | |
866 | # $p_dir : "parent directory" | |
867 | # preconditions: | |
868 | # chdir (if not no_chdir) to dir | |
869 | ||
870 | sub _find_dir_symlnk($$$) { | |
7e47e6ff | 871 | my ($wanted, $dir_loc, $p_dir) = @_; # $dir_loc is the absolute directory |
81793b90 GS |
872 | my @Stack; |
873 | my @filenames; | |
874 | my $new_loc; | |
7e47e6ff | 875 | my $updir_loc = $dir_loc; # untainted parent directory |
81793b90 GS |
876 | my $SE = []; |
877 | my $dir_name = $p_dir; | |
7e47e6ff JH |
878 | my $dir_pref; |
879 | my $loc_pref; | |
39e79f6b | 880 | my $dir_rel = $File::Find::current_dir; |
7e47e6ff JH |
881 | my $byd_flag; # flag for pending stack entry if $bydepth |
882 | my $tainted = 0; | |
883 | my $ok = 1; | |
884 | ||
862f843b NC |
885 | $dir_pref = ( $p_dir eq '/' ? '/' : "$p_dir/" ); |
886 | $loc_pref = ( $dir_loc eq '/' ? '/' : "$dir_loc/" ); | |
81793b90 GS |
887 | |
888 | local ($dir, $name, $fullname, $prune, *DIR); | |
7e47e6ff JH |
889 | |
890 | unless ($no_chdir) { | |
891 | # untaint the topdir | |
892 | if (( $untaint ) && (is_tainted($dir_loc) )) { | |
893 | ( $updir_loc ) = $dir_loc =~ m|$untaint_pat|; # parent dir, now untainted | |
894 | # once untainted, $updir_loc is pushed on the stack (as parent directory); | |
3555aed3 SP |
895 | # hence, we don't need to untaint the parent directory every time we chdir |
896 | # to it later | |
7e47e6ff | 897 | unless (defined $updir_loc) { |
81793b90 GS |
898 | if ($untaint_skip == 0) { |
899 | die "directory $dir_loc is still tainted"; | |
900 | } | |
901 | else { | |
902 | return; | |
903 | } | |
904 | } | |
905 | } | |
7e47e6ff JH |
906 | $ok = chdir($updir_loc) unless ($p_dir eq $File::Find::current_dir); |
907 | unless ($ok) { | |
cd68ec93 | 908 | warnings::warnif "Can't cd to $updir_loc: $!\n"; |
81793b90 GS |
909 | return; |
910 | } | |
911 | } | |
912 | ||
7e47e6ff JH |
913 | push @Stack,[$dir_loc,$updir_loc,$p_dir,$dir_rel,-1] if $bydepth; |
914 | ||
81793b90 GS |
915 | while (defined $SE) { |
916 | ||
917 | unless ($bydepth) { | |
7e47e6ff | 918 | # change (back) to parent directory (always untainted) |
704ea872 | 919 | unless ($no_chdir) { |
7e47e6ff | 920 | unless (chdir $updir_loc) { |
cd68ec93 | 921 | warnings::warnif "Can't cd to $updir_loc: $!\n"; |
704ea872 GS |
922 | next; |
923 | } | |
924 | } | |
7e47e6ff JH |
925 | $dir= $p_dir; # $File::Find::dir |
926 | $name= $dir_name; # $File::Find::name | |
927 | $_= ($no_chdir ? $dir_name : $dir_rel ); # $_ | |
928 | $fullname= $dir_loc; # $File::Find::fullname | |
81793b90 | 929 | # prune may happen here |
7e47e6ff | 930 | $prune= 0; |
704ea872 | 931 | lstat($_); # make sure file tests with '_' work |
abfdd623 | 932 | { $wanted_callback->() }; # protect against wild "next" |
7e47e6ff | 933 | next if $prune; |
81793b90 GS |
934 | } |
935 | ||
936 | # change to that directory | |
7e47e6ff JH |
937 | unless ($no_chdir || ($dir_rel eq $File::Find::current_dir)) { |
938 | $updir_loc = $dir_loc; | |
939 | if ( ($untaint) && (($tainted) || ($tainted = is_tainted($dir_loc) )) ) { | |
3555aed3 | 940 | # untaint $dir_loc, what will be pushed on the stack as (untainted) parent dir |
7e47e6ff JH |
941 | ( $updir_loc ) = $dir_loc =~ m|$untaint_pat|; |
942 | unless (defined $updir_loc) { | |
81793b90 GS |
943 | if ($untaint_skip == 0) { |
944 | die "directory $dir_loc is still tainted"; | |
a0d0e21e | 945 | } |
237437d0 | 946 | else { |
81793b90 | 947 | next; |
237437d0 | 948 | } |
a0d0e21e LW |
949 | } |
950 | } | |
7e47e6ff | 951 | unless (chdir $updir_loc) { |
cd68ec93 | 952 | warnings::warnif "Can't cd to $updir_loc: $!\n"; |
81793b90 GS |
953 | next; |
954 | } | |
955 | } | |
956 | ||
7e47e6ff | 957 | $dir = $dir_name; # $File::Find::dir |
81793b90 GS |
958 | |
959 | # Get the list of files in the current directory. | |
7e47e6ff | 960 | unless (opendir DIR, ($no_chdir ? $dir_loc : $File::Find::current_dir)) { |
cd68ec93 | 961 | warnings::warnif "Can't opendir($dir_loc): $!\n"; |
81793b90 GS |
962 | next; |
963 | } | |
964 | @filenames = readdir DIR; | |
965 | closedir(DIR); | |
966 | ||
967 | for my $FN (@filenames) { | |
a1ccf0c4 JM |
968 | if ($Is_VMS) { |
969 | # Big hammer here - Compensate for VMS trailing . and .dir | |
970 | # No win situation until this is changed, but this | |
971 | # will handle the majority of the cases with breaking the fewest. | |
972 | ||
973 | $FN =~ s/\.dir\z//i; | |
974 | $FN =~ s#\.$## if ($FN ne '.'); | |
975 | } | |
7e47e6ff | 976 | next if $FN =~ $File::Find::skip_pattern; |
81793b90 GS |
977 | |
978 | # follow symbolic links / do an lstat | |
07867069 | 979 | $new_loc = Follow_SymLink($loc_pref.$FN); |
81793b90 GS |
980 | |
981 | # ignore if invalid symlink | |
3555aed3 | 982 | unless (defined $new_loc) { |
fab43c1b | 983 | if (!defined -l _ && $dangling_symlinks) { |
615a2b9b SP |
984 | if (ref $dangling_symlinks eq 'CODE') { |
985 | $dangling_symlinks->($FN, $dir_pref); | |
986 | } else { | |
987 | warnings::warnif "$dir_pref$FN is a dangling symbolic link\n"; | |
988 | } | |
989 | } | |
990 | ||
991 | $fullname = undef; | |
992 | $name = $dir_pref . $FN; | |
993 | $_ = ($no_chdir ? $name : $FN); | |
994 | { $wanted_callback->() }; | |
995 | next; | |
996 | } | |
7e47e6ff | 997 | |
81793b90 | 998 | if (-d _) { |
a1ccf0c4 JM |
999 | if ($Is_VMS) { |
1000 | $FN =~ s/\.dir\z//i; | |
1001 | $FN =~ s#\.$## if ($FN ne '.'); | |
1002 | $new_loc =~ s/\.dir\z//i; | |
1003 | $new_loc =~ s#\.$## if ($new_loc ne '.'); | |
1004 | } | |
7e47e6ff | 1005 | push @Stack,[$new_loc,$updir_loc,$dir_name,$FN,1]; |
81793b90 GS |
1006 | } |
1007 | else { | |
3555aed3 | 1008 | $fullname = $new_loc; # $File::Find::fullname |
7e47e6ff JH |
1009 | $name = $dir_pref . $FN; # $File::Find::name |
1010 | $_ = ($no_chdir ? $name : $FN); # $_ | |
abfdd623 | 1011 | { $wanted_callback->() }; # protect against wild "next" |
81793b90 GS |
1012 | } |
1013 | } | |
1014 | ||
81793b90 GS |
1015 | } |
1016 | continue { | |
57e73c4b | 1017 | while (defined($SE = pop @Stack)) { |
7e47e6ff | 1018 | ($dir_loc, $updir_loc, $p_dir, $dir_rel, $byd_flag) = @$SE; |
862f843b NC |
1019 | $dir_name = ($p_dir eq '/' ? "/$dir_rel" : "$p_dir/$dir_rel"); |
1020 | $dir_pref = "$dir_name/"; | |
1021 | $loc_pref = "$dir_loc/"; | |
7e47e6ff JH |
1022 | if ( $byd_flag < 0 ) { # must be finddepth, report dirname now |
1023 | unless ($no_chdir || ($dir_rel eq $File::Find::current_dir)) { | |
3555aed3 | 1024 | unless (chdir $updir_loc) { # $updir_loc (parent dir) is always untainted |
cd68ec93 | 1025 | warnings::warnif "Can't cd to $updir_loc: $!\n"; |
7e47e6ff JH |
1026 | next; |
1027 | } | |
1028 | } | |
1029 | $fullname = $dir_loc; # $File::Find::fullname | |
1030 | $name = $dir_name; # $File::Find::name | |
862f843b NC |
1031 | if ( substr($name,-2) eq '/.' ) { |
1032 | substr($name, length($name) == 2 ? -1 : -2) = ''; # $File::Find::name | |
7e47e6ff | 1033 | } |
862f843b NC |
1034 | $dir = $p_dir; # $File::Find::dir |
1035 | $_ = ($no_chdir ? $dir_name : $dir_rel); # $_ | |
1036 | if ( substr($_,-2) eq '/.' ) { | |
1037 | substr($_, length($_) == 2 ? -1 : -2) = ''; | |
7e47e6ff JH |
1038 | } |
1039 | ||
1040 | lstat($_); # make sure file tests with '_' work | |
abfdd623 | 1041 | { $wanted_callback->() }; # protect against wild "next" |
7e47e6ff JH |
1042 | } |
1043 | else { | |
1044 | push @Stack,[$dir_loc, $updir_loc, $p_dir, $dir_rel,-1] if $bydepth; | |
1045 | last; | |
1046 | } | |
a0d0e21e LW |
1047 | } |
1048 | } | |
1049 | } | |
1050 | ||
81793b90 | 1051 | |
20408e3c | 1052 | sub wrap_wanted { |
81793b90 GS |
1053 | my $wanted = shift; |
1054 | if ( ref($wanted) eq 'HASH' ) { | |
4c90698d RB |
1055 | unless( exists $wanted->{wanted} and ref( $wanted->{wanted} ) eq 'CODE' ) { |
1056 | die 'no &wanted subroutine given'; | |
1057 | } | |
81793b90 GS |
1058 | if ( $wanted->{follow} || $wanted->{follow_fast}) { |
1059 | $wanted->{follow_skip} = 1 unless defined $wanted->{follow_skip}; | |
1060 | } | |
1061 | if ( $wanted->{untaint} ) { | |
3555aed3 | 1062 | $wanted->{untaint_pattern} = $File::Find::untaint_pattern |
81793b90 GS |
1063 | unless defined $wanted->{untaint_pattern}; |
1064 | $wanted->{untaint_skip} = 0 unless defined $wanted->{untaint_skip}; | |
1065 | } | |
1066 | return $wanted; | |
1067 | } | |
4c90698d | 1068 | elsif( ref( $wanted ) eq 'CODE' ) { |
81793b90 GS |
1069 | return { wanted => $wanted }; |
1070 | } | |
4c90698d RB |
1071 | else { |
1072 | die 'no &wanted subroutine given'; | |
1073 | } | |
a0d0e21e LW |
1074 | } |
1075 | ||
20408e3c | 1076 | sub find { |
81793b90 GS |
1077 | my $wanted = shift; |
1078 | _find_opt(wrap_wanted($wanted), @_); | |
a0d0e21e LW |
1079 | } |
1080 | ||
55d729e4 | 1081 | sub finddepth { |
81793b90 GS |
1082 | my $wanted = wrap_wanted(shift); |
1083 | $wanted->{bydepth} = 1; | |
1084 | _find_opt($wanted, @_); | |
20408e3c | 1085 | } |
6280b799 | 1086 | |
7e47e6ff JH |
1087 | # default |
1088 | $File::Find::skip_pattern = qr/^\.{1,2}\z/; | |
1089 | $File::Find::untaint_pattern = qr|^([-+@\w./]+)$|; | |
1090 | ||
6280b799 | 1091 | # These are hard-coded for now, but may move to hint files. |
10eba763 | 1092 | if ($^O eq 'VMS') { |
81793b90 | 1093 | $Is_VMS = 1; |
7e47e6ff JH |
1094 | $File::Find::dont_use_nlink = 1; |
1095 | } | |
b242981a AD |
1096 | elsif ($^O eq 'MSWin32') { |
1097 | $Is_Win32 = 1; | |
1098 | } | |
748a9306 | 1099 | |
7e47e6ff JH |
1100 | # this _should_ work properly on all platforms |
1101 | # where File::Find can be expected to work | |
1102 | $File::Find::current_dir = File::Spec->curdir || '.'; | |
1103 | ||
81793b90 | 1104 | $File::Find::dont_use_nlink = 1 |
b242981a | 1105 | if $^O eq 'os2' || $^O eq 'dos' || $^O eq 'amigaos' || $Is_Win32 || |
0c52c6a9 | 1106 | $^O eq 'interix' || $^O eq 'cygwin' || $^O eq 'epoc' || $^O eq 'qnx' || |
1119cb72 | 1107 | $^O eq 'nto'; |
6280b799 | 1108 | |
20408e3c GS |
1109 | # Set dont_use_nlink in your hint file if your system's stat doesn't |
1110 | # report the number of links in a directory as an indication | |
1111 | # of the number of files. | |
1112 | # See, e.g. hints/machten.sh for MachTen 2.2. | |
81793b90 GS |
1113 | unless ($File::Find::dont_use_nlink) { |
1114 | require Config; | |
1115 | $File::Find::dont_use_nlink = 1 if ($Config::Config{'dont_use_nlink'}); | |
20408e3c GS |
1116 | } |
1117 | ||
3555aed3 SP |
1118 | # We need a function that checks if a scalar is tainted. Either use the |
1119 | # Scalar::Util module's tainted() function or our (slower) pure Perl | |
7e47e6ff JH |
1120 | # fallback is_tainted_pp() |
1121 | { | |
1122 | local $@; | |
1123 | eval { require Scalar::Util }; | |
1124 | *is_tainted = $@ ? \&is_tainted_pp : \&Scalar::Util::tainted; | |
1125 | } | |
1126 | ||
a0d0e21e | 1127 | 1; |