Commit | Line | Data |
---|---|---|
920dae64 AT |
1 | #!perl -w |
2 | ||
3 | # use strict fails | |
4 | #Can't use string ("main::glob") as a symbol ref while "strict refs" in use at /usr/lib/perl5/5.005/File/DosGlob.pm line 191. | |
5 | ||
6 | # | |
7 | # Documentation at the __END__ | |
8 | # | |
9 | ||
10 | package File::DosGlob; | |
11 | ||
12 | our $VERSION = '1.00'; | |
13 | use strict; | |
14 | use warnings; | |
15 | ||
16 | sub doglob { | |
17 | my $cond = shift; | |
18 | my @retval = (); | |
19 | #print "doglob: ", join('|', @_), "\n"; | |
20 | OUTER: | |
21 | for my $pat (@_) { | |
22 | my @matched = (); | |
23 | my @globdirs = (); | |
24 | my $head = '.'; | |
25 | my $sepchr = '/'; | |
26 | my $tail; | |
27 | next OUTER unless defined $pat and $pat ne ''; | |
28 | # if arg is within quotes strip em and do no globbing | |
29 | if ($pat =~ /^"(.*)"\z/s) { | |
30 | $pat = $1; | |
31 | if ($cond eq 'd') { push(@retval, $pat) if -d $pat } | |
32 | else { push(@retval, $pat) if -e $pat } | |
33 | next OUTER; | |
34 | } | |
35 | # wildcards with a drive prefix such as h:*.pm must be changed | |
36 | # to h:./*.pm to expand correctly | |
37 | if ($pat =~ m|^([A-Za-z]:)[^/\\]|s) { | |
38 | substr($_,0,2) = $1 . "./"; | |
39 | } | |
40 | if ($pat =~ m|^(.*)([\\/])([^\\/]*)\z|s) { | |
41 | ($head, $sepchr, $tail) = ($1,$2,$3); | |
42 | #print "div: |$head|$sepchr|$tail|\n"; | |
43 | push (@retval, $pat), next OUTER if $tail eq ''; | |
44 | if ($head =~ /[*?]/) { | |
45 | @globdirs = doglob('d', $head); | |
46 | push(@retval, doglob($cond, map {"$_$sepchr$tail"} @globdirs)), | |
47 | next OUTER if @globdirs; | |
48 | } | |
49 | $head .= $sepchr if $head eq '' or $head =~ /^[A-Za-z]:\z/s; | |
50 | $pat = $tail; | |
51 | } | |
52 | # | |
53 | # If file component has no wildcards, we can avoid opendir | |
54 | unless ($pat =~ /[*?]/) { | |
55 | $head = '' if $head eq '.'; | |
56 | $head .= $sepchr unless $head eq '' or substr($head,-1) eq $sepchr; | |
57 | $head .= $pat; | |
58 | if ($cond eq 'd') { push(@retval,$head) if -d $head } | |
59 | else { push(@retval,$head) if -e $head } | |
60 | next OUTER; | |
61 | } | |
62 | opendir(D, $head) or next OUTER; | |
63 | my @leaves = readdir D; | |
64 | closedir D; | |
65 | $head = '' if $head eq '.'; | |
66 | $head .= $sepchr unless $head eq '' or substr($head,-1) eq $sepchr; | |
67 | ||
68 | # escape regex metachars but not glob chars | |
69 | $pat =~ s:([].+^\-\${}[|]):\\$1:g; | |
70 | # and convert DOS-style wildcards to regex | |
71 | $pat =~ s/\*/.*/g; | |
72 | $pat =~ s/\?/.?/g; | |
73 | ||
74 | #print "regex: '$pat', head: '$head'\n"; | |
75 | my $matchsub = sub { $_[0] =~ m|^$pat\z|is }; | |
76 | INNER: | |
77 | for my $e (@leaves) { | |
78 | next INNER if $e eq '.' or $e eq '..'; | |
79 | next INNER if $cond eq 'd' and ! -d "$head$e"; | |
80 | push(@matched, "$head$e"), next INNER if &$matchsub($e); | |
81 | # | |
82 | # [DOS compatibility special case] | |
83 | # Failed, add a trailing dot and try again, but only | |
84 | # if name does not have a dot in it *and* pattern | |
85 | # has a dot *and* name is shorter than 9 chars. | |
86 | # | |
87 | if (index($e,'.') == -1 and length($e) < 9 | |
88 | and index($pat,'\\.') != -1) { | |
89 | push(@matched, "$head$e"), next INNER if &$matchsub("$e."); | |
90 | } | |
91 | } | |
92 | push @retval, @matched if @matched; | |
93 | } | |
94 | return @retval; | |
95 | } | |
96 | ||
97 | ||
98 | # | |
99 | # Do DOS-like globbing on Mac OS | |
100 | # | |
101 | sub doglob_Mac { | |
102 | my $cond = shift; | |
103 | my @retval = (); | |
104 | ||
105 | #print "doglob_Mac: ", join('|', @_), "\n"; | |
106 | OUTER: | |
107 | for my $arg (@_) { | |
108 | local $_ = $arg; | |
109 | my @matched = (); | |
110 | my @globdirs = (); | |
111 | my $head = ':'; | |
112 | my $not_esc_head = $head; | |
113 | my $sepchr = ':'; | |
114 | next OUTER unless defined $_ and $_ ne ''; | |
115 | # if arg is within quotes strip em and do no globbing | |
116 | if (/^"(.*)"\z/s) { | |
117 | $_ = $1; | |
118 | # $_ may contain escaped metachars '\*', '\?' and '\' | |
119 | my $not_esc_arg = $_; | |
120 | $not_esc_arg =~ s/\\([*?\\])/$1/g; | |
121 | if ($cond eq 'd') { push(@retval, $not_esc_arg) if -d $not_esc_arg } | |
122 | else { push(@retval, $not_esc_arg) if -e $not_esc_arg } | |
123 | next OUTER; | |
124 | } | |
125 | ||
126 | if (m|^(.*?)(:+)([^:]*)\z|s) { # note: $1 is not greedy | |
127 | my $tail; | |
128 | ($head, $sepchr, $tail) = ($1,$2,$3); | |
129 | #print "div: |$head|$sepchr|$tail|\n"; | |
130 | push (@retval, $_), next OUTER if $tail eq ''; | |
131 | # | |
132 | # $head may contain escaped metachars '\*' and '\?' | |
133 | ||
134 | my $tmp_head = $head; | |
135 | # if a '*' or '?' is preceded by an odd count of '\', temporary delete | |
136 | # it (and its preceding backslashes), i.e. don't treat '\*' and '\?' as | |
137 | # wildcards | |
138 | $tmp_head =~ s/(\\*)([*?])/$2 x ((length($1) + 1) % 2)/eg; | |
139 | ||
140 | if ($tmp_head =~ /[*?]/) { # if there are wildcards ... | |
141 | @globdirs = doglob_Mac('d', $head); | |
142 | push(@retval, doglob_Mac($cond, map {"$_$sepchr$tail"} @globdirs)), | |
143 | next OUTER if @globdirs; | |
144 | } | |
145 | ||
146 | $head .= $sepchr; | |
147 | $not_esc_head = $head; | |
148 | # unescape $head for file operations | |
149 | $not_esc_head =~ s/\\([*?\\])/$1/g; | |
150 | $_ = $tail; | |
151 | } | |
152 | # | |
153 | # If file component has no wildcards, we can avoid opendir | |
154 | ||
155 | my $tmp_tail = $_; | |
156 | # if a '*' or '?' is preceded by an odd count of '\', temporary delete | |
157 | # it (and its preceding backslashes), i.e. don't treat '\*' and '\?' as | |
158 | # wildcards | |
159 | $tmp_tail =~ s/(\\*)([*?])/$2 x ((length($1) + 1) % 2)/eg; | |
160 | ||
161 | unless ($tmp_tail =~ /[*?]/) { # if there are wildcards ... | |
162 | $not_esc_head = $head = '' if $head eq ':'; | |
163 | my $not_esc_tail = $_; | |
164 | # unescape $head and $tail for file operations | |
165 | $not_esc_tail =~ s/\\([*?\\])/$1/g; | |
166 | $head .= $_; | |
167 | $not_esc_head .= $not_esc_tail; | |
168 | if ($cond eq 'd') { push(@retval,$head) if -d $not_esc_head } | |
169 | else { push(@retval,$head) if -e $not_esc_head } | |
170 | next OUTER; | |
171 | } | |
172 | #print "opendir($not_esc_head)\n"; | |
173 | opendir(D, $not_esc_head) or next OUTER; | |
174 | my @leaves = readdir D; | |
175 | closedir D; | |
176 | ||
177 | # escape regex metachars but not '\' and glob chars '*', '?' | |
178 | $_ =~ s:([].+^\-\${}[|]):\\$1:g; | |
179 | # and convert DOS-style wildcards to regex, | |
180 | # but only if they are not escaped | |
181 | $_ =~ s/(\\*)([*?])/$1 . ('.' x ((length($1) + 1) % 2)) . $2/eg; | |
182 | ||
183 | #print "regex: '$_', head: '$head', unescaped head: '$not_esc_head'\n"; | |
184 | my $matchsub = eval 'sub { $_[0] =~ m|^' . $_ . '\\z|ios }'; | |
185 | warn($@), next OUTER if $@; | |
186 | INNER: | |
187 | for my $e (@leaves) { | |
188 | next INNER if $e eq '.' or $e eq '..'; | |
189 | next INNER if $cond eq 'd' and ! -d "$not_esc_head$e"; | |
190 | ||
191 | if (&$matchsub($e)) { | |
192 | my $leave = (($not_esc_head eq ':') && (-f "$not_esc_head$e")) ? | |
193 | "$e" : "$not_esc_head$e"; | |
194 | # | |
195 | # On Mac OS, the two glob metachars '*' and '?' and the escape | |
196 | # char '\' are valid characters for file and directory names. | |
197 | # We have to escape and treat them specially. | |
198 | $leave =~ s|([*?\\])|\\$1|g; | |
199 | push(@matched, $leave); | |
200 | next INNER; | |
201 | } | |
202 | } | |
203 | push @retval, @matched if @matched; | |
204 | } | |
205 | return @retval; | |
206 | } | |
207 | ||
208 | # | |
209 | # _expand_volume() will only be used on Mac OS (Classic): | |
210 | # Takes an array of original patterns as argument and returns an array of | |
211 | # possibly modified patterns. Each original pattern is processed like | |
212 | # that: | |
213 | # + If there's a volume name in the pattern, we push a separate pattern | |
214 | # for each mounted volume that matches (with '*', '?' and '\' escaped). | |
215 | # + If there's no volume name in the original pattern, it is pushed | |
216 | # unchanged. | |
217 | # Note that the returned array of patterns may be empty. | |
218 | # | |
219 | sub _expand_volume { | |
220 | ||
221 | require MacPerl; # to be verbose | |
222 | ||
223 | my @pat = @_; | |
224 | my @new_pat = (); | |
225 | my @FSSpec_Vols = MacPerl::Volumes(); | |
226 | my @mounted_volumes = (); | |
227 | ||
228 | foreach my $spec_vol (@FSSpec_Vols) { | |
229 | # push all mounted volumes into array | |
230 | push @mounted_volumes, MacPerl::MakePath($spec_vol); | |
231 | } | |
232 | #print "mounted volumes: |@mounted_volumes|\n"; | |
233 | ||
234 | while (@pat) { | |
235 | my $pat = shift @pat; | |
236 | if ($pat =~ /^([^:]+:)(.*)\z/) { # match a volume name? | |
237 | my $vol_pat = $1; | |
238 | my $tail = $2; | |
239 | # | |
240 | # escape regex metachars but not '\' and glob chars '*', '?' | |
241 | $vol_pat =~ s:([].+^\-\${}[|]):\\$1:g; | |
242 | # and convert DOS-style wildcards to regex, | |
243 | # but only if they are not escaped | |
244 | $vol_pat =~ s/(\\*)([*?])/$1 . ('.' x ((length($1) + 1) % 2)) . $2/eg; | |
245 | #print "volume regex: '$vol_pat' \n"; | |
246 | ||
247 | foreach my $volume (@mounted_volumes) { | |
248 | if ($volume =~ m|^$vol_pat\z|ios) { | |
249 | # | |
250 | # On Mac OS, the two glob metachars '*' and '?' and the | |
251 | # escape char '\' are valid characters for volume names. | |
252 | # We have to escape and treat them specially. | |
253 | $volume =~ s|([*?\\])|\\$1|g; | |
254 | push @new_pat, $volume . $tail; | |
255 | } | |
256 | } | |
257 | } else { # no volume name in pattern, push original pattern | |
258 | push @new_pat, $pat; | |
259 | } | |
260 | } | |
261 | return @new_pat; | |
262 | } | |
263 | ||
264 | ||
265 | # | |
266 | # _preprocess_pattern() will only be used on Mac OS (Classic): | |
267 | # Resolves any updirs in the pattern. Removes a single trailing colon | |
268 | # from the pattern, unless it's a volume name pattern like "*HD:" | |
269 | # | |
270 | sub _preprocess_pattern { | |
271 | my @pat = @_; | |
272 | ||
273 | foreach my $p (@pat) { | |
274 | my $proceed; | |
275 | # resolve any updirs, e.g. "*HD:t?p::a*" -> "*HD:a*" | |
276 | do { | |
277 | $proceed = ($p =~ s/^(.*):[^:]+::(.*?)\z/$1:$2/); | |
278 | } while ($proceed); | |
279 | # remove a single trailing colon, e.g. ":*:" -> ":*" | |
280 | $p =~ s/:([^:]+):\z/:$1/; | |
281 | } | |
282 | return @pat; | |
283 | } | |
284 | ||
285 | ||
286 | # | |
287 | # _un_escape() will only be used on Mac OS (Classic): | |
288 | # Unescapes a list of arguments which may contain escaped | |
289 | # metachars '*', '?' and '\'. | |
290 | # | |
291 | sub _un_escape { | |
292 | foreach (@_) { | |
293 | s/\\([*?\\])/$1/g; | |
294 | } | |
295 | return @_; | |
296 | } | |
297 | ||
298 | # | |
299 | # this can be used to override CORE::glob in a specific | |
300 | # package by saying C<use File::DosGlob 'glob';> in that | |
301 | # namespace. | |
302 | # | |
303 | ||
304 | # context (keyed by second cxix arg provided by core) | |
305 | my %iter; | |
306 | my %entries; | |
307 | ||
308 | sub glob { | |
309 | my($pat,$cxix) = @_; | |
310 | my @pat; | |
311 | ||
312 | # glob without args defaults to $_ | |
313 | $pat = $_ unless defined $pat; | |
314 | ||
315 | # extract patterns | |
316 | if ($pat =~ /\s/) { | |
317 | require Text::ParseWords; | |
318 | @pat = Text::ParseWords::parse_line('\s+',0,$pat); | |
319 | } | |
320 | else { | |
321 | push @pat, $pat; | |
322 | } | |
323 | ||
324 | # Mike Mestnik: made to do abc{1,2,3} == abc1 abc2 abc3. | |
325 | # abc3 will be the original {3} (and drop the {}). | |
326 | # abc1 abc2 will be put in @appendpat. | |
327 | # This was just the esiest way, not nearly the best. | |
328 | REHASH: { | |
329 | my @appendpat = (); | |
330 | for (@pat) { | |
331 | # There must be a "," I.E. abc{efg} is not what we want. | |
332 | while ( /^(.*)(?<!\\)\{(.*?)(?<!\\)\,.*?(?<!\\)\}(.*)$/ ) { | |
333 | my ($start, $match, $end) = ($1, $2, $3); | |
334 | #print "Got: \n\t$start\n\t$match\n\t$end\n"; | |
335 | my $tmp = "$start$match$end"; | |
336 | while ( $tmp =~ s/^(.*?)(?<!\\)\{(?:.*(?<!\\)\,)?(.*\Q$match\E.*?)(?:(?<!\\)\,.*)?(?<!\\)\}(.*)$/$1$2$3/ ) { | |
337 | #print "Striped: $tmp\n"; | |
338 | # these expanshions will be preformed by the original, | |
339 | # when we call REHASH. | |
340 | } | |
341 | push @appendpat, ("$tmp"); | |
342 | s/^\Q$start\E(?<!\\)\{\Q$match\E(?<!\\)\,/$start\{/; | |
343 | if ( /^\Q$start\E(?<!\\)\{(?!.*?(?<!\\)\,.*?\Q$end\E$)(.*)(?<!\\)\}\Q$end\E$/ ) { | |
344 | $match = $1; | |
345 | #print "GOT: \n\t$start\n\t$match\n\t$end\n\n"; | |
346 | $_ = "$start$match$end"; | |
347 | } | |
348 | } | |
349 | #print "Sould have "GOT" vs "Got"!\n"; | |
350 | #FIXME: There should be checking for this. | |
351 | # How or what should be done about failure is beond me. | |
352 | } | |
353 | if ( $#appendpat != -1 | |
354 | ) { | |
355 | #print "LOOP\n"; | |
356 | #FIXME: Max loop, no way! :") | |
357 | for ( @appendpat ) { | |
358 | push @pat, $_; | |
359 | } | |
360 | goto REHASH; | |
361 | } | |
362 | } | |
363 | for ( @pat ) { | |
364 | s/\\{/{/g; | |
365 | s/\\}/}/g; | |
366 | s/\\,/,/g; | |
367 | } | |
368 | #print join ("\n", @pat). "\n"; | |
369 | ||
370 | # assume global context if not provided one | |
371 | $cxix = '_G_' unless defined $cxix; | |
372 | $iter{$cxix} = 0 unless exists $iter{$cxix}; | |
373 | ||
374 | # if we're just beginning, do it all first | |
375 | if ($iter{$cxix} == 0) { | |
376 | if ($^O eq 'MacOS') { | |
377 | # first, take care of updirs and trailing colons | |
378 | @pat = _preprocess_pattern(@pat); | |
379 | # expand volume names | |
380 | @pat = _expand_volume(@pat); | |
381 | $entries{$cxix} = (@pat) ? [_un_escape( doglob_Mac(1,@pat) )] : [()]; | |
382 | } else { | |
383 | $entries{$cxix} = [doglob(1,@pat)]; | |
384 | } | |
385 | } | |
386 | ||
387 | # chuck it all out, quick or slow | |
388 | if (wantarray) { | |
389 | delete $iter{$cxix}; | |
390 | return @{delete $entries{$cxix}}; | |
391 | } | |
392 | else { | |
393 | if ($iter{$cxix} = scalar @{$entries{$cxix}}) { | |
394 | return shift @{$entries{$cxix}}; | |
395 | } | |
396 | else { | |
397 | # return undef for EOL | |
398 | delete $iter{$cxix}; | |
399 | delete $entries{$cxix}; | |
400 | return undef; | |
401 | } | |
402 | } | |
403 | } | |
404 | ||
405 | { | |
406 | no strict 'refs'; | |
407 | ||
408 | sub import { | |
409 | my $pkg = shift; | |
410 | return unless @_; | |
411 | my $sym = shift; | |
412 | my $callpkg = ($sym =~ s/^GLOBAL_//s ? 'CORE::GLOBAL' : caller(0)); | |
413 | *{$callpkg.'::'.$sym} = \&{$pkg.'::'.$sym} if $sym eq 'glob'; | |
414 | } | |
415 | } | |
416 | 1; | |
417 | ||
418 | __END__ | |
419 | ||
420 | =head1 NAME | |
421 | ||
422 | File::DosGlob - DOS like globbing and then some | |
423 | ||
424 | =head1 SYNOPSIS | |
425 | ||
426 | require 5.004; | |
427 | ||
428 | # override CORE::glob in current package | |
429 | use File::DosGlob 'glob'; | |
430 | ||
431 | # override CORE::glob in ALL packages (use with extreme caution!) | |
432 | use File::DosGlob 'GLOBAL_glob'; | |
433 | ||
434 | @perlfiles = glob "..\\pe?l/*.p?"; | |
435 | print <..\\pe?l/*.p?>; | |
436 | ||
437 | # from the command line (overrides only in main::) | |
438 | > perl -MFile::DosGlob=glob -e "print <../pe*/*p?>" | |
439 | ||
440 | =head1 DESCRIPTION | |
441 | ||
442 | A module that implements DOS-like globbing with a few enhancements. | |
443 | It is largely compatible with perlglob.exe (the M$ setargv.obj | |
444 | version) in all but one respect--it understands wildcards in | |
445 | directory components. | |
446 | ||
447 | For example, C<<..\\l*b\\file/*glob.p?>> will work as expected (in | |
448 | that it will find something like '..\lib\File/DosGlob.pm' alright). | |
449 | Note that all path components are case-insensitive, and that | |
450 | backslashes and forward slashes are both accepted, and preserved. | |
451 | You may have to double the backslashes if you are putting them in | |
452 | literally, due to double-quotish parsing of the pattern by perl. | |
453 | ||
454 | Spaces in the argument delimit distinct patterns, so | |
455 | C<glob('*.exe *.dll')> globs all filenames that end in C<.exe> | |
456 | or C<.dll>. If you want to put in literal spaces in the glob | |
457 | pattern, you can escape them with either double quotes, or backslashes. | |
458 | e.g. C<glob('c:/"Program Files"/*/*.dll')>, or | |
459 | C<glob('c:/Program\ Files/*/*.dll')>. The argument is tokenized using | |
460 | C<Text::ParseWords::parse_line()>, so see L<Text::ParseWords> for details | |
461 | of the quoting rules used. | |
462 | ||
463 | Extending it to csh patterns is left as an exercise to the reader. | |
464 | ||
465 | =head1 NOTES | |
466 | ||
467 | =over 4 | |
468 | ||
469 | =item * | |
470 | ||
471 | Mac OS (Classic) users should note a few differences. The specification | |
472 | of pathnames in glob patterns adheres to the usual Mac OS conventions: | |
473 | The path separator is a colon ':', not a slash '/' or backslash '\'. A | |
474 | full path always begins with a volume name. A relative pathname on Mac | |
475 | OS must always begin with a ':', except when specifying a file or | |
476 | directory name in the current working directory, where the leading colon | |
477 | is optional. If specifying a volume name only, a trailing ':' is | |
478 | required. Due to these rules, a glob like E<lt>*:E<gt> will find all | |
479 | mounted volumes, while a glob like E<lt>*E<gt> or E<lt>:*E<gt> will find | |
480 | all files and directories in the current directory. | |
481 | ||
482 | Note that updirs in the glob pattern are resolved before the matching begins, | |
483 | i.e. a pattern like "*HD:t?p::a*" will be matched as "*HD:a*". Note also, | |
484 | that a single trailing ':' in the pattern is ignored (unless it's a volume | |
485 | name pattern like "*HD:"), i.e. a glob like <:*:> will find both directories | |
486 | I<and> files (and not, as one might expect, only directories). | |
487 | ||
488 | The metachars '*', '?' and the escape char '\' are valid characters in | |
489 | volume, directory and file names on Mac OS. Hence, if you want to match | |
490 | a '*', '?' or '\' literally, you have to escape these characters. Due to | |
491 | perl's quoting rules, things may get a bit complicated, when you want to | |
492 | match a string like '\*' literally, or when you want to match '\' literally, | |
493 | but treat the immediately following character '*' as metachar. So, here's a | |
494 | rule of thumb (applies to both single- and double-quoted strings): escape | |
495 | each '*' or '?' or '\' with a backslash, if you want to treat them literally, | |
496 | and then double each backslash and your are done. E.g. | |
497 | ||
498 | - Match '\*' literally | |
499 | ||
500 | escape both '\' and '*' : '\\\*' | |
501 | double the backslashes : '\\\\\\*' | |
502 | ||
503 | (Internally, the glob routine sees a '\\\*', which means that both '\' and | |
504 | '*' are escaped.) | |
505 | ||
506 | ||
507 | - Match '\' literally, treat '*' as metachar | |
508 | ||
509 | escape '\' but not '*' : '\\*' | |
510 | double the backslashes : '\\\\*' | |
511 | ||
512 | (Internally, the glob routine sees a '\\*', which means that '\' is escaped and | |
513 | '*' is not.) | |
514 | ||
515 | Note that you also have to quote literal spaces in the glob pattern, as described | |
516 | above. | |
517 | ||
518 | =back | |
519 | ||
520 | =head1 EXPORTS (by request only) | |
521 | ||
522 | glob() | |
523 | ||
524 | =head1 BUGS | |
525 | ||
526 | Should probably be built into the core, and needs to stop | |
527 | pandering to DOS habits. Needs a dose of optimizium too. | |
528 | ||
529 | =head1 AUTHOR | |
530 | ||
531 | Gurusamy Sarathy <gsar@activestate.com> | |
532 | ||
533 | =head1 HISTORY | |
534 | ||
535 | =over 4 | |
536 | ||
537 | =item * | |
538 | ||
539 | Support for globally overriding glob() (GSAR 3-JUN-98) | |
540 | ||
541 | =item * | |
542 | ||
543 | Scalar context, independent iterator context fixes (GSAR 15-SEP-97) | |
544 | ||
545 | =item * | |
546 | ||
547 | A few dir-vs-file optimizations result in glob importation being | |
548 | 10 times faster than using perlglob.exe, and using perlglob.bat is | |
549 | only twice as slow as perlglob.exe (GSAR 28-MAY-97) | |
550 | ||
551 | =item * | |
552 | ||
553 | Several cleanups prompted by lack of compatible perlglob.exe | |
554 | under Borland (GSAR 27-MAY-97) | |
555 | ||
556 | =item * | |
557 | ||
558 | Initial version (GSAR 20-FEB-97) | |
559 | ||
560 | =back | |
561 | ||
562 | =head1 SEE ALSO | |
563 | ||
564 | perl | |
565 | ||
566 | perlglob.bat | |
567 | ||
568 | Text::ParseWords | |
569 | ||
570 | =cut | |
571 |