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; |
c00d3472 |
6 | our $VERSION = '1.06'; |
a0d0e21e |
7 | require Exporter; |
6280b799 |
8 | require Cwd; |
a0d0e21e |
9 | |
7bd31527 |
10 | # |
11 | # Modified to ensure sub-directory traversal order is not inverded by stack |
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 |
16 | =head1 NAME |
17 | |
abfdd623 |
18 | File::Find - Traverse a directory tree. |
f06db76b |
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 |
30 | use File::Find; |
31 | find({ wanted => \&process, follow => 1 }, '.'); |
f06db76b |
32 | |
33 | =head1 DESCRIPTION |
34 | |
abfdd623 |
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 |
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 |
53 | |
54 | =item B<finddepth> |
55 | |
56 | finddepth(\&wanted, @directories); |
57 | finddepth(\%options, @directories); |
58 | |
6eb87ff8 |
59 | C<finddepth()> works just like C<find()> except that is invokes the |
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 |
64 | |
65 | =back |
66 | |
67 | =head2 %options |
68 | |
95e23d19 |
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 |
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 |
81 | described in L<The wanted function> below. |
81793b90 |
82 | |
83 | =item C<bydepth> |
84 | |
85 | Reports the name of a directory only AFTER all its entries |
95e23d19 |
86 | have been reported. Entry point C<finddepth()> is a shortcut for |
87 | specifying C<<{ bydepth => 1 }>> in the first argument of C<find()>. |
81793b90 |
88 | |
719c805e |
89 | =item C<preprocess> |
90 | |
7e47e6ff |
91 | The value should be a code reference. This code reference is used to |
95e23d19 |
92 | preprocess the current directory. The name of the currently processed |
93 | directory is in C<$File::Find::dir>. Your preprocessing function is |
94 | called after C<readdir()>, but before the loop that calls the C<wanted()> |
7e47e6ff |
95 | function. It is called with a list of strings (actually file/directory |
96 | names) and is expected to return a list of strings. The code can be |
97 | used to sort the file/directory names alphabetically, numerically, |
98 | or to filter out directory entries based on their name alone. When |
99 | I<follow> or I<follow_fast> are in effect, C<preprocess> is a no-op. |
719c805e |
100 | |
101 | =item C<postprocess> |
102 | |
7e47e6ff |
103 | The value should be a code reference. It is invoked just before leaving |
104 | the currently processed directory. It is called in void context with no |
95e23d19 |
105 | arguments. The name of the current directory is in C<$File::Find::dir>. This |
7e47e6ff |
106 | hook is handy for summarizing a directory, such as calculating its disk |
3fa6e24b |
107 | usage. When I<follow> or I<follow_fast> are in effect, C<postprocess> is a |
7e47e6ff |
108 | no-op. |
719c805e |
109 | |
81793b90 |
110 | =item C<follow> |
111 | |
112 | Causes symbolic links to be followed. Since directory trees with symbolic |
113 | links (followed) may contain files more than once and may even have |
114 | cycles, a hash has to be built up with an entry for each file. |
115 | This might be expensive both in space and time for a large |
116 | directory tree. See I<follow_fast> and I<follow_skip> below. |
117 | If either I<follow> or I<follow_fast> is in effect: |
118 | |
119 | =over 6 |
120 | |
a45bd81d |
121 | =item * |
81793b90 |
122 | |
f10e1564 |
123 | It is guaranteed that an I<lstat> has been called before the user's |
95e23d19 |
124 | C<wanted()> function is called. This enables fast file checks involving S< _>. |
81793b90 |
125 | |
a45bd81d |
126 | =item * |
81793b90 |
127 | |
128 | There is a variable C<$File::Find::fullname> which holds the absolute |
129 | pathname of the file with all symbolic links resolved |
130 | |
131 | =back |
132 | |
133 | =item C<follow_fast> |
134 | |
f10e1564 |
135 | This is similar to I<follow> except that it may report some files more |
136 | than once. It does detect cycles, however. Since only symbolic links |
137 | have to be hashed, this is much cheaper both in space and time. If |
95e23d19 |
138 | processing a file more than once (by the user's C<wanted()> function) |
81793b90 |
139 | is worse than just taking time, the option I<follow> should be used. |
140 | |
141 | =item C<follow_skip> |
142 | |
143 | C<follow_skip==1>, which is the default, causes all files which are |
144 | neither directories nor symbolic links to be ignored if they are about |
145 | to be processed a second time. If a directory or a symbolic link |
146 | are about to be processed a second time, File::Find dies. |
95e23d19 |
147 | |
81793b90 |
148 | C<follow_skip==0> causes File::Find to die if any file is about to be |
149 | processed a second time. |
95e23d19 |
150 | |
81793b90 |
151 | C<follow_skip==2> causes File::Find to ignore any duplicate files and |
7e47e6ff |
152 | directories but to proceed normally otherwise. |
20408e3c |
153 | |
80e52b73 |
154 | =item C<dangling_symlinks> |
155 | |
156 | If true and a code reference, will be called with the symbolic link |
157 | name and the directory it lives in as arguments. Otherwise, if true |
158 | and warnings are on, warning "symbolic_link_name is a dangling |
159 | symbolic link\n" will be issued. If false, the dangling symbolic link |
160 | will be silently ignored. |
f06db76b |
161 | |
81793b90 |
162 | =item C<no_chdir> |
163 | |
95e23d19 |
164 | Does not C<chdir()> to each directory as it recurses. The C<wanted()> |
81793b90 |
165 | function will need to be aware of this, of course. In this case, |
166 | C<$_> will be the same as C<$File::Find::name>. |
167 | |
168 | =item C<untaint> |
169 | |
170 | If find is used in taint-mode (-T command line switch or if EUID != UID |
171 | or if EGID != GID) then internally directory names have to be untainted |
7e47e6ff |
172 | before they can be chdir'ed to. Therefore they are checked against a regular |
173 | expression I<untaint_pattern>. Note that all names passed to the user's |
174 | I<wanted()> function are still tainted. If this option is used while |
175 | not in taint-mode, C<untaint> is a no-op. |
81793b90 |
176 | |
177 | =item C<untaint_pattern> |
178 | |
179 | See above. This should be set using the C<qr> quoting operator. |
180 | The default is set to C<qr|^([-+@\w./]+)$|>. |
1cffc1dd |
181 | Note that the parentheses are vital. |
81793b90 |
182 | |
183 | =item C<untaint_skip> |
184 | |
7e47e6ff |
185 | If set, a directory which fails the I<untaint_pattern> is skipped, |
186 | including all its sub-directories. The default is to 'die' in such a case. |
81793b90 |
187 | |
188 | =back |
189 | |
abfdd623 |
190 | =head2 The wanted function |
191 | |
95e23d19 |
192 | The C<wanted()> function does whatever verifications you want on |
193 | each file and directory. Note that despite its name, the C<wanted()> |
194 | function is a generic callback function, and does B<not> tell |
195 | File::Find if a file is "wanted" or not. In fact, its return value |
196 | is ignored. |
197 | |
198 | The wanted function takes no arguments but rather does its work |
abfdd623 |
199 | through a collection of variables. |
200 | |
201 | =over 4 |
202 | |
f837ebe2 |
203 | =item C<$File::Find::dir> is the current directory name, |
abfdd623 |
204 | |
205 | =item C<$_> is the current filename within that directory |
206 | |
f837ebe2 |
207 | =item C<$File::Find::name> is the complete pathname to the file. |
abfdd623 |
208 | |
209 | =back |
210 | |
f837ebe2 |
211 | Don't modify these variables. |
212 | |
95e23d19 |
213 | For example, when examining the file F</some/path/foo.ext> you will have: |
abfdd623 |
214 | |
215 | $File::Find::dir = /some/path/ |
216 | $_ = foo.ext |
217 | $File::Find::name = /some/path/foo.ext |
218 | |
219 | You are chdir()'d toC<$File::Find::dir> when the function is called, |
220 | unless C<no_chdir> was specified. Note that when changing to |
221 | directories is in effect the root directory (F</>) is a somewhat |
222 | special case inasmuch as the concatenation of C<$File::Find::dir>, |
223 | C<'/'> and C<$_> is not literally equal to C<$File::Find::name>. The |
224 | table below summarizes all variants: |
5cf0a2f2 |
225 | |
226 | $File::Find::name $File::Find::dir $_ |
227 | default / / . |
228 | no_chdir=>0 /etc / etc |
229 | /etc/x /etc x |
abfdd623 |
230 | |
5cf0a2f2 |
231 | no_chdir=>1 / / / |
232 | /etc / /etc |
233 | /etc/x /etc /etc/x |
234 | |
235 | |
236 | When <follow> or <follow_fast> are in effect, there is |
f10e1564 |
237 | also a C<$File::Find::fullname>. The function may set |
238 | C<$File::Find::prune> to prune the tree unless C<bydepth> was |
239 | specified. Unless C<follow> or C<follow_fast> is specified, for |
240 | compatibility reasons (find.pl, find2perl) there are in addition the |
241 | following globals available: C<$File::Find::topdir>, |
242 | C<$File::Find::topdev>, C<$File::Find::topino>, |
e7b91b67 |
243 | C<$File::Find::topmode> and C<$File::Find::topnlink>. |
47a735e8 |
244 | |
20408e3c |
245 | This library is useful for the C<find2perl> tool, which when fed, |
f06db76b |
246 | |
247 | find2perl / -name .nfs\* -mtime +7 \ |
81793b90 |
248 | -exec rm -f {} \; -o -fstype nfs -prune |
f06db76b |
249 | |
250 | produces something like: |
251 | |
252 | sub wanted { |
c7b9dd21 |
253 | /^\.nfs.*\z/s && |
81793b90 |
254 | (($dev, $ino, $mode, $nlink, $uid, $gid) = lstat($_)) && |
f06db76b |
255 | int(-M _) > 7 && |
256 | unlink($_) |
257 | || |
81793b90 |
258 | ($nlink || (($dev, $ino, $mode, $nlink, $uid, $gid) = lstat($_))) && |
f06db76b |
259 | $dev < 0 && |
6280b799 |
260 | ($File::Find::prune = 1); |
f06db76b |
261 | } |
262 | |
43dece2a |
263 | Notice the C<_> in the above C<int(-M _)>: the C<_> is a magical |
264 | filehandle that caches the information from the preceding |
95e23d19 |
265 | C<stat()>, C<lstat()>, or filetest. |
43dece2a |
266 | |
1cffc1dd |
267 | Here's another interesting wanted function. It will find all symbolic |
268 | links that don't resolve: |
f06db76b |
269 | |
270 | sub wanted { |
81793b90 |
271 | -l && !-e && print "bogus link: $File::Find::name\n"; |
237437d0 |
272 | } |
f06db76b |
273 | |
81793b90 |
274 | See also the script C<pfind> on CPAN for a nice application of this |
275 | module. |
276 | |
cd68ec93 |
277 | =head1 WARNINGS |
278 | |
279 | If you run your program with the C<-w> switch, or if you use the |
280 | C<warnings> pragma, File::Find will report warnings for several weird |
281 | situations. You can disable these warnings by putting the statement |
282 | |
283 | no warnings 'File::Find'; |
284 | |
285 | in the appropriate scope. See L<perllexwarn> for more info about lexical |
286 | warnings. |
287 | |
81793b90 |
288 | =head1 CAVEAT |
289 | |
5fa2bf2b |
290 | =over 2 |
291 | |
292 | =item $dont_use_nlink |
293 | |
294 | You can set the variable C<$File::Find::dont_use_nlink> to 1, if you want to |
6cf3b067 |
295 | force File::Find to always stat directories. This was used for file systems |
296 | that do not have an C<nlink> count matching the number of sub-directories. |
297 | Examples are ISO-9660 (CD-ROM), AFS, HPFS (OS/2 file system), FAT (DOS file |
298 | system) and a couple of others. |
5fa2bf2b |
299 | |
6cf3b067 |
300 | You shouldn't need to set this variable, since File::Find should now detect |
301 | such file systems on-the-fly and switch itself to using stat. This works even |
302 | for parts of your file system, like a mounted CD-ROM. |
5fa2bf2b |
303 | |
6cf3b067 |
304 | If you do set C<$File::Find::dont_use_nlink> to 1, you will notice slow-downs. |
5fa2bf2b |
305 | |
306 | =item symlinks |
307 | |
f10e1564 |
308 | Be aware that the option to follow symbolic links can be dangerous. |
81793b90 |
309 | Depending on the structure of the directory tree (including symbolic |
310 | links to directories) you might traverse a given (physical) directory |
311 | more than once (only if C<follow_fast> is in effect). |
312 | Furthermore, deleting or changing files in a symbolically linked directory |
313 | might cause very unpleasant surprises, since you delete or change files |
314 | in an unknown directory. |
0530a6c4 |
315 | |
5fa2bf2b |
316 | =back |
317 | |
7e47e6ff |
318 | =head1 NOTES |
319 | |
320 | =over 4 |
321 | |
322 | =item * |
323 | |
324 | Mac OS (Classic) users should note a few differences: |
325 | |
326 | =over 4 |
327 | |
328 | =item * |
329 | |
330 | The path separator is ':', not '/', and the current directory is denoted |
331 | as ':', not '.'. You should be careful about specifying relative pathnames. |
332 | While a full path always begins with a volume name, a relative pathname |
333 | should always begin with a ':'. If specifying a volume name only, a |
334 | trailing ':' is required. |
335 | |
336 | =item * |
337 | |
338 | C<$File::Find::dir> is guaranteed to end with a ':'. If C<$_> |
339 | contains the name of a directory, that name may or may not end with a |
340 | ':'. Likewise, C<$File::Find::name>, which contains the complete |
341 | pathname to that directory, and C<$File::Find::fullname>, which holds |
342 | the absolute pathname of that directory with all symbolic links resolved, |
343 | may or may not end with a ':'. |
344 | |
345 | =item * |
346 | |
347 | The default C<untaint_pattern> (see above) on Mac OS is set to |
348 | C<qr|^(.+)$|>. Note that the parentheses are vital. |
349 | |
350 | =item * |
351 | |
352 | The invisible system file "Icon\015" is ignored. While this file may |
353 | appear in every directory, there are some more invisible system files |
354 | on every volume, which are all located at the volume root level (i.e. |
355 | "MacintoshHD:"). These system files are B<not> excluded automatically. |
356 | Your filter may use the following code to recognize invisible files or |
357 | directories (requires Mac::Files): |
358 | |
359 | use Mac::Files; |
360 | |
361 | # invisible() -- returns 1 if file/directory is invisible, |
1cffc1dd |
362 | # 0 if it's visible or undef if an error occurred |
7e47e6ff |
363 | |
364 | sub invisible($) { |
365 | my $file = shift; |
366 | my ($fileCat, $fileInfo); |
367 | my $invisible_flag = 1 << 14; |
368 | |
369 | if ( $fileCat = FSpGetCatInfo($file) ) { |
370 | if ($fileInfo = $fileCat->ioFlFndrInfo() ) { |
371 | return (($fileInfo->fdFlags & $invisible_flag) && 1); |
372 | } |
373 | } |
374 | return undef; |
375 | } |
376 | |
377 | Generally, invisible files are system files, unless an odd application |
378 | decides to use invisible files for its own purposes. To distinguish |
379 | such files from system files, you have to look at the B<type> and B<creator> |
380 | file attributes. The MacPerl built-in functions C<GetFileInfo(FILE)> and |
381 | C<SetFileInfo(CREATOR, TYPE, FILES)> offer access to these attributes |
382 | (see MacPerl.pm for details). |
383 | |
384 | Files that appear on the desktop actually reside in an (hidden) directory |
385 | named "Desktop Folder" on the particular disk volume. Note that, although |
386 | all desktop files appear to be on the same "virtual" desktop, each disk |
387 | volume actually maintains its own "Desktop Folder" directory. |
388 | |
389 | =back |
390 | |
391 | =back |
0530a6c4 |
392 | |
6eb87ff8 |
393 | =head1 BUGS AND CAVEATS |
394 | |
395 | Despite the name of the C<finddepth()> function, both C<find()> and |
396 | C<finddepth()> perform a depth-first search of the directory |
397 | hierarchy. |
398 | |
a85af077 |
399 | =head1 HISTORY |
400 | |
401 | File::Find used to produce incorrect results if called recursively. |
402 | During the development of perl 5.8 this bug was fixed. |
403 | The first fixed version of File::Find was 1.01. |
404 | |
f06db76b |
405 | =cut |
406 | |
b75c8c73 |
407 | our @ISA = qw(Exporter); |
408 | our @EXPORT = qw(find finddepth); |
6280b799 |
409 | |
a0d0e21e |
410 | |
81793b90 |
411 | use strict; |
412 | my $Is_VMS; |
7e47e6ff |
413 | my $Is_MacOS; |
81793b90 |
414 | |
415 | require File::Basename; |
7e47e6ff |
416 | require File::Spec; |
81793b90 |
417 | |
9f826d6a |
418 | # Should ideally be my() not our() but local() currently |
419 | # refuses to operate on lexicals |
420 | |
421 | our %SLnkSeen; |
422 | our ($wanted_callback, $avoid_nlink, $bydepth, $no_chdir, $follow, |
719c805e |
423 | $follow_skip, $full_check, $untaint, $untaint_skip, $untaint_pat, |
80e52b73 |
424 | $pre_process, $post_process, $dangling_symlinks); |
81793b90 |
425 | |
426 | sub contract_name { |
427 | my ($cdir,$fn) = @_; |
428 | |
7e47e6ff |
429 | return substr($cdir,0,rindex($cdir,'/')) if $fn eq $File::Find::current_dir; |
81793b90 |
430 | |
431 | $cdir = substr($cdir,0,rindex($cdir,'/')+1); |
432 | |
433 | $fn =~ s|^\./||; |
434 | |
435 | my $abs_name= $cdir . $fn; |
436 | |
437 | if (substr($fn,0,3) eq '../') { |
fecbda2b |
438 | 1 while $abs_name =~ s!/[^/]*/\.\./!/!; |
81793b90 |
439 | } |
440 | |
441 | return $abs_name; |
442 | } |
443 | |
7e47e6ff |
444 | # return the absolute name of a directory or file |
445 | sub contract_name_Mac { |
446 | my ($cdir,$fn) = @_; |
447 | my $abs_name; |
448 | |
449 | if ($fn =~ /^(:+)(.*)$/) { # valid pathname starting with a ':' |
450 | |
451 | my $colon_count = length ($1); |
452 | if ($colon_count == 1) { |
453 | $abs_name = $cdir . $2; |
454 | return $abs_name; |
455 | } |
456 | else { |
457 | # need to move up the tree, but |
458 | # only if it's not a volume name |
459 | for (my $i=1; $i<$colon_count; $i++) { |
460 | unless ($cdir =~ /^[^:]+:$/) { # volume name |
461 | $cdir =~ s/[^:]+:$//; |
462 | } |
463 | else { |
464 | return undef; |
465 | } |
466 | } |
467 | $abs_name = $cdir . $2; |
468 | return $abs_name; |
469 | } |
470 | |
471 | } |
472 | else { |
473 | |
474 | # $fn may be a valid path to a directory or file or (dangling) |
475 | # symlink, without a leading ':' |
476 | if ( (-e $fn) || (-l $fn) ) { |
477 | if ($fn =~ /^[^:]+:/) { # a volume name like DataHD:* |
478 | return $fn; # $fn is already an absolute path |
479 | } |
480 | else { |
481 | $abs_name = $cdir . $fn; |
482 | return $abs_name; |
483 | } |
484 | } |
485 | else { # argh!, $fn is not a valid directory/file |
486 | return undef; |
487 | } |
488 | } |
489 | } |
81793b90 |
490 | |
491 | sub PathCombine($$) { |
492 | my ($Base,$Name) = @_; |
493 | my $AbsName; |
494 | |
7e47e6ff |
495 | if ($Is_MacOS) { |
496 | # $Name is the resolved symlink (always a full path on MacOS), |
497 | # i.e. there's no need to call contract_name_Mac() |
498 | $AbsName = $Name; |
499 | |
500 | # (simple) check for recursion |
501 | if ( ( $Base =~ /^$AbsName/) && (-d $AbsName) ) { # recursion |
502 | return undef; |
503 | } |
81793b90 |
504 | } |
505 | else { |
7e47e6ff |
506 | if (substr($Name,0,1) eq '/') { |
507 | $AbsName= $Name; |
508 | } |
509 | else { |
510 | $AbsName= contract_name($Base,$Name); |
511 | } |
81793b90 |
512 | |
7e47e6ff |
513 | # (simple) check for recursion |
514 | my $newlen= length($AbsName); |
515 | if ($newlen <= length($Base)) { |
516 | if (($newlen == length($Base) || substr($Base,$newlen,1) eq '/') |
517 | && $AbsName eq substr($Base,0,$newlen)) |
518 | { |
519 | return undef; |
520 | } |
81793b90 |
521 | } |
522 | } |
523 | return $AbsName; |
524 | } |
525 | |
526 | sub Follow_SymLink($) { |
527 | my ($AbsName) = @_; |
528 | |
529 | my ($NewName,$DEV, $INO); |
530 | ($DEV, $INO)= lstat $AbsName; |
531 | |
532 | while (-l _) { |
533 | if ($SLnkSeen{$DEV, $INO}++) { |
534 | if ($follow_skip < 2) { |
535 | die "$AbsName is encountered a second time"; |
a0d0e21e |
536 | } |
537 | else { |
81793b90 |
538 | return undef; |
a0d0e21e |
539 | } |
540 | } |
81793b90 |
541 | $NewName= PathCombine($AbsName, readlink($AbsName)); |
542 | unless(defined $NewName) { |
543 | if ($follow_skip < 2) { |
544 | die "$AbsName is a recursive symbolic link"; |
545 | } |
546 | else { |
547 | return undef; |
a0d0e21e |
548 | } |
81793b90 |
549 | } |
550 | else { |
551 | $AbsName= $NewName; |
552 | } |
553 | ($DEV, $INO) = lstat($AbsName); |
554 | return undef unless defined $DEV; # dangling symbolic link |
555 | } |
556 | |
cd68ec93 |
557 | if ($full_check && defined $DEV && $SLnkSeen{$DEV, $INO}++) { |
7e47e6ff |
558 | if ( ($follow_skip < 1) || ((-d _) && ($follow_skip < 2)) ) { |
81793b90 |
559 | die "$AbsName encountered a second time"; |
560 | } |
561 | else { |
562 | return undef; |
563 | } |
564 | } |
565 | |
566 | return $AbsName; |
567 | } |
568 | |
17f410f9 |
569 | our($dir, $name, $fullname, $prune); |
81793b90 |
570 | sub _find_dir_symlnk($$$); |
571 | sub _find_dir($$$); |
572 | |
7e47e6ff |
573 | # check whether or not a scalar variable is tainted |
574 | # (code straight from the Camel, 3rd ed., page 561) |
575 | sub is_tainted_pp { |
576 | my $arg = shift; |
577 | my $nada = substr($arg, 0, 0); # zero-length |
578 | local $@; |
579 | eval { eval "# $nada" }; |
580 | return length($@) != 0; |
581 | } |
582 | |
81793b90 |
583 | sub _find_opt { |
584 | my $wanted = shift; |
585 | die "invalid top directory" unless defined $_[0]; |
586 | |
9f826d6a |
587 | # This function must local()ize everything because callbacks may |
588 | # call find() or finddepth() |
589 | |
590 | local %SLnkSeen; |
591 | local ($wanted_callback, $avoid_nlink, $bydepth, $no_chdir, $follow, |
592 | $follow_skip, $full_check, $untaint, $untaint_skip, $untaint_pat, |
80e52b73 |
593 | $pre_process, $post_process, $dangling_symlinks); |
17ab9c14 |
594 | local($dir, $name, $fullname, $prune, $_); |
9f826d6a |
595 | |
a0c9c202 |
596 | my $cwd = $wanted->{bydepth} ? Cwd::fastcwd() : Cwd::getcwd(); |
80e52b73 |
597 | my $cwd_untainted = $cwd; |
598 | my $check_t_cwd = 1; |
599 | $wanted_callback = $wanted->{wanted}; |
600 | $bydepth = $wanted->{bydepth}; |
601 | $pre_process = $wanted->{preprocess}; |
602 | $post_process = $wanted->{postprocess}; |
603 | $no_chdir = $wanted->{no_chdir}; |
604 | $full_check = $wanted->{follow}; |
605 | $follow = $full_check || $wanted->{follow_fast}; |
606 | $follow_skip = $wanted->{follow_skip}; |
607 | $untaint = $wanted->{untaint}; |
608 | $untaint_pat = $wanted->{untaint_pattern}; |
609 | $untaint_skip = $wanted->{untaint_skip}; |
610 | $dangling_symlinks = $wanted->{dangling_symlinks}; |
81793b90 |
611 | |
1cffc1dd |
612 | # for compatibility reasons (find.pl, find2perl) |
9f826d6a |
613 | local our ($topdir, $topdev, $topino, $topmode, $topnlink); |
81793b90 |
614 | |
615 | # a symbolic link to a directory doesn't increase the link count |
616 | $avoid_nlink = $follow || $File::Find::dont_use_nlink; |
617 | |
e7b91b67 |
618 | my ($abs_dir, $Is_Dir); |
81793b90 |
619 | |
620 | Proc_Top_Item: |
621 | foreach my $TOP (@_) { |
7e47e6ff |
622 | my $top_item = $TOP; |
623 | |
624 | if ($Is_MacOS) { |
625 | ($topdev,$topino,$topmode,$topnlink) = $follow ? stat $top_item : lstat $top_item; |
626 | $top_item = ":$top_item" |
3fa6e24b |
627 | if ( (-d _) && ( $top_item !~ /:/ ) ); |
7e47e6ff |
628 | } |
629 | else { |
630 | $top_item =~ s|/\z|| unless $top_item eq '/'; |
631 | ($topdev,$topino,$topmode,$topnlink) = $follow ? stat $top_item : lstat $top_item; |
632 | } |
633 | |
634 | $Is_Dir= 0; |
635 | |
636 | if ($follow) { |
637 | |
638 | if ($Is_MacOS) { |
639 | $cwd = "$cwd:" unless ($cwd =~ /:$/); # for safety |
640 | |
641 | if ($top_item eq $File::Find::current_dir) { |
642 | $abs_dir = $cwd; |
643 | } |
644 | else { |
645 | $abs_dir = contract_name_Mac($cwd, $top_item); |
646 | unless (defined $abs_dir) { |
cd68ec93 |
647 | warnings::warnif "Can't determine absolute path for $top_item (No such file or directory)\n"; |
7e47e6ff |
648 | next Proc_Top_Item; |
649 | } |
650 | } |
651 | |
652 | } |
653 | else { |
654 | if (substr($top_item,0,1) eq '/') { |
655 | $abs_dir = $top_item; |
656 | } |
657 | elsif ($top_item eq $File::Find::current_dir) { |
658 | $abs_dir = $cwd; |
659 | } |
660 | else { # care about any ../ |
661 | $abs_dir = contract_name("$cwd/",$top_item); |
662 | } |
663 | } |
664 | $abs_dir= Follow_SymLink($abs_dir); |
665 | unless (defined $abs_dir) { |
80e52b73 |
666 | if ($dangling_symlinks) { |
667 | if (ref $dangling_symlinks eq 'CODE') { |
668 | $dangling_symlinks->($top_item, $cwd); |
669 | } else { |
cd68ec93 |
670 | warnings::warnif "$top_item is a dangling symbolic link\n"; |
80e52b73 |
671 | } |
672 | } |
81793b90 |
673 | next Proc_Top_Item; |
7e47e6ff |
674 | } |
675 | |
676 | if (-d _) { |
81793b90 |
677 | _find_dir_symlnk($wanted, $abs_dir, $top_item); |
678 | $Is_Dir= 1; |
7e47e6ff |
679 | } |
680 | } |
81793b90 |
681 | else { # no follow |
7e47e6ff |
682 | $topdir = $top_item; |
683 | unless (defined $topnlink) { |
cd68ec93 |
684 | warnings::warnif "Can't stat $top_item: $!\n"; |
7e47e6ff |
685 | next Proc_Top_Item; |
686 | } |
687 | if (-d _) { |
544ff7a7 |
688 | $top_item =~ s/\.dir\z//i if $Is_VMS; |
e7b91b67 |
689 | _find_dir($wanted, $top_item, $topnlink); |
81793b90 |
690 | $Is_Dir= 1; |
7e47e6ff |
691 | } |
237437d0 |
692 | else { |
81793b90 |
693 | $abs_dir= $top_item; |
7e47e6ff |
694 | } |
695 | } |
81793b90 |
696 | |
7e47e6ff |
697 | unless ($Is_Dir) { |
81793b90 |
698 | unless (($_,$dir) = File::Basename::fileparse($abs_dir)) { |
7e47e6ff |
699 | if ($Is_MacOS) { |
700 | ($dir,$_) = (':', $top_item); # $File::Find::dir, $_ |
701 | } |
702 | else { |
703 | ($dir,$_) = ('./', $top_item); |
704 | } |
81793b90 |
705 | } |
706 | |
7e47e6ff |
707 | $abs_dir = $dir; |
708 | if (( $untaint ) && (is_tainted($dir) )) { |
709 | ( $abs_dir ) = $dir =~ m|$untaint_pat|; |
81793b90 |
710 | unless (defined $abs_dir) { |
711 | if ($untaint_skip == 0) { |
7e47e6ff |
712 | die "directory $dir is still tainted"; |
81793b90 |
713 | } |
714 | else { |
715 | next Proc_Top_Item; |
716 | } |
717 | } |
7e47e6ff |
718 | } |
81793b90 |
719 | |
7e47e6ff |
720 | unless ($no_chdir || chdir $abs_dir) { |
cd68ec93 |
721 | warnings::warnif "Couldn't chdir $abs_dir: $!\n"; |
7e47e6ff |
722 | next Proc_Top_Item; |
723 | } |
719911cc |
724 | |
7e47e6ff |
725 | $name = $abs_dir . $_; # $File::Find::name |
3bb6d3e5 |
726 | $_ = $name if $no_chdir; |
719911cc |
727 | |
abfdd623 |
728 | { $wanted_callback->() }; # protect against wild "next" |
81793b90 |
729 | |
7e47e6ff |
730 | } |
81793b90 |
731 | |
7e47e6ff |
732 | unless ( $no_chdir ) { |
733 | if ( ($check_t_cwd) && (($untaint) && (is_tainted($cwd) )) ) { |
734 | ( $cwd_untainted ) = $cwd =~ m|$untaint_pat|; |
735 | unless (defined $cwd_untainted) { |
736 | die "insecure cwd in find(depth)"; |
737 | } |
738 | $check_t_cwd = 0; |
739 | } |
740 | unless (chdir $cwd_untainted) { |
741 | die "Can't cd to $cwd: $!\n"; |
742 | } |
743 | } |
81793b90 |
744 | } |
745 | } |
746 | |
747 | # API: |
748 | # $wanted |
749 | # $p_dir : "parent directory" |
750 | # $nlink : what came back from the stat |
751 | # preconditions: |
752 | # chdir (if not no_chdir) to dir |
753 | |
754 | sub _find_dir($$$) { |
755 | my ($wanted, $p_dir, $nlink) = @_; |
756 | my ($CdLvl,$Level) = (0,0); |
757 | my @Stack; |
758 | my @filenames; |
759 | my ($subcount,$sub_nlink); |
760 | my $SE= []; |
761 | my $dir_name= $p_dir; |
7e47e6ff |
762 | my $dir_pref; |
39e79f6b |
763 | my $dir_rel = $File::Find::current_dir; |
7e47e6ff |
764 | my $tainted = 0; |
5fa2bf2b |
765 | my $no_nlink; |
7e47e6ff |
766 | |
767 | if ($Is_MacOS) { |
768 | $dir_pref= ($p_dir =~ /:$/) ? $p_dir : "$p_dir:"; # preface |
7e47e6ff |
769 | } |
770 | else { |
771 | $dir_pref= ( $p_dir eq '/' ? '/' : "$p_dir/" ); |
7e47e6ff |
772 | } |
81793b90 |
773 | |
774 | local ($dir, $name, $prune, *DIR); |
7e47e6ff |
775 | |
776 | unless ( $no_chdir || ($p_dir eq $File::Find::current_dir)) { |
81793b90 |
777 | my $udir = $p_dir; |
7e47e6ff |
778 | if (( $untaint ) && (is_tainted($p_dir) )) { |
779 | ( $udir ) = $p_dir =~ m|$untaint_pat|; |
81793b90 |
780 | unless (defined $udir) { |
781 | if ($untaint_skip == 0) { |
782 | die "directory $p_dir is still tainted"; |
783 | } |
784 | else { |
785 | return; |
786 | } |
237437d0 |
787 | } |
a0d0e21e |
788 | } |
8d8eebbf |
789 | unless (chdir ($Is_VMS && $udir !~ /[\/\[<]+/ ? "./$udir" : $udir)) { |
cd68ec93 |
790 | warnings::warnif "Can't cd to $udir: $!\n"; |
81793b90 |
791 | return; |
792 | } |
793 | } |
7e47e6ff |
794 | |
795 | # push the starting directory |
57e73c4b |
796 | push @Stack,[$CdLvl,$p_dir,$dir_rel,-1] if $bydepth; |
81793b90 |
797 | |
7e47e6ff |
798 | if ($Is_MacOS) { |
799 | $p_dir = $dir_pref; # ensure trailing ':' |
800 | } |
801 | |
81793b90 |
802 | while (defined $SE) { |
803 | unless ($bydepth) { |
7e47e6ff |
804 | $dir= $p_dir; # $File::Find::dir |
805 | $name= $dir_name; # $File::Find::name |
806 | $_= ($no_chdir ? $dir_name : $dir_rel ); # $_ |
81793b90 |
807 | # prune may happen here |
7e47e6ff |
808 | $prune= 0; |
abfdd623 |
809 | { $wanted_callback->() }; # protect against wild "next" |
7e47e6ff |
810 | next if $prune; |
81793b90 |
811 | } |
7e47e6ff |
812 | |
81793b90 |
813 | # change to that directory |
7e47e6ff |
814 | unless ($no_chdir || ($dir_rel eq $File::Find::current_dir)) { |
81793b90 |
815 | my $udir= $dir_rel; |
7e47e6ff |
816 | if ( ($untaint) && (($tainted) || ($tainted = is_tainted($dir_rel) )) ) { |
817 | ( $udir ) = $dir_rel =~ m|$untaint_pat|; |
81793b90 |
818 | unless (defined $udir) { |
819 | if ($untaint_skip == 0) { |
7e47e6ff |
820 | if ($Is_MacOS) { |
821 | die "directory ($p_dir) $dir_rel is still tainted"; |
822 | } |
823 | else { |
824 | die "directory (" . ($p_dir ne '/' ? $p_dir : '') . "/) $dir_rel is still tainted"; |
825 | } |
826 | } else { # $untaint_skip == 1 |
827 | next; |
81793b90 |
828 | } |
829 | } |
830 | } |
8d8eebbf |
831 | unless (chdir ($Is_VMS && $udir !~ /[\/\[<]+/ ? "./$udir" : $udir)) { |
7e47e6ff |
832 | if ($Is_MacOS) { |
cd68ec93 |
833 | warnings::warnif "Can't cd to ($p_dir) $udir: $!\n"; |
7e47e6ff |
834 | } |
835 | else { |
cd68ec93 |
836 | warnings::warnif "Can't cd to (" . |
837 | ($p_dir ne '/' ? $p_dir : '') . "/) $udir: $!\n"; |
7e47e6ff |
838 | } |
81793b90 |
839 | next; |
840 | } |
841 | $CdLvl++; |
842 | } |
843 | |
7e47e6ff |
844 | if ($Is_MacOS) { |
845 | $dir_name = "$dir_name:" unless ($dir_name =~ /:$/); |
846 | } |
847 | |
848 | $dir= $dir_name; # $File::Find::dir |
81793b90 |
849 | |
850 | # Get the list of files in the current directory. |
7e47e6ff |
851 | unless (opendir DIR, ($no_chdir ? $dir_name : $File::Find::current_dir)) { |
cd68ec93 |
852 | warnings::warnif "Can't opendir($dir_name): $!\n"; |
81793b90 |
853 | next; |
854 | } |
855 | @filenames = readdir DIR; |
856 | closedir(DIR); |
abfdd623 |
857 | @filenames = $pre_process->(@filenames) if $pre_process; |
719c805e |
858 | push @Stack,[$CdLvl,$dir_name,"",-2] if $post_process; |
81793b90 |
859 | |
5fa2bf2b |
860 | # default: use whatever was specifid |
861 | # (if $nlink >= 2, and $avoid_nlink == 0, this will switch back) |
862 | $no_nlink = $avoid_nlink; |
863 | # if dir has wrong nlink count, force switch to slower stat method |
864 | $no_nlink = 1 if ($nlink < 2); |
865 | |
866 | if ($nlink == 2 && !$no_nlink) { |
81793b90 |
867 | # This dir has no subdirectories. |
868 | for my $FN (@filenames) { |
7e47e6ff |
869 | next if $FN =~ $File::Find::skip_pattern; |
81793b90 |
870 | |
7e47e6ff |
871 | $name = $dir_pref . $FN; # $File::Find::name |
872 | $_ = ($no_chdir ? $name : $FN); # $_ |
abfdd623 |
873 | { $wanted_callback->() }; # protect against wild "next" |
81793b90 |
874 | } |
875 | |
876 | } |
877 | else { |
878 | # This dir has subdirectories. |
879 | $subcount = $nlink - 2; |
880 | |
7bd31527 |
881 | # HACK: insert directories at this position. so as to preserve |
882 | # the user pre-processed ordering of files. |
883 | # EG: directory traversal is in user sorted order, not at random. |
884 | my $stack_top = @Stack; |
885 | |
81793b90 |
886 | for my $FN (@filenames) { |
7e47e6ff |
887 | next if $FN =~ $File::Find::skip_pattern; |
5fa2bf2b |
888 | if ($subcount > 0 || $no_nlink) { |
81793b90 |
889 | # Seen all the subdirs? |
890 | # check for directoriness. |
891 | # stat is faster for a file in the current directory |
07867069 |
892 | $sub_nlink = (lstat ($no_chdir ? $dir_pref . $FN : $FN))[3]; |
81793b90 |
893 | |
894 | if (-d _) { |
895 | --$subcount; |
544ff7a7 |
896 | $FN =~ s/\.dir\z//i if $Is_VMS; |
7bd31527 |
897 | # HACK: replace push to preserve dir traversal order |
898 | #push @Stack,[$CdLvl,$dir_name,$FN,$sub_nlink]; |
899 | splice @Stack, $stack_top, 0, |
900 | [$CdLvl,$dir_name,$FN,$sub_nlink]; |
81793b90 |
901 | } |
902 | else { |
7e47e6ff |
903 | $name = $dir_pref . $FN; # $File::Find::name |
904 | $_= ($no_chdir ? $name : $FN); # $_ |
abfdd623 |
905 | { $wanted_callback->() }; # protect against wild "next" |
81793b90 |
906 | } |
907 | } |
07867069 |
908 | else { |
7e47e6ff |
909 | $name = $dir_pref . $FN; # $File::Find::name |
910 | $_= ($no_chdir ? $name : $FN); # $_ |
abfdd623 |
911 | { $wanted_callback->() }; # protect against wild "next" |
81793b90 |
912 | } |
913 | } |
914 | } |
17b275ff |
915 | } |
916 | continue { |
57e73c4b |
917 | while ( defined ($SE = pop @Stack) ) { |
81793b90 |
918 | ($Level, $p_dir, $dir_rel, $nlink) = @$SE; |
919 | if ($CdLvl > $Level && !$no_chdir) { |
7e47e6ff |
920 | my $tmp; |
921 | if ($Is_MacOS) { |
922 | $tmp = (':' x ($CdLvl-$Level)) . ':'; |
923 | } |
924 | else { |
925 | $tmp = join('/',('..') x ($CdLvl-$Level)); |
926 | } |
927 | die "Can't cd to $dir_name" . $tmp |
928 | unless chdir ($tmp); |
81793b90 |
929 | $CdLvl = $Level; |
930 | } |
7e47e6ff |
931 | |
932 | if ($Is_MacOS) { |
933 | # $pdir always has a trailing ':', except for the starting dir, |
934 | # where $dir_rel eq ':' |
935 | $dir_name = "$p_dir$dir_rel"; |
936 | $dir_pref = "$dir_name:"; |
937 | } |
938 | else { |
939 | $dir_name = ($p_dir eq '/' ? "/$dir_rel" : "$p_dir/$dir_rel"); |
940 | $dir_pref = "$dir_name/"; |
941 | } |
942 | |
719c805e |
943 | if ( $nlink == -2 ) { |
7e47e6ff |
944 | $name = $dir = $p_dir; # $File::Find::name / dir |
39e79f6b |
945 | $_ = $File::Find::current_dir; |
abfdd623 |
946 | $post_process->(); # End-of-directory processing |
7e47e6ff |
947 | } |
948 | elsif ( $nlink < 0 ) { # must be finddepth, report dirname now |
949 | $name = $dir_name; |
950 | if ($Is_MacOS) { |
951 | if ($dir_rel eq ':') { # must be the top dir, where we started |
952 | $name =~ s|:$||; # $File::Find::name |
953 | $p_dir = "$p_dir:" unless ($p_dir =~ /:$/); |
954 | } |
955 | $dir = $p_dir; # $File::Find::dir |
956 | $_ = ($no_chdir ? $name : $dir_rel); # $_ |
957 | } |
958 | else { |
959 | if ( substr($name,-2) eq '/.' ) { |
5cf0a2f2 |
960 | substr($name, length($name) == 2 ? -1 : -2) = ''; |
7e47e6ff |
961 | } |
962 | $dir = $p_dir; |
963 | $_ = ($no_chdir ? $dir_name : $dir_rel ); |
964 | if ( substr($_,-2) eq '/.' ) { |
5cf0a2f2 |
965 | substr($_, length($_) == 2 ? -1 : -2) = ''; |
7e47e6ff |
966 | } |
967 | } |
abfdd623 |
968 | { $wanted_callback->() }; # protect against wild "next" |
7e47e6ff |
969 | } |
970 | else { |
971 | push @Stack,[$CdLvl,$p_dir,$dir_rel,-1] if $bydepth; |
972 | last; |
973 | } |
81793b90 |
974 | } |
a0d0e21e |
975 | } |
976 | } |
977 | |
81793b90 |
978 | |
979 | # API: |
980 | # $wanted |
981 | # $dir_loc : absolute location of a dir |
982 | # $p_dir : "parent directory" |
983 | # preconditions: |
984 | # chdir (if not no_chdir) to dir |
985 | |
986 | sub _find_dir_symlnk($$$) { |
7e47e6ff |
987 | my ($wanted, $dir_loc, $p_dir) = @_; # $dir_loc is the absolute directory |
81793b90 |
988 | my @Stack; |
989 | my @filenames; |
990 | my $new_loc; |
7e47e6ff |
991 | my $updir_loc = $dir_loc; # untainted parent directory |
81793b90 |
992 | my $SE = []; |
993 | my $dir_name = $p_dir; |
7e47e6ff |
994 | my $dir_pref; |
995 | my $loc_pref; |
39e79f6b |
996 | my $dir_rel = $File::Find::current_dir; |
7e47e6ff |
997 | my $byd_flag; # flag for pending stack entry if $bydepth |
998 | my $tainted = 0; |
999 | my $ok = 1; |
1000 | |
1001 | if ($Is_MacOS) { |
1002 | $dir_pref = ($p_dir =~ /:$/) ? "$p_dir" : "$p_dir:"; |
1003 | $loc_pref = ($dir_loc =~ /:$/) ? "$dir_loc" : "$dir_loc:"; |
7e47e6ff |
1004 | } else { |
1005 | $dir_pref = ( $p_dir eq '/' ? '/' : "$p_dir/" ); |
1006 | $loc_pref = ( $dir_loc eq '/' ? '/' : "$dir_loc/" ); |
7e47e6ff |
1007 | } |
81793b90 |
1008 | |
1009 | local ($dir, $name, $fullname, $prune, *DIR); |
7e47e6ff |
1010 | |
1011 | unless ($no_chdir) { |
1012 | # untaint the topdir |
1013 | if (( $untaint ) && (is_tainted($dir_loc) )) { |
1014 | ( $updir_loc ) = $dir_loc =~ m|$untaint_pat|; # parent dir, now untainted |
1015 | # once untainted, $updir_loc is pushed on the stack (as parent directory); |
1016 | # hence, we don't need to untaint the parent directory every time we chdir |
1017 | # to it later |
1018 | unless (defined $updir_loc) { |
81793b90 |
1019 | if ($untaint_skip == 0) { |
1020 | die "directory $dir_loc is still tainted"; |
1021 | } |
1022 | else { |
1023 | return; |
1024 | } |
1025 | } |
1026 | } |
7e47e6ff |
1027 | $ok = chdir($updir_loc) unless ($p_dir eq $File::Find::current_dir); |
1028 | unless ($ok) { |
cd68ec93 |
1029 | warnings::warnif "Can't cd to $updir_loc: $!\n"; |
81793b90 |
1030 | return; |
1031 | } |
1032 | } |
1033 | |
7e47e6ff |
1034 | push @Stack,[$dir_loc,$updir_loc,$p_dir,$dir_rel,-1] if $bydepth; |
1035 | |
1036 | if ($Is_MacOS) { |
1037 | $p_dir = $dir_pref; # ensure trailing ':' |
1038 | } |
57e73c4b |
1039 | |
81793b90 |
1040 | while (defined $SE) { |
1041 | |
1042 | unless ($bydepth) { |
7e47e6ff |
1043 | # change (back) to parent directory (always untainted) |
704ea872 |
1044 | unless ($no_chdir) { |
7e47e6ff |
1045 | unless (chdir $updir_loc) { |
cd68ec93 |
1046 | warnings::warnif "Can't cd to $updir_loc: $!\n"; |
704ea872 |
1047 | next; |
1048 | } |
1049 | } |
7e47e6ff |
1050 | $dir= $p_dir; # $File::Find::dir |
1051 | $name= $dir_name; # $File::Find::name |
1052 | $_= ($no_chdir ? $dir_name : $dir_rel ); # $_ |
1053 | $fullname= $dir_loc; # $File::Find::fullname |
81793b90 |
1054 | # prune may happen here |
7e47e6ff |
1055 | $prune= 0; |
704ea872 |
1056 | lstat($_); # make sure file tests with '_' work |
abfdd623 |
1057 | { $wanted_callback->() }; # protect against wild "next" |
7e47e6ff |
1058 | next if $prune; |
81793b90 |
1059 | } |
1060 | |
1061 | # change to that directory |
7e47e6ff |
1062 | unless ($no_chdir || ($dir_rel eq $File::Find::current_dir)) { |
1063 | $updir_loc = $dir_loc; |
1064 | if ( ($untaint) && (($tainted) || ($tainted = is_tainted($dir_loc) )) ) { |
1065 | # untaint $dir_loc, what will be pushed on the stack as (untainted) parent dir |
1066 | ( $updir_loc ) = $dir_loc =~ m|$untaint_pat|; |
1067 | unless (defined $updir_loc) { |
81793b90 |
1068 | if ($untaint_skip == 0) { |
1069 | die "directory $dir_loc is still tainted"; |
a0d0e21e |
1070 | } |
237437d0 |
1071 | else { |
81793b90 |
1072 | next; |
237437d0 |
1073 | } |
a0d0e21e |
1074 | } |
1075 | } |
7e47e6ff |
1076 | unless (chdir $updir_loc) { |
cd68ec93 |
1077 | warnings::warnif "Can't cd to $updir_loc: $!\n"; |
81793b90 |
1078 | next; |
1079 | } |
1080 | } |
1081 | |
7e47e6ff |
1082 | if ($Is_MacOS) { |
1083 | $dir_name = "$dir_name:" unless ($dir_name =~ /:$/); |
1084 | } |
1085 | |
1086 | $dir = $dir_name; # $File::Find::dir |
81793b90 |
1087 | |
1088 | # Get the list of files in the current directory. |
7e47e6ff |
1089 | unless (opendir DIR, ($no_chdir ? $dir_loc : $File::Find::current_dir)) { |
cd68ec93 |
1090 | warnings::warnif "Can't opendir($dir_loc): $!\n"; |
81793b90 |
1091 | next; |
1092 | } |
1093 | @filenames = readdir DIR; |
1094 | closedir(DIR); |
1095 | |
1096 | for my $FN (@filenames) { |
7e47e6ff |
1097 | next if $FN =~ $File::Find::skip_pattern; |
81793b90 |
1098 | |
1099 | # follow symbolic links / do an lstat |
07867069 |
1100 | $new_loc = Follow_SymLink($loc_pref.$FN); |
81793b90 |
1101 | |
1102 | # ignore if invalid symlink |
1103 | next unless defined $new_loc; |
7e47e6ff |
1104 | |
81793b90 |
1105 | if (-d _) { |
7e47e6ff |
1106 | push @Stack,[$new_loc,$updir_loc,$dir_name,$FN,1]; |
81793b90 |
1107 | } |
1108 | else { |
7e47e6ff |
1109 | $fullname = $new_loc; # $File::Find::fullname |
1110 | $name = $dir_pref . $FN; # $File::Find::name |
1111 | $_ = ($no_chdir ? $name : $FN); # $_ |
abfdd623 |
1112 | { $wanted_callback->() }; # protect against wild "next" |
81793b90 |
1113 | } |
1114 | } |
1115 | |
81793b90 |
1116 | } |
1117 | continue { |
57e73c4b |
1118 | while (defined($SE = pop @Stack)) { |
7e47e6ff |
1119 | ($dir_loc, $updir_loc, $p_dir, $dir_rel, $byd_flag) = @$SE; |
1120 | if ($Is_MacOS) { |
1121 | # $p_dir always has a trailing ':', except for the starting dir, |
1122 | # where $dir_rel eq ':' |
1123 | $dir_name = "$p_dir$dir_rel"; |
1124 | $dir_pref = "$dir_name:"; |
1125 | $loc_pref = ($dir_loc =~ /:$/) ? $dir_loc : "$dir_loc:"; |
1126 | } |
1127 | else { |
1128 | $dir_name = ($p_dir eq '/' ? "/$dir_rel" : "$p_dir/$dir_rel"); |
1129 | $dir_pref = "$dir_name/"; |
1130 | $loc_pref = "$dir_loc/"; |
1131 | } |
1132 | if ( $byd_flag < 0 ) { # must be finddepth, report dirname now |
1133 | unless ($no_chdir || ($dir_rel eq $File::Find::current_dir)) { |
1134 | unless (chdir $updir_loc) { # $updir_loc (parent dir) is always untainted |
cd68ec93 |
1135 | warnings::warnif "Can't cd to $updir_loc: $!\n"; |
7e47e6ff |
1136 | next; |
1137 | } |
1138 | } |
1139 | $fullname = $dir_loc; # $File::Find::fullname |
1140 | $name = $dir_name; # $File::Find::name |
1141 | if ($Is_MacOS) { |
1142 | if ($dir_rel eq ':') { # must be the top dir, where we started |
1143 | $name =~ s|:$||; # $File::Find::name |
1144 | $p_dir = "$p_dir:" unless ($p_dir =~ /:$/); |
1145 | } |
1146 | $dir = $p_dir; # $File::Find::dir |
1147 | $_ = ($no_chdir ? $name : $dir_rel); # $_ |
1148 | } |
1149 | else { |
1150 | if ( substr($name,-2) eq '/.' ) { |
f801979b |
1151 | substr($name, length($name) == 2 ? -1 : -2) = ''; # $File::Find::name |
7e47e6ff |
1152 | } |
1153 | $dir = $p_dir; # $File::Find::dir |
1154 | $_ = ($no_chdir ? $dir_name : $dir_rel); # $_ |
1155 | if ( substr($_,-2) eq '/.' ) { |
f801979b |
1156 | substr($_, length($_) == 2 ? -1 : -2) = ''; |
7e47e6ff |
1157 | } |
1158 | } |
1159 | |
1160 | lstat($_); # make sure file tests with '_' work |
abfdd623 |
1161 | { $wanted_callback->() }; # protect against wild "next" |
7e47e6ff |
1162 | } |
1163 | else { |
1164 | push @Stack,[$dir_loc, $updir_loc, $p_dir, $dir_rel,-1] if $bydepth; |
1165 | last; |
1166 | } |
a0d0e21e |
1167 | } |
1168 | } |
1169 | } |
1170 | |
81793b90 |
1171 | |
20408e3c |
1172 | sub wrap_wanted { |
81793b90 |
1173 | my $wanted = shift; |
1174 | if ( ref($wanted) eq 'HASH' ) { |
1175 | if ( $wanted->{follow} || $wanted->{follow_fast}) { |
1176 | $wanted->{follow_skip} = 1 unless defined $wanted->{follow_skip}; |
1177 | } |
1178 | if ( $wanted->{untaint} ) { |
7e47e6ff |
1179 | $wanted->{untaint_pattern} = $File::Find::untaint_pattern |
81793b90 |
1180 | unless defined $wanted->{untaint_pattern}; |
1181 | $wanted->{untaint_skip} = 0 unless defined $wanted->{untaint_skip}; |
1182 | } |
1183 | return $wanted; |
1184 | } |
1185 | else { |
1186 | return { wanted => $wanted }; |
1187 | } |
a0d0e21e |
1188 | } |
1189 | |
20408e3c |
1190 | sub find { |
81793b90 |
1191 | my $wanted = shift; |
1192 | _find_opt(wrap_wanted($wanted), @_); |
a0d0e21e |
1193 | } |
1194 | |
55d729e4 |
1195 | sub finddepth { |
81793b90 |
1196 | my $wanted = wrap_wanted(shift); |
1197 | $wanted->{bydepth} = 1; |
1198 | _find_opt($wanted, @_); |
20408e3c |
1199 | } |
6280b799 |
1200 | |
7e47e6ff |
1201 | # default |
1202 | $File::Find::skip_pattern = qr/^\.{1,2}\z/; |
1203 | $File::Find::untaint_pattern = qr|^([-+@\w./]+)$|; |
1204 | |
6280b799 |
1205 | # These are hard-coded for now, but may move to hint files. |
10eba763 |
1206 | if ($^O eq 'VMS') { |
81793b90 |
1207 | $Is_VMS = 1; |
7e47e6ff |
1208 | $File::Find::dont_use_nlink = 1; |
1209 | } |
1210 | elsif ($^O eq 'MacOS') { |
1211 | $Is_MacOS = 1; |
1212 | $File::Find::dont_use_nlink = 1; |
1213 | $File::Find::skip_pattern = qr/^Icon\015\z/; |
1214 | $File::Find::untaint_pattern = qr|^(.+)$|; |
748a9306 |
1215 | } |
1216 | |
7e47e6ff |
1217 | # this _should_ work properly on all platforms |
1218 | # where File::Find can be expected to work |
1219 | $File::Find::current_dir = File::Spec->curdir || '.'; |
1220 | |
81793b90 |
1221 | $File::Find::dont_use_nlink = 1 |
497711e7 |
1222 | if $^O eq 'os2' || $^O eq 'dos' || $^O eq 'amigaos' || $^O eq 'MSWin32' || |
1119cb72 |
1223 | $^O eq 'cygwin' || $^O eq 'epoc' || $^O eq 'qnx' || |
1224 | $^O eq 'nto'; |
6280b799 |
1225 | |
20408e3c |
1226 | # Set dont_use_nlink in your hint file if your system's stat doesn't |
1227 | # report the number of links in a directory as an indication |
1228 | # of the number of files. |
1229 | # See, e.g. hints/machten.sh for MachTen 2.2. |
81793b90 |
1230 | unless ($File::Find::dont_use_nlink) { |
1231 | require Config; |
1232 | $File::Find::dont_use_nlink = 1 if ($Config::Config{'dont_use_nlink'}); |
20408e3c |
1233 | } |
1234 | |
7e47e6ff |
1235 | # We need a function that checks if a scalar is tainted. Either use the |
1236 | # Scalar::Util module's tainted() function or our (slower) pure Perl |
1237 | # fallback is_tainted_pp() |
1238 | { |
1239 | local $@; |
1240 | eval { require Scalar::Util }; |
1241 | *is_tainted = $@ ? \&is_tainted_pp : \&Scalar::Util::tainted; |
1242 | } |
1243 | |
a0d0e21e |
1244 | 1; |