1 | #!/usr/local/bin/perl -w
|
---|
2 | 'di';
|
---|
3 | 'ig00';
|
---|
4 | ##############################################################################
|
---|
5 | ##
|
---|
6 | ## search
|
---|
7 | ##
|
---|
8 | ## Jeffrey Friedl (jfriedl@omron.co.jp), Dec 1994.
|
---|
9 | ## Copyright 19.... ah hell, just take it.
|
---|
10 | ##
|
---|
11 | ## BLURB:
|
---|
12 | ## A combo of find and grep -- more or less do a 'grep' on a whole
|
---|
13 | ## directory tree. Fast, with lots of options. Much more powerful than
|
---|
14 | ## the simple "find ... | xargs grep ....". Has a full man page.
|
---|
15 | ## Powerfully customizable.
|
---|
16 | ##
|
---|
17 | ## This file is big, but mostly comments and man page.
|
---|
18 | ##
|
---|
19 | ## See man page for usage info.
|
---|
20 | ## Return value: 2=error, 1=nothing found, 0=something found.
|
---|
21 | ##
|
---|
22 |
|
---|
23 | $version = "950918.5";
|
---|
24 | ##
|
---|
25 | ## "950918.5";
|
---|
26 | ## Changed all 'sysread' to 'read' because Linux perl's don't seem
|
---|
27 | ## to like sysread()
|
---|
28 | ##
|
---|
29 | ## "941227.4";
|
---|
30 | ## Added -n, -u
|
---|
31 | ##
|
---|
32 | ## "941222.3"
|
---|
33 | ## Added -nice (due to Lionel Cons <Lionel.Cons@cern.ch>)
|
---|
34 | ## Removed any leading "./" from name.
|
---|
35 | ## Added default flags for ~/.search, including TTY, -nice, -list, etc.
|
---|
36 | ## Program name now has path removed when printed in diagnostics.
|
---|
37 | ## Added simple tilde-expansion to -dir arg.
|
---|
38 | ## Added -dskip, etc. Fixed -iregex bug.
|
---|
39 | ## Changed -dir to be additive, adding -ddir.
|
---|
40 | ## Now screen out devices, pipes, and sockets.
|
---|
41 | ## More tidying and lots of expanding of the man page
|
---|
42 | ##
|
---|
43 | ##
|
---|
44 | ## "941217.2";
|
---|
45 | ## initial release.
|
---|
46 |
|
---|
47 | $stripped=0;
|
---|
48 |
|
---|
49 | &init;
|
---|
50 | $rc_file = join('/', $ENV{'HOME'}, ".search");
|
---|
51 |
|
---|
52 | &check_args;
|
---|
53 |
|
---|
54 | ## Make sure we've got a regex.
|
---|
55 | ## Don't need one if -find or -showrc was specified.
|
---|
56 | $!=2, die "expecting regex arguments.\n"
|
---|
57 | if $FIND_ONLY == 0 && $showrc == 0 && @ARGV == 0;
|
---|
58 |
|
---|
59 | &prepare_to_search($rc_file);
|
---|
60 |
|
---|
61 | &import_program if !defined &dodir; ## BIG key to speed.
|
---|
62 |
|
---|
63 | ## do search while there are directories to be done.
|
---|
64 | &dodir(shift(@todo)) while @todo;
|
---|
65 |
|
---|
66 | &clear_message if $VERBOSE && $STDERR_IS_TTY;
|
---|
67 | exit($retval);
|
---|
68 | ###############################################################################
|
---|
69 |
|
---|
70 | sub init
|
---|
71 | {
|
---|
72 | ## initialize variables that might be reset by command-line args
|
---|
73 | $DOREP=0; ## set true by -dorep (redo multi-hardlink files)
|
---|
74 | $DOREP=1 if $^O eq 'MSWin32';
|
---|
75 | $DO_SORT=0; ## set by -sort (sort files in a dir before checking)
|
---|
76 | $FIND_ONLY=0; ## set by -find (don't search files)
|
---|
77 | $LIST_ONLY=0; ## set true by -l (list filenames only)
|
---|
78 | $NEWER=0; ## set by -newer, "-mtime -###"
|
---|
79 | $NICE=0; ## set by -nice (print human-readable output)
|
---|
80 | $NOLINKS=0; ## set true by -nolinks (don't follow symlinks)
|
---|
81 | $OLDER=0; ## set by -older, "-mtime ###"
|
---|
82 | $PREPEND_FILENAME=1; ## set false by -h (don't prefix lines with filename)
|
---|
83 | $REPORT_LINENUM=0; ## set true by -n (show line numbers)
|
---|
84 | $VERBOSE=0; ## set to a value by -v, -vv, etc. (verbose messages)
|
---|
85 | $WHY=0; ## set true by -why, -vvv+ (report why skipped)
|
---|
86 | $XDEV=0; ## set true by -xdev (stay on one filesystem)
|
---|
87 | $all=0; ## set true by -all (don't skip many kinds of files)
|
---|
88 | $iflag = ''; ## set to 'i' by -i (ignore case);
|
---|
89 | $norc=0; ## set by -norc (don't load rc file)
|
---|
90 | $showrc=0; ## set by -showrc (show what happens with rc file)
|
---|
91 | $underlineOK=0; ## set true by -u (watch for underline stuff)
|
---|
92 | $words=0; ## set true by -w (match whole-words only)
|
---|
93 | $DELAY=0; ## inter-file delay (seconds)
|
---|
94 | $retval=1; ## will set to 0 if we find anything.
|
---|
95 |
|
---|
96 | ## various elements of stat() that we might access
|
---|
97 | $STAT_DEV = 1;
|
---|
98 | $STAT_INODE = 2;
|
---|
99 | $STAT_MTIME = 9;
|
---|
100 |
|
---|
101 | $VV_PRINT_COUNT = 50; ## with -vv, print every VV_PRINT_COUNT files, or...
|
---|
102 | $VV_SIZE = 1024*1024; ## ...every VV_SIZE bytes searched
|
---|
103 | $vv_print = $vv_size = 0; ## running totals.
|
---|
104 |
|
---|
105 | ## set default options, in case the rc file wants them
|
---|
106 | $opt{'TTY'}= 1 if -t STDOUT;
|
---|
107 |
|
---|
108 | ## want to know this for debugging message stuff
|
---|
109 | $STDERR_IS_TTY = -t STDERR ? 1 : 0;
|
---|
110 | $STDERR_SCREWS_STDOUT = ($STDERR_IS_TTY && -t STDOUT) ? 1 : 0;
|
---|
111 |
|
---|
112 | $0 =~ s,.*/,,; ## clean up $0 for any diagnostics we'll be printing.
|
---|
113 | }
|
---|
114 |
|
---|
115 | ##
|
---|
116 | ## Check arguments.
|
---|
117 | ##
|
---|
118 | sub check_args
|
---|
119 | {
|
---|
120 | while (@ARGV && $ARGV[0] =~ m/^-/)
|
---|
121 | {
|
---|
122 | $arg = shift(@ARGV);
|
---|
123 |
|
---|
124 | if ($arg eq '-version' || ($VERBOSE && $arg eq '-help')) {
|
---|
125 | print qq/Jeffrey's file search, version "$version".\n/;
|
---|
126 | exit(0) unless $arg eq '-help';
|
---|
127 | }
|
---|
128 | if ($arg eq '-help') {
|
---|
129 | print <<INLINE_LITERAL_TEXT;
|
---|
130 | usage: $0 [options] [-e] [PerlRegex ....]
|
---|
131 | OPTIONS TELLING *WHERE* TO SEARCH:
|
---|
132 | -dir DIR start search at the named directory (default is current dir).
|
---|
133 | -xdev stay on starting file system.
|
---|
134 | -sort sort the files in each directory before processing.
|
---|
135 | -nolinks don't follow symbolic links.
|
---|
136 | OPTIONS TELLING WHICH FILES TO EVEN CONSIDER:
|
---|
137 | -mtime # consider files modified > # days ago (-# for < # days old)
|
---|
138 | -newer FILE consider files modified more recently than FILE (also -older)
|
---|
139 | -name GLOB consider files whose name matches pattern (also -regex).
|
---|
140 | -skip GLOB opposite of -name: identifies files to not consider.
|
---|
141 | -path GLOB like -name, but for files whose whole path is described.
|
---|
142 | -dpath/-dregex/-dskip versions for selecting or pruning directories.
|
---|
143 | -all don't skip any files marked to be skipped by the startup file.
|
---|
144 | -x<SPECIAL> (see manual, and/or try -showrc).
|
---|
145 | -why report why a file isn't checked (also implied by -vvvv).
|
---|
146 | OPTIONS TELLING WHAT TO DO WITH FILES THAT WILL BE CONSIDERED:
|
---|
147 | -f | -find just list files (PerlRegex ignored). Default is to grep them.
|
---|
148 | -ff | -ffind Does a faster -find (implies -find -all -dorep)
|
---|
149 | OPTIONS CONTROLLING HOW THE SEARCH IS DONE (AND WHAT IS PRINTED):
|
---|
150 | -l | -list only list files with matches, not the lines themselves.
|
---|
151 | -nice | -nnice print more "human readable" output.
|
---|
152 | -n prefix each output line with its line number in the file.
|
---|
153 | -h don't prefix output lines with file name.
|
---|
154 | -u also look "inside" manpage-style underlined text
|
---|
155 | -i do case-insensitive searching.
|
---|
156 | -w match words only (as defined by perl's \\b).
|
---|
157 | OTHER OPTIONS:
|
---|
158 | -v, -vv, -vvv various levels of message verbosity.
|
---|
159 | -e end of options (in case a regex looks like an option).
|
---|
160 | -showrc show what the rc file sets, then exit.
|
---|
161 | -norc don't load the rc file.
|
---|
162 | -dorep check files with multiple hard links multiple times.
|
---|
163 | INLINE_LITERAL_TEXT
|
---|
164 | print "Use -v -help for more verbose help.\n" unless $VERBOSE;
|
---|
165 | print "This script file is also a man page.\n" unless $stripped;
|
---|
166 | print <<INLINE_LITERAL_TEXT if $VERBOSE;
|
---|
167 |
|
---|
168 | If -f (or -find) given, PerlRegex is optional and ignored.
|
---|
169 | Otherwise, will search for files with lines matching any of the given regexes.
|
---|
170 |
|
---|
171 | Combining things like -name and -mtime implies boolean AND.
|
---|
172 | However, duplicating things (such as -name '*.c' -name '*.txt') implies OR.
|
---|
173 |
|
---|
174 | -mtime may be given floating point (i.e. 1.5 is a day and a half).
|
---|
175 | -iskip/-idskip/-ipath/... etc are case-insensitive versions.
|
---|
176 |
|
---|
177 | If any letter in -newer/-older is upper case, "or equal" is
|
---|
178 | inserted into the test.
|
---|
179 |
|
---|
180 | You can always find the latest version on the World Wide Web in
|
---|
181 | http://www.wg.omron.co.jp/~jfriedl/perl/
|
---|
182 | INLINE_LITERAL_TEXT
|
---|
183 | exit(0);
|
---|
184 | }
|
---|
185 | $DOREP=1, next if $arg eq '-dorep'; ## do repeats
|
---|
186 | $DO_SORT=1, next if $arg eq '-sort'; ## sort files
|
---|
187 | $NOLINKS=1, next if $arg eq '-nolinks'; ## no sym. links
|
---|
188 | $PREPEND_FILENAME=0, next if $arg eq '-h'; ## no filename prefix
|
---|
189 | $REPORT_LINENUM=1, next if $arg eq '-n'; ## show line numbers
|
---|
190 | $WHY=1, next if $arg eq '-why'; ## tell why skipped
|
---|
191 | $XDEV=1, next if $arg eq '-xdev'; ## don't leave F.S.
|
---|
192 | $all=1,$opt{'-all'}=1,next if $arg eq '-all'; ## don't skip *.Z, etc
|
---|
193 | $iflag='i', next if $arg eq '-i'; ## ignore case
|
---|
194 | $norc=1, next if $arg eq '-norc'; ## don't load rc file
|
---|
195 | $showrc=1, next if $arg eq '-showrc'; ## show rc file
|
---|
196 | $underlineOK=1, next if $arg eq '-u'; ## look throuh underln.
|
---|
197 | $words=1, next if $arg eq '-w'; ## match "words" only
|
---|
198 | &strip if $arg eq '-strip'; ## dump this program
|
---|
199 | last if $arg eq '-e';
|
---|
200 | $DELAY=$1, next if $arg =~ m/-delay(\d+)/;
|
---|
201 |
|
---|
202 | $FIND_ONLY=1, next if $arg =~/^-f(ind)?$/;## do "find" only
|
---|
203 |
|
---|
204 | $FIND_ONLY=1, $DOREP=1, $all=1,
|
---|
205 | next if $arg =~/^-ff(ind)?$/;## fast -find
|
---|
206 | $LIST_ONLY=1,$opt{'-list'}=1,
|
---|
207 | next if $arg =~/^-l(ist)?$/;## only list files
|
---|
208 |
|
---|
209 | if ($arg =~ m/^-(v+)$/) { ## verbosity
|
---|
210 | $VERBOSE =length($1);
|
---|
211 | foreach $len (1..$VERBOSE) { $opt{'-'.('v' x $len)}=1 }
|
---|
212 | next;
|
---|
213 | }
|
---|
214 | if ($arg =~ m/^-(n+)ice$/) { ## "nice" output
|
---|
215 | $NICE =length($1);
|
---|
216 | foreach $len (1..$NICE) { $opt{'-'.('n' x $len).'ice'}=1 }
|
---|
217 | next;
|
---|
218 | }
|
---|
219 |
|
---|
220 | if ($arg =~ m/^-(i?)(d?)skip$/) {
|
---|
221 | local($i) = $1 eq 'i';
|
---|
222 | local($d) = $2 eq 'd';
|
---|
223 | $! = 2, die qq/$0: expecting glob arg to -$arg\n/ unless @ARGV;
|
---|
224 | foreach (split(/\s+/, shift @ARGV)) {
|
---|
225 | if ($d) {
|
---|
226 | $idskip{$_}=1 if $i;
|
---|
227 | $dskip{$_}=1;
|
---|
228 | } else {
|
---|
229 | $iskip{$_}=1 if $i;
|
---|
230 | $skip{$_}=1;
|
---|
231 | }
|
---|
232 | }
|
---|
233 | next;
|
---|
234 | }
|
---|
235 |
|
---|
236 |
|
---|
237 | if ($arg =~ m/^-(i?)(d?)(regex|path|name)$/) {
|
---|
238 | local($i) = $1 eq 'i';
|
---|
239 | $! = 2, die qq/$0: expecting arg to -$arg\n/ unless @ARGV;
|
---|
240 | foreach (split(/\s+/, shift @ARGV)) {
|
---|
241 | $iname{join(',', $arg, $_)}=1 if $i;
|
---|
242 | $name{join(',', $arg, $_)}=1;
|
---|
243 | }
|
---|
244 | next;
|
---|
245 | }
|
---|
246 |
|
---|
247 | if ($arg =~ m/^-d?dir$/) {
|
---|
248 | $opt{'-dir'}=1;
|
---|
249 | $! = 2, die qq/$0: expecting filename arg to -$arg\n/ unless @ARGV;
|
---|
250 | $start = shift(@ARGV);
|
---|
251 | $start =~ s#^~(/+|$)#$ENV{'HOME'}$1# if defined $ENV{'HOME'};
|
---|
252 | $! = 2, die qq/$0: can't find ${arg}'s "$start"\n/ unless -e $start;
|
---|
253 | $! = 2, die qq/$0: ${arg}'s "$start" not a directory.\n/ unless -d _;
|
---|
254 | undef(@todo), $opt{'-ddir'}=1 if $arg eq '-ddir';
|
---|
255 | push(@todo, $start);
|
---|
256 | next;
|
---|
257 | }
|
---|
258 |
|
---|
259 | if ($arg =~ m/^-(new|old)er$/i) {
|
---|
260 | $! = 2, die "$0: expecting filename arg to -$arg\n" unless @ARGV;
|
---|
261 | local($file, $time) = shift(@ARGV);
|
---|
262 | $! = 2, die qq/$0: can't stat -${arg}'s "$file"./
|
---|
263 | unless $time = (stat($file))[$STAT_MTIME];
|
---|
264 | local($upper) = $arg =~ tr/A-Z//;
|
---|
265 | if ($arg =~ m/new/i) {
|
---|
266 | $time++ unless $upper;
|
---|
267 | $NEWER = $time if $NEWER < $time;
|
---|
268 | } else {
|
---|
269 | $time-- unless $upper;
|
---|
270 | $OLDER = $time if $OLDER == 0 || $OLDER > $time;
|
---|
271 | }
|
---|
272 | next;
|
---|
273 | }
|
---|
274 |
|
---|
275 | if ($arg =~ m/-mtime/) {
|
---|
276 | $! = 2, die "$0: expecting numerical arg to -$arg\n" unless @ARGV;
|
---|
277 | local($days) = shift(@ARGV);
|
---|
278 | $! = 2, die qq/$0: inappropriate arg ($days) to $arg\n/ if $days==0;
|
---|
279 | $days *= 3600 * 24;
|
---|
280 | if ($days < 0) {
|
---|
281 | local($time) = $^T + $days;
|
---|
282 | $NEWER = $time if $NEWER < $time;
|
---|
283 | } else {
|
---|
284 | local($time) = $^T - $days;
|
---|
285 | $OLDER = $time if $OLDER == 0 || $OLDER > $time;
|
---|
286 | }
|
---|
287 | next;
|
---|
288 | }
|
---|
289 |
|
---|
290 | ## special user options
|
---|
291 | if ($arg =~ m/^-x(.+)/) {
|
---|
292 | foreach (split(/[\s,]+/, $1)) { $user_opt{$_} = $opt{$_}= 1; }
|
---|
293 | next;
|
---|
294 | }
|
---|
295 |
|
---|
296 | $! = 2, die "$0: unknown arg [$arg]\n";
|
---|
297 | }
|
---|
298 | }
|
---|
299 |
|
---|
300 | ##
|
---|
301 | ## Given a filename glob, return a regex.
|
---|
302 | ## If the glob has no globbing chars (no * ? or [..]), then
|
---|
303 | ## prepend an effective '*' to it.
|
---|
304 | ##
|
---|
305 | sub glob_to_regex
|
---|
306 | {
|
---|
307 | local($glob) = @_;
|
---|
308 | local(@parts) = $glob =~ m/\\.|[*?]|\[]?[^]]*]|[^[\\*?]+/g;
|
---|
309 | local($trueglob)=0;
|
---|
310 | foreach (@parts) {
|
---|
311 | if ($_ eq '*' || $_ eq '?') {
|
---|
312 | $_ = ".$_";
|
---|
313 | $trueglob=1; ## * and ? are a real glob
|
---|
314 | } elsif (substr($_, 0, 1) eq '[') {
|
---|
315 | $trueglob=1; ## [..] is a real glob
|
---|
316 | } else {
|
---|
317 | s/^\\//; ## remove any leading backslash;
|
---|
318 | s/\W/\\$&/g; ## now quote anything dangerous;
|
---|
319 | }
|
---|
320 | }
|
---|
321 | unshift(@parts, '.*') unless $trueglob;
|
---|
322 | join('', '^', @parts, '$');
|
---|
323 | }
|
---|
324 |
|
---|
325 | sub prepare_to_search
|
---|
326 | {
|
---|
327 | local($rc_file) = @_;
|
---|
328 |
|
---|
329 | $HEADER_BYTES=0; ## Might be set nonzero in &read_rc;
|
---|
330 | $last_message_length = 0; ## For &message and &clear_message.
|
---|
331 |
|
---|
332 | &read_rc($rc_file, $showrc) unless $norc;
|
---|
333 | exit(0) if $showrc;
|
---|
334 |
|
---|
335 | $NEXT_DIR_ENTRY = $DO_SORT ? 'shift @files' : 'readdir(DIR)';
|
---|
336 | $WHY = 1 if $VERBOSE > 3; ## Arg -vvvv or above implies -why.
|
---|
337 | @todo = ('.') if @todo == 0; ## Where we'll start looking
|
---|
338 |
|
---|
339 | ## see if any user options were specified that weren't accounted for
|
---|
340 | foreach $opt (keys %user_opt) {
|
---|
341 | next if defined $seen_opt{$opt};
|
---|
342 | warn "warning: -x$opt never considered.\n";
|
---|
343 | }
|
---|
344 |
|
---|
345 | die "$0: multiple time constraints exclude all possible files.\n"
|
---|
346 | if ($NEWER && $OLDER) && ($NEWER > $OLDER);
|
---|
347 |
|
---|
348 | ##
|
---|
349 | ## Process any -skip/-iskip args that had been given
|
---|
350 | ##
|
---|
351 | local(@skip_test);
|
---|
352 | foreach $glob (keys %skip) {
|
---|
353 | $i = defined($iskip{$glob}) ? 'i': '';
|
---|
354 | push(@skip_test, '$name =~ m/'. &glob_to_regex($glob). "/$i");
|
---|
355 | }
|
---|
356 | if (@skip_test) {
|
---|
357 | $SKIP_TEST = join('||',@skip_test);
|
---|
358 | $DO_SKIP_TEST = 1;
|
---|
359 | } else {
|
---|
360 | $DO_SKIP_TEST = $SKIP_TEST = 0;
|
---|
361 | }
|
---|
362 |
|
---|
363 | ##
|
---|
364 | ## Process any -dskip/-idskip args that had been given
|
---|
365 | ##
|
---|
366 | local(@dskip_test);
|
---|
367 | foreach $glob (keys %dskip) {
|
---|
368 | $i = defined($idskip{$glob}) ? 'i': '';
|
---|
369 | push(@dskip_test, '$name =~ m/'. &glob_to_regex($glob). "/$i");
|
---|
370 | }
|
---|
371 | if (@dskip_test) {
|
---|
372 | $DSKIP_TEST = join('||',@dskip_test);
|
---|
373 | $DO_DSKIP_TEST = 1;
|
---|
374 | } else {
|
---|
375 | $DO_DSKIP_TEST = $DSKIP_TEST = 0;
|
---|
376 | }
|
---|
377 |
|
---|
378 |
|
---|
379 | ##
|
---|
380 | ## Process any -name, -path, -regex, etc. args that had been given.
|
---|
381 | ##
|
---|
382 | undef @name_test;
|
---|
383 | undef @dname_test;
|
---|
384 | foreach $key (keys %name) {
|
---|
385 | local($type, $pat) = split(/,/, $key, 2);
|
---|
386 | local($i) = defined($iname{$key}) ? 'i' : '';
|
---|
387 | if ($type =~ /regex/) {
|
---|
388 | $pat =~ s/!/\\!/g;
|
---|
389 | $test = "\$name =~ m!^$pat\$!$i";
|
---|
390 | } else {
|
---|
391 | local($var) = $type eq 'name' ? '$name' : '$file';
|
---|
392 | $test = "$var =~ m/". &glob_to_regex($pat). "/$i";
|
---|
393 | }
|
---|
394 | if ($type =~ m/^-i?d/) {
|
---|
395 | push(@dname_test, $test);
|
---|
396 | } else {
|
---|
397 | push(@name_test, $test);
|
---|
398 | }
|
---|
399 | }
|
---|
400 | if (@name_test) {
|
---|
401 | $GLOB_TESTS = join('||', @name_test);
|
---|
402 |
|
---|
403 | $DO_GLOB_TESTS = 1;
|
---|
404 | } else {
|
---|
405 | $GLOB_TESTS = $DO_GLOB_TESTS = 0;
|
---|
406 | }
|
---|
407 | if (@dname_test) {
|
---|
408 | $DGLOB_TESTS = join('||', @dname_test);
|
---|
409 | $DO_DGLOB_TESTS = 1;
|
---|
410 | } else {
|
---|
411 | $DGLOB_TESTS = $DO_DGLOB_TESTS = 0;
|
---|
412 | }
|
---|
413 |
|
---|
414 |
|
---|
415 | ##
|
---|
416 | ## Process any 'magic' things from the startup file.
|
---|
417 | ##
|
---|
418 | if (@magic_tests && $HEADER_BYTES) {
|
---|
419 | ## the $magic' one is for when &dodir is not inlined
|
---|
420 | $tests = join('||',@magic_tests);
|
---|
421 | $MAGIC_TESTS = " { package magic; \$val = ($tests) }";
|
---|
422 | $DO_MAGIC_TESTS = 1;
|
---|
423 | } else {
|
---|
424 | $MAGIC_TESTS = 1;
|
---|
425 | $DO_MAGIC_TESTS = 0;
|
---|
426 | }
|
---|
427 |
|
---|
428 | ##
|
---|
429 | ## Prepare regular expressions.
|
---|
430 | ##
|
---|
431 | {
|
---|
432 | local(@regex_tests);
|
---|
433 |
|
---|
434 | if ($LIST_ONLY) {
|
---|
435 | $mflag = '';
|
---|
436 | ## need to have $* set, but perl5 just won''t shut up about it.
|
---|
437 | if ($] >= 5) {
|
---|
438 | $mflag = 'm';
|
---|
439 | } else {
|
---|
440 | eval ' $* = 1 ';
|
---|
441 | }
|
---|
442 | }
|
---|
443 |
|
---|
444 | ##
|
---|
445 | ## Until I figure out a better way to deal with it,
|
---|
446 | ## We have to worry about a regex like [^xyz] when doing $LIST_ONLY.
|
---|
447 | ## Such a regex *will* match \n, and if I'm pulling in multiple
|
---|
448 | ## lines, it can allow lines to match that would otherwise not match.
|
---|
449 | ##
|
---|
450 | ## Therefore, if there is a '[^' in a regex, we can NOT take a chance
|
---|
451 | ## an use the fast listonly.
|
---|
452 | ##
|
---|
453 | $CAN_USE_FAST_LISTONLY = $LIST_ONLY;
|
---|
454 |
|
---|
455 | local(@extra);
|
---|
456 | local($underline_glue) = ($] >= 5) ? '(:?_\cH)?' : '(_\cH)?';
|
---|
457 | while (@ARGV) {
|
---|
458 | $regex = shift(@ARGV);
|
---|
459 | ##
|
---|
460 | ## If watching for underlined things too, add another regex.
|
---|
461 | ##
|
---|
462 | if ($underlineOK) {
|
---|
463 | if ($regex =~ m/[?*+{}()\\.|^\$[]/) {
|
---|
464 | warn "$0: warning, can't underline-safe ``$regex''.\n";
|
---|
465 | } else {
|
---|
466 | $regex = join($underline_glue, split(//, $regex));
|
---|
467 | }
|
---|
468 | }
|
---|
469 |
|
---|
470 | ## If nothing special in the regex, just use index...
|
---|
471 | ## is quite a bit faster.
|
---|
472 | if (($iflag eq '') && ($words == 0) &&
|
---|
473 | $regex !~ m/[?*+{}()\\.|^\$[]/)
|
---|
474 | {
|
---|
475 | push(@regex_tests, "(index(\$_, q+$regex+)>=0)");
|
---|
476 |
|
---|
477 | } else {
|
---|
478 | $regex =~ s#[\$\@\/]\w#\\$&#;
|
---|
479 | if ($words) {
|
---|
480 | if ($regex =~ m/\|/) {
|
---|
481 | ## could be dangerous -- see if we can wrap in parens.
|
---|
482 | if ($regex =~ m/\\\d/) {
|
---|
483 | warn "warning: -w and a | in a regex is dangerous.\n"
|
---|
484 | } else {
|
---|
485 | $regex = join($regex, '(', ')');
|
---|
486 | }
|
---|
487 | }
|
---|
488 | $regex = join($regex, '\b', '\b');
|
---|
489 | }
|
---|
490 | $CAN_USE_FAST_LISTONLY = 0 if substr($regex, "[^") >= 0;
|
---|
491 | push(@regex_tests, "m/$regex/$iflag$mflag");
|
---|
492 | }
|
---|
493 |
|
---|
494 | ## If we're done, but still have @extra to do, get set for that.
|
---|
495 | if (@ARGV == 0 && @extra) {
|
---|
496 | @ARGV = @extra; ## now deal with the extra stuff.
|
---|
497 | $underlineOK = 0; ## but no more of this.
|
---|
498 | undef @extra; ## or this.
|
---|
499 | }
|
---|
500 | }
|
---|
501 | if (@regex_tests) {
|
---|
502 | $REGEX_TEST = join('||', @regex_tests);
|
---|
503 | ## print STDERR $REGEX_TEST, "\n"; exit;
|
---|
504 | } else {
|
---|
505 | ## must be doing -find -- just give something syntactically correct.
|
---|
506 | $REGEX_TEST = 1;
|
---|
507 | }
|
---|
508 | }
|
---|
509 |
|
---|
510 | ##
|
---|
511 | ## Make sure we can read the first item(s).
|
---|
512 | ##
|
---|
513 | foreach $start (@todo) {
|
---|
514 | $! = 2, die qq/$0: can't stat "$start"\n/
|
---|
515 | unless ($dev,$inode) = (stat($start))[$STAT_DEV,$STAT_INODE];
|
---|
516 |
|
---|
517 | if (defined $dir_done{"$dev,$inode"}) {
|
---|
518 | ## ignore the repeat.
|
---|
519 | warn(qq/ignoring "$start" (same as "$dir_done{"$dev,$inode"}").\n/)
|
---|
520 | if $VERBOSE;
|
---|
521 | next;
|
---|
522 | }
|
---|
523 |
|
---|
524 | ## if -xdev was given, remember the device.
|
---|
525 | $xdev{$dev} = 1 if $XDEV;
|
---|
526 |
|
---|
527 | ## Note that we won't want to do it again
|
---|
528 | $dir_done{"$dev,$inode"} = $start;
|
---|
529 | }
|
---|
530 | }
|
---|
531 |
|
---|
532 |
|
---|
533 | ##
|
---|
534 | ## See the comment above the __END__ above the 'sub dodir' below.
|
---|
535 | ##
|
---|
536 | sub import_program
|
---|
537 | {
|
---|
538 | sub bad {
|
---|
539 | print STDERR "$0: internal error (@_)\n";
|
---|
540 | exit 2;
|
---|
541 | }
|
---|
542 |
|
---|
543 | ## Read from data, up to next __END__. This will be &dodir.
|
---|
544 | local($/) = "\n__END__";
|
---|
545 | $prog = <DATA>;
|
---|
546 | close(DATA);
|
---|
547 |
|
---|
548 | $prog =~ s/\beval\b//g; ## remove any 'eval'
|
---|
549 |
|
---|
550 | ## Inline uppercase $-variables by their current values.
|
---|
551 | if ($] >= 5) {
|
---|
552 | $prog =~ s/\$([A-Z][A-Z0-9_]{2,}\b)/
|
---|
553 | &bad($1) if !defined ${$main::{$1}}; ${$main::{$1}};/eg;
|
---|
554 | } else {
|
---|
555 | $prog =~ s/\$([A-Z][A-Z0-9_]{2,}\b)/local(*VAR) = $_main{$1};
|
---|
556 | &bad($1) if !defined $VAR; $VAR;/eg;
|
---|
557 | }
|
---|
558 |
|
---|
559 | eval $prog; ## now do it. This will define &dodir;
|
---|
560 | $!=2, die "$0 internal error: $@\n" if $@;
|
---|
561 | }
|
---|
562 |
|
---|
563 | ###########################################################################
|
---|
564 |
|
---|
565 | ##
|
---|
566 | ## Read the .search file:
|
---|
567 | ## Blank lines and lines that are only #-comments ignored.
|
---|
568 | ## Newlines may be escaped to create long lines
|
---|
569 | ## Other lines are directives.
|
---|
570 | ##
|
---|
571 | ## A directive may begin with an optional tag in the form <...>
|
---|
572 | ## Things inside the <...> are evaluated as with:
|
---|
573 | ## <(this || that) && must>
|
---|
574 | ## will be true if
|
---|
575 | ## -xmust -xthis or -xmust -xthat
|
---|
576 | ## were specified on the command line (order doesn't matter, though)
|
---|
577 | ## A directive is not done if there is a tag and it's false.
|
---|
578 | ## Any characters but whitespace and &|()>,! may appear after an -x
|
---|
579 | ## (although "-xdev" is special). -xmust,this is the same as -xmust -xthis.
|
---|
580 | ## Something like -x~ would make <~> true, and <!~> false.
|
---|
581 | ##
|
---|
582 | ## Directives are in the form:
|
---|
583 | ## option: STRING
|
---|
584 | ## magic : NUMBYTES : EXPR
|
---|
585 | ##
|
---|
586 | ## With option:
|
---|
587 | ## The STRING is parsed like a Bourne shell command line, and the
|
---|
588 | ## options are used as if given on the command line.
|
---|
589 | ## No comments are allowed on 'option' lines.
|
---|
590 | ## Examples:
|
---|
591 | ## # skip objects and libraries
|
---|
592 | ## option: -skip '.o .a'
|
---|
593 | ## # skip emacs *~ and *# files, unless -x~ given:
|
---|
594 | ## <!~> option: -skip '~ #'
|
---|
595 | ##
|
---|
596 | ## With magic:
|
---|
597 | ## EXPR can be pretty much any perl (comments allowed!).
|
---|
598 | ## If it evaluates to true for any particular file, it is skipped.
|
---|
599 | ## The only info you'll have about a file is the variable $H, which
|
---|
600 | ## will have at least the first NUMBYTES of the file (less if the file
|
---|
601 | ## is shorter than that, of course, and maybe more). You'll also have
|
---|
602 | ## any variables you set in previous 'magic' lines.
|
---|
603 | ## Examples:
|
---|
604 | ## magic: 6 : ($x6 = substr($H, 0, 6)) eq 'GIF87a'
|
---|
605 | ## magic: 6 : $x6 eq 'GIF89a'
|
---|
606 | ##
|
---|
607 | ## magic: 6 : (($x6 = substr($H, 0, 6)) eq 'GIF87a' ## old gif \
|
---|
608 | ## || $x6 eq 'GIF89a' ## new gif
|
---|
609 | ## (the above two sets are the same)
|
---|
610 | ## ## Check the first 32 bytes for "binarish" looking bytes.
|
---|
611 | ## ## Don't blindly dump on any high-bit set, as non-ASCII text
|
---|
612 | ## ## often has them set. \x80 and \xff seem to be special, though.
|
---|
613 | ## ## Require two in a row to not get things like perl's $^T.
|
---|
614 | ## ## This is known to get *.Z, *.gz, pkzip, *.elc and about any
|
---|
615 | ## ## executable you'll find.
|
---|
616 | ## magic: 32 : $H =~ m/[\x00-\x06\x10-\x1a\x1c-\x1f\x80\xff]{2}/
|
---|
617 | ##
|
---|
618 | sub read_rc
|
---|
619 | {
|
---|
620 | local($file, $show) = @_;
|
---|
621 | local($line_num, $ln, $tag) = 0;
|
---|
622 | local($use_default, @default) = 0;
|
---|
623 |
|
---|
624 | { package magic; $ = 0; } ## turn off warnings for when we run EXPR's
|
---|
625 |
|
---|
626 | unless (open(RC, "$file")) {
|
---|
627 | $use_default=1;
|
---|
628 | $file = "<internal default startup file>";
|
---|
629 | ## no RC file -- use this default.
|
---|
630 | @default = split(/\n/,<<'--------INLINE_LITERAL_TEXT');
|
---|
631 | magic: 32 : $H =~ m/[\x00-\x06\x10-\x1a\x1c-\x1f\x80\xff]{2}/
|
---|
632 | option: -skip '.a .COM .elc .EXE .gz .o .pbm .xbm .dvi'
|
---|
633 | option: -iskip '.tarz .zip .z .lzh .jpg .jpeg .gif .uu'
|
---|
634 | <!~> option: -skip '~ #'
|
---|
635 | --------INLINE_LITERAL_TEXT
|
---|
636 | }
|
---|
637 |
|
---|
638 | ##
|
---|
639 | ## Make an eval error pretty.
|
---|
640 | ##
|
---|
641 | sub clean_eval_error {
|
---|
642 | local($_) = @_;
|
---|
643 | s/ in file \(eval\) at line \d+,//g; ## perl4-style error
|
---|
644 | s/ at \(eval \d+\) line \d+,//g; ## perl5-style error
|
---|
645 | $_ = $` if m/\n/; ## remove all but first line
|
---|
646 | "$_\n";
|
---|
647 | }
|
---|
648 |
|
---|
649 | print "reading RC file: $file\n" if $show;
|
---|
650 |
|
---|
651 | while (defined($_ = ($use_default ? shift(@default) : <RC>))) {
|
---|
652 | $ln = ++$line_num; ## note starting line num.
|
---|
653 | $_ .= <RC>, $line_num++ while s/\\\n?$/\n/; ## allow continuations
|
---|
654 | next if /^\s*(#.*)?$/; ## skip blank or comment-only lines.
|
---|
655 | $do = '';
|
---|
656 |
|
---|
657 | ## look for an initial <...> tag.
|
---|
658 | if (s/^\s*<([^>]*)>//) {
|
---|
659 | ## This simple s// will make the tag ready to eval.
|
---|
660 | ($tag = $msg = $1) =~
|
---|
661 | s/[^\s&|(!)]+/
|
---|
662 | $seen_opt{$&}=1; ## note seen option
|
---|
663 | "defined(\$opt{q>$&>})" ## (q>> is safe quoting here)
|
---|
664 | /eg;
|
---|
665 |
|
---|
666 | ## see if the tag is true or not, abort this line if not.
|
---|
667 | $dothis = (eval $tag);
|
---|
668 | $!=2, die "$file $ln <$msg>: $_".&clean_eval_error($@) if $@;
|
---|
669 |
|
---|
670 | if ($show) {
|
---|
671 | $msg =~ s/[^\s&|(!)]+/-x$&/;
|
---|
672 | $msg =~ s/\s*!\s*/ no /g;
|
---|
673 | $msg =~ s/\s*&&\s*/ and /g;
|
---|
674 | $msg =~ s/\s*\|\|\s*/ or /g;
|
---|
675 | $msg =~ s/^\s+//; $msg =~ s/\s+$//;
|
---|
676 | $do = $dothis ? "(doing because $msg)" :
|
---|
677 | "(do if $msg)";
|
---|
678 | } elsif (!$dothis) {
|
---|
679 | next;
|
---|
680 | }
|
---|
681 | }
|
---|
682 |
|
---|
683 | if (m/^\s*option\s*:\s*/) {
|
---|
684 | next if $all && !$show; ## -all turns off these checks;
|
---|
685 | local($_) = $';
|
---|
686 | s/\n$//;
|
---|
687 | local($orig) = $_;
|
---|
688 | print " $do option: $_\n" if $show;
|
---|
689 | local($0) = "$0 ($file)"; ## for any error message.
|
---|
690 | local(@ARGV);
|
---|
691 | local($this);
|
---|
692 | ##
|
---|
693 | ## Parse $_ as a Bourne shell line -- fill @ARGV
|
---|
694 | ##
|
---|
695 | while (length) {
|
---|
696 | if (s/^\s+//) {
|
---|
697 | push(@ARGV, $this) if defined $this;
|
---|
698 | undef $this;
|
---|
699 | next;
|
---|
700 | }
|
---|
701 | $this = '' if !defined $this;
|
---|
702 | $this .= $1 while s/^'([^']*)'// ||
|
---|
703 | s/^"([^"]*)"// ||
|
---|
704 | s/^([^'"\s\\]+)//||
|
---|
705 | s/^(\\[\D\d])//;
|
---|
706 | die "$file $ln: error parsing $orig at $_\n" if m/^\S/;
|
---|
707 | }
|
---|
708 | push(@ARGV, $this) if defined $this;
|
---|
709 | &check_args;
|
---|
710 | die qq/$file $ln: unused arg "@ARGV".\n/ if @ARGV;
|
---|
711 | next;
|
---|
712 | }
|
---|
713 |
|
---|
714 | if (m/^\s*magic\s*:\s*(\d+)\s*:\s*/) {
|
---|
715 | next if $all && !$show; ## -all turns off these checks;
|
---|
716 | local($bytes, $check) = ($1, $');
|
---|
717 |
|
---|
718 | if ($show) {
|
---|
719 | $check =~ s/\n?$/\n/;
|
---|
720 | print " $do contents: $check";
|
---|
721 | }
|
---|
722 | ## Check to make sure the thing at least compiles.
|
---|
723 | eval "package magic; (\$H = '1'x \$main'bytes) && (\n$check\n)\n";
|
---|
724 | $! = 2, die "$file $ln: ".&clean_eval_error($@) if $@;
|
---|
725 |
|
---|
726 | $HEADER_BYTES = $bytes if $bytes > $HEADER_BYTES;
|
---|
727 | push(@magic_tests, "(\n$check\n)");
|
---|
728 | next;
|
---|
729 | }
|
---|
730 | $! = 2, die "$file $ln: unknown command\n";
|
---|
731 | }
|
---|
732 | close(RC);
|
---|
733 | }
|
---|
734 |
|
---|
735 | sub message
|
---|
736 | {
|
---|
737 | if (!$STDERR_IS_TTY) {
|
---|
738 | print STDERR $_[0], "\n";
|
---|
739 | } else {
|
---|
740 | local($text) = @_;
|
---|
741 | $thislength = length($text);
|
---|
742 | if ($thislength >= $last_message_length) {
|
---|
743 | print STDERR $text, "\r";
|
---|
744 | } else {
|
---|
745 | print STDERR $text, ' 'x ($last_message_length-$thislength),"\r";
|
---|
746 | }
|
---|
747 | $last_message_length = $thislength;
|
---|
748 | }
|
---|
749 | }
|
---|
750 |
|
---|
751 | sub clear_message
|
---|
752 | {
|
---|
753 | print STDERR ' ' x $last_message_length, "\r" if $last_message_length;
|
---|
754 | $vv_print = $vv_size = $last_message_length = 0;
|
---|
755 | }
|
---|
756 |
|
---|
757 | ##
|
---|
758 | ## Output a copy of this program with comments, extra whitespace, and
|
---|
759 | ## the trailing man page removed. On an ultra slow machine, such a copy
|
---|
760 | ## might load faster (but I can't tell any difference on my machine).
|
---|
761 | ##
|
---|
762 | sub strip {
|
---|
763 | seek(DATA, 0, 0) || die "$0: can't reset internal pointer.\n";
|
---|
764 | while(<DATA>) {
|
---|
765 | print, next if /INLINE_LITERAL_TEXT/.../INLINE_LITERAL_TEXT/;
|
---|
766 | ## must mention INLINE_LITERAL_TEXT on this line!
|
---|
767 | s/\#\#.*|^\s+|\s+$//; ## remove cruft
|
---|
768 | last if $_ eq '.00;';
|
---|
769 | next if ($_ eq '') || ($_ eq "'di'") || ($_ eq "'ig00'");
|
---|
770 | s/\$stripped=0;/\$stripped=1;/;
|
---|
771 | s/\s\s+/ /; ## squish multiple whitespaces down to one.
|
---|
772 | print $_, "\n";
|
---|
773 | }
|
---|
774 | exit(0);
|
---|
775 | }
|
---|
776 |
|
---|
777 | ##
|
---|
778 | ## Just to shut up -w. Never executed.
|
---|
779 | ##
|
---|
780 | sub dummy {
|
---|
781 |
|
---|
782 | 1 || &dummy || &dir_done || &bad || &message || $NEXT_DIR_ENTRY ||
|
---|
783 | $DELAY || $VV_SIZE || $VV_PRINT_COUNT || $STDERR_SCREWS_STDOUT ||
|
---|
784 | @files || @files || $magic'H || $magic'H || $xdev{''} || &clear_message;
|
---|
785 |
|
---|
786 | }
|
---|
787 |
|
---|
788 | ##
|
---|
789 | ## If the following __END__ is in place, what follows will be
|
---|
790 | ## inlined when the program first starts up. Any $ variable name
|
---|
791 | ## all in upper case, specifically, any string matching
|
---|
792 | ## \$([A-Z][A-Z0-9_]{2,}\b
|
---|
793 | ## will have the true value for that variable inlined. Also, any 'eval' is
|
---|
794 | ## removed
|
---|
795 | ##
|
---|
796 | ## The idea is that when the whole thing is then eval'ed to define &dodir,
|
---|
797 | ## the perl optimizer will make all the decisions that are based upon
|
---|
798 | ## command-line options (such as $VERBOSE), since they'll be inlined as
|
---|
799 | ## constants
|
---|
800 | ##
|
---|
801 | ## Also, and here's the big win, the tests for matching the regex, and a
|
---|
802 | ## few others, are all inlined. Should be blinding speed here.
|
---|
803 | ##
|
---|
804 | ## See the read from <DATA> above for where all this takes place.
|
---|
805 | ## But all-in-all, you *want* the __END__ here. Comment it out only for
|
---|
806 | ## debugging....
|
---|
807 | ##
|
---|
808 |
|
---|
809 | __END__
|
---|
810 |
|
---|
811 | ##
|
---|
812 | ## Given a directory, check all "appropriate" files in it.
|
---|
813 | ## Shove any subdirectories into the global @todo, so they'll be done
|
---|
814 | ## later.
|
---|
815 | ##
|
---|
816 | ## Be careful about adding any upper-case variables, as they are subject
|
---|
817 | ## to being inlined. See comments above the __END__ above.
|
---|
818 | ##
|
---|
819 | sub dodir
|
---|
820 | {
|
---|
821 | local($dir) = @_;
|
---|
822 | $dir =~ s,/+$,,; ## remove any trailing slash.
|
---|
823 | unless (opendir(DIR, "$dir/.")) {
|
---|
824 | &clear_message if $VERBOSE && $STDERR_SCREWS_STDOUT;
|
---|
825 | warn qq($0: can't opendir "$dir/".\n);
|
---|
826 | return;
|
---|
827 | }
|
---|
828 |
|
---|
829 | if ($VERBOSE) {
|
---|
830 | &message($dir);
|
---|
831 | $vv_print = $vv_size = 0;
|
---|
832 | }
|
---|
833 |
|
---|
834 | @files = sort readdir(DIR) if $DO_SORT;
|
---|
835 |
|
---|
836 | while (defined($name = eval $NEXT_DIR_ENTRY))
|
---|
837 | {
|
---|
838 | next if $name eq '.' || $name eq '..'; ## never follow these.
|
---|
839 |
|
---|
840 | ## create full relative pathname.
|
---|
841 | $file = $dir eq '.' ? $name : "$dir/$name";
|
---|
842 |
|
---|
843 | ## if link and skipping them, do so.
|
---|
844 | if ($NOLINKS && -l $file) {
|
---|
845 | warn qq/skip (symlink): $file\n/ if $WHY;
|
---|
846 | next;
|
---|
847 | }
|
---|
848 |
|
---|
849 | ## skip things unless files or directories
|
---|
850 | unless (-f $file || -d _) {
|
---|
851 | if ($WHY) {
|
---|
852 | $why = (-S _ && "socket") ||
|
---|
853 | (-p _ && "pipe") ||
|
---|
854 | (-b _ && "block special")||
|
---|
855 | (-c _ && "char special") || "somekinda special";
|
---|
856 | warn qq/skip ($why): $file\n/;
|
---|
857 | }
|
---|
858 | next;
|
---|
859 | }
|
---|
860 |
|
---|
861 | ## skip things we can't read
|
---|
862 | unless (-r _) {
|
---|
863 | if ($WHY) {
|
---|
864 | $why = (-l $file) ? "follow" : "read";
|
---|
865 | warn qq/skip (can't $why): $file\n/;
|
---|
866 | }
|
---|
867 | next;
|
---|
868 | }
|
---|
869 |
|
---|
870 | ## skip things that are empty
|
---|
871 | unless (-s _ || -d _) {
|
---|
872 | warn qq/skip (empty): $file\n/ if $WHY;
|
---|
873 | next;
|
---|
874 | }
|
---|
875 |
|
---|
876 | ## Note file device & inode. If -xdev, skip if appropriate.
|
---|
877 | ($dev, $inode) = (stat(_))[$STAT_DEV, $STAT_INODE];
|
---|
878 | if ($XDEV && defined $xdev{$dev}) {
|
---|
879 | warn qq/skip (other device): $file\n/ if $WHY;
|
---|
880 | next;
|
---|
881 | }
|
---|
882 | $id = "$dev,$inode";
|
---|
883 |
|
---|
884 | ## special work for a directory
|
---|
885 | if (-d _) {
|
---|
886 | ## Do checks for directory file endings.
|
---|
887 | if ($DO_DSKIP_TEST && (eval $DSKIP_TEST)) {
|
---|
888 | warn qq/skip (-dskip): $file\n/ if $WHY;
|
---|
889 | next;
|
---|
890 | }
|
---|
891 | ## do checks for -name/-regex/-path tests
|
---|
892 | if ($DO_DGLOB_TESTS && !(eval $DGLOB_TESTS)) {
|
---|
893 | warn qq/skip (dirname): $file\n/ if $WHY;
|
---|
894 | next;
|
---|
895 | }
|
---|
896 |
|
---|
897 | ## _never_ redo a directory
|
---|
898 | if (defined $dir_done{$id} and $^O ne 'MSWin32') {
|
---|
899 | warn qq/skip (did as "$dir_done{$id}"): $file\n/ if $WHY;
|
---|
900 | next;
|
---|
901 | }
|
---|
902 | $dir_done{$id} = $file; ## mark it done.
|
---|
903 | unshift(@todo, $file); ## add to the list to do.
|
---|
904 | next;
|
---|
905 | }
|
---|
906 | if ($WHY == 0 && $VERBOSE > 1) {
|
---|
907 | if ($VERBOSE>2||$vv_print++>$VV_PRINT_COUNT||($vv_size+=-s _)>$VV_SIZE){
|
---|
908 | &message($file);
|
---|
909 | $vv_print = $vv_size = 0;
|
---|
910 | }
|
---|
911 | }
|
---|
912 |
|
---|
913 | ## do time-related tests
|
---|
914 | if ($NEWER || $OLDER) {
|
---|
915 | $_ = (stat(_))[$STAT_MTIME];
|
---|
916 | if ($NEWER && $_ < $NEWER) {
|
---|
917 | warn qq/skip (too old): $file\n/ if $WHY;
|
---|
918 | next;
|
---|
919 | }
|
---|
920 | if ($OLDER && $_ > $OLDER) {
|
---|
921 | warn qq/skip (too new): $file\n/ if $WHY;
|
---|
922 | next;
|
---|
923 | }
|
---|
924 | }
|
---|
925 |
|
---|
926 | ## do checks for file endings
|
---|
927 | if ($DO_SKIP_TEST && (eval $SKIP_TEST)) {
|
---|
928 | warn qq/skip (-skip): $file\n/ if $WHY;
|
---|
929 | next;
|
---|
930 | }
|
---|
931 |
|
---|
932 | ## do checks for -name/-regex/-path tests
|
---|
933 | if ($DO_GLOB_TESTS && !(eval $GLOB_TESTS)) {
|
---|
934 | warn qq/skip (filename): $file\n/ if $WHY;
|
---|
935 | next;
|
---|
936 | }
|
---|
937 |
|
---|
938 |
|
---|
939 | ## If we're not repeating files,
|
---|
940 | ## skip this one if we've done it, or note we're doing it.
|
---|
941 | unless ($DOREP) {
|
---|
942 | if (defined $file_done{$id}) {
|
---|
943 | warn qq/skip (did as "$file_done{$id}"): $file\n/ if $WHY;
|
---|
944 | next;
|
---|
945 | }
|
---|
946 | $file_done{$id} = $file;
|
---|
947 | }
|
---|
948 |
|
---|
949 | if ($DO_MAGIC_TESTS) {
|
---|
950 | if (!open(FILE_IN, $file)) {
|
---|
951 | &clear_message if $VERBOSE && $STDERR_SCREWS_STDOUT;
|
---|
952 | warn qq/$0: can't open: $file\n/;
|
---|
953 | next;
|
---|
954 | }
|
---|
955 | unless (read(FILE_IN, $magic'H, $HEADER_BYTES)) {
|
---|
956 | &clear_message if $VERBOSE && $STDERR_SCREWS_STDOUT;
|
---|
957 | warn qq/$0: can't read from "$file"\n"/;
|
---|
958 | close(FILE_IN);
|
---|
959 | next;
|
---|
960 | }
|
---|
961 |
|
---|
962 | eval $MAGIC_TESTS;
|
---|
963 | if ($magic'val) {
|
---|
964 | close(FILE_IN);
|
---|
965 | warn qq/skip (magic): $file\n/ if $WHY;
|
---|
966 | next;
|
---|
967 | }
|
---|
968 | seek(FILE_IN, 0, 0); ## reset for later <FILE_IN>
|
---|
969 | }
|
---|
970 |
|
---|
971 | if ($WHY != 0 && $VERBOSE > 1) {
|
---|
972 | if ($VERBOSE>2||$vv_print++>$VV_PRINT_COUNT||($vv_size+=-s _)>$VV_SIZE){
|
---|
973 | &message($file);
|
---|
974 | $vv_print = $vv_size = 0;
|
---|
975 | }
|
---|
976 | }
|
---|
977 |
|
---|
978 | if ($DELAY) {
|
---|
979 | sleep($DELAY);
|
---|
980 | }
|
---|
981 |
|
---|
982 | if ($FIND_ONLY) {
|
---|
983 | &clear_message if $VERBOSE && $STDERR_SCREWS_STDOUT;
|
---|
984 | print $file, "\n";
|
---|
985 | $retval=0; ## we've found something
|
---|
986 | close(FILE_IN) if $DO_MAGIC_TESTS;
|
---|
987 | next;
|
---|
988 | } else {
|
---|
989 | ## if we weren't doing magic tests, file won't be open yet...
|
---|
990 | if (!$DO_MAGIC_TESTS && !open(FILE_IN, $file)) {
|
---|
991 | &clear_message if $VERBOSE && $STDERR_SCREWS_STDOUT;
|
---|
992 | warn qq/$0: can't open: $file\n/;
|
---|
993 | next;
|
---|
994 | }
|
---|
995 | if ($LIST_ONLY && $CAN_USE_FAST_LISTONLY) {
|
---|
996 | ##
|
---|
997 | ## This is rather complex, but buys us a LOT when we're just
|
---|
998 | ## listing files and not the individual internal lines.
|
---|
999 | ##
|
---|
1000 | local($size) = 4096; ## block-size in which to do reads
|
---|
1001 | local($nl); ## will point to $_'s ending newline.
|
---|
1002 | local($read); ## will be how many bytes read.
|
---|
1003 | local($_) = ''; ## Starts out empty
|
---|
1004 | local($hold); ## (see below)
|
---|
1005 |
|
---|
1006 | while (($read = read(FILE_IN,$_,$size,length($_)))||length($_))
|
---|
1007 | {
|
---|
1008 | undef @parts;
|
---|
1009 | ## if read a full block, but no newline, need to read more.
|
---|
1010 | while ($read == $size && ($nl = rindex($_, "\n")) < 0) {
|
---|
1011 | push(@parts, $_); ## save that part
|
---|
1012 | $read = read(FILE_IN, $_, $size); ## keep trying
|
---|
1013 | }
|
---|
1014 |
|
---|
1015 | ##
|
---|
1016 | ## If we had to save parts, must now combine them together.
|
---|
1017 | ## adjusting $nl to reflect the now-larger $_. This should
|
---|
1018 | ## be a lot more efficient than using any kind of .= in the
|
---|
1019 | ## loop above.
|
---|
1020 | ##
|
---|
1021 | if (@parts) {
|
---|
1022 | local($lastlen) = length($_); #only need if $nl >= 0
|
---|
1023 | $_ = join('', @parts, $_);
|
---|
1024 | $nl = length($_) - ($lastlen - $nl) if $nl >= 0;
|
---|
1025 | }
|
---|
1026 |
|
---|
1027 | ##
|
---|
1028 | ## If we're at the end of the file, then we can use $_ as
|
---|
1029 | ## is. Otherwise, we need to remove the final partial-line
|
---|
1030 | ## and save it so that it'll be at the beginning of the
|
---|
1031 | ## next read (where the rest of the line will be layed in
|
---|
1032 | ## right after it). $hold will be what we should save
|
---|
1033 | ## until next time.
|
---|
1034 | ##
|
---|
1035 | if ($read != $size || $nl < 0) {
|
---|
1036 | $hold = '';
|
---|
1037 | } else {
|
---|
1038 | $hold = substr($_, $nl + 1);
|
---|
1039 | substr($_, $nl + 1) = '';
|
---|
1040 | }
|
---|
1041 |
|
---|
1042 | ##
|
---|
1043 | ## Now have a bunch of full lines in $_. Use it.
|
---|
1044 | ##
|
---|
1045 | if (eval $REGEX_TEST) {
|
---|
1046 | &clear_message if $VERBOSE && $STDERR_SCREWS_STDOUT;
|
---|
1047 | print $file, "\n";
|
---|
1048 | $retval=0; ## we've found something
|
---|
1049 |
|
---|
1050 | last;
|
---|
1051 | }
|
---|
1052 |
|
---|
1053 | ## Prepare for next read....
|
---|
1054 | $_ = $hold;
|
---|
1055 | }
|
---|
1056 |
|
---|
1057 | } else { ## else not using faster block scanning.....
|
---|
1058 |
|
---|
1059 | $lines_printed = 0 if $NICE;
|
---|
1060 | while (<FILE_IN>) {
|
---|
1061 | study;
|
---|
1062 | next unless (eval $REGEX_TEST);
|
---|
1063 |
|
---|
1064 | ##
|
---|
1065 | ## We found a matching line.
|
---|
1066 | ##
|
---|
1067 | $retval=0;
|
---|
1068 | &clear_message if $VERBOSE && $STDERR_SCREWS_STDOUT;
|
---|
1069 | if ($LIST_ONLY) {
|
---|
1070 | print $file, "\n";
|
---|
1071 | last;
|
---|
1072 | } else {
|
---|
1073 | ## prepare to print line.
|
---|
1074 | if ($NICE && $lines_printed++ == 0) {
|
---|
1075 | print '-' x 70, "\n" if $NICE > 1;
|
---|
1076 | print $file, ":\n";
|
---|
1077 | }
|
---|
1078 |
|
---|
1079 | ##
|
---|
1080 | ## Print all the prelim stuff. This looks less efficient
|
---|
1081 | ## than it needs to be, but that's so that when the eval
|
---|
1082 | ## is compiled (and the tests are optimized away), the
|
---|
1083 | ## result will be less actual PRINTs than the more natural
|
---|
1084 | ## way of doing these tests....
|
---|
1085 | ##
|
---|
1086 | if ($NICE) {
|
---|
1087 | if ($REPORT_LINENUM) {
|
---|
1088 | print " line $.: ";
|
---|
1089 | } else {
|
---|
1090 | print " ";
|
---|
1091 | }
|
---|
1092 | } elsif ($REPORT_LINENUM && $PREPEND_FILENAME) {
|
---|
1093 | print "$file,:$.: ";
|
---|
1094 | } elsif ($PREPEND_FILENAME) {
|
---|
1095 | print "$file: ";
|
---|
1096 | } elsif ($REPORT_LINENUM) {
|
---|
1097 | print "$.: ";
|
---|
1098 | }
|
---|
1099 | print $_;
|
---|
1100 | print "\n" unless m/\n$/;
|
---|
1101 | }
|
---|
1102 | }
|
---|
1103 | print "\n" if ($NICE > 1) && $lines_printed;
|
---|
1104 | }
|
---|
1105 | close(FILE_IN);
|
---|
1106 | }
|
---|
1107 | }
|
---|
1108 | closedir(DIR);
|
---|
1109 | }
|
---|
1110 |
|
---|
1111 | __END__
|
---|
1112 | .00; ## finish .ig
|
---|
1113 |
|
---|
1114 | 'di \" finish diversion--previous line must be blank
|
---|
1115 | .nr nl 0-1 \" fake up transition to first page again
|
---|
1116 | .nr % 0 \" start at page 1
|
---|
1117 | .\"__________________NORMAL_MAN_PAGE_BELOW_________________
|
---|
1118 | .ll+10n
|
---|
1119 | .TH search 1 "Dec 17, 1994"
|
---|
1120 | .SH SEARCH
|
---|
1121 | search \- search files (a'la grep) in a whole directory tree.
|
---|
1122 | .SH SYNOPSIS
|
---|
1123 | search [ grep-like and find-like options] [regex ....]
|
---|
1124 | .SH DESCRIPTION
|
---|
1125 | .I Search
|
---|
1126 | is more or less a combo of 'find' and 'grep' (although the regular
|
---|
1127 | expression flavor is that of the perl being used, which is closer to
|
---|
1128 | egrep's than grep's).
|
---|
1129 |
|
---|
1130 | .I Search
|
---|
1131 | does generally the same kind of thing that
|
---|
1132 | .nf
|
---|
1133 | find <blah blah> | xargs egrep <blah blah>
|
---|
1134 | .fi
|
---|
1135 | does, but is
|
---|
1136 | .I much
|
---|
1137 | more powerful and efficient (and intuitive, I think).
|
---|
1138 |
|
---|
1139 | This manual describes
|
---|
1140 | .I search
|
---|
1141 | as of version "941227.4". You can always find the latest version at
|
---|
1142 | .nf
|
---|
1143 | http://www.wg.omron.co.jp/~jfriedl/perl/index.html
|
---|
1144 | .fi
|
---|
1145 |
|
---|
1146 | .SH "QUICK EXAMPLE"
|
---|
1147 | Basic use is simple:
|
---|
1148 | .nf
|
---|
1149 | % search jeff
|
---|
1150 | .fi
|
---|
1151 | will search files in the current directory, and all sub directories, for
|
---|
1152 | files that have "jeff" in them. The lines will be listed with the
|
---|
1153 | containing file's name prepended.
|
---|
1154 | .PP
|
---|
1155 | If you list more than one regex, such as with
|
---|
1156 | .nf
|
---|
1157 | % search jeff Larry Randal+ 'Stoc?k' 'C.*son'
|
---|
1158 | .fi
|
---|
1159 | then a line containing any of the regexes will be listed.
|
---|
1160 | This makes it effectively the same as
|
---|
1161 | .nf
|
---|
1162 | % search 'jeff|Larry|Randal+|Stoc?k|C.*son'
|
---|
1163 | .fi
|
---|
1164 | However, listing them separately is much more efficient (and is easier
|
---|
1165 | to type).
|
---|
1166 | .PP
|
---|
1167 | Note that in the case of these examples, the
|
---|
1168 | .B \-w
|
---|
1169 | (list whole-words only) option would be useful.
|
---|
1170 | .PP
|
---|
1171 | Normally, various kinds of files are automatically removed from consideration.
|
---|
1172 | If it has has a certain ending (such as ".tar", ".Z", ".o", .etc), or if
|
---|
1173 | the beginning of the file looks like a binary, it'll be excluded.
|
---|
1174 | You can control exactly how this works -- see below. One quick way to
|
---|
1175 | override this is to use the
|
---|
1176 | .B \-all
|
---|
1177 | option, which means to consider all the files that would normally be
|
---|
1178 | automatically excluded.
|
---|
1179 | Or, if you're curious, you can use
|
---|
1180 | .B \-why
|
---|
1181 | to have notes about what files are skipped (and why) printed to stderr.
|
---|
1182 |
|
---|
1183 | .SH "BASIC OVERVIEW"
|
---|
1184 | Normally, the search starts in the current directory, considering files in
|
---|
1185 | all subdirectories.
|
---|
1186 |
|
---|
1187 | You can use the
|
---|
1188 | .I ~/.search
|
---|
1189 | file to control ways to automatically exclude files.
|
---|
1190 | If you don't have this file, a default one will kick in, which automatically
|
---|
1191 | add
|
---|
1192 | .nf
|
---|
1193 | -skip .o .Z .gif
|
---|
1194 | .fi
|
---|
1195 | (among others) to exclude those kinds of files (which you probably want to
|
---|
1196 | skip when searching for text, as is normal).
|
---|
1197 | Files that look to be be binary will also be excluded.
|
---|
1198 |
|
---|
1199 | Files ending with "#" and "~" will also be excluded unless the
|
---|
1200 | .B -x~
|
---|
1201 | option is given.
|
---|
1202 |
|
---|
1203 | You can use
|
---|
1204 | .B -showrc
|
---|
1205 | to show what kinds of files will normally be skipped.
|
---|
1206 | See the section on the startup file
|
---|
1207 | for more info.
|
---|
1208 |
|
---|
1209 | You can use the
|
---|
1210 | .B -all
|
---|
1211 | option to indicate you want to consider all files that would otherwise be
|
---|
1212 | skipped by the startup file.
|
---|
1213 |
|
---|
1214 | Based upon various other flags (see "WHICH FILES TO CONSIDER" below),
|
---|
1215 | more files might be removed from consideration. For example
|
---|
1216 | .nf
|
---|
1217 | -mtime 3
|
---|
1218 | .fi
|
---|
1219 | will exclude files that aren't at least three days old (change the 3 to -3
|
---|
1220 | to exclude files that are more than three days old), while
|
---|
1221 | .nf
|
---|
1222 | -skip .*
|
---|
1223 | .fi
|
---|
1224 | would exclude any file beginning with a dot (of course, '.' and '..' are
|
---|
1225 | special and always excluded).
|
---|
1226 |
|
---|
1227 | If you'd like to see what files are being excluded, and why, you can get the
|
---|
1228 | list via the
|
---|
1229 | .B \-why
|
---|
1230 | option.
|
---|
1231 |
|
---|
1232 | If a file makes it past all the checks, it is then "considered".
|
---|
1233 | This usually means it is greped for the regular expressions you gave
|
---|
1234 | on the command line.
|
---|
1235 |
|
---|
1236 | If any of the regexes match a line, the line is printed.
|
---|
1237 | However, if
|
---|
1238 | .B -list
|
---|
1239 | is given, just the filename is printed. Or, if
|
---|
1240 | .B -nice
|
---|
1241 | is given, a somewhat more (human-)readable output is generated.
|
---|
1242 |
|
---|
1243 | If you're searching a huge tree and want to keep informed about how
|
---|
1244 | the search is progressing,
|
---|
1245 | .B -v
|
---|
1246 | will print (to stderr) the current directory being searched.
|
---|
1247 | Using
|
---|
1248 | .B -vv
|
---|
1249 | will also print the current file "every so often", which could be useful
|
---|
1250 | if a directory is huge. Using
|
---|
1251 | .B -vvv
|
---|
1252 | will print the update with every file.
|
---|
1253 |
|
---|
1254 | Below is the full listing of options.
|
---|
1255 |
|
---|
1256 | .SH "OPTIONS TELLING *WHERE* TO SEARCH"
|
---|
1257 | .TP
|
---|
1258 | .BI -dir " DIR"
|
---|
1259 | Start searching at the named directory instead of the current directory.
|
---|
1260 | If multiple
|
---|
1261 | .B -dir
|
---|
1262 | arguments are given, multiple trees will be searched.
|
---|
1263 | .TP
|
---|
1264 | .BI -ddir " DIR"
|
---|
1265 | Like
|
---|
1266 | .B -dir
|
---|
1267 | except it flushes any previous
|
---|
1268 | .B -dir
|
---|
1269 | directories (i.e. "-dir A -dir B -dir C" will search A, B, and C, while
|
---|
1270 | "-dir A -ddir B -dir C" will search only B and C. This might be of use
|
---|
1271 | in the startup file (see that section below).
|
---|
1272 | .TP
|
---|
1273 | .B -xdev
|
---|
1274 | Stay on the same filesystem as the starting directory/directories.
|
---|
1275 | .TP
|
---|
1276 | .B -sort
|
---|
1277 | Sort the items in a directory before processing them.
|
---|
1278 | Normally they are processed in whatever order they happen to be read from
|
---|
1279 | the directory.
|
---|
1280 | .TP
|
---|
1281 | .B -nolinks
|
---|
1282 | Don't follow symbolic links. Normally they're followed.
|
---|
1283 |
|
---|
1284 | .SH "OPTIONS CONTROLLING WHICH FILES TO CONSIDER AND EXCLUDE"
|
---|
1285 | .TP
|
---|
1286 | .BI -mtime " NUM"
|
---|
1287 | Only consider files that were last changed more than
|
---|
1288 | .I NUM
|
---|
1289 | days ago
|
---|
1290 | (less than
|
---|
1291 | .I NUM
|
---|
1292 | days if
|
---|
1293 | .I NUM
|
---|
1294 | has '-' prepended, i.e. "-mtime -2.5" means to consider files that
|
---|
1295 | have been changed in the last two and a half days).
|
---|
1296 | .TP
|
---|
1297 | .B -older FILE
|
---|
1298 | Only consider files that have not changed since
|
---|
1299 | .I FILE
|
---|
1300 | was last changed.
|
---|
1301 | If there is any upper case in the "-older", "or equal" is added to the sense
|
---|
1302 | of the test. Therefore, "search -older ./file regex" will never consider
|
---|
1303 | "./file", while "search -Older ./file regex" will.
|
---|
1304 |
|
---|
1305 | If a file is a symbolic link, the time used is that of the file and not the
|
---|
1306 | link.
|
---|
1307 | .TP
|
---|
1308 | .BI -newer " FILE"
|
---|
1309 | Opposite of
|
---|
1310 | .BR -older .
|
---|
1311 | .TP
|
---|
1312 | .BI -name " GLOB"
|
---|
1313 | Only consider files that match the shell filename pattern
|
---|
1314 | .IR GLOB .
|
---|
1315 | The check is only done on a file's name (use
|
---|
1316 | .B -path
|
---|
1317 | to check the whole path, and use
|
---|
1318 | .B -dname
|
---|
1319 | to check directory names).
|
---|
1320 |
|
---|
1321 | Multiple specifications can be given by separating them with spaces, a'la
|
---|
1322 | .nf
|
---|
1323 | -name '*.c *.h'
|
---|
1324 | .fi
|
---|
1325 | to consider C source and header files.
|
---|
1326 | If
|
---|
1327 | .I GLOB
|
---|
1328 | doesn't contain any special pattern characters, a '*' is prepended.
|
---|
1329 | This last example could have been given as
|
---|
1330 | .nf
|
---|
1331 | -name '.c .h'
|
---|
1332 | .fi
|
---|
1333 | It could also be given as
|
---|
1334 | .nf
|
---|
1335 | -name .c -name .h
|
---|
1336 | .fi
|
---|
1337 | or
|
---|
1338 | .nf
|
---|
1339 | -name '*.c' -name '*.h'
|
---|
1340 | .fi
|
---|
1341 | or
|
---|
1342 | .nf
|
---|
1343 | -name '*.[ch]'
|
---|
1344 | .fi
|
---|
1345 | (among others)
|
---|
1346 | but in this last case, you have to be sure to supply the leading '*'.
|
---|
1347 | .TP
|
---|
1348 | .BI -path " GLOB"
|
---|
1349 | Like
|
---|
1350 | .B -name
|
---|
1351 | except the entire path is checked against the pattern.
|
---|
1352 | .TP
|
---|
1353 | .B -regex " REGEX"
|
---|
1354 | Considers files whose names (not paths) match the given perl regex
|
---|
1355 | exactly.
|
---|
1356 | .TP
|
---|
1357 | .BI -iname " GLOB"
|
---|
1358 | Case-insensitive version of
|
---|
1359 | .BR -name .
|
---|
1360 | .TP
|
---|
1361 | .BI -ipath " GLOB"
|
---|
1362 | Case-insensitive version of
|
---|
1363 | .BR -path .
|
---|
1364 | .TP
|
---|
1365 | .BI -iregex " REGEX"
|
---|
1366 | Case-insensitive version of
|
---|
1367 | .BR -regex .
|
---|
1368 |
|
---|
1369 | .TP
|
---|
1370 | .BI -dpath " GLOB"
|
---|
1371 | Only search down directories whose path matches the given pattern (this
|
---|
1372 | doesn't apply to the initial directory given by
|
---|
1373 | .BI -dir ,
|
---|
1374 | of course).
|
---|
1375 | Something like
|
---|
1376 | .nf
|
---|
1377 | -dir /usr/man -dpath /usr/man/man*
|
---|
1378 | .fi
|
---|
1379 | would completely skip
|
---|
1380 | "/usr/man/cat1", "/usr/man/cat2", etc.
|
---|
1381 | .TP
|
---|
1382 | .BI -dskip " GLOB"
|
---|
1383 | Skips directories whose name (not path) matches the given pattern.
|
---|
1384 | Something like
|
---|
1385 | .nf
|
---|
1386 | -dir /usr/man -dskip cat*
|
---|
1387 | .fi
|
---|
1388 | would completely skip any directory in the tree whose name begins with "cat"
|
---|
1389 | (including "/usr/man/cat1", "/usr/man/cat2", etc.).
|
---|
1390 | .TP
|
---|
1391 | .BI -dregex " REGEX"
|
---|
1392 | Like
|
---|
1393 | .BI -dpath ,
|
---|
1394 | but the pattern is a full perl regex. Note that this quite different
|
---|
1395 | from
|
---|
1396 | .B -regex
|
---|
1397 | which considers only file names (not paths). This option considers
|
---|
1398 | full directory paths (not just names). It's much more useful this way.
|
---|
1399 | Sorry if it's confusing.
|
---|
1400 | .TP
|
---|
1401 | .BI -dpath " GLOB"
|
---|
1402 | This option exists, but is probably not very useful. It probably wants to
|
---|
1403 | be like the '-below' or something I mention in the "TODO" section.
|
---|
1404 | .TP
|
---|
1405 | .BI -idpath " GLOB"
|
---|
1406 | Case-insensitive version of
|
---|
1407 | .BR -dpath .
|
---|
1408 | .TP
|
---|
1409 | .BI -idskip " GLOB"
|
---|
1410 | Case-insensitive version of
|
---|
1411 | .BR -dskip .
|
---|
1412 | .TP
|
---|
1413 | .BI -idregex " REGEX"
|
---|
1414 | Case-insensitive version of
|
---|
1415 | .BR -dregex .
|
---|
1416 | .TP
|
---|
1417 | .B -all
|
---|
1418 | Ignore any 'magic' or 'option' lines in the startup file.
|
---|
1419 | The effect is that all files that would otherwise be automatically
|
---|
1420 | excluded are considered.
|
---|
1421 | .TP
|
---|
1422 | .BI -x SPECIAL
|
---|
1423 | Arguments starting with
|
---|
1424 | .B -x
|
---|
1425 | (except
|
---|
1426 | .BR -xdev ,
|
---|
1427 | explained elsewhere) do special interaction with the
|
---|
1428 | .I ~/.search
|
---|
1429 | startup file. Something like
|
---|
1430 | .nf
|
---|
1431 | -xflag1 -xflag2
|
---|
1432 | .fi
|
---|
1433 | will turn on "flag1" and "flag2" in the startup file (and is
|
---|
1434 | the same as "-xflag1,flag2"). You can use this to write your own
|
---|
1435 | rules for what kinds of files are to be considered.
|
---|
1436 |
|
---|
1437 | For example, the internal-default startup file contains the line
|
---|
1438 | .nf
|
---|
1439 | <!~> option: -skip '~ #'
|
---|
1440 | .fi
|
---|
1441 | This means that if the
|
---|
1442 | .B -x~
|
---|
1443 | flag is
|
---|
1444 | .I not
|
---|
1445 | seen, the option
|
---|
1446 | .nf
|
---|
1447 | -skip '~ #'
|
---|
1448 | .fi
|
---|
1449 | should be done.
|
---|
1450 | The effect is that emacs temp and backup files are not normally
|
---|
1451 | considered, but you can included them with the -x~ flag.
|
---|
1452 |
|
---|
1453 | You can write your own rules to customize
|
---|
1454 | .I search
|
---|
1455 | in powerful ways. See the STARTUP FILE section below.
|
---|
1456 | .TP
|
---|
1457 | .B -why
|
---|
1458 | Print a message (to stderr) when and why a file is not considered.
|
---|
1459 |
|
---|
1460 | .SH "OPTIONS TELLING WHAT TO DO WITH FILES THAT WILL BE CONSIDERED"
|
---|
1461 | .TP
|
---|
1462 | .B -find
|
---|
1463 | (you can use
|
---|
1464 | .B -f
|
---|
1465 | as well).
|
---|
1466 | This option changes the basic action of
|
---|
1467 | .IR search .
|
---|
1468 |
|
---|
1469 | Normally, if a file is considered, it is searched
|
---|
1470 | for the regular expressions as described earlier. However, if this option
|
---|
1471 | is given, the filename is printed and no searching takes place. This turns
|
---|
1472 | .I search
|
---|
1473 | into a 'find' of some sorts.
|
---|
1474 |
|
---|
1475 | In this case, no regular expressions are needed on the command line
|
---|
1476 | (any that are there are silently ignored).
|
---|
1477 |
|
---|
1478 | This is not intended to be a replacement for the 'find' program,
|
---|
1479 | but to aid
|
---|
1480 | you in understanding just what files are getting past the exclusion checks.
|
---|
1481 | If you really want to use it as a sort of replacement for the 'find' program,
|
---|
1482 | you might want to use
|
---|
1483 | .B -all
|
---|
1484 | so that it doesn't waste time checking to see if the file is binary, etc
|
---|
1485 | (unless you really want that, of course).
|
---|
1486 |
|
---|
1487 | If you use
|
---|
1488 | .BR -find ,
|
---|
1489 | none of the "GREP-LIKE OPTIONS" (below) matter.
|
---|
1490 |
|
---|
1491 | As a replacement for 'find',
|
---|
1492 | .I search
|
---|
1493 | is probably a bit slower (or in the case of GNU find, a lot slower --
|
---|
1494 | GNU find is
|
---|
1495 | .I unbelievably
|
---|
1496 | fast).
|
---|
1497 | However, "search -ffind"
|
---|
1498 | might be more useful than 'find' when options such as
|
---|
1499 | .B -skip
|
---|
1500 | are used (at least until 'find' gets such functionality).
|
---|
1501 | .TP
|
---|
1502 | .B -ffind
|
---|
1503 | (or
|
---|
1504 | .BR -ff )
|
---|
1505 | A faster more 'find'-like find. Does
|
---|
1506 | .nf
|
---|
1507 | -find -all -dorep
|
---|
1508 | .fi
|
---|
1509 | .SH "GREP-LIKE OPTIONS"
|
---|
1510 | These options control how a searched file is accessed,
|
---|
1511 | and how things are printed.
|
---|
1512 | .TP
|
---|
1513 | .B -i
|
---|
1514 | Ignore letter case when matching.
|
---|
1515 | .TP
|
---|
1516 | .B -w
|
---|
1517 | Consider only whole-word matches ("whole word" as defined by perl's "\\b"
|
---|
1518 | regex).
|
---|
1519 | .TP
|
---|
1520 | .B -u
|
---|
1521 | If the regex(es) is/are simple, try to modify them so that they'll work
|
---|
1522 | in manpage-like underlined text (i.e. like _^Ht_^Hh_^Hi_^Hs).
|
---|
1523 | This is very rudimentary at the moment.
|
---|
1524 | .TP
|
---|
1525 | .B -list
|
---|
1526 | (you can use
|
---|
1527 | .B -l
|
---|
1528 | too).
|
---|
1529 | Don't print matching lines, but the names of files that contain matching
|
---|
1530 | lines. This will likely be *much* faster, as special optimizations are
|
---|
1531 | made -- particularly with large files.
|
---|
1532 | .TP
|
---|
1533 | .B -n
|
---|
1534 | Pepfix each line by its line number.
|
---|
1535 | .TP
|
---|
1536 | .B -nice
|
---|
1537 | Not a grep-like option, but similar to
|
---|
1538 | .BR -list ,
|
---|
1539 | so included here.
|
---|
1540 | .B -nice
|
---|
1541 | will have the output be a bit more human-readable, with matching lines printed
|
---|
1542 | slightly indented after the filename, a'la
|
---|
1543 | .nf
|
---|
1544 |
|
---|
1545 | % search foo
|
---|
1546 | somedir/somefile: line with foo in it
|
---|
1547 | somedir/somefile: some food for thought
|
---|
1548 | anotherdir/x: don't be a buffoon!
|
---|
1549 | %
|
---|
1550 |
|
---|
1551 | .fi
|
---|
1552 | will become
|
---|
1553 | .nf
|
---|
1554 |
|
---|
1555 | % search -nice foo
|
---|
1556 | somedir/somefile:
|
---|
1557 | line with foo in it
|
---|
1558 | some food for thought
|
---|
1559 | anotherdir/x:
|
---|
1560 | don't be a buffoon!
|
---|
1561 | %
|
---|
1562 |
|
---|
1563 | .fi
|
---|
1564 | This option due to Lionel Cons.
|
---|
1565 | .TP
|
---|
1566 | .B -nnice
|
---|
1567 | Be a bit nicer than
|
---|
1568 | .BR -nice .
|
---|
1569 | Prefix each file's output by a rule line, and follow with an extra blank line.
|
---|
1570 | .TP
|
---|
1571 | .B -h
|
---|
1572 | Don't prepend each output line with the name of the file
|
---|
1573 | (meaningless when
|
---|
1574 | .B -find
|
---|
1575 | or
|
---|
1576 | .B -l
|
---|
1577 | are given).
|
---|
1578 |
|
---|
1579 | .SH "OTHER OPTIONS"
|
---|
1580 | .TP
|
---|
1581 | .B -help
|
---|
1582 | Print the usage information.
|
---|
1583 | .TP
|
---|
1584 | .B -version
|
---|
1585 | Print the version information and quit.
|
---|
1586 | .TP
|
---|
1587 | .B -v
|
---|
1588 | Set the level of message verbosity.
|
---|
1589 | .B -v
|
---|
1590 | will print a note whenever a new directory is entered.
|
---|
1591 | .B -vv
|
---|
1592 | will also print a note "every so often". This can be useful to see
|
---|
1593 | what's happening when searching huge directories.
|
---|
1594 | .B -vvv
|
---|
1595 | will print a new with every file.
|
---|
1596 | .B -vvvv
|
---|
1597 | is
|
---|
1598 | -vvv
|
---|
1599 | plus
|
---|
1600 | .BR -why .
|
---|
1601 | .TP
|
---|
1602 | .B -e
|
---|
1603 | This ends the options, and can be useful if the regex begins with '-'.
|
---|
1604 | .TP
|
---|
1605 | .B -showrc
|
---|
1606 | Shows what is being considered in the startup file, then exits.
|
---|
1607 | .TP
|
---|
1608 | .B -dorep
|
---|
1609 | Normally, an identical file won't be checked twice (even with multiple
|
---|
1610 | hard or symbolic links). If you're just trying to do a fast
|
---|
1611 | .BR -find ,
|
---|
1612 | the bookkeeping to remember which files have been seen is not desirable,
|
---|
1613 | so you can eliminate the bookkeeping with this flag.
|
---|
1614 |
|
---|
1615 | .SH "STARTUP FILE"
|
---|
1616 | When
|
---|
1617 | .I search
|
---|
1618 | starts up, it processes the directives in
|
---|
1619 | .IR ~/.search .
|
---|
1620 | If no such file exists, a default
|
---|
1621 | internal version is used.
|
---|
1622 |
|
---|
1623 | The internal version looks like:
|
---|
1624 | .nf
|
---|
1625 |
|
---|
1626 | magic: 32 : $H =~ m/[\ex00-\ex06\ex10-\ex1a\ex1c-\ex1f\ex80\exff]{2}/
|
---|
1627 | option: -skip '.a .COM .elc .EXE .gz .o .pbm .xbm .dvi'
|
---|
1628 | option: -iskip '.tarz .zip .z .lzh .jpg .jpeg .gif .uu'
|
---|
1629 | <!~> option: -skip '~ #'
|
---|
1630 |
|
---|
1631 | .fi
|
---|
1632 | If you wish to create your own "~/.search",
|
---|
1633 | you might consider copying the above, and then working from there.
|
---|
1634 |
|
---|
1635 | There are two kinds of directives in a startup file: "magic" and "option".
|
---|
1636 | .RS 0n
|
---|
1637 | .TP
|
---|
1638 | OPTION
|
---|
1639 | Option lines will automatically do the command-line options given.
|
---|
1640 | For example, the line
|
---|
1641 | .nf
|
---|
1642 | option: -v
|
---|
1643 | .fi
|
---|
1644 | in you startup file will turn on -v every time, without needing to type it
|
---|
1645 | on the command line.
|
---|
1646 |
|
---|
1647 | The text on the line after the "option:" directive is processed
|
---|
1648 | like the Bourne shell, so make sure to pay attention to quoting.
|
---|
1649 | .nf
|
---|
1650 | option: -skip .exe .com
|
---|
1651 | .fi
|
---|
1652 | will give an error (".com" by itself isn't a valid option), while
|
---|
1653 | .nf
|
---|
1654 | option: -skip ".exe .com"
|
---|
1655 | .fi
|
---|
1656 | will properly include it as part of -skip's argument.
|
---|
1657 |
|
---|
1658 | .TP
|
---|
1659 | MAGIC
|
---|
1660 | Magic lines are used to determine if a file should be considered a binary
|
---|
1661 | or not (the term "magic" refers to checking a file's magic number). These
|
---|
1662 | are described in more detail below.
|
---|
1663 | .RE
|
---|
1664 |
|
---|
1665 | Blank lines and comments (lines beginning with '#') are allowed.
|
---|
1666 |
|
---|
1667 | If a line begins with <...>, then it's a check to see if the
|
---|
1668 | directive on the line should be done or not. The stuff inside the <...>
|
---|
1669 | can contain perl's && (and), || (or), ! (not), and parens for grouping,
|
---|
1670 | along with "flags" that might be indicated by the user with
|
---|
1671 | .BI -x flag
|
---|
1672 | options.
|
---|
1673 |
|
---|
1674 | For example, using "-xfoo" will cause "foo" to be true inside the <...>
|
---|
1675 | blocks. Therefore, a line beginning with "<foo>" would be done only when
|
---|
1676 | "-xfoo" had been specified, while a line beginning with "<!foo>" would be
|
---|
1677 | done only when "-xfoo" is not specified (of course, a line without any <...>
|
---|
1678 | is done in either case).
|
---|
1679 |
|
---|
1680 | A realistic example might be
|
---|
1681 | .nf
|
---|
1682 | <!v> -vv
|
---|
1683 | .fi
|
---|
1684 | This will cause -vv messages to be the default, but allow "-xv" to override.
|
---|
1685 |
|
---|
1686 | There are a few flags that are set automatically:
|
---|
1687 | .RS
|
---|
1688 | .TP
|
---|
1689 | .B TTY
|
---|
1690 | true if the output is to the screen (as opposed to being redirected to a file).
|
---|
1691 | You can force this (as with all the other automatic flags) with -xTTY.
|
---|
1692 | .TP
|
---|
1693 | .B -v
|
---|
1694 | True if -v was specified. If -vv was specified, both
|
---|
1695 | .B -v
|
---|
1696 | and
|
---|
1697 | .B -vv
|
---|
1698 | flags are true (and so on).
|
---|
1699 | .TP
|
---|
1700 | .B -nice
|
---|
1701 | True if -nice was specified. Same thing about -nnice as for -vv.
|
---|
1702 | .PP
|
---|
1703 | .TP
|
---|
1704 | .B -list
|
---|
1705 | true if -list (or -l) was given.
|
---|
1706 | .TP
|
---|
1707 | .B -dir
|
---|
1708 | true if -dir was given.
|
---|
1709 | .RE
|
---|
1710 |
|
---|
1711 | Using this info, you might change the last example to
|
---|
1712 | .nf
|
---|
1713 |
|
---|
1714 | <!v && !-v> option: -vv
|
---|
1715 |
|
---|
1716 | .fi
|
---|
1717 | The added "&& !-v" means "and if the '-v' option not given".
|
---|
1718 | This will allow you to use "-v" alone on the command line, and not
|
---|
1719 | have this directive add the more verbose "-vv" automatically.
|
---|
1720 |
|
---|
1721 | .RS 0
|
---|
1722 | Some other examples:
|
---|
1723 | .TP
|
---|
1724 | <!-dir && !here> option: -dir ~/
|
---|
1725 | Effectively make the default directory your home directory (instead of the
|
---|
1726 | current directory). Using -dir or -xhere will undo this.
|
---|
1727 | .TP
|
---|
1728 | <tex> option: -name .tex -dir ~/pub
|
---|
1729 | Create '-xtex' to search only "*.tex" files in your ~/pub directory tree.
|
---|
1730 | Actually, this could be made a bit better. If you combine '-xtex' and '-dir'
|
---|
1731 | on the command line, this directive will add ~/pub to the list, when you
|
---|
1732 | probably want to use the -dir directory only. You could do
|
---|
1733 | .nf
|
---|
1734 |
|
---|
1735 | <tex> option: -name .tex
|
---|
1736 | <tex && !-dir> option: -dir ~/pub
|
---|
1737 | .fi
|
---|
1738 |
|
---|
1739 | to will allow '-xtex' to work as before, but allow a command-line "-dir"
|
---|
1740 | to take precedence with respect to ~/pub.
|
---|
1741 | .TP
|
---|
1742 | <fluff> option: -nnice -sort -i -vvv
|
---|
1743 | Combine a few user-friendly options into one '-xfluff' option.
|
---|
1744 | .TP
|
---|
1745 | <man> option: -ddir /usr/man -v -w
|
---|
1746 | When the '-xman' option is given, search "/usr/man" for whole-words
|
---|
1747 | (of whatever regex or regexes are given on the command line), with -v.
|
---|
1748 | .RE
|
---|
1749 |
|
---|
1750 | The lines in the startup file are executed from top to bottom, so something
|
---|
1751 | like
|
---|
1752 | .nf
|
---|
1753 |
|
---|
1754 | <both> option: -xflag1 -xflag2
|
---|
1755 | <flag1> option: ...whatever...
|
---|
1756 | <flag2> option: ...whatever...
|
---|
1757 |
|
---|
1758 | .fi
|
---|
1759 | will allow '-xboth' to be the same as '-xflag1 -xflag2' (or '-xflag1,flag2'
|
---|
1760 | for that matter). However, if you put the "<both>" line below the others,
|
---|
1761 | they will not be true when encountered, so the result would be different
|
---|
1762 | (and probably undesired).
|
---|
1763 |
|
---|
1764 | The "magic" directives are used to determine if a file looks to be binary
|
---|
1765 | or not. The form of a magic line is
|
---|
1766 | .nf
|
---|
1767 | magic: \fISIZE\fP : \fIPERLCODE\fP
|
---|
1768 | .fi
|
---|
1769 | where
|
---|
1770 | .I SIZE
|
---|
1771 | is the number of bytes of the file you need to check, and
|
---|
1772 | .I PERLCODE
|
---|
1773 | is the code to do the check. Within
|
---|
1774 | .IR PERLCODE ,
|
---|
1775 | the variable $H will hold at least the first
|
---|
1776 | .I SIZE
|
---|
1777 | bytes of the file (unless the file is shorter than that, of course).
|
---|
1778 | It might hold more bytes. The perl should evaluate to true if the file
|
---|
1779 | should be considered a binary.
|
---|
1780 |
|
---|
1781 | An example might be
|
---|
1782 | .nf
|
---|
1783 | magic: 6 : substr($H, 0, 6) eq 'GIF87a'
|
---|
1784 | .fi
|
---|
1785 | to test for a GIF ("-iskip .gif" is better, but this might be useful
|
---|
1786 | if you have images in files without the ".gif" extension).
|
---|
1787 |
|
---|
1788 | Since the startup file is checked from top to bottom, you can be a bit
|
---|
1789 | efficient:
|
---|
1790 | .nf
|
---|
1791 | magic: 6 : ($x6 = substr($H, 0, 6)) eq 'GIF87a'
|
---|
1792 | magic: 6 : $x6 eq 'GIF89a'
|
---|
1793 | .fi
|
---|
1794 | You could also write the same thing as
|
---|
1795 | .nf
|
---|
1796 | magic: 6 : (($x6 = substr($H, 0, 6)) eq 'GIF87a') || ## an old gif, or.. \e
|
---|
1797 | $x6 eq 'GIF89a' ## .. a new one.
|
---|
1798 | .fi
|
---|
1799 | since newlines may be escaped.
|
---|
1800 |
|
---|
1801 | The default internal startup file includes
|
---|
1802 | .nf
|
---|
1803 | magic: 32 : $H =~ m/[\ex00-\ex06\ex10-\ex1a\ex1c-\ex1f\ex80\exff]{2}/
|
---|
1804 | .fi
|
---|
1805 | which checks for certain non-printable characters, and catches a large
|
---|
1806 | number of binary files, including most system's executables, linkable
|
---|
1807 | objects, compressed, tarred, and otherwise folded, spindled, and mutilated
|
---|
1808 | files.
|
---|
1809 |
|
---|
1810 | Another example might be
|
---|
1811 | .nf
|
---|
1812 | ## an archive library
|
---|
1813 | magic: 17 : substr($H, 0, 17) eq "!<arch>\en__.SYMDEF"
|
---|
1814 | .fi
|
---|
1815 |
|
---|
1816 | .SH "RETURN VALUE"
|
---|
1817 | .I Search
|
---|
1818 | returns zero if lines (or files, if appropriate) were found,
|
---|
1819 | or if no work was requested (such as with
|
---|
1820 | .BR -help ).
|
---|
1821 | Returns 1 if no lines (or files) were found.
|
---|
1822 | Returns 2 on error.
|
---|
1823 |
|
---|
1824 | .SH TODO
|
---|
1825 | Things I'd like to add some day:
|
---|
1826 | .nf
|
---|
1827 | + show surrounding lines (context).
|
---|
1828 | + highlight matched portions of lines.
|
---|
1829 | + add '-and', which can go between regexes to override
|
---|
1830 | the default logical or of the regexes.
|
---|
1831 | + add something like
|
---|
1832 | -below GLOB
|
---|
1833 | which will examine a tree and only consider files that
|
---|
1834 | lie in a directory deeper than one named by the pattern.
|
---|
1835 | + add 'warning' and 'error' directives.
|
---|
1836 | + add 'help' directive.
|
---|
1837 | .fi
|
---|
1838 | .SH BUGS
|
---|
1839 | If -xdev and multiple -dir arguments are given, any file in any of the
|
---|
1840 | target filesystems are allowed. It would be better to allow each filesystem
|
---|
1841 | for each separate tree.
|
---|
1842 |
|
---|
1843 | Multiple -dir args might also cause some confusing effects. Doing
|
---|
1844 | .nf
|
---|
1845 | -dir some/dir -dir other
|
---|
1846 | .fi
|
---|
1847 | will search "some/dir" completely, then search "other" completely. This
|
---|
1848 | is good. However, something like
|
---|
1849 | .nf
|
---|
1850 | -dir some/dir -dir some/dir/more/specific
|
---|
1851 | .fi
|
---|
1852 | will search "some/dir" completely *except for* "some/dir/more/specific",
|
---|
1853 | after which it will return and be searched. Not really a bug, but just sort
|
---|
1854 | of odd.
|
---|
1855 |
|
---|
1856 | File times (for -newer, etc.) of symbolic links are for the file, not the
|
---|
1857 | link. This could cause some misunderstandings.
|
---|
1858 |
|
---|
1859 | Probably more. Please let me know.
|
---|
1860 | .SH AUTHOR
|
---|
1861 | Jeffrey Friedl, Omron Corp (jfriedl@omron.co.jp)
|
---|
1862 | .br
|
---|
1863 | http://www.wg.omron.co.jp/cgi-bin/j-e/jfriedl.html
|
---|
1864 |
|
---|
1865 | .SH "LATEST SOURCE"
|
---|
1866 | See http://www.wg.omron.co.jp/~jfriedl/perl/index.html
|
---|