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; |
be56ab1f |
6 | our $VERSION = '1.07'; |
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); |
4c621faf |
594 | local($dir, $name, $fullname, $prune); |
bc125c03 |
595 | local *_ = \my $a; |
9f826d6a |
596 | |
a0c9c202 |
597 | my $cwd = $wanted->{bydepth} ? Cwd::fastcwd() : Cwd::getcwd(); |
80e52b73 |
598 | my $cwd_untainted = $cwd; |
599 | my $check_t_cwd = 1; |
600 | $wanted_callback = $wanted->{wanted}; |
601 | $bydepth = $wanted->{bydepth}; |
602 | $pre_process = $wanted->{preprocess}; |
603 | $post_process = $wanted->{postprocess}; |
604 | $no_chdir = $wanted->{no_chdir}; |
605 | $full_check = $wanted->{follow}; |
606 | $follow = $full_check || $wanted->{follow_fast}; |
607 | $follow_skip = $wanted->{follow_skip}; |
608 | $untaint = $wanted->{untaint}; |
609 | $untaint_pat = $wanted->{untaint_pattern}; |
610 | $untaint_skip = $wanted->{untaint_skip}; |
611 | $dangling_symlinks = $wanted->{dangling_symlinks}; |
81793b90 |
612 | |
1cffc1dd |
613 | # for compatibility reasons (find.pl, find2perl) |
9f826d6a |
614 | local our ($topdir, $topdev, $topino, $topmode, $topnlink); |
81793b90 |
615 | |
616 | # a symbolic link to a directory doesn't increase the link count |
617 | $avoid_nlink = $follow || $File::Find::dont_use_nlink; |
618 | |
e7b91b67 |
619 | my ($abs_dir, $Is_Dir); |
81793b90 |
620 | |
621 | Proc_Top_Item: |
4c621faf |
622 | foreach my $TOP (@_) { |
7e47e6ff |
623 | my $top_item = $TOP; |
624 | |
625 | if ($Is_MacOS) { |
626 | ($topdev,$topino,$topmode,$topnlink) = $follow ? stat $top_item : lstat $top_item; |
627 | $top_item = ":$top_item" |
3fa6e24b |
628 | if ( (-d _) && ( $top_item !~ /:/ ) ); |
7e47e6ff |
629 | } |
630 | else { |
631 | $top_item =~ s|/\z|| unless $top_item eq '/'; |
632 | ($topdev,$topino,$topmode,$topnlink) = $follow ? stat $top_item : lstat $top_item; |
633 | } |
634 | |
635 | $Is_Dir= 0; |
636 | |
637 | if ($follow) { |
638 | |
639 | if ($Is_MacOS) { |
640 | $cwd = "$cwd:" unless ($cwd =~ /:$/); # for safety |
641 | |
642 | if ($top_item eq $File::Find::current_dir) { |
643 | $abs_dir = $cwd; |
644 | } |
645 | else { |
646 | $abs_dir = contract_name_Mac($cwd, $top_item); |
647 | unless (defined $abs_dir) { |
cd68ec93 |
648 | warnings::warnif "Can't determine absolute path for $top_item (No such file or directory)\n"; |
7e47e6ff |
649 | next Proc_Top_Item; |
650 | } |
651 | } |
652 | |
653 | } |
654 | else { |
655 | if (substr($top_item,0,1) eq '/') { |
656 | $abs_dir = $top_item; |
657 | } |
658 | elsif ($top_item eq $File::Find::current_dir) { |
659 | $abs_dir = $cwd; |
660 | } |
661 | else { # care about any ../ |
662 | $abs_dir = contract_name("$cwd/",$top_item); |
663 | } |
664 | } |
665 | $abs_dir= Follow_SymLink($abs_dir); |
666 | unless (defined $abs_dir) { |
80e52b73 |
667 | if ($dangling_symlinks) { |
668 | if (ref $dangling_symlinks eq 'CODE') { |
669 | $dangling_symlinks->($top_item, $cwd); |
670 | } else { |
cd68ec93 |
671 | warnings::warnif "$top_item is a dangling symbolic link\n"; |
80e52b73 |
672 | } |
673 | } |
81793b90 |
674 | next Proc_Top_Item; |
7e47e6ff |
675 | } |
676 | |
677 | if (-d _) { |
81793b90 |
678 | _find_dir_symlnk($wanted, $abs_dir, $top_item); |
679 | $Is_Dir= 1; |
7e47e6ff |
680 | } |
681 | } |
81793b90 |
682 | else { # no follow |
7e47e6ff |
683 | $topdir = $top_item; |
684 | unless (defined $topnlink) { |
cd68ec93 |
685 | warnings::warnif "Can't stat $top_item: $!\n"; |
7e47e6ff |
686 | next Proc_Top_Item; |
687 | } |
688 | if (-d _) { |
544ff7a7 |
689 | $top_item =~ s/\.dir\z//i if $Is_VMS; |
e7b91b67 |
690 | _find_dir($wanted, $top_item, $topnlink); |
81793b90 |
691 | $Is_Dir= 1; |
7e47e6ff |
692 | } |
237437d0 |
693 | else { |
81793b90 |
694 | $abs_dir= $top_item; |
7e47e6ff |
695 | } |
696 | } |
81793b90 |
697 | |
7e47e6ff |
698 | unless ($Is_Dir) { |
81793b90 |
699 | unless (($_,$dir) = File::Basename::fileparse($abs_dir)) { |
7e47e6ff |
700 | if ($Is_MacOS) { |
701 | ($dir,$_) = (':', $top_item); # $File::Find::dir, $_ |
702 | } |
703 | else { |
704 | ($dir,$_) = ('./', $top_item); |
705 | } |
81793b90 |
706 | } |
707 | |
7e47e6ff |
708 | $abs_dir = $dir; |
709 | if (( $untaint ) && (is_tainted($dir) )) { |
710 | ( $abs_dir ) = $dir =~ m|$untaint_pat|; |
81793b90 |
711 | unless (defined $abs_dir) { |
712 | if ($untaint_skip == 0) { |
7e47e6ff |
713 | die "directory $dir is still tainted"; |
81793b90 |
714 | } |
715 | else { |
716 | next Proc_Top_Item; |
717 | } |
718 | } |
7e47e6ff |
719 | } |
81793b90 |
720 | |
7e47e6ff |
721 | unless ($no_chdir || chdir $abs_dir) { |
cd68ec93 |
722 | warnings::warnif "Couldn't chdir $abs_dir: $!\n"; |
7e47e6ff |
723 | next Proc_Top_Item; |
724 | } |
719911cc |
725 | |
7e47e6ff |
726 | $name = $abs_dir . $_; # $File::Find::name |
3bb6d3e5 |
727 | $_ = $name if $no_chdir; |
719911cc |
728 | |
abfdd623 |
729 | { $wanted_callback->() }; # protect against wild "next" |
81793b90 |
730 | |
7e47e6ff |
731 | } |
81793b90 |
732 | |
7e47e6ff |
733 | unless ( $no_chdir ) { |
734 | if ( ($check_t_cwd) && (($untaint) && (is_tainted($cwd) )) ) { |
735 | ( $cwd_untainted ) = $cwd =~ m|$untaint_pat|; |
736 | unless (defined $cwd_untainted) { |
737 | die "insecure cwd in find(depth)"; |
738 | } |
739 | $check_t_cwd = 0; |
740 | } |
741 | unless (chdir $cwd_untainted) { |
742 | die "Can't cd to $cwd: $!\n"; |
743 | } |
744 | } |
81793b90 |
745 | } |
746 | } |
747 | |
748 | # API: |
749 | # $wanted |
750 | # $p_dir : "parent directory" |
751 | # $nlink : what came back from the stat |
752 | # preconditions: |
753 | # chdir (if not no_chdir) to dir |
754 | |
755 | sub _find_dir($$$) { |
756 | my ($wanted, $p_dir, $nlink) = @_; |
757 | my ($CdLvl,$Level) = (0,0); |
758 | my @Stack; |
759 | my @filenames; |
760 | my ($subcount,$sub_nlink); |
761 | my $SE= []; |
762 | my $dir_name= $p_dir; |
7e47e6ff |
763 | my $dir_pref; |
39e79f6b |
764 | my $dir_rel = $File::Find::current_dir; |
7e47e6ff |
765 | my $tainted = 0; |
5fa2bf2b |
766 | my $no_nlink; |
7e47e6ff |
767 | |
768 | if ($Is_MacOS) { |
769 | $dir_pref= ($p_dir =~ /:$/) ? $p_dir : "$p_dir:"; # preface |
7e47e6ff |
770 | } |
771 | else { |
772 | $dir_pref= ( $p_dir eq '/' ? '/' : "$p_dir/" ); |
7e47e6ff |
773 | } |
81793b90 |
774 | |
775 | local ($dir, $name, $prune, *DIR); |
7e47e6ff |
776 | |
777 | unless ( $no_chdir || ($p_dir eq $File::Find::current_dir)) { |
81793b90 |
778 | my $udir = $p_dir; |
7e47e6ff |
779 | if (( $untaint ) && (is_tainted($p_dir) )) { |
780 | ( $udir ) = $p_dir =~ m|$untaint_pat|; |
81793b90 |
781 | unless (defined $udir) { |
782 | if ($untaint_skip == 0) { |
783 | die "directory $p_dir is still tainted"; |
784 | } |
785 | else { |
786 | return; |
787 | } |
237437d0 |
788 | } |
a0d0e21e |
789 | } |
8d8eebbf |
790 | unless (chdir ($Is_VMS && $udir !~ /[\/\[<]+/ ? "./$udir" : $udir)) { |
cd68ec93 |
791 | warnings::warnif "Can't cd to $udir: $!\n"; |
81793b90 |
792 | return; |
793 | } |
794 | } |
7e47e6ff |
795 | |
796 | # push the starting directory |
57e73c4b |
797 | push @Stack,[$CdLvl,$p_dir,$dir_rel,-1] if $bydepth; |
81793b90 |
798 | |
7e47e6ff |
799 | if ($Is_MacOS) { |
800 | $p_dir = $dir_pref; # ensure trailing ':' |
801 | } |
802 | |
81793b90 |
803 | while (defined $SE) { |
804 | unless ($bydepth) { |
7e47e6ff |
805 | $dir= $p_dir; # $File::Find::dir |
806 | $name= $dir_name; # $File::Find::name |
807 | $_= ($no_chdir ? $dir_name : $dir_rel ); # $_ |
81793b90 |
808 | # prune may happen here |
7e47e6ff |
809 | $prune= 0; |
abfdd623 |
810 | { $wanted_callback->() }; # protect against wild "next" |
7e47e6ff |
811 | next if $prune; |
81793b90 |
812 | } |
7e47e6ff |
813 | |
81793b90 |
814 | # change to that directory |
7e47e6ff |
815 | unless ($no_chdir || ($dir_rel eq $File::Find::current_dir)) { |
81793b90 |
816 | my $udir= $dir_rel; |
7e47e6ff |
817 | if ( ($untaint) && (($tainted) || ($tainted = is_tainted($dir_rel) )) ) { |
818 | ( $udir ) = $dir_rel =~ m|$untaint_pat|; |
81793b90 |
819 | unless (defined $udir) { |
820 | if ($untaint_skip == 0) { |
7e47e6ff |
821 | if ($Is_MacOS) { |
822 | die "directory ($p_dir) $dir_rel is still tainted"; |
823 | } |
824 | else { |
825 | die "directory (" . ($p_dir ne '/' ? $p_dir : '') . "/) $dir_rel is still tainted"; |
826 | } |
827 | } else { # $untaint_skip == 1 |
828 | next; |
81793b90 |
829 | } |
830 | } |
831 | } |
8d8eebbf |
832 | unless (chdir ($Is_VMS && $udir !~ /[\/\[<]+/ ? "./$udir" : $udir)) { |
7e47e6ff |
833 | if ($Is_MacOS) { |
cd68ec93 |
834 | warnings::warnif "Can't cd to ($p_dir) $udir: $!\n"; |
7e47e6ff |
835 | } |
836 | else { |
cd68ec93 |
837 | warnings::warnif "Can't cd to (" . |
838 | ($p_dir ne '/' ? $p_dir : '') . "/) $udir: $!\n"; |
7e47e6ff |
839 | } |
81793b90 |
840 | next; |
841 | } |
842 | $CdLvl++; |
843 | } |
844 | |
7e47e6ff |
845 | if ($Is_MacOS) { |
846 | $dir_name = "$dir_name:" unless ($dir_name =~ /:$/); |
847 | } |
848 | |
849 | $dir= $dir_name; # $File::Find::dir |
81793b90 |
850 | |
851 | # Get the list of files in the current directory. |
7e47e6ff |
852 | unless (opendir DIR, ($no_chdir ? $dir_name : $File::Find::current_dir)) { |
cd68ec93 |
853 | warnings::warnif "Can't opendir($dir_name): $!\n"; |
81793b90 |
854 | next; |
855 | } |
856 | @filenames = readdir DIR; |
857 | closedir(DIR); |
abfdd623 |
858 | @filenames = $pre_process->(@filenames) if $pre_process; |
719c805e |
859 | push @Stack,[$CdLvl,$dir_name,"",-2] if $post_process; |
81793b90 |
860 | |
5fa2bf2b |
861 | # default: use whatever was specifid |
862 | # (if $nlink >= 2, and $avoid_nlink == 0, this will switch back) |
863 | $no_nlink = $avoid_nlink; |
864 | # if dir has wrong nlink count, force switch to slower stat method |
865 | $no_nlink = 1 if ($nlink < 2); |
866 | |
867 | if ($nlink == 2 && !$no_nlink) { |
81793b90 |
868 | # This dir has no subdirectories. |
869 | for my $FN (@filenames) { |
7e47e6ff |
870 | next if $FN =~ $File::Find::skip_pattern; |
81793b90 |
871 | |
7e47e6ff |
872 | $name = $dir_pref . $FN; # $File::Find::name |
873 | $_ = ($no_chdir ? $name : $FN); # $_ |
abfdd623 |
874 | { $wanted_callback->() }; # protect against wild "next" |
81793b90 |
875 | } |
876 | |
877 | } |
878 | else { |
879 | # This dir has subdirectories. |
880 | $subcount = $nlink - 2; |
881 | |
7bd31527 |
882 | # HACK: insert directories at this position. so as to preserve |
883 | # the user pre-processed ordering of files. |
884 | # EG: directory traversal is in user sorted order, not at random. |
885 | my $stack_top = @Stack; |
886 | |
81793b90 |
887 | for my $FN (@filenames) { |
7e47e6ff |
888 | next if $FN =~ $File::Find::skip_pattern; |
5fa2bf2b |
889 | if ($subcount > 0 || $no_nlink) { |
81793b90 |
890 | # Seen all the subdirs? |
891 | # check for directoriness. |
892 | # stat is faster for a file in the current directory |
07867069 |
893 | $sub_nlink = (lstat ($no_chdir ? $dir_pref . $FN : $FN))[3]; |
81793b90 |
894 | |
895 | if (-d _) { |
896 | --$subcount; |
544ff7a7 |
897 | $FN =~ s/\.dir\z//i if $Is_VMS; |
7bd31527 |
898 | # HACK: replace push to preserve dir traversal order |
899 | #push @Stack,[$CdLvl,$dir_name,$FN,$sub_nlink]; |
900 | splice @Stack, $stack_top, 0, |
901 | [$CdLvl,$dir_name,$FN,$sub_nlink]; |
81793b90 |
902 | } |
903 | else { |
7e47e6ff |
904 | $name = $dir_pref . $FN; # $File::Find::name |
905 | $_= ($no_chdir ? $name : $FN); # $_ |
abfdd623 |
906 | { $wanted_callback->() }; # protect against wild "next" |
81793b90 |
907 | } |
908 | } |
07867069 |
909 | else { |
7e47e6ff |
910 | $name = $dir_pref . $FN; # $File::Find::name |
911 | $_= ($no_chdir ? $name : $FN); # $_ |
abfdd623 |
912 | { $wanted_callback->() }; # protect against wild "next" |
81793b90 |
913 | } |
914 | } |
915 | } |
17b275ff |
916 | } |
917 | continue { |
57e73c4b |
918 | while ( defined ($SE = pop @Stack) ) { |
81793b90 |
919 | ($Level, $p_dir, $dir_rel, $nlink) = @$SE; |
920 | if ($CdLvl > $Level && !$no_chdir) { |
7e47e6ff |
921 | my $tmp; |
922 | if ($Is_MacOS) { |
923 | $tmp = (':' x ($CdLvl-$Level)) . ':'; |
924 | } |
925 | else { |
926 | $tmp = join('/',('..') x ($CdLvl-$Level)); |
927 | } |
928 | die "Can't cd to $dir_name" . $tmp |
929 | unless chdir ($tmp); |
81793b90 |
930 | $CdLvl = $Level; |
931 | } |
7e47e6ff |
932 | |
933 | if ($Is_MacOS) { |
934 | # $pdir always has a trailing ':', except for the starting dir, |
935 | # where $dir_rel eq ':' |
936 | $dir_name = "$p_dir$dir_rel"; |
937 | $dir_pref = "$dir_name:"; |
938 | } |
939 | else { |
940 | $dir_name = ($p_dir eq '/' ? "/$dir_rel" : "$p_dir/$dir_rel"); |
941 | $dir_pref = "$dir_name/"; |
942 | } |
943 | |
719c805e |
944 | if ( $nlink == -2 ) { |
7e47e6ff |
945 | $name = $dir = $p_dir; # $File::Find::name / dir |
39e79f6b |
946 | $_ = $File::Find::current_dir; |
abfdd623 |
947 | $post_process->(); # End-of-directory processing |
7e47e6ff |
948 | } |
949 | elsif ( $nlink < 0 ) { # must be finddepth, report dirname now |
950 | $name = $dir_name; |
951 | if ($Is_MacOS) { |
952 | if ($dir_rel eq ':') { # must be the top dir, where we started |
953 | $name =~ s|:$||; # $File::Find::name |
954 | $p_dir = "$p_dir:" unless ($p_dir =~ /:$/); |
955 | } |
956 | $dir = $p_dir; # $File::Find::dir |
957 | $_ = ($no_chdir ? $name : $dir_rel); # $_ |
958 | } |
959 | else { |
960 | if ( substr($name,-2) eq '/.' ) { |
5cf0a2f2 |
961 | substr($name, length($name) == 2 ? -1 : -2) = ''; |
7e47e6ff |
962 | } |
963 | $dir = $p_dir; |
964 | $_ = ($no_chdir ? $dir_name : $dir_rel ); |
965 | if ( substr($_,-2) eq '/.' ) { |
5cf0a2f2 |
966 | substr($_, length($_) == 2 ? -1 : -2) = ''; |
7e47e6ff |
967 | } |
968 | } |
abfdd623 |
969 | { $wanted_callback->() }; # protect against wild "next" |
7e47e6ff |
970 | } |
971 | else { |
972 | push @Stack,[$CdLvl,$p_dir,$dir_rel,-1] if $bydepth; |
973 | last; |
974 | } |
81793b90 |
975 | } |
a0d0e21e |
976 | } |
977 | } |
978 | |
81793b90 |
979 | |
980 | # API: |
981 | # $wanted |
982 | # $dir_loc : absolute location of a dir |
983 | # $p_dir : "parent directory" |
984 | # preconditions: |
985 | # chdir (if not no_chdir) to dir |
986 | |
987 | sub _find_dir_symlnk($$$) { |
7e47e6ff |
988 | my ($wanted, $dir_loc, $p_dir) = @_; # $dir_loc is the absolute directory |
81793b90 |
989 | my @Stack; |
990 | my @filenames; |
991 | my $new_loc; |
7e47e6ff |
992 | my $updir_loc = $dir_loc; # untainted parent directory |
81793b90 |
993 | my $SE = []; |
994 | my $dir_name = $p_dir; |
7e47e6ff |
995 | my $dir_pref; |
996 | my $loc_pref; |
39e79f6b |
997 | my $dir_rel = $File::Find::current_dir; |
7e47e6ff |
998 | my $byd_flag; # flag for pending stack entry if $bydepth |
999 | my $tainted = 0; |
1000 | my $ok = 1; |
1001 | |
1002 | if ($Is_MacOS) { |
1003 | $dir_pref = ($p_dir =~ /:$/) ? "$p_dir" : "$p_dir:"; |
1004 | $loc_pref = ($dir_loc =~ /:$/) ? "$dir_loc" : "$dir_loc:"; |
7e47e6ff |
1005 | } else { |
1006 | $dir_pref = ( $p_dir eq '/' ? '/' : "$p_dir/" ); |
1007 | $loc_pref = ( $dir_loc eq '/' ? '/' : "$dir_loc/" ); |
7e47e6ff |
1008 | } |
81793b90 |
1009 | |
1010 | local ($dir, $name, $fullname, $prune, *DIR); |
7e47e6ff |
1011 | |
1012 | unless ($no_chdir) { |
1013 | # untaint the topdir |
1014 | if (( $untaint ) && (is_tainted($dir_loc) )) { |
1015 | ( $updir_loc ) = $dir_loc =~ m|$untaint_pat|; # parent dir, now untainted |
1016 | # once untainted, $updir_loc is pushed on the stack (as parent directory); |
1017 | # hence, we don't need to untaint the parent directory every time we chdir |
1018 | # to it later |
1019 | unless (defined $updir_loc) { |
81793b90 |
1020 | if ($untaint_skip == 0) { |
1021 | die "directory $dir_loc is still tainted"; |
1022 | } |
1023 | else { |
1024 | return; |
1025 | } |
1026 | } |
1027 | } |
7e47e6ff |
1028 | $ok = chdir($updir_loc) unless ($p_dir eq $File::Find::current_dir); |
1029 | unless ($ok) { |
cd68ec93 |
1030 | warnings::warnif "Can't cd to $updir_loc: $!\n"; |
81793b90 |
1031 | return; |
1032 | } |
1033 | } |
1034 | |
7e47e6ff |
1035 | push @Stack,[$dir_loc,$updir_loc,$p_dir,$dir_rel,-1] if $bydepth; |
1036 | |
1037 | if ($Is_MacOS) { |
1038 | $p_dir = $dir_pref; # ensure trailing ':' |
1039 | } |
57e73c4b |
1040 | |
81793b90 |
1041 | while (defined $SE) { |
1042 | |
1043 | unless ($bydepth) { |
7e47e6ff |
1044 | # change (back) to parent directory (always untainted) |
704ea872 |
1045 | unless ($no_chdir) { |
7e47e6ff |
1046 | unless (chdir $updir_loc) { |
cd68ec93 |
1047 | warnings::warnif "Can't cd to $updir_loc: $!\n"; |
704ea872 |
1048 | next; |
1049 | } |
1050 | } |
7e47e6ff |
1051 | $dir= $p_dir; # $File::Find::dir |
1052 | $name= $dir_name; # $File::Find::name |
1053 | $_= ($no_chdir ? $dir_name : $dir_rel ); # $_ |
1054 | $fullname= $dir_loc; # $File::Find::fullname |
81793b90 |
1055 | # prune may happen here |
7e47e6ff |
1056 | $prune= 0; |
704ea872 |
1057 | lstat($_); # make sure file tests with '_' work |
abfdd623 |
1058 | { $wanted_callback->() }; # protect against wild "next" |
7e47e6ff |
1059 | next if $prune; |
81793b90 |
1060 | } |
1061 | |
1062 | # change to that directory |
7e47e6ff |
1063 | unless ($no_chdir || ($dir_rel eq $File::Find::current_dir)) { |
1064 | $updir_loc = $dir_loc; |
1065 | if ( ($untaint) && (($tainted) || ($tainted = is_tainted($dir_loc) )) ) { |
1066 | # untaint $dir_loc, what will be pushed on the stack as (untainted) parent dir |
1067 | ( $updir_loc ) = $dir_loc =~ m|$untaint_pat|; |
1068 | unless (defined $updir_loc) { |
81793b90 |
1069 | if ($untaint_skip == 0) { |
1070 | die "directory $dir_loc is still tainted"; |
a0d0e21e |
1071 | } |
237437d0 |
1072 | else { |
81793b90 |
1073 | next; |
237437d0 |
1074 | } |
a0d0e21e |
1075 | } |
1076 | } |
7e47e6ff |
1077 | unless (chdir $updir_loc) { |
cd68ec93 |
1078 | warnings::warnif "Can't cd to $updir_loc: $!\n"; |
81793b90 |
1079 | next; |
1080 | } |
1081 | } |
1082 | |
7e47e6ff |
1083 | if ($Is_MacOS) { |
1084 | $dir_name = "$dir_name:" unless ($dir_name =~ /:$/); |
1085 | } |
1086 | |
1087 | $dir = $dir_name; # $File::Find::dir |
81793b90 |
1088 | |
1089 | # Get the list of files in the current directory. |
7e47e6ff |
1090 | unless (opendir DIR, ($no_chdir ? $dir_loc : $File::Find::current_dir)) { |
cd68ec93 |
1091 | warnings::warnif "Can't opendir($dir_loc): $!\n"; |
81793b90 |
1092 | next; |
1093 | } |
1094 | @filenames = readdir DIR; |
1095 | closedir(DIR); |
1096 | |
1097 | for my $FN (@filenames) { |
7e47e6ff |
1098 | next if $FN =~ $File::Find::skip_pattern; |
81793b90 |
1099 | |
1100 | # follow symbolic links / do an lstat |
07867069 |
1101 | $new_loc = Follow_SymLink($loc_pref.$FN); |
81793b90 |
1102 | |
1103 | # ignore if invalid symlink |
1104 | next unless defined $new_loc; |
7e47e6ff |
1105 | |
81793b90 |
1106 | if (-d _) { |
7e47e6ff |
1107 | push @Stack,[$new_loc,$updir_loc,$dir_name,$FN,1]; |
81793b90 |
1108 | } |
1109 | else { |
7e47e6ff |
1110 | $fullname = $new_loc; # $File::Find::fullname |
1111 | $name = $dir_pref . $FN; # $File::Find::name |
1112 | $_ = ($no_chdir ? $name : $FN); # $_ |
abfdd623 |
1113 | { $wanted_callback->() }; # protect against wild "next" |
81793b90 |
1114 | } |
1115 | } |
1116 | |
81793b90 |
1117 | } |
1118 | continue { |
57e73c4b |
1119 | while (defined($SE = pop @Stack)) { |
7e47e6ff |
1120 | ($dir_loc, $updir_loc, $p_dir, $dir_rel, $byd_flag) = @$SE; |
1121 | if ($Is_MacOS) { |
1122 | # $p_dir always has a trailing ':', except for the starting dir, |
1123 | # where $dir_rel eq ':' |
1124 | $dir_name = "$p_dir$dir_rel"; |
1125 | $dir_pref = "$dir_name:"; |
1126 | $loc_pref = ($dir_loc =~ /:$/) ? $dir_loc : "$dir_loc:"; |
1127 | } |
1128 | else { |
1129 | $dir_name = ($p_dir eq '/' ? "/$dir_rel" : "$p_dir/$dir_rel"); |
1130 | $dir_pref = "$dir_name/"; |
1131 | $loc_pref = "$dir_loc/"; |
1132 | } |
1133 | if ( $byd_flag < 0 ) { # must be finddepth, report dirname now |
1134 | unless ($no_chdir || ($dir_rel eq $File::Find::current_dir)) { |
1135 | unless (chdir $updir_loc) { # $updir_loc (parent dir) is always untainted |
cd68ec93 |
1136 | warnings::warnif "Can't cd to $updir_loc: $!\n"; |
7e47e6ff |
1137 | next; |
1138 | } |
1139 | } |
1140 | $fullname = $dir_loc; # $File::Find::fullname |
1141 | $name = $dir_name; # $File::Find::name |
1142 | if ($Is_MacOS) { |
1143 | if ($dir_rel eq ':') { # must be the top dir, where we started |
1144 | $name =~ s|:$||; # $File::Find::name |
1145 | $p_dir = "$p_dir:" unless ($p_dir =~ /:$/); |
1146 | } |
1147 | $dir = $p_dir; # $File::Find::dir |
1148 | $_ = ($no_chdir ? $name : $dir_rel); # $_ |
1149 | } |
1150 | else { |
1151 | if ( substr($name,-2) eq '/.' ) { |
f801979b |
1152 | substr($name, length($name) == 2 ? -1 : -2) = ''; # $File::Find::name |
7e47e6ff |
1153 | } |
1154 | $dir = $p_dir; # $File::Find::dir |
1155 | $_ = ($no_chdir ? $dir_name : $dir_rel); # $_ |
1156 | if ( substr($_,-2) eq '/.' ) { |
f801979b |
1157 | substr($_, length($_) == 2 ? -1 : -2) = ''; |
7e47e6ff |
1158 | } |
1159 | } |
1160 | |
1161 | lstat($_); # make sure file tests with '_' work |
abfdd623 |
1162 | { $wanted_callback->() }; # protect against wild "next" |
7e47e6ff |
1163 | } |
1164 | else { |
1165 | push @Stack,[$dir_loc, $updir_loc, $p_dir, $dir_rel,-1] if $bydepth; |
1166 | last; |
1167 | } |
a0d0e21e |
1168 | } |
1169 | } |
1170 | } |
1171 | |
81793b90 |
1172 | |
20408e3c |
1173 | sub wrap_wanted { |
81793b90 |
1174 | my $wanted = shift; |
1175 | if ( ref($wanted) eq 'HASH' ) { |
1176 | if ( $wanted->{follow} || $wanted->{follow_fast}) { |
1177 | $wanted->{follow_skip} = 1 unless defined $wanted->{follow_skip}; |
1178 | } |
1179 | if ( $wanted->{untaint} ) { |
7e47e6ff |
1180 | $wanted->{untaint_pattern} = $File::Find::untaint_pattern |
81793b90 |
1181 | unless defined $wanted->{untaint_pattern}; |
1182 | $wanted->{untaint_skip} = 0 unless defined $wanted->{untaint_skip}; |
1183 | } |
1184 | return $wanted; |
1185 | } |
1186 | else { |
1187 | return { wanted => $wanted }; |
1188 | } |
a0d0e21e |
1189 | } |
1190 | |
20408e3c |
1191 | sub find { |
81793b90 |
1192 | my $wanted = shift; |
1193 | _find_opt(wrap_wanted($wanted), @_); |
a0d0e21e |
1194 | } |
1195 | |
55d729e4 |
1196 | sub finddepth { |
81793b90 |
1197 | my $wanted = wrap_wanted(shift); |
1198 | $wanted->{bydepth} = 1; |
1199 | _find_opt($wanted, @_); |
20408e3c |
1200 | } |
6280b799 |
1201 | |
7e47e6ff |
1202 | # default |
1203 | $File::Find::skip_pattern = qr/^\.{1,2}\z/; |
1204 | $File::Find::untaint_pattern = qr|^([-+@\w./]+)$|; |
1205 | |
6280b799 |
1206 | # These are hard-coded for now, but may move to hint files. |
10eba763 |
1207 | if ($^O eq 'VMS') { |
81793b90 |
1208 | $Is_VMS = 1; |
7e47e6ff |
1209 | $File::Find::dont_use_nlink = 1; |
1210 | } |
1211 | elsif ($^O eq 'MacOS') { |
1212 | $Is_MacOS = 1; |
1213 | $File::Find::dont_use_nlink = 1; |
1214 | $File::Find::skip_pattern = qr/^Icon\015\z/; |
1215 | $File::Find::untaint_pattern = qr|^(.+)$|; |
748a9306 |
1216 | } |
1217 | |
7e47e6ff |
1218 | # this _should_ work properly on all platforms |
1219 | # where File::Find can be expected to work |
1220 | $File::Find::current_dir = File::Spec->curdir || '.'; |
1221 | |
81793b90 |
1222 | $File::Find::dont_use_nlink = 1 |
497711e7 |
1223 | if $^O eq 'os2' || $^O eq 'dos' || $^O eq 'amigaos' || $^O eq 'MSWin32' || |
1119cb72 |
1224 | $^O eq 'cygwin' || $^O eq 'epoc' || $^O eq 'qnx' || |
1225 | $^O eq 'nto'; |
6280b799 |
1226 | |
20408e3c |
1227 | # Set dont_use_nlink in your hint file if your system's stat doesn't |
1228 | # report the number of links in a directory as an indication |
1229 | # of the number of files. |
1230 | # See, e.g. hints/machten.sh for MachTen 2.2. |
81793b90 |
1231 | unless ($File::Find::dont_use_nlink) { |
1232 | require Config; |
1233 | $File::Find::dont_use_nlink = 1 if ($Config::Config{'dont_use_nlink'}); |
20408e3c |
1234 | } |
1235 | |
7e47e6ff |
1236 | # We need a function that checks if a scalar is tainted. Either use the |
1237 | # Scalar::Util module's tainted() function or our (slower) pure Perl |
1238 | # fallback is_tainted_pp() |
1239 | { |
1240 | local $@; |
1241 | eval { require Scalar::Util }; |
1242 | *is_tainted = $@ ? \&is_tainted_pp : \&Scalar::Util::tainted; |
1243 | } |
1244 | |
a0d0e21e |
1245 | 1; |