[537] | 1 | ###########################################################################
|
---|
| 2 | #
|
---|
| 3 | # util.pm -- various useful utilities
|
---|
| 4 | # A component of the Greenstone digital library software
|
---|
| 5 | # from the New Zealand Digital Library Project at the
|
---|
| 6 | # University of Waikato, New Zealand.
|
---|
| 7 | #
|
---|
| 8 | # Copyright (C) 1999 New Zealand Digital Library Project
|
---|
| 9 | #
|
---|
| 10 | # This program is free software; you can redistribute it and/or modify
|
---|
| 11 | # it under the terms of the GNU General Public License as published by
|
---|
| 12 | # the Free Software Foundation; either version 2 of the License, or
|
---|
| 13 | # (at your option) any later version.
|
---|
| 14 | #
|
---|
| 15 | # This program is distributed in the hope that it will be useful,
|
---|
| 16 | # but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
| 17 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
| 18 | # GNU General Public License for more details.
|
---|
| 19 | #
|
---|
| 20 | # You should have received a copy of the GNU General Public License
|
---|
| 21 | # along with this program; if not, write to the Free Software
|
---|
| 22 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
| 23 | #
|
---|
| 24 | ###########################################################################
|
---|
[4] | 25 |
|
---|
| 26 | package util;
|
---|
| 27 |
|
---|
| 28 | use File::Copy;
|
---|
[619] | 29 | use File::Basename;
|
---|
[4] | 30 |
|
---|
[14926] | 31 | use strict;
|
---|
[14365] | 32 |
|
---|
[14926] | 33 |
|
---|
[4] | 34 | # removes files (but not directories)
|
---|
| 35 | sub rm {
|
---|
| 36 | my (@files) = @_;
|
---|
[18469] | 37 |
|
---|
[4] | 38 | my @filefiles = ();
|
---|
| 39 |
|
---|
| 40 | # make sure the files we want to delete exist
|
---|
| 41 | # and are regular files
|
---|
[10046] | 42 | foreach my $file (@files) {
|
---|
[4] | 43 | if (!-e $file) {
|
---|
| 44 | print STDERR "util::rm $file does not exist\n";
|
---|
[721] | 45 | } elsif ((!-f $file) && (!-l $file)) {
|
---|
| 46 | print STDERR "util::rm $file is not a regular (or symbolic) file\n";
|
---|
[4] | 47 | } else {
|
---|
| 48 | push (@filefiles, $file);
|
---|
| 49 | }
|
---|
| 50 | }
|
---|
| 51 |
|
---|
| 52 | # remove the files
|
---|
| 53 | my $numremoved = unlink @filefiles;
|
---|
| 54 |
|
---|
| 55 | # check to make sure all of them were removed
|
---|
| 56 | if ($numremoved != scalar(@filefiles)) {
|
---|
| 57 | print STDERR "util::rm Not all files were removed\n";
|
---|
| 58 | }
|
---|
| 59 | }
|
---|
| 60 |
|
---|
| 61 |
|
---|
[10211] | 62 |
|
---|
[4] | 63 | # recursive removal
|
---|
[10211] | 64 | sub filtered_rm_r {
|
---|
| 65 | my ($files,$file_accept_re,$file_reject_re) = @_;
|
---|
[4] | 66 |
|
---|
[10211] | 67 | my @files_array = (ref $files eq "ARRAY") ? @$files : ($files);
|
---|
| 68 |
|
---|
[4] | 69 | # recursively remove the files
|
---|
[10211] | 70 | foreach my $file (@files_array) {
|
---|
[4] | 71 | $file =~ s/[\/\\]+$//; # remove trailing slashes
|
---|
| 72 |
|
---|
| 73 | if (!-e $file) {
|
---|
[10211] | 74 | print STDERR "util::filtered_rm_r $file does not exist\n";
|
---|
[4] | 75 |
|
---|
[721] | 76 | } elsif ((-d $file) && (!-l $file)) { # don't recurse down symbolic link
|
---|
[4] | 77 | # get the contents of this directory
|
---|
| 78 | if (!opendir (INDIR, $file)) {
|
---|
[10211] | 79 | print STDERR "util::filtered_rm_r could not open directory $file\n";
|
---|
[4] | 80 | } else {
|
---|
| 81 | my @filedir = grep (!/^\.\.?$/, readdir (INDIR));
|
---|
| 82 | closedir (INDIR);
|
---|
[10211] | 83 |
|
---|
[4] | 84 | # remove all the files in this directory
|
---|
[10211] | 85 | map {$_="$file/$_";} @filedir;
|
---|
| 86 | &filtered_rm_r (\@filedir,$file_accept_re,$file_reject_re);
|
---|
[4] | 87 |
|
---|
[10211] | 88 | if (!defined $file_accept_re && !defined $file_reject_re) {
|
---|
| 89 | # remove this directory
|
---|
| 90 | if (!rmdir $file) {
|
---|
| 91 | print STDERR "util::filtered_rm_r couldn't remove directory $file\n";
|
---|
| 92 | }
|
---|
[4] | 93 | }
|
---|
| 94 | }
|
---|
[10211] | 95 | } else {
|
---|
| 96 | next if (defined $file_reject_re && ($file =~ m/$file_reject_re/));
|
---|
[4] | 97 |
|
---|
[10211] | 98 | if ((!defined $file_accept_re) || ($file =~ m/$file_accept_re/)) {
|
---|
| 99 | # remove this file
|
---|
| 100 | &rm ($file);
|
---|
| 101 | }
|
---|
[4] | 102 | }
|
---|
| 103 | }
|
---|
| 104 | }
|
---|
| 105 |
|
---|
[10211] | 106 |
|
---|
| 107 | # recursive removal
|
---|
| 108 | sub rm_r {
|
---|
| 109 | my (@files) = @_;
|
---|
| 110 |
|
---|
| 111 | # use the more general (but reterospectively written function
|
---|
| 112 | # filtered_rm_r function()
|
---|
| 113 |
|
---|
| 114 | filtered_rm_r(\@files,undef,undef); # no accept or reject expressions
|
---|
| 115 | }
|
---|
| 116 |
|
---|
| 117 |
|
---|
| 118 |
|
---|
| 119 |
|
---|
[721] | 120 | # moves a file or a group of files
|
---|
| 121 | sub mv {
|
---|
| 122 | my $dest = pop (@_);
|
---|
| 123 | my (@srcfiles) = @_;
|
---|
[4] | 124 |
|
---|
[721] | 125 | # remove trailing slashes from source and destination files
|
---|
| 126 | $dest =~ s/[\\\/]+$//;
|
---|
| 127 | map {$_ =~ s/[\\\/]+$//;} @srcfiles;
|
---|
| 128 |
|
---|
| 129 | # a few sanity checks
|
---|
| 130 | if (scalar (@srcfiles) == 0) {
|
---|
| 131 | print STDERR "util::mv no destination directory given\n";
|
---|
| 132 | return;
|
---|
| 133 | } elsif ((scalar (@srcfiles) > 1) && (!-d $dest)) {
|
---|
| 134 | print STDERR "util::mv if multiple source files are given the ".
|
---|
| 135 | "destination must be a directory\n";
|
---|
| 136 | return;
|
---|
| 137 | }
|
---|
| 138 |
|
---|
| 139 | # move the files
|
---|
[8716] | 140 | foreach my $file (@srcfiles) {
|
---|
[721] | 141 | my $tempdest = $dest;
|
---|
| 142 | if (-d $tempdest) {
|
---|
| 143 | my ($filename) = $file =~ /([^\\\/]+)$/;
|
---|
| 144 | $tempdest .= "/$filename";
|
---|
| 145 | }
|
---|
| 146 | if (!-e $file) {
|
---|
| 147 | print STDERR "util::mv $file does not exist\n";
|
---|
| 148 | } else {
|
---|
| 149 | rename ($file, $tempdest);
|
---|
| 150 | }
|
---|
| 151 | }
|
---|
| 152 | }
|
---|
| 153 |
|
---|
| 154 |
|
---|
[4] | 155 | # copies a file or a group of files
|
---|
| 156 | sub cp {
|
---|
| 157 | my $dest = pop (@_);
|
---|
| 158 | my (@srcfiles) = @_;
|
---|
| 159 |
|
---|
| 160 | # remove trailing slashes from source and destination files
|
---|
| 161 | $dest =~ s/[\\\/]+$//;
|
---|
| 162 | map {$_ =~ s/[\\\/]+$//;} @srcfiles;
|
---|
| 163 |
|
---|
| 164 | # a few sanity checks
|
---|
| 165 | if (scalar (@srcfiles) == 0) {
|
---|
| 166 | print STDERR "util::cp no destination directory given\n";
|
---|
| 167 | return;
|
---|
| 168 | } elsif ((scalar (@srcfiles) > 1) && (!-d $dest)) {
|
---|
| 169 | print STDERR "util::cp if multiple source files are given the ".
|
---|
| 170 | "destination must be a directory\n";
|
---|
| 171 | return;
|
---|
| 172 | }
|
---|
| 173 |
|
---|
| 174 | # copy the files
|
---|
[8716] | 175 | foreach my $file (@srcfiles) {
|
---|
[4] | 176 | my $tempdest = $dest;
|
---|
| 177 | if (-d $tempdest) {
|
---|
| 178 | my ($filename) = $file =~ /([^\\\/]+)$/;
|
---|
| 179 | $tempdest .= "/$filename";
|
---|
| 180 | }
|
---|
| 181 | if (!-e $file) {
|
---|
| 182 | print STDERR "util::cp $file does not exist\n";
|
---|
| 183 | } elsif (!-f $file) {
|
---|
| 184 | print STDERR "util::cp $file is not a plain file\n";
|
---|
| 185 | } else {
|
---|
| 186 | &File::Copy::copy ($file, $tempdest);
|
---|
| 187 | }
|
---|
| 188 | }
|
---|
| 189 | }
|
---|
| 190 |
|
---|
| 191 |
|
---|
[721] | 192 |
|
---|
[4] | 193 | # recursively copies a file or group of files
|
---|
[1454] | 194 | # syntax: cp_r (sourcefiles, destination directory)
|
---|
| 195 | # destination must be a directory - to copy one file to
|
---|
| 196 | # another use cp instead
|
---|
[4] | 197 | sub cp_r {
|
---|
| 198 | my $dest = pop (@_);
|
---|
| 199 | my (@srcfiles) = @_;
|
---|
| 200 |
|
---|
| 201 | # a few sanity checks
|
---|
| 202 | if (scalar (@srcfiles) == 0) {
|
---|
[1454] | 203 | print STDERR "util::cp_r no destination directory given\n";
|
---|
[4] | 204 | return;
|
---|
[1454] | 205 | } elsif (-f $dest) {
|
---|
| 206 | print STDERR "util::cp_r destination must be a directory\n";
|
---|
[4] | 207 | return;
|
---|
| 208 | }
|
---|
| 209 |
|
---|
[1454] | 210 | # create destination directory if it doesn't exist already
|
---|
| 211 | if (! -d $dest) {
|
---|
| 212 | my $store_umask = umask(0002);
|
---|
| 213 | mkdir ($dest, 0777);
|
---|
| 214 | umask($store_umask);
|
---|
| 215 | }
|
---|
| 216 |
|
---|
[4] | 217 | # copy the files
|
---|
[8716] | 218 | foreach my $file (@srcfiles) {
|
---|
[4] | 219 |
|
---|
| 220 | if (!-e $file) {
|
---|
[1454] | 221 | print STDERR "util::cp_r $file does not exist\n";
|
---|
[4] | 222 |
|
---|
| 223 | } elsif (-d $file) {
|
---|
[1586] | 224 | # make the new directory
|
---|
| 225 | my ($filename) = $file =~ /([^\\\/]*)$/;
|
---|
| 226 | $dest = &util::filename_cat ($dest, $filename);
|
---|
| 227 | my $store_umask = umask(0002);
|
---|
| 228 | mkdir ($dest, 0777);
|
---|
| 229 | umask($store_umask);
|
---|
[836] | 230 |
|
---|
[4] | 231 | # get the contents of this directory
|
---|
| 232 | if (!opendir (INDIR, $file)) {
|
---|
| 233 | print STDERR "util::cp_r could not open directory $file\n";
|
---|
| 234 | } else {
|
---|
[1454] | 235 | my @filedir = readdir (INDIR);
|
---|
[4] | 236 | closedir (INDIR);
|
---|
[8716] | 237 | foreach my $f (@filedir) {
|
---|
[1454] | 238 | next if $f =~ /^\.\.?$/;
|
---|
| 239 | # copy all the files in this directory
|
---|
| 240 | my $ff = &util::filename_cat ($file, $f);
|
---|
| 241 | &cp_r ($ff, $dest);
|
---|
| 242 | }
|
---|
[4] | 243 | }
|
---|
| 244 |
|
---|
| 245 | } else {
|
---|
[1454] | 246 | &cp($file, $dest);
|
---|
[4] | 247 | }
|
---|
| 248 | }
|
---|
| 249 | }
|
---|
[21762] | 250 | # recursively copies a file or group of files
|
---|
| 251 | # syntax: cp_r (sourcefiles, destination directory)
|
---|
| 252 | # destination must be a directory - to copy one file to
|
---|
| 253 | # another use cp instead
|
---|
| 254 | sub cp_r_nosvn {
|
---|
| 255 | my $dest = pop (@_);
|
---|
| 256 | my (@srcfiles) = @_;
|
---|
[4] | 257 |
|
---|
[21762] | 258 | # a few sanity checks
|
---|
| 259 | if (scalar (@srcfiles) == 0) {
|
---|
| 260 | print STDERR "util::cp_r no destination directory given\n";
|
---|
| 261 | return;
|
---|
| 262 | } elsif (-f $dest) {
|
---|
| 263 | print STDERR "util::cp_r destination must be a directory\n";
|
---|
| 264 | return;
|
---|
| 265 | }
|
---|
| 266 |
|
---|
| 267 | # create destination directory if it doesn't exist already
|
---|
| 268 | if (! -d $dest) {
|
---|
| 269 | my $store_umask = umask(0002);
|
---|
| 270 | mkdir ($dest, 0777);
|
---|
| 271 | umask($store_umask);
|
---|
| 272 | }
|
---|
| 273 |
|
---|
| 274 | # copy the files
|
---|
| 275 | foreach my $file (@srcfiles) {
|
---|
| 276 |
|
---|
| 277 | if (!-e $file) {
|
---|
| 278 | print STDERR "util::cp_r $file does not exist\n";
|
---|
| 279 |
|
---|
| 280 | } elsif (-d $file) {
|
---|
| 281 | # make the new directory
|
---|
| 282 | my ($filename) = $file =~ /([^\\\/]*)$/;
|
---|
| 283 | $dest = &util::filename_cat ($dest, $filename);
|
---|
| 284 | my $store_umask = umask(0002);
|
---|
| 285 | mkdir ($dest, 0777);
|
---|
| 286 | umask($store_umask);
|
---|
| 287 |
|
---|
| 288 | # get the contents of this directory
|
---|
| 289 | if (!opendir (INDIR, $file)) {
|
---|
| 290 | print STDERR "util::cp_r could not open directory $file\n";
|
---|
| 291 | } else {
|
---|
| 292 | my @filedir = readdir (INDIR);
|
---|
| 293 | closedir (INDIR);
|
---|
| 294 | foreach my $f (@filedir) {
|
---|
| 295 | next if $f =~ /^\.\.?$/;
|
---|
| 296 | next if $f =~ /^\.svn$/;
|
---|
| 297 | # copy all the files in this directory
|
---|
| 298 | my $ff = &util::filename_cat ($file, $f);
|
---|
| 299 | &cp_r ($ff, $dest);
|
---|
| 300 | }
|
---|
| 301 | }
|
---|
| 302 |
|
---|
| 303 | } else {
|
---|
| 304 | &cp($file, $dest);
|
---|
| 305 | }
|
---|
| 306 | }
|
---|
| 307 | }
|
---|
| 308 |
|
---|
[11179] | 309 | # copies a directory and its contents, excluding subdirectories, into a new directory
|
---|
| 310 | sub cp_r_toplevel {
|
---|
| 311 | my $dest = pop (@_);
|
---|
| 312 | my (@srcfiles) = @_;
|
---|
[4] | 313 |
|
---|
[11179] | 314 | # a few sanity checks
|
---|
| 315 | if (scalar (@srcfiles) == 0) {
|
---|
| 316 | print STDERR "util::cp_r no destination directory given\n";
|
---|
| 317 | return;
|
---|
| 318 | } elsif (-f $dest) {
|
---|
| 319 | print STDERR "util::cp_r destination must be a directory\n";
|
---|
| 320 | return;
|
---|
| 321 | }
|
---|
| 322 |
|
---|
| 323 | # create destination directory if it doesn't exist already
|
---|
| 324 | if (! -d $dest) {
|
---|
| 325 | my $store_umask = umask(0002);
|
---|
| 326 | mkdir ($dest, 0777);
|
---|
| 327 | umask($store_umask);
|
---|
| 328 | }
|
---|
| 329 |
|
---|
| 330 | # copy the files
|
---|
| 331 | foreach my $file (@srcfiles) {
|
---|
| 332 |
|
---|
| 333 | if (!-e $file) {
|
---|
| 334 | print STDERR "util::cp_r $file does not exist\n";
|
---|
| 335 |
|
---|
| 336 | } elsif (-d $file) {
|
---|
| 337 | # make the new directory
|
---|
| 338 | my ($filename) = $file =~ /([^\\\/]*)$/;
|
---|
| 339 | $dest = &util::filename_cat ($dest, $filename);
|
---|
| 340 | my $store_umask = umask(0002);
|
---|
| 341 | mkdir ($dest, 0777);
|
---|
| 342 | umask($store_umask);
|
---|
| 343 |
|
---|
| 344 | # get the contents of this directory
|
---|
| 345 | if (!opendir (INDIR, $file)) {
|
---|
| 346 | print STDERR "util::cp_r could not open directory $file\n";
|
---|
| 347 | } else {
|
---|
| 348 | my @filedir = readdir (INDIR);
|
---|
| 349 | closedir (INDIR);
|
---|
| 350 | foreach my $f (@filedir) {
|
---|
| 351 | next if $f =~ /^\.\.?$/;
|
---|
| 352 |
|
---|
| 353 | # copy all the files in this directory, but not directories
|
---|
| 354 | my $ff = &util::filename_cat ($file, $f);
|
---|
| 355 | if (-f $ff) {
|
---|
| 356 | &cp($ff, $dest);
|
---|
| 357 | #&cp_r ($ff, $dest);
|
---|
| 358 | }
|
---|
| 359 | }
|
---|
| 360 | }
|
---|
| 361 |
|
---|
| 362 | } else {
|
---|
| 363 | &cp($file, $dest);
|
---|
| 364 | }
|
---|
| 365 | }
|
---|
| 366 | }
|
---|
| 367 |
|
---|
[721] | 368 | sub mk_dir {
|
---|
| 369 | my ($dir) = @_;
|
---|
| 370 |
|
---|
[836] | 371 | my $store_umask = umask(0002);
|
---|
| 372 | my $mkdir_ok = mkdir ($dir, 0777);
|
---|
| 373 | umask($store_umask);
|
---|
| 374 |
|
---|
| 375 | if (!$mkdir_ok)
|
---|
| 376 | {
|
---|
[721] | 377 | print STDERR "util::mk_dir could not create directory $dir\n";
|
---|
| 378 | return;
|
---|
| 379 | }
|
---|
| 380 | }
|
---|
| 381 |
|
---|
[1046] | 382 | # in case anyone cares - I did some testing (using perls Benchmark module)
|
---|
| 383 | # on this subroutine against File::Path::mkpath (). mk_all_dir() is apparently
|
---|
| 384 | # slightly faster (surprisingly) - Stefan.
|
---|
[4] | 385 | sub mk_all_dir {
|
---|
| 386 | my ($dir) = @_;
|
---|
| 387 |
|
---|
| 388 | # use / for the directory separator, remove duplicate and
|
---|
| 389 | # trailing slashes
|
---|
| 390 | $dir=~s/[\\\/]+/\//g;
|
---|
| 391 | $dir=~s/[\\\/]+$//;
|
---|
| 392 |
|
---|
| 393 | # make sure the cache directory exists
|
---|
| 394 | my $dirsofar = "";
|
---|
| 395 | my $first = 1;
|
---|
[8716] | 396 | foreach my $dirname (split ("/", $dir)) {
|
---|
[4] | 397 | $dirsofar .= "/" unless $first;
|
---|
| 398 | $first = 0;
|
---|
| 399 |
|
---|
| 400 | $dirsofar .= $dirname;
|
---|
| 401 |
|
---|
| 402 | next if $dirname =~ /^(|[a-z]:)$/i;
|
---|
[836] | 403 | if (!-e $dirsofar)
|
---|
| 404 | {
|
---|
| 405 | my $store_umask = umask(0002);
|
---|
| 406 | my $mkdir_ok = mkdir ($dirsofar, 0777);
|
---|
| 407 | umask($store_umask);
|
---|
| 408 | if (!$mkdir_ok)
|
---|
| 409 | {
|
---|
| 410 | print STDERR "util::mk_all_dir could not create directory $dirsofar\n";
|
---|
| 411 | return;
|
---|
| 412 | }
|
---|
| 413 | }
|
---|
[4] | 414 | }
|
---|
| 415 | }
|
---|
| 416 |
|
---|
[619] | 417 | # make hard link to file if supported by OS, otherwise copy the file
|
---|
| 418 | sub hard_link {
|
---|
[18463] | 419 | my ($src, $dest, $verbosity) = @_;
|
---|
[4] | 420 |
|
---|
[619] | 421 | # remove trailing slashes from source and destination files
|
---|
| 422 | $src =~ s/[\\\/]+$//;
|
---|
| 423 | $dest =~ s/[\\\/]+$//;
|
---|
| 424 |
|
---|
| 425 | # a few sanity checks
|
---|
[812] | 426 | if (-e $dest) {
|
---|
| 427 | # destination file already exists
|
---|
| 428 | return;
|
---|
| 429 | }
|
---|
| 430 | elsif (!-e $src) {
|
---|
[619] | 431 | print STDERR "util::hard_link source file $src does not exist\n";
|
---|
[3628] | 432 | return 1;
|
---|
[619] | 433 | }
|
---|
| 434 | elsif (-d $src) {
|
---|
| 435 | print STDERR "util::hard_link source $src is a directory\n";
|
---|
[3628] | 436 | return 1;
|
---|
[619] | 437 | }
|
---|
| 438 |
|
---|
| 439 | my $dest_dir = &File::Basename::dirname($dest);
|
---|
| 440 | mk_all_dir($dest_dir) if (!-e $dest_dir);
|
---|
| 441 |
|
---|
[14365] | 442 |
|
---|
[22119] | 443 | if (!link($src, $dest)) {
|
---|
[18463] | 444 | if ((!defined $verbosity) || ($verbosity>2)) {
|
---|
| 445 | print STDERR "util::hard_link: unable to create hard link. ";
|
---|
| 446 | print STDERR " Copying file: $src -> $dest\n";
|
---|
| 447 | }
|
---|
[14365] | 448 | &File::Copy::copy ($src, $dest);
|
---|
[619] | 449 | }
|
---|
[3628] | 450 | return 0;
|
---|
[619] | 451 | }
|
---|
| 452 |
|
---|
[2193] | 453 | # make soft link to file if supported by OS, otherwise copy file
|
---|
[721] | 454 | sub soft_link {
|
---|
[15165] | 455 | my ($src, $dest, $ensure_paths_absolute) = @_;
|
---|
[619] | 456 |
|
---|
[721] | 457 | # remove trailing slashes from source and destination files
|
---|
| 458 | $src =~ s/[\\\/]+$//;
|
---|
| 459 | $dest =~ s/[\\\/]+$//;
|
---|
[619] | 460 |
|
---|
[15165] | 461 | # Ensure file paths are absolute IF requested to do so
|
---|
| 462 | # Soft_linking didn't work for relative paths
|
---|
| 463 | if(defined $ensure_paths_absolute && $ensure_paths_absolute) {
|
---|
| 464 | # We need to ensure that the src file is the absolute path
|
---|
| 465 | # See http://perldoc.perl.org/File/Spec.html
|
---|
| 466 | if(!File::Spec->file_name_is_absolute( $src )) { # it's relative
|
---|
| 467 | $src = File::Spec->rel2abs($src); # make absolute
|
---|
| 468 | }
|
---|
| 469 | # Might as well ensure that the destination file's absolute path is used
|
---|
| 470 | if(!File::Spec->file_name_is_absolute( $dest )) {
|
---|
| 471 | $dest = File::Spec->rel2abs($dest); # make absolute
|
---|
| 472 | }
|
---|
| 473 | }
|
---|
| 474 |
|
---|
[721] | 475 | # a few sanity checks
|
---|
| 476 | if (!-e $src) {
|
---|
| 477 | print STDERR "util::soft_link source file $src does not exist\n";
|
---|
| 478 | return 0;
|
---|
| 479 | }
|
---|
[619] | 480 |
|
---|
[721] | 481 | my $dest_dir = &File::Basename::dirname($dest);
|
---|
| 482 | mk_all_dir($dest_dir) if (!-e $dest_dir);
|
---|
[14365] | 483 |
|
---|
[2193] | 484 | if ($ENV{'GSDLOS'} =~ /^windows$/i) {
|
---|
[14365] | 485 | # symlink not supported on windows
|
---|
| 486 | &File::Copy::copy ($src, $dest);
|
---|
[2193] | 487 |
|
---|
| 488 | } elsif (!eval {symlink($src, $dest)}) {
|
---|
[2974] | 489 | print STDERR "util::soft_link: unable to create soft link.\n";
|
---|
[721] | 490 | return 0;
|
---|
| 491 | }
|
---|
| 492 |
|
---|
| 493 | return 1;
|
---|
| 494 | }
|
---|
| 495 |
|
---|
| 496 |
|
---|
| 497 |
|
---|
| 498 |
|
---|
[4] | 499 | # updates a copy of a directory in some other part of the filesystem
|
---|
| 500 | # verbosity settings are: 0=low, 1=normal, 2=high
|
---|
| 501 | # both $fromdir and $todir should be absolute paths
|
---|
| 502 | sub cachedir {
|
---|
| 503 | my ($fromdir, $todir, $verbosity) = @_;
|
---|
| 504 | $verbosity = 1 unless defined $verbosity;
|
---|
| 505 |
|
---|
| 506 | # use / for the directory separator, remove duplicate and
|
---|
| 507 | # trailing slashes
|
---|
| 508 | $fromdir=~s/[\\\/]+/\//g;
|
---|
| 509 | $fromdir=~s/[\\\/]+$//;
|
---|
| 510 | $todir=~s/[\\\/]+/\//g;
|
---|
| 511 | $todir=~s/[\\\/]+$//;
|
---|
| 512 |
|
---|
| 513 | &mk_all_dir ($todir);
|
---|
| 514 |
|
---|
| 515 | # get the directories in ascending order
|
---|
| 516 | if (!opendir (FROMDIR, $fromdir)) {
|
---|
| 517 | print STDERR "util::cachedir could not read directory $fromdir\n";
|
---|
| 518 | return;
|
---|
| 519 | }
|
---|
| 520 | my @fromdir = grep (!/^\.\.?$/, sort(readdir (FROMDIR)));
|
---|
| 521 | closedir (FROMDIR);
|
---|
| 522 |
|
---|
| 523 | if (!opendir (TODIR, $todir)) {
|
---|
| 524 | print STDERR "util::cacedir could not read directory $todir\n";
|
---|
| 525 | return;
|
---|
| 526 | }
|
---|
| 527 | my @todir = grep (!/^\.\.?$/, sort(readdir (TODIR)));
|
---|
| 528 | closedir (TODIR);
|
---|
| 529 |
|
---|
| 530 | my $fromi = 0;
|
---|
| 531 | my $toi = 0;
|
---|
| 532 |
|
---|
| 533 | while ($fromi < scalar(@fromdir) || $toi < scalar(@todir)) {
|
---|
| 534 | # print "fromi: $fromi toi: $toi\n";
|
---|
| 535 |
|
---|
| 536 | # see if we should delete a file/directory
|
---|
| 537 | # this should happen if the file/directory
|
---|
| 538 | # is not in the from list or if its a different
|
---|
| 539 | # size, or has an older timestamp
|
---|
| 540 | if ($toi < scalar(@todir)) {
|
---|
| 541 | if (($fromi >= scalar(@fromdir)) ||
|
---|
| 542 | ($todir[$toi] lt $fromdir[$fromi] ||
|
---|
| 543 | ($todir[$toi] eq $fromdir[$fromi] &&
|
---|
| 544 | &differentfiles("$fromdir/$fromdir[$fromi]","$todir/$todir[$toi]",
|
---|
| 545 | $verbosity)))) {
|
---|
| 546 |
|
---|
| 547 | # the files are different
|
---|
| 548 | &rm_r("$todir/$todir[$toi]");
|
---|
| 549 | splice(@todir, $toi, 1); # $toi stays the same
|
---|
| 550 |
|
---|
| 551 | } elsif ($todir[$toi] eq $fromdir[$fromi]) {
|
---|
| 552 | # the files are the same
|
---|
| 553 | # if it is a directory, check its contents
|
---|
| 554 | if (-d "$todir/$todir[$toi]") {
|
---|
| 555 | &cachedir ("$fromdir/$fromdir[$fromi]",
|
---|
| 556 | "$todir/$todir[$toi]", $verbosity);
|
---|
| 557 | }
|
---|
| 558 |
|
---|
| 559 | $toi++;
|
---|
| 560 | $fromi++;
|
---|
| 561 | next;
|
---|
| 562 | }
|
---|
| 563 | }
|
---|
| 564 |
|
---|
| 565 | # see if we should insert a file/directory
|
---|
| 566 | # we should insert a file/directory if there
|
---|
| 567 | # is no tofiles left or if the tofile does not exist
|
---|
| 568 | if ($fromi < scalar(@fromdir) && ($toi >= scalar(@todir) ||
|
---|
| 569 | $todir[$toi] gt $fromdir[$fromi])) {
|
---|
| 570 | &cp_r ("$fromdir/$fromdir[$fromi]", "$todir/$fromdir[$fromi]");
|
---|
| 571 | splice (@todir, $toi, 0, $fromdir[$fromi]);
|
---|
| 572 |
|
---|
| 573 | $toi++;
|
---|
| 574 | $fromi++;
|
---|
| 575 | }
|
---|
| 576 | }
|
---|
| 577 | }
|
---|
| 578 |
|
---|
| 579 | # this function returns -1 if either file is not found
|
---|
| 580 | # assumes that $file1 and $file2 are absolute file names or
|
---|
| 581 | # in the current directory
|
---|
| 582 | # $file2 is allowed to be newer than $file1
|
---|
| 583 | sub differentfiles {
|
---|
| 584 | my ($file1, $file2, $verbosity) = @_;
|
---|
| 585 | $verbosity = 1 unless defined $verbosity;
|
---|
| 586 |
|
---|
| 587 | $file1 =~ s/\/+$//;
|
---|
| 588 | $file2 =~ s/\/+$//;
|
---|
| 589 |
|
---|
| 590 | my ($file1name) = $file1 =~ /\/([^\/]*)$/;
|
---|
| 591 | my ($file2name) = $file2 =~ /\/([^\/]*)$/;
|
---|
| 592 |
|
---|
| 593 | return -1 unless (-e $file1 && -e $file2);
|
---|
| 594 | if ($file1name ne $file2name) {
|
---|
| 595 | print STDERR "filenames are not the same\n" if ($verbosity >= 2);
|
---|
| 596 | return 1;
|
---|
| 597 | }
|
---|
| 598 |
|
---|
[8716] | 599 | my @file1stat = stat ($file1);
|
---|
| 600 | my @file2stat = stat ($file2);
|
---|
[4] | 601 |
|
---|
| 602 | if (-d $file1) {
|
---|
| 603 | if (! -d $file2) {
|
---|
| 604 | print STDERR "one file is a directory\n" if ($verbosity >= 2);
|
---|
| 605 | return 1;
|
---|
| 606 | }
|
---|
| 607 | return 0;
|
---|
| 608 | }
|
---|
| 609 |
|
---|
| 610 | # both must be regular files
|
---|
| 611 | unless (-f $file1 && -f $file2) {
|
---|
| 612 | print STDERR "one file is not a regular file\n" if ($verbosity >= 2);
|
---|
| 613 | return 1;
|
---|
| 614 | }
|
---|
| 615 |
|
---|
| 616 | # the size of the files must be the same
|
---|
| 617 | if ($file1stat[7] != $file2stat[7]) {
|
---|
| 618 | print STDERR "different sized files\n" if ($verbosity >= 2);
|
---|
| 619 | return 1;
|
---|
| 620 | }
|
---|
| 621 |
|
---|
| 622 | # the second file cannot be older than the first
|
---|
| 623 | if ($file1stat[9] > $file2stat[9]) {
|
---|
| 624 | print STDERR "file is older\n" if ($verbosity >= 2);
|
---|
| 625 | return 1;
|
---|
| 626 | }
|
---|
| 627 |
|
---|
| 628 | return 0;
|
---|
| 629 | }
|
---|
| 630 |
|
---|
| 631 |
|
---|
[16266] | 632 | sub get_tmp_filename
|
---|
| 633 | {
|
---|
| 634 | my $file_ext = shift(@_) || undef;
|
---|
| 635 |
|
---|
[22438] | 636 | my $opt_dot_file_ext = "";
|
---|
| 637 | if (defined $file_ext) {
|
---|
| 638 | if ($file_ext !~ m/\./) {
|
---|
| 639 | # no dot, so needs one added in at start
|
---|
| 640 | $opt_dot_file_ext = ".$file_ext"
|
---|
| 641 | }
|
---|
| 642 | else {
|
---|
| 643 | # allow for "extensions" such as _metadata.txt to be handled
|
---|
| 644 | # gracefully
|
---|
| 645 | $opt_dot_file_ext = $file_ext;
|
---|
| 646 | }
|
---|
| 647 | }
|
---|
[16266] | 648 |
|
---|
[2795] | 649 | my $tmpdir = filename_cat($ENV{'GSDLHOME'}, "tmp");
|
---|
[4] | 650 | &mk_all_dir ($tmpdir) unless -e $tmpdir;
|
---|
| 651 |
|
---|
| 652 | my $count = 1000;
|
---|
| 653 | my $rand = int(rand $count);
|
---|
[16266] | 654 | my $full_tmp_filename = &filename_cat($tmpdir, "F$rand$opt_dot_file_ext");
|
---|
| 655 |
|
---|
| 656 | while (-e $full_tmp_filename) {
|
---|
[4] | 657 | $rand = int(rand $count);
|
---|
[16266] | 658 | $full_tmp_filename = &filename_cat($tmpdir, "F$rand$opt_dot_file_ext");
|
---|
[4] | 659 | $count++;
|
---|
| 660 | }
|
---|
[16266] | 661 |
|
---|
| 662 | return $full_tmp_filename;
|
---|
[4] | 663 | }
|
---|
| 664 |
|
---|
[21218] | 665 | sub get_toplevel_tmp_dir
|
---|
| 666 | {
|
---|
| 667 | return filename_cat($ENV{'GSDLHOME'}, "tmp");
|
---|
| 668 | }
|
---|
| 669 |
|
---|
| 670 |
|
---|
[17512] | 671 | sub filename_to_regex {
|
---|
| 672 | my $filename = shift (@_);
|
---|
[4] | 673 |
|
---|
[17512] | 674 | # need to put single backslash back to double so that regex works
|
---|
| 675 | if ($ENV{'GSDLOS'} =~ /^windows$/i) {
|
---|
| 676 | $filename =~ s/\\/\\\\/g;
|
---|
| 677 | }
|
---|
| 678 | return $filename;
|
---|
| 679 | }
|
---|
| 680 |
|
---|
[4] | 681 | sub filename_cat {
|
---|
[7507] | 682 | my $first_file = shift(@_);
|
---|
[4] | 683 | my (@filenames) = @_;
|
---|
[10146] | 684 |
|
---|
[16266] | 685 | # Useful for debugging
|
---|
| 686 | # -- might make sense to call caller(0) rather than (1)??
|
---|
| 687 | # my ($cpackage,$cfilename,$cline,$csubr,$chas_args,$cwantarray) = caller(1);
|
---|
[15113] | 688 | # print STDERR "Calling method; $cfilename:$cline $cpackage->$csubr\n";
|
---|
[18913] | 689 |
|
---|
| 690 | # If first_file is not null or empty, then add it back into the list
|
---|
| 691 | if (defined $first_file && $first_file =~ /\S/) {
|
---|
[7507] | 692 | unshift(@filenames, $first_file);
|
---|
| 693 | }
|
---|
| 694 |
|
---|
[4] | 695 | my $filename = join("/", @filenames);
|
---|
| 696 |
|
---|
| 697 | # remove duplicate slashes and remove the last slash
|
---|
[488] | 698 | if ($ENV{'GSDLOS'} =~ /^windows$/i) {
|
---|
| 699 | $filename =~ s/[\\\/]+/\\/g;
|
---|
| 700 | } else {
|
---|
[836] | 701 | $filename =~ s/[\/]+/\//g;
|
---|
| 702 | # DB: want a filename abc\de.html to remain like this
|
---|
[488] | 703 | }
|
---|
| 704 | $filename =~ s/[\\\/]$//;
|
---|
[4] | 705 |
|
---|
| 706 | return $filename;
|
---|
| 707 | }
|
---|
| 708 |
|
---|
[21413] | 709 |
|
---|
| 710 | sub pathname_cat {
|
---|
| 711 | my $first_path = shift(@_);
|
---|
| 712 | my (@pathnames) = @_;
|
---|
| 713 |
|
---|
| 714 | # If first_path is not null or empty, then add it back into the list
|
---|
| 715 | if (defined $first_path && $first_path =~ /\S/) {
|
---|
| 716 | unshift(@pathnames, $first_path);
|
---|
| 717 | }
|
---|
| 718 |
|
---|
[21425] | 719 | my $join_char;
|
---|
[21413] | 720 | if ($ENV{'GSDLOS'} =~ /^windows$/i) {
|
---|
| 721 | $join_char = ";";
|
---|
| 722 | } else {
|
---|
| 723 | $join_char = ":";
|
---|
| 724 | }
|
---|
| 725 |
|
---|
| 726 | my $pathname = join($join_char, @pathnames);
|
---|
| 727 |
|
---|
| 728 | # remove duplicate slashes
|
---|
| 729 | if ($ENV{'GSDLOS'} =~ /^windows$/i) {
|
---|
| 730 | $pathname =~ s/[\\\/]+/\\/g;
|
---|
| 731 | } else {
|
---|
| 732 | $pathname =~ s/[\/]+/\//g;
|
---|
| 733 | # DB: want a pathname abc\de.html to remain like this
|
---|
| 734 | }
|
---|
| 735 |
|
---|
| 736 | return $pathname;
|
---|
| 737 | }
|
---|
| 738 |
|
---|
| 739 |
|
---|
[19616] | 740 | sub tidy_up_oid {
|
---|
| 741 | my ($OID) = @_;
|
---|
| 742 | if ($OID =~ /\./) {
|
---|
| 743 | print STDERR "Warning, identifier $OID contains periods (.), removing them\n";
|
---|
| 744 | $OID =~ s/\.//g; #remove any periods
|
---|
| 745 | }
|
---|
| 746 | if ($OID =~ /^\s.*\s$/) {
|
---|
| 747 | print STDERR "Warning, identifier $OID starts or ends with whitespace. Removing it\n";
|
---|
| 748 | # remove starting and trailing whitespace
|
---|
| 749 | $OID =~ s/^\s+//;
|
---|
| 750 | $OID =~ s/\s+$//;
|
---|
| 751 | }
|
---|
| 752 | if ($OID =~ /^[\d]*$/) {
|
---|
| 753 | print STDERR "Warning, identifier $OID contains only digits. Prepending 'D'.\n";
|
---|
| 754 | $OID = "D" . $OID;
|
---|
| 755 | }
|
---|
| 756 |
|
---|
| 757 | return $OID;
|
---|
| 758 | }
|
---|
[10212] | 759 | sub envvar_prepend {
|
---|
| 760 | my ($var,$val) = @_;
|
---|
| 761 |
|
---|
[16404] | 762 | # do not prepend any value/path that's already in the environment variable
|
---|
[16442] | 763 | if ($ENV{'GSDLOS'} =~ /^windows$/i)
|
---|
| 764 | {
|
---|
| 765 | my $escaped_val = $val;
|
---|
| 766 | $escaped_val =~ s/\\/\\\\/g; # escape any Windows backslashes for upcoming regex
|
---|
[22386] | 767 | if (!defined($ENV{$var})) {
|
---|
| 768 | $ENV{$var} = "$val";
|
---|
| 769 | }
|
---|
| 770 | elsif($ENV{$var} !~ m/$escaped_val/) {
|
---|
[16442] | 771 | $ENV{$var} = "$val;".$ENV{$var};
|
---|
[16404] | 772 | }
|
---|
[16442] | 773 | }
|
---|
| 774 | else {
|
---|
[22386] | 775 | if (!defined($ENV{$var})) {
|
---|
| 776 | $ENV{$var} = "$val";
|
---|
| 777 | }
|
---|
| 778 | elsif($ENV{$var} !~ m/$val/) {
|
---|
[16442] | 779 | $ENV{$var} = "$val:".$ENV{$var};
|
---|
[16404] | 780 | }
|
---|
[10212] | 781 | }
|
---|
| 782 | }
|
---|
| 783 |
|
---|
| 784 | sub envvar_append {
|
---|
| 785 | my ($var,$val) = @_;
|
---|
| 786 |
|
---|
[16404] | 787 | # do not append any value/path that's already in the environment variable
|
---|
[16442] | 788 | if ($ENV{'GSDLOS'} =~ /^windows$/i)
|
---|
| 789 | {
|
---|
| 790 | my $escaped_val = $val;
|
---|
| 791 | $escaped_val =~ s/\\/\\\\/g; # escape any Windows backslashes for upcoming regex
|
---|
[22386] | 792 | if (!defined($ENV{$var})) {
|
---|
| 793 | $ENV{$var} = "$val";
|
---|
| 794 | }
|
---|
| 795 | elsif($ENV{$var} !~ m/$escaped_val/) {
|
---|
[16404] | 796 | $ENV{$var} .= ";$val";
|
---|
| 797 | }
|
---|
[16442] | 798 | }
|
---|
| 799 | else {
|
---|
[22386] | 800 | if (!defined($ENV{$var})) {
|
---|
| 801 | $ENV{$var} = "$val";
|
---|
| 802 | }
|
---|
| 803 | elsif($ENV{$var} !~ m/$val/) {
|
---|
[16404] | 804 | $ENV{$var} .= ":$val";
|
---|
| 805 | }
|
---|
[16442] | 806 | }
|
---|
[10212] | 807 | }
|
---|
| 808 |
|
---|
[16442] | 809 |
|
---|
[16380] | 810 | # splits a filename into a prefix and a tail extension using the tail_re, or
|
---|
| 811 | # if that fails, splits on the file_extension . (dot)
|
---|
| 812 | sub get_prefix_and_tail_by_regex {
|
---|
[10212] | 813 |
|
---|
[16380] | 814 | my ($filename,$tail_re) = @_;
|
---|
| 815 |
|
---|
| 816 | my ($file_prefix,$file_ext) = ($filename =~ m/^(.*?)($tail_re)$/);
|
---|
| 817 | if ((!defined $file_prefix) || (!defined $file_ext)) {
|
---|
| 818 | ($file_prefix,$file_ext) = ($filename =~ m/^(.*)(\..*?)$/);
|
---|
| 819 | }
|
---|
| 820 |
|
---|
| 821 | return ($file_prefix,$file_ext);
|
---|
| 822 | }
|
---|
| 823 |
|
---|
| 824 | # get full path and file only path from a base_dir (which may be empty) and
|
---|
| 825 | # file (which may contain directories)
|
---|
| 826 | sub get_full_filenames {
|
---|
| 827 | my ($base_dir, $file) = @_;
|
---|
| 828 |
|
---|
| 829 | my $filename_full_path = $file;
|
---|
| 830 | # add on directory if present
|
---|
| 831 | $filename_full_path = &util::filename_cat ($base_dir, $file) if $base_dir =~ /\S/;
|
---|
| 832 |
|
---|
| 833 | my $filename_no_path = $file;
|
---|
| 834 |
|
---|
| 835 | # remove directory if present
|
---|
| 836 | $filename_no_path =~ s/^.*[\/\\]//;
|
---|
| 837 | return ($filename_full_path, $filename_no_path);
|
---|
| 838 | }
|
---|
| 839 |
|
---|
[8682] | 840 | # returns the path of a file without the filename -- ie. the directory the file is in
|
---|
| 841 | sub filename_head {
|
---|
| 842 | my $filename = shift(@_);
|
---|
| 843 |
|
---|
| 844 | if ($ENV{'GSDLOS'} =~ /^windows$/i) {
|
---|
| 845 | $filename =~ s/[^\\\\]*$//;
|
---|
| 846 | }
|
---|
| 847 | else {
|
---|
| 848 | $filename =~ s/[^\\\/]*$//;
|
---|
| 849 | }
|
---|
| 850 |
|
---|
| 851 | return $filename;
|
---|
| 852 | }
|
---|
| 853 |
|
---|
| 854 |
|
---|
[1454] | 855 | # returns 1 if filename1 and filename2 point to the same
|
---|
| 856 | # file or directory
|
---|
| 857 | sub filenames_equal {
|
---|
| 858 | my ($filename1, $filename2) = @_;
|
---|
| 859 |
|
---|
| 860 | # use filename_cat to clean up trailing slashes and
|
---|
| 861 | # multiple slashes
|
---|
| 862 | $filename1 = filename_cat ($filename1);
|
---|
[2516] | 863 | $filename2 = filename_cat ($filename2);
|
---|
[1454] | 864 |
|
---|
| 865 | # filenames not case sensitive on windows
|
---|
| 866 | if ($ENV{'GSDLOS'} =~ /^windows$/i) {
|
---|
| 867 | $filename1 =~ tr/[A-Z]/[a-z]/;
|
---|
| 868 | $filename2 =~ tr/[A-Z]/[a-z]/;
|
---|
| 869 | }
|
---|
| 870 | return 1 if $filename1 eq $filename2;
|
---|
| 871 | return 0;
|
---|
| 872 | }
|
---|
| 873 |
|
---|
[10281] | 874 | sub filename_within_collection
|
---|
| 875 | {
|
---|
| 876 | my ($filename) = @_;
|
---|
| 877 |
|
---|
| 878 | my $collect_dir = $ENV{'GSDLCOLLECTDIR'};
|
---|
| 879 |
|
---|
| 880 | if (defined $collect_dir) {
|
---|
| 881 | my $dirsep = &util::get_dirsep();
|
---|
| 882 | if ($collect_dir !~ m/$dirsep$/) {
|
---|
| 883 | $collect_dir .= $dirsep;
|
---|
| 884 | }
|
---|
| 885 |
|
---|
| 886 | $collect_dir =~ s/\\/\\\\/g; # escape DOS style file separator
|
---|
| 887 |
|
---|
[15875] | 888 | # if from within GSDLCOLLECTDIR, then remove directory prefix
|
---|
| 889 | # so source_filename is realative to it. This is done to aid
|
---|
| 890 | # portability, i.e. the collection can be moved to somewhere
|
---|
| 891 | # else on the file system and the archives directory will still
|
---|
| 892 | # work. This is needed, for example in the applet version of
|
---|
| 893 | # GLI where GSDLHOME/collect on the server will be different to
|
---|
| 894 | # the collect directory of the remove user. Of course,
|
---|
| 895 | # GSDLCOLLECTDIR subsequently needs to be put back on to turn
|
---|
| 896 | # it back into a full pathname.
|
---|
| 897 |
|
---|
[10281] | 898 | if ($filename =~ /^$collect_dir(.*)$/) {
|
---|
| 899 | $filename = $1;
|
---|
| 900 | }
|
---|
| 901 | }
|
---|
| 902 |
|
---|
| 903 | return $filename;
|
---|
| 904 | }
|
---|
| 905 |
|
---|
[18441] | 906 | sub filename_is_absolute
|
---|
| 907 | {
|
---|
| 908 | my ($filename) = @_;
|
---|
| 909 |
|
---|
| 910 | if ($ENV{'GSDLOS'} =~ /^windows$/i) {
|
---|
| 911 | return ($filename =~ m/^(\w:)?\\/);
|
---|
| 912 | }
|
---|
| 913 | else {
|
---|
| 914 | return ($filename =~ m/^\//);
|
---|
| 915 | }
|
---|
| 916 | }
|
---|
| 917 |
|
---|
| 918 |
|
---|
[17572] | 919 | ## @method make_absolute()
|
---|
| 920 | #
|
---|
| 921 | # Ensure the given file path is absolute in respect to the given base path.
|
---|
| 922 | #
|
---|
| 923 | # @param $base_dir A string denoting the base path the given dir must be
|
---|
| 924 | # absolute to.
|
---|
| 925 | # @param $dir The directory to be made absolute as a string. Note that the
|
---|
| 926 | # dir may already be absolute, in which case it will remain
|
---|
| 927 | # unchanged.
|
---|
| 928 | # @return The now absolute form of the directory as a string.
|
---|
| 929 | #
|
---|
| 930 | # @author John Thompson, DL Consulting Ltd.
|
---|
| 931 | # @copy 2006 DL Consulting Ltd.
|
---|
| 932 | #
|
---|
| 933 | #used in buildcol.pl, doesn't work for all cases --kjdon
|
---|
| 934 | sub make_absolute {
|
---|
| 935 |
|
---|
| 936 | my ($base_dir, $dir) = @_;
|
---|
[18441] | 937 | ### print STDERR "dir = $dir\n";
|
---|
[17572] | 938 | $dir =~ s/[\\\/]+/\//g;
|
---|
| 939 | $dir = $base_dir . "/$dir" unless ($dir =~ m|^(\w:)?/|);
|
---|
| 940 | $dir =~ s|^/tmp_mnt||;
|
---|
| 941 | 1 while($dir =~ s|/[^/]*/\.\./|/|g);
|
---|
| 942 | $dir =~ s|/[.][.]?/|/|g;
|
---|
| 943 | $dir =~ tr|/|/|s;
|
---|
[18441] | 944 | ### print STDERR "dir = $dir\n";
|
---|
[17572] | 945 |
|
---|
| 946 | return $dir;
|
---|
| 947 | }
|
---|
| 948 | ## make_absolute() ##
|
---|
[10281] | 949 |
|
---|
[7929] | 950 | sub get_dirsep {
|
---|
| 951 |
|
---|
| 952 | if ($ENV{'GSDLOS'} =~ /^windows$/i) {
|
---|
| 953 | return "\\";
|
---|
| 954 | } else {
|
---|
| 955 | return "\/";
|
---|
| 956 | }
|
---|
| 957 | }
|
---|
| 958 |
|
---|
[619] | 959 | sub get_os_dirsep {
|
---|
[4] | 960 |
|
---|
[619] | 961 | if ($ENV{'GSDLOS'} =~ /^windows$/i) {
|
---|
| 962 | return "\\\\";
|
---|
| 963 | } else {
|
---|
| 964 | return "\\\/";
|
---|
| 965 | }
|
---|
| 966 | }
|
---|
| 967 |
|
---|
| 968 | sub get_re_dirsep {
|
---|
| 969 |
|
---|
| 970 | return "\\\\|\\\/";
|
---|
| 971 | }
|
---|
| 972 |
|
---|
| 973 |
|
---|
[15003] | 974 | sub get_dirsep_tail {
|
---|
| 975 | my ($filename) = @_;
|
---|
| 976 |
|
---|
| 977 | # returns last part of directory or filename
|
---|
| 978 | # On unix e.g. a/b.d => b.d
|
---|
| 979 | # a/b/c => c
|
---|
| 980 |
|
---|
[15088] | 981 | my $dirsep = get_re_dirsep();
|
---|
| 982 | my @dirs = split (/$dirsep/, $filename);
|
---|
| 983 | my $tail = pop @dirs;
|
---|
[15003] | 984 |
|
---|
[15088] | 985 | # - caused problems under windows
|
---|
| 986 | #my ($tail) = ($filename =~ m/^(?:.*?$dirsep)?(.*?)$/);
|
---|
| 987 |
|
---|
[15003] | 988 | return $tail;
|
---|
| 989 | }
|
---|
| 990 |
|
---|
| 991 |
|
---|
[4] | 992 | # if this is running on windows we want binaries to end in
|
---|
| 993 | # .exe, otherwise they don't have to end in any extension
|
---|
| 994 | sub get_os_exe {
|
---|
| 995 | return ".exe" if $ENV{'GSDLOS'} =~ /^windows$/i;
|
---|
| 996 | return "";
|
---|
| 997 | }
|
---|
| 998 |
|
---|
| 999 |
|
---|
[86] | 1000 | # test to see whether this is a big or little endian machine
|
---|
[15713] | 1001 | sub is_little_endian
|
---|
| 1002 | {
|
---|
| 1003 | # To determine the name of the operating system, the variable $^O is a cheap alternative to pulling it out of the Config module;
|
---|
| 1004 | # If it is a Macintosh machine (i.e. the Darwin operating system), regardless if it's running on the IBM power-pc cpu or the x86 Intel-based chip with a power-pc emulator running on top of it, it's big-endian
|
---|
| 1005 | # Otherwise, it's little endian
|
---|
| 1006 |
|
---|
| 1007 | #return 0 if $^O =~ /^darwin$/i;
|
---|
[17714] | 1008 | #return 0 if $ENV{'GSDLOS'} =~ /^darwin$/i;
|
---|
| 1009 |
|
---|
| 1010 | # Going back to stating exactly whether the machine is little endian
|
---|
| 1011 | # or big endian, without any special case for Macs. Since for rata it comes
|
---|
| 1012 | # back with little endian and for shuttle with bigendian.
|
---|
[15713] | 1013 | return (ord(substr(pack("s",1), 0, 1)) == 1);
|
---|
[86] | 1014 | }
|
---|
[4] | 1015 |
|
---|
[86] | 1016 |
|
---|
[135] | 1017 | # will return the collection name if successful, "" otherwise
|
---|
| 1018 | sub use_collection {
|
---|
[1454] | 1019 | my ($collection, $collectdir) = @_;
|
---|
[135] | 1020 |
|
---|
[1454] | 1021 | if (!defined $collectdir || $collectdir eq "") {
|
---|
| 1022 | $collectdir = &filename_cat ($ENV{'GSDLHOME'}, "collect");
|
---|
| 1023 | }
|
---|
| 1024 |
|
---|
[135] | 1025 | # get and check the collection
|
---|
| 1026 | if (!defined($collection) || $collection eq "") {
|
---|
| 1027 | if (defined $ENV{'GSDLCOLLECTION'}) {
|
---|
| 1028 | $collection = $ENV{'GSDLCOLLECTION'};
|
---|
| 1029 | } else {
|
---|
[2359] | 1030 | print STDOUT "No collection specified\n";
|
---|
[135] | 1031 | return "";
|
---|
| 1032 | }
|
---|
| 1033 | }
|
---|
| 1034 |
|
---|
| 1035 | if ($collection eq "modelcol") {
|
---|
[2359] | 1036 | print STDOUT "You can't use modelcol.\n";
|
---|
[135] | 1037 | return "";
|
---|
| 1038 | }
|
---|
| 1039 |
|
---|
| 1040 | # make sure the environment variables GSDLCOLLECTION and GSDLCOLLECTDIR
|
---|
| 1041 | # are defined
|
---|
[17204] | 1042 | $ENV{'GSDLCOLLECTION'} = $collection;
|
---|
[1454] | 1043 | $ENV{'GSDLCOLLECTDIR'} = &filename_cat ($collectdir, $collection);
|
---|
[135] | 1044 |
|
---|
| 1045 | # make sure this collection exists
|
---|
| 1046 | if (!-e $ENV{'GSDLCOLLECTDIR'}) {
|
---|
[2359] | 1047 | print STDOUT "Invalid collection ($collection).\n";
|
---|
[135] | 1048 | return "";
|
---|
| 1049 | }
|
---|
| 1050 |
|
---|
| 1051 | # everything is ready to go
|
---|
| 1052 | return $collection;
|
---|
| 1053 | }
|
---|
| 1054 |
|
---|
[21207] | 1055 | sub get_current_collection_name {
|
---|
| 1056 | return $ENV{'GSDLCOLLECTION'};
|
---|
| 1057 | }
|
---|
[14926] | 1058 |
|
---|
| 1059 |
|
---|
| 1060 | # will return the collection name if successful, "" otherwise.
|
---|
| 1061 | # Like use_collection (above) but for greenstone 3 (taking account of site level)
|
---|
| 1062 |
|
---|
| 1063 | sub use_site_collection {
|
---|
| 1064 | my ($site, $collection, $collectdir) = @_;
|
---|
| 1065 |
|
---|
| 1066 | if (!defined $collectdir || $collectdir eq "") {
|
---|
| 1067 | die "GSDL3HOME not set.\n" unless defined $ENV{'GSDL3HOME'};
|
---|
| 1068 | $collectdir = &filename_cat ($ENV{'GSDL3HOME'}, "sites", $site, "collect");
|
---|
| 1069 | }
|
---|
| 1070 |
|
---|
| 1071 | # collectdir explicitly set by this point (using $site variable if required).
|
---|
| 1072 | # Can call "old" gsdl2 use_collection now.
|
---|
| 1073 |
|
---|
| 1074 | return use_collection($collection,$collectdir);
|
---|
| 1075 | }
|
---|
| 1076 |
|
---|
| 1077 |
|
---|
| 1078 |
|
---|
[15018] | 1079 | sub locate_config_file
|
---|
| 1080 | {
|
---|
| 1081 | my ($file) = @_;
|
---|
| 1082 |
|
---|
| 1083 | my $locations = locate_config_files($file);
|
---|
| 1084 |
|
---|
| 1085 | return shift @$locations; # returns undef if 'locations' is empty
|
---|
| 1086 | }
|
---|
| 1087 |
|
---|
| 1088 |
|
---|
| 1089 | sub locate_config_files
|
---|
| 1090 | {
|
---|
| 1091 | my ($file) = @_;
|
---|
| 1092 |
|
---|
| 1093 | my @locations = ();
|
---|
| 1094 |
|
---|
| 1095 | if (-e $file) {
|
---|
| 1096 | # Clearly specified (most likely full filename)
|
---|
| 1097 | # No need to hunt in 'etc' directories, return value unchanged
|
---|
| 1098 | push(@locations,$file);
|
---|
| 1099 | }
|
---|
| 1100 | else {
|
---|
| 1101 | # Check for collection specific one before looking in global GSDL 'etc'
|
---|
[16969] | 1102 | if (defined $ENV{'GSDLCOLLECTDIR'} && $ENV{'GSDLCOLLECTDIR'} ne "") {
|
---|
| 1103 | my $test_collect_etc_filename
|
---|
| 1104 | = &util::filename_cat($ENV{'GSDLCOLLECTDIR'},"etc", $file);
|
---|
| 1105 |
|
---|
| 1106 | if (-e $test_collect_etc_filename) {
|
---|
| 1107 | push(@locations,$test_collect_etc_filename);
|
---|
| 1108 | }
|
---|
[15018] | 1109 | }
|
---|
| 1110 | my $test_main_etc_filename
|
---|
| 1111 | = &util::filename_cat($ENV{'GSDLHOME'},"etc", $file);
|
---|
| 1112 | if (-e $test_main_etc_filename) {
|
---|
| 1113 | push(@locations,$test_main_etc_filename);
|
---|
| 1114 | }
|
---|
| 1115 | }
|
---|
| 1116 |
|
---|
| 1117 | return \@locations;
|
---|
| 1118 | }
|
---|
| 1119 |
|
---|
| 1120 |
|
---|
[9955] | 1121 | sub hyperlink_text
|
---|
| 1122 | {
|
---|
| 1123 | my ($text) = @_;
|
---|
| 1124 |
|
---|
| 1125 | $text =~ s/(http:\/\/[^\s]+)/<a href=\"$1\">$1<\/a>/mg;
|
---|
| 1126 | $text =~ s/(^|\s+)(www\.(\w|\.)+)/<a href=\"http:\/\/$2\">$2<\/a>/mg;
|
---|
| 1127 |
|
---|
| 1128 | return $text;
|
---|
| 1129 | }
|
---|
| 1130 |
|
---|
| 1131 |
|
---|
[16436] | 1132 | # A method to check if a directory is empty (note that an empty directory still has non-zero size!!!)
|
---|
| 1133 | # Code is from http://episteme.arstechnica.com/eve/forums/a/tpc/f/6330927813/m/436007700831
|
---|
| 1134 | sub is_dir_empty
|
---|
| 1135 | {
|
---|
| 1136 | my ($path) = @_;
|
---|
| 1137 | opendir DIR, $path;
|
---|
| 1138 | while(my $entry = readdir DIR) {
|
---|
| 1139 | next if($entry =~ /^\.\.?$/);
|
---|
| 1140 | closedir DIR;
|
---|
| 1141 | return 0;
|
---|
| 1142 | }
|
---|
| 1143 | closedir DIR;
|
---|
| 1144 | return 1;
|
---|
| 1145 | }
|
---|
| 1146 |
|
---|
[18337] | 1147 | # Returns the given filename converted using either URL encoding or base64
|
---|
| 1148 | # encoding, as specified by $rename_method. If the given filename has no suffix
|
---|
[20413] | 1149 | # (if it is just the tailname), then $no_suffix should be some defined value.
|
---|
| 1150 | # rename_method can be url, none, base64
|
---|
[18319] | 1151 | sub rename_file {
|
---|
[18337] | 1152 | my ($filename, $rename_method, $no_suffix) = @_;
|
---|
[18329] | 1153 |
|
---|
[18337] | 1154 | if(!$filename) { # undefined or empty string
|
---|
[18329] | 1155 | return $filename;
|
---|
| 1156 | }
|
---|
[18319] | 1157 |
|
---|
[20413] | 1158 | if (!$rename_method) {
|
---|
| 1159 | print STDERR "WARNING: no file renaming method specified. Defaulting to using URL encoding...\n";
|
---|
| 1160 | # Debugging information
|
---|
| 1161 | my ($cpackage,$cfilename,$cline,$csubr,$chas_args,$cwantarray) = caller(1);
|
---|
| 1162 | print STDERR "Called from method: $cfilename:$cline $cpackage->$csubr\n";
|
---|
| 1163 | $rename_method = "url";
|
---|
| 1164 | } elsif($rename_method eq "none") {
|
---|
| 1165 | return $filename; # would have already been renamed
|
---|
| 1166 | }
|
---|
| 1167 |
|
---|
[19762] | 1168 | # No longer replace spaces with underscores, since underscores mess with incremental rebuild
|
---|
| 1169 | ### Replace spaces with underscore. Do this first else it can go wrong below when getting tailname
|
---|
| 1170 | ###$filename =~ s/ /_/g;
|
---|
[18337] | 1171 |
|
---|
| 1172 | my ($tailname,$dirname,$suffix);
|
---|
| 1173 | if($no_suffix) { # given a tailname, no suffix
|
---|
| 1174 | ($tailname,$dirname) = File::Basename::fileparse($filename);
|
---|
| 1175 | }
|
---|
| 1176 | else {
|
---|
| 1177 | ($tailname,$dirname,$suffix) = File::Basename::fileparse($filename, "\\.(?:[^\\.]+?)\$");
|
---|
| 1178 | }
|
---|
[18400] | 1179 | $suffix = "" if !$suffix;
|
---|
[18337] | 1180 |
|
---|
[20413] | 1181 | if ($rename_method eq "url") {
|
---|
[18319] | 1182 | $tailname = &unicode::url_encode($tailname);
|
---|
| 1183 | }
|
---|
| 1184 | elsif ($rename_method eq "base64") {
|
---|
[18341] | 1185 | $tailname = &unicode::base64_encode($tailname);
|
---|
[18319] | 1186 | $tailname =~ s/\s*//sg; # for some reason it adds spaces not just at end but also in middle
|
---|
| 1187 | }
|
---|
[18326] | 1188 |
|
---|
[18319] | 1189 | $filename = "$tailname$suffix";
|
---|
[18326] | 1190 | $filename = "$dirname$filename" if ($dirname ne "./" && $dirname ne ".\\");
|
---|
[18319] | 1191 |
|
---|
| 1192 | return $filename;
|
---|
| 1193 | }
|
---|
| 1194 |
|
---|
[21616] | 1195 |
|
---|
| 1196 | # BACKWARDS COMPATIBILITY: Just in case there are old .ldb/.bdb files
|
---|
[21664] | 1197 | sub rename_ldb_or_bdb_file {
|
---|
[18657] | 1198 | my ($filename_no_ext) = @_;
|
---|
| 1199 |
|
---|
| 1200 | my $new_filename = "$filename_no_ext.gdb";
|
---|
[21615] | 1201 | return if (-f $new_filename); # if the file has the right extension, don't need to do anything
|
---|
[18657] | 1202 | # try ldb
|
---|
| 1203 | my $old_filename = "$filename_no_ext.ldb";
|
---|
| 1204 |
|
---|
| 1205 | if (-f $old_filename) {
|
---|
[19056] | 1206 | print STDERR "Renaming $old_filename to $new_filename\n";
|
---|
| 1207 | rename ($old_filename, $new_filename)
|
---|
| 1208 | || print STDERR "Rename failed: $!\n";
|
---|
[18657] | 1209 | return;
|
---|
| 1210 | }
|
---|
| 1211 | # try bdb
|
---|
| 1212 | $old_filename = "$filename_no_ext.bdb";
|
---|
| 1213 | if (-f $old_filename) {
|
---|
[19056] | 1214 | print STDERR "Renaming $old_filename to $new_filename\n";
|
---|
| 1215 | rename ($old_filename, $new_filename)
|
---|
| 1216 | || print STDERR "Rename failed: $!\n";
|
---|
[18657] | 1217 | return;
|
---|
| 1218 | }
|
---|
| 1219 | }
|
---|
| 1220 |
|
---|
| 1221 |
|
---|
[21719] | 1222 | # Returns the greenstone URL prefix extracted from the appropriate GS2/GS3 config file.
|
---|
| 1223 | # By default, /greenstone3 for GS3 or /greenstone for GS2.
|
---|
| 1224 | sub get_greenstone_url_prefix() {
|
---|
| 1225 | # if already set on a previous occasion, just return that
|
---|
| 1226 | # (Don't want to keep repeating this: cost of re-opening and scanning files.)
|
---|
| 1227 | return $ENV{'GREENSTONE_URL_PREFIX'} if($ENV{'GREENSTONE_URL_PREFIX'});
|
---|
[18657] | 1228 |
|
---|
[21719] | 1229 | my ($configfile, $urlprefix, $defaultUrlprefix);
|
---|
| 1230 | my @propertynames = ();
|
---|
| 1231 |
|
---|
| 1232 | if($ENV{'GSDL3SRCHOME'}) {
|
---|
| 1233 | $defaultUrlprefix = "/greenstone3";
|
---|
| 1234 | $configfile = &util::filename_cat($ENV{'GSDL3SRCHOME'}, "packages", "tomcat", "conf", "Catalina", "localhost", "greenstone3.xml");
|
---|
| 1235 | push(@propertynames, qw/path\s*\=/);
|
---|
| 1236 | } else {
|
---|
| 1237 | $defaultUrlprefix = "/greenstone";
|
---|
| 1238 | $configfile = &util::filename_cat($ENV{'GSDLHOME'}, "cgi-bin", "gsdlsite.cfg");
|
---|
| 1239 | push(@propertynames, (qw/\nhttpprefix/, qw/\ngwcgi/)); # inspect one property then the other
|
---|
| 1240 | }
|
---|
| 1241 |
|
---|
| 1242 | $urlprefix = &extract_propvalue_from_file($configfile, \@propertynames);
|
---|
| 1243 |
|
---|
| 1244 | if(!$urlprefix) { # no values found for URL prefix, use default values
|
---|
| 1245 | $urlprefix = $defaultUrlprefix;
|
---|
| 1246 | } else {
|
---|
| 1247 | #gwcgi can contain more than the wanted prefix, we split on / to get the first "directory" level
|
---|
| 1248 | $urlprefix =~ s/^\///; # remove the starting slash
|
---|
| 1249 | my @dirs = split(/(\\|\/)/, $urlprefix);
|
---|
| 1250 | $urlprefix = shift(@dirs);
|
---|
| 1251 |
|
---|
| 1252 | if($urlprefix !~ m/^\//) { # in all cases: ensure the required forward slash is at the front
|
---|
| 1253 | $urlprefix = "/$urlprefix";
|
---|
| 1254 | }
|
---|
| 1255 | }
|
---|
| 1256 |
|
---|
| 1257 | # set for the future
|
---|
| 1258 | $ENV{'GREENSTONE_URL_PREFIX'} = $urlprefix;
|
---|
| 1259 | # print STDERR "*** in get_greenstone_url_prefix(): $urlprefix\n\n";
|
---|
| 1260 | return $urlprefix;
|
---|
| 1261 | }
|
---|
| 1262 |
|
---|
| 1263 |
|
---|
| 1264 | # Given a config file (xml or java properties file) and a list/array of regular expressions
|
---|
| 1265 | # that represent property names to match on, this function will return the value for the 1st
|
---|
| 1266 | # matching property name. If the return value is undefined, no matching property was found.
|
---|
| 1267 | sub extract_propvalue_from_file() {
|
---|
| 1268 | my ($configfile, $propertynames) = @_;
|
---|
| 1269 |
|
---|
| 1270 | my $value;
|
---|
| 1271 | unless(open(FIN, "<$configfile")) {
|
---|
| 1272 | print STDERR "extract_propvalue_from_file(): Unable to open $configfile. $!\n";
|
---|
| 1273 | return $value; # not initialised
|
---|
| 1274 | }
|
---|
| 1275 |
|
---|
| 1276 | # Read the entire file at once, as one single line, then close it
|
---|
| 1277 | my $filecontents;
|
---|
| 1278 | {
|
---|
| 1279 | local $/ = undef;
|
---|
| 1280 | $filecontents = <FIN>;
|
---|
| 1281 | }
|
---|
| 1282 | close(FIN);
|
---|
| 1283 |
|
---|
| 1284 | foreach my $regex (@$propertynames) {
|
---|
| 1285 | ($value) = $filecontents=~ m/$regex\s*(\S*)/s; # read value of the property given by regex up to the 1st space
|
---|
| 1286 | if($value) {
|
---|
| 1287 | $value =~ s/^\"//; # remove any startquotes
|
---|
| 1288 | $value =~ s/\".*$//; # remove the 1st endquotes (if any) followed by any xml
|
---|
| 1289 | last; # found value for a matching property, break from loop
|
---|
| 1290 | }
|
---|
| 1291 | }
|
---|
| 1292 |
|
---|
| 1293 | return $value;
|
---|
| 1294 | }
|
---|
| 1295 |
|
---|
| 1296 |
|
---|
[4] | 1297 | 1;
|
---|