[1031] | 1 | #!/usr/bin/perl -w
|
---|
[4] | 2 |
|
---|
[538] | 3 | ###########################################################################
|
---|
| 4 | #
|
---|
| 5 | # buildcol.pl -- This program will build a particular collection
|
---|
| 6 | # A component of the Greenstone digital library software
|
---|
| 7 | # from the New Zealand Digital Library Project at the
|
---|
| 8 | # University of Waikato, New Zealand.
|
---|
| 9 | #
|
---|
| 10 | # Copyright (C) 1999 New Zealand Digital Library Project
|
---|
| 11 | #
|
---|
| 12 | # This program is free software; you can redistribute it and/or modify
|
---|
| 13 | # it under the terms of the GNU General Public License as published by
|
---|
| 14 | # the Free Software Foundation; either version 2 of the License, or
|
---|
| 15 | # (at your option) any later version.
|
---|
| 16 | #
|
---|
| 17 | # This program is distributed in the hope that it will be useful,
|
---|
| 18 | # but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
| 19 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
| 20 | # GNU General Public License for more details.
|
---|
| 21 | #
|
---|
| 22 | # You should have received a copy of the GNU General Public License
|
---|
| 23 | # along with this program; if not, write to the Free Software
|
---|
| 24 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
| 25 | #
|
---|
| 26 | ###########################################################################
|
---|
[4] | 27 |
|
---|
| 28 | BEGIN {
|
---|
| 29 | die "GSDLHOME not set\n" unless defined $ENV{'GSDLHOME'};
|
---|
| 30 | die "GSDLOS not set\n" unless defined $ENV{'GSDLOS'};
|
---|
[8] | 31 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib");
|
---|
| 32 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib/plugins");
|
---|
[946] | 33 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib/classify");
|
---|
[4] | 34 | }
|
---|
| 35 |
|
---|
[130] | 36 | use colcfg;
|
---|
[4] | 37 | use parsargv;
|
---|
| 38 | use util;
|
---|
| 39 |
|
---|
| 40 | &main();
|
---|
| 41 |
|
---|
| 42 | sub print_usage {
|
---|
| 43 | print STDERR "\n usage: $0 [options] collection-name\n\n";
|
---|
| 44 | print STDERR " options:\n";
|
---|
| 45 | print STDERR " -verbosity number 0=none, 3=lots\n";
|
---|
| 46 | print STDERR " -archivedir directory Where the archives live\n";
|
---|
| 47 | print STDERR " -cachedir directory Where to cache the archives\n";
|
---|
[313] | 48 | print STDERR " -builddir directory Where to put the built indexes\n";
|
---|
| 49 | print STDERR " -maxdocs number Maximum number of documents to build\n";
|
---|
[784] | 50 | print STDERR " -debug Print output to STDOUT\n";
|
---|
[783] | 51 | print STDERR " -mode all|compress_text|build_index|infodb\n";
|
---|
[784] | 52 | print STDERR " -index indexname Index to build (will build all in\n";
|
---|
| 53 | print STDERR " config file if not set)\n";
|
---|
| 54 | print STDERR " -keepold will not destroy the current contents of the\n";
|
---|
| 55 | print STDERR " building directory\n";
|
---|
[1279] | 56 | print STDERR " -allclassifications Don't remove empty classifications\n";
|
---|
| 57 | print STDERR " -create_images Attempt to create default images for new\n";
|
---|
| 58 | print STDERR " collection. This relies on the Gimp being\n";
|
---|
| 59 | print STDERR " installed along with relevant perl modules\n";
|
---|
| 60 | print STDERR " to allow scripting from perl\n\n";
|
---|
[4] | 61 | }
|
---|
| 62 |
|
---|
| 63 |
|
---|
| 64 | sub main
|
---|
| 65 | {
|
---|
[783] | 66 | my ($verbosity, $archivedir, $cachedir, $builddir, $maxdocs,
|
---|
[1279] | 67 | $debug, $mode, $indexname, $keepold, $allclassifications,
|
---|
| 68 | $create_images);
|
---|
[4] | 69 | if (!parsargv::parse(\@ARGV,
|
---|
| 70 | 'verbosity/\d+/2', \$verbosity,
|
---|
| 71 | 'archivedir/.*/', \$archivedir,
|
---|
| 72 | 'cachedir/.*/', \$cachedir,
|
---|
[313] | 73 | 'builddir/.*/', \$builddir,
|
---|
[814] | 74 | 'maxdocs/^\-?\d+/-1', \$maxdocs,
|
---|
[783] | 75 | 'debug', \$debug,
|
---|
| 76 | 'mode/^(all|compress_text|build_index|infodb)$/all', \$mode,
|
---|
| 77 | 'index/.*/', \$indexname,
|
---|
[784] | 78 | 'keepold', \$keepold,
|
---|
[1279] | 79 | 'allclassifications', \$allclassifications,
|
---|
| 80 | 'create_images', \$create_images)) {
|
---|
[4] | 81 | &print_usage();
|
---|
| 82 | die "\n";
|
---|
| 83 | }
|
---|
| 84 |
|
---|
| 85 | # get and check the collection
|
---|
[130] | 86 | if (($collection = &util::use_collection(@ARGV)) eq "") {
|
---|
[4] | 87 | &print_usage();
|
---|
| 88 | die "\n";
|
---|
| 89 | }
|
---|
[130] | 90 |
|
---|
| 91 | # read the configuration file
|
---|
| 92 | $textindex = "section:text";
|
---|
[1279] | 93 | $configfilename = &util::filename_cat ($ENV{'GSDLCOLLECTDIR'}, "etc", "collect.cfg");
|
---|
[130] | 94 | if (-e $configfilename) {
|
---|
| 95 | $collectcfg = &colcfg::read_collect_cfg ($configfilename);
|
---|
| 96 | if (defined $collectcfg->{'archivedir'} && $archivedir eq "") {
|
---|
| 97 | $archivedir = $collectcfg->{'archivedir'};
|
---|
| 98 | }
|
---|
| 99 | if (defined $collectcfg->{'cachedir'} && $cachedir eq "") {
|
---|
| 100 | $cachedir = $collectcfg->{'cachedir'};
|
---|
| 101 | }
|
---|
| 102 | if (defined $collectcfg->{'builddir'} && $builddir eq "") {
|
---|
| 103 | $builddir = $collectcfg->{'builddir'};
|
---|
| 104 | }
|
---|
| 105 | } else {
|
---|
| 106 | die "Couldn't find the configuration file $configfilename\n";
|
---|
[4] | 107 | }
|
---|
[130] | 108 |
|
---|
[1279] | 109 | # create default images if required
|
---|
| 110 | if ($create_images) {
|
---|
| 111 | my $collection_name = $collection;
|
---|
| 112 | $collection_name = $collectcfg->{'collectionmeta'}->{'collectionname'}
|
---|
| 113 | if defined $collectcfg->{'collectionmeta'}->{'collectionname'};
|
---|
| 114 |
|
---|
| 115 | &create_images ($collection_name);
|
---|
| 116 | }
|
---|
| 117 |
|
---|
[4] | 118 | # fill in the default archives and building directories if none
|
---|
| 119 | # were supplied, turn all \ into / and remove trailing /
|
---|
[130] | 120 | $archivedir = "$ENV{'GSDLCOLLECTDIR'}/archives" if $archivedir eq "";
|
---|
[4] | 121 | $archivedir =~ s/[\\\/]+/\//g;
|
---|
| 122 | $archivedir =~ s/\/$//;
|
---|
[130] | 123 | $builddir = "$ENV{'GSDLCOLLECTDIR'}/building" if $builddir eq "";
|
---|
[4] | 124 | $builddir =~ s/[\\\/]+/\//g;
|
---|
| 125 | $builddir =~ s/\/$//;
|
---|
| 126 |
|
---|
| 127 | # update the archive cache if needed
|
---|
| 128 | if ($cachedir) {
|
---|
| 129 | print STDERR "Updating archive cache\n" if ($verbosity >= 1);
|
---|
| 130 |
|
---|
| 131 | $cachedir =~ s/[\\\/]+$//;
|
---|
| 132 | $cachedir .= "/collect/$collection" unless
|
---|
| 133 | $cachedir =~ /collect\/$collection/;
|
---|
| 134 |
|
---|
| 135 | $realarchivedir = "$cachedir/archives";
|
---|
| 136 | $realbuilddir = "$cachedir/building";
|
---|
| 137 | &util::mk_all_dir ($realarchivedir);
|
---|
| 138 | &util::mk_all_dir ($realbuilddir);
|
---|
| 139 | &util::cachedir ($archivedir, $realarchivedir, $verbosity);
|
---|
| 140 |
|
---|
| 141 | } else {
|
---|
| 142 | $realarchivedir = $archivedir;
|
---|
| 143 | $realbuilddir = $builddir;
|
---|
| 144 | }
|
---|
| 145 |
|
---|
| 146 | # build it in realbuilddir
|
---|
| 147 | &util::mk_all_dir ($realbuilddir);
|
---|
| 148 |
|
---|
| 149 |
|
---|
| 150 | # if a builder class has been created for this collection, use it
|
---|
| 151 | # otherwise, use the mg builder
|
---|
[130] | 152 | if (-e "$ENV{'GSDLCOLLECTDIR'}/perllib/${collection}builder.pm") {
|
---|
| 153 | $builderdir = "$ENV{'GSDLCOLLECTDIR'}/perllib";
|
---|
[4] | 154 | $buildertype = "${collection}builder";
|
---|
| 155 | } else {
|
---|
[8] | 156 | $builderdir = "$ENV{'GSDLHOME'}/perllib";
|
---|
[4] | 157 | $buildertype = "mgbuilder";
|
---|
| 158 | }
|
---|
| 159 |
|
---|
| 160 | require "$builderdir/$buildertype.pm";
|
---|
| 161 |
|
---|
| 162 | eval("\$builder = new $buildertype(\$collection, " .
|
---|
[313] | 163 | "\$realarchivedir, \$realbuilddir, \$verbosity, " .
|
---|
[784] | 164 | "\$maxdocs, \$debug, \$keepold, \$allclassifications)");
|
---|
[4] | 165 | die "$@" if $@;
|
---|
| 166 |
|
---|
| 167 | $builder->init();
|
---|
| 168 |
|
---|
[783] | 169 | if ($mode =~ /^all$/i) {
|
---|
| 170 | $builder->compress_text($textindex);
|
---|
| 171 | $builder->build_indexes($indexname);
|
---|
| 172 | $builder->make_infodatabase();
|
---|
| 173 | $builder->collect_specific();
|
---|
| 174 | } elsif ($mode =~ /^compress_text$/i) {
|
---|
| 175 | $builder->compress_text($textindex);
|
---|
| 176 | } elsif ($mode =~ /^build_index$/i) {
|
---|
| 177 | $builder->build_indexes($indexname);
|
---|
| 178 | } elsif ($mode =~ /^infodb$/i) {
|
---|
| 179 | $builder->make_infodatabase();
|
---|
| 180 | } else {
|
---|
| 181 | die "unknown mode: $mode\n";
|
---|
| 182 | }
|
---|
[4] | 183 |
|
---|
[783] | 184 | $builder->make_auxiliary_files() if !$debug;
|
---|
| 185 | $builder->deinit();
|
---|
| 186 |
|
---|
| 187 | if (($realbuilddir ne $builddir) && !$debug) {
|
---|
[4] | 188 | print STDERR "Copying back the cached build\n" if ($verbosity >= 1);
|
---|
| 189 | &util::rm_r ($builddir);
|
---|
| 190 | &util::cp_r ($realbuilddir, $builddir);
|
---|
| 191 | }
|
---|
| 192 | }
|
---|
| 193 |
|
---|
[1279] | 194 | sub create_images {
|
---|
| 195 | my ($collection_name) = @_;
|
---|
[4] | 196 |
|
---|
[1279] | 197 | my $image_script = &util::filename_cat ($ENV{'GSDLHOME'}, "bin", "script", "gimp", "title_icon.pl");
|
---|
| 198 | if (!-e $image_script) {
|
---|
| 199 | print STDERR "WARNING: Image making script ($image_script) could not be found\n";
|
---|
| 200 | print STDERR " Default images will not be generated\n\n";
|
---|
| 201 | return;
|
---|
| 202 | }
|
---|
| 203 |
|
---|
| 204 | my $imagedir = &util::filename_cat ($ENV{'GSDLCOLLECTDIR'}, "images");
|
---|
| 205 |
|
---|
| 206 | &util::mk_all_dir ($imagedir);
|
---|
| 207 |
|
---|
| 208 | # create the images
|
---|
| 209 | system ("$image_script -size 1.5 -image_dir \"$imagedir\" -filename $collection.gif -text \"$collection_name\"");
|
---|
| 210 | system ("$image_script -image_dir \"$imagedir\" -filename ${collection}sm.gif -text \"$collection_name\"");
|
---|
| 211 |
|
---|
| 212 | # update the collect.cfg configuration file (this will need
|
---|
| 213 | # to be changed when the config file format changes)
|
---|
| 214 | if (!open (CFGFILE, $configfilename)) {
|
---|
| 215 | print STDERR "WARNING: Couldn't open config file ($configfilename)\n";
|
---|
| 216 | print STDERR " for updating so collection images may not be linked correctly\n";
|
---|
| 217 | return;
|
---|
| 218 | }
|
---|
| 219 |
|
---|
| 220 | my $line = ""; my $file = "";
|
---|
| 221 | my $found = 0; my $foundsm = 0;
|
---|
| 222 | while (defined ($line = <CFGFILE>)) {
|
---|
| 223 | if ($line =~ /collectionmeta\s+iconcollection\s+/) {
|
---|
| 224 | $line = "collectionmeta iconcollection _httprefix_/collect/$collection/images/$collection.gif\n";
|
---|
| 225 | $found = 1;
|
---|
| 226 | } elsif ($line =~ /collectionmeta\s+iconcollectionsmall\s+/) {
|
---|
| 227 | $line = "collectionmeta iconcollectionsmall _httprefix_/collect/$collection/images/${collection}sm.gif\n";
|
---|
| 228 | $foundsm = 1;
|
---|
| 229 | }
|
---|
| 230 | $file .= $line;
|
---|
| 231 | }
|
---|
| 232 | close CFGFILE;
|
---|
| 233 |
|
---|
| 234 | $file .= "collectionmeta iconcollection _httprefix_/collect/$collection/images/$collection.gif\n" if !$found;
|
---|
| 235 | $file .= "collectionmeta iconcollectionsmall _httprefix_/collect/$collection/images/${collection}sm.gif\n" if !$foundsm;
|
---|
| 236 |
|
---|
| 237 | if (!open (CFGFILE, ">$configfilename")) {
|
---|
| 238 | print STDERR "WARNING: Couldn't open config file ($configfilename)\n";
|
---|
| 239 | print STDERR " for updating so collection images may not be linked correctly\n";
|
---|
| 240 | return;
|
---|
| 241 | }
|
---|
| 242 | print CFGFILE $file;
|
---|
| 243 | close CFGFILE;
|
---|
| 244 | }
|
---|