root/main/trunk/greenstone2/bin/script/build @ 24192

Revision 24192, 26.3 KB (checked in by ak19, 8 years ago)

Sam discovered that using dollar-Config{perlpath} in place of dollar-hat-X is the better way to obtain the path to the perl that is being used. We hope this will not be a relative path on the Mac as dollar-hat-x was on Professor Witten's Mac when we tried it there today.

  • Property svn:executable set to *
  • Property svn:keywords set to Author Date Id Revision
Line 
1#!/usr/bin/perl -w
2
3###########################################################################
4#
5# build --
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) 2000 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###########################################################################
27
28# This perl script may be called directly or by running build.bat on
29# windows (build.bat is in bin\windows)
30
31# Note that this script has grown over time and now has many options for
32# use when called from within the collector. If it appears to
33# over-complicate things a little, that's why.
34
35# I think that currently (Nov 2010) its only officially used from the collector
36# and depositor
37
38package build;
39
40use strict;
41no strict 'refs';
42
43use FileHandle;
44use File::Copy;
45use Config; # for getting the perlpath in the recommended way
46
47BEGIN {
48
49    die "GSDLHOME not set - did you remember to source setup.bash (unix) or " .
50    "run setup.bat (windows)?\n" unless defined $ENV{'GSDLHOME'};
51    die "GSDLOS not set - did you remember to source setup.bash (unix) or " .
52    "run setup.bat (windows)?\n" unless defined $ENV{'GSDLOS'};
53    unshift (@INC, "$ENV{'GSDLHOME'}/perllib");
54
55    STDOUT->autoflush(1);
56    STDERR->autoflush(1);
57}
58
59use lib qq($ENV{'GSDLHOME'}/perllib/cpan);
60use Mail::Sendmail;
61use parsargv;
62use util;
63use cfgread;
64use colcfg;
65use dbutil;
66
67# set up greenstone environment
68&util::setup_greenstone_env($ENV{'GSDLHOME'}, $ENV{'GSDLOS'});
69# is the following needed with the previous line??
70# set up path - this allows for paths not to be supplied to system calls
71# and overcomes problems when GSDLHOME contains spaces (double quoting
72# the call doesn't work on win2k and probably other variants of winnt)
73my $path_separator = ":";
74$path_separator = ";" if $ENV{'GSDLOS'} =~ /^windows$/;
75$ENV{'PATH'} = &util::filename_cat($ENV{'GSDLHOME'}, "bin", $ENV{'GSDLOS'}) .
76    $path_separator . &util::filename_cat($ENV{'GSDLHOME'}, "bin", "script") .
77    $path_separator . $ENV{'PATH'};
78
79# all the input option variables
80my ($optionfile, $indextype, $append, $manifest, $remove_archives, $remove_import, $buildtype, $infodbtype, $maxdocs, @download, $collectdir, $site, $dontinstall, $save_archives, $out, $make_writable, $log_events, $event_log_file, $email_events, $mail_server, $statsfile, $event_header);
81
82&parse_args (\@ARGV);
83
84my ($collection) = @ARGV;
85
86if (!defined $collection || $collection !~ /\w/) {
87    print STDERR "You must specify a collection to build\n";
88    &print_usage();
89    die "\n";
90}
91
92if ($optionfile =~ /\w/) {
93    open (OPTIONS, $optionfile) || die "Couldn't open $optionfile\n";
94    my $line = [];
95    my $options = [];
96    while (defined ($line = &cfgread::read_cfg_line ('build::OPTIONS'))) {
97    push (@$options, @$line);
98    }
99    close OPTIONS;
100    &parse_args ($options);
101}
102
103if ($maxdocs == -1) {
104    $maxdocs = "";
105} else {
106    $maxdocs = "-maxdocs $maxdocs";
107}
108
109my $cdir = $collectdir;
110my $gs_mode;
111if (defined $site && $site =~ /\w/)
112{
113    die "GSDL3HOME not set." unless $ENV{'GSDL3HOME'};
114    $cdir = &util::filename_cat ($ENV{'GSDL3HOME'}, "sites", $site, "collect") unless $collectdir =~ /\w/;
115    $gs_mode = "gs3";
116}
117else
118{
119    $cdir = &util::filename_cat ($ENV{'GSDLHOME'}, "collect") unless $collectdir =~ /\w/;
120    $gs_mode = "gs2";
121}
122
123my $importdir = &util::filename_cat ($cdir, $collection, "import");
124my $archivedir = &util::filename_cat ($cdir, $collection, "archives");
125my $buildingdir = &util::filename_cat ($cdir, $collection, "building");
126my $indexdir = &util::filename_cat ($cdir, $collection, "index");
127my $etcdir = &util::filename_cat ($cdir, $collection, "etc");
128my $bindir = &util::filename_cat ($ENV{'GSDLHOME'}, "bin");
129
130my $use_out = 0;
131my $outfile = $out;
132if ($out !~ /^(STDERR|STDOUT)$/i) {
133    open (OUT, ">$out") || die "Couldn't open output file $out\n";
134    $out = "OUT";
135   
136    # delete any existing .final file
137    &util::rm ("$outfile.final") if -e "$outfile.final";
138
139    $use_out = 1;
140}
141$out->autoflush(1);
142
143# delete any .kill file left laying around from a previously aborted build
144if (-e &util::filename_cat ($cdir, $collection, ".kill")) {
145    &util::rm (&util::filename_cat ($cdir, $collection, ".kill"));
146}
147
148# get maintainer email address from main.cfg
149my $maintainer = "NULL";
150my $main_cfg = &util::filename_cat ($ENV{'GSDLHOME'}, "etc", "main.cfg");
151my $cfgdata = &cfgread::read_cfg_file ($main_cfg, "maintainer");
152if (defined $cfgdata->{'maintainer'} && $cfgdata->{'maintainer'} =~ /\w/) {
153    $maintainer = $cfgdata->{'maintainer'};
154}
155# if maintainer is "NULL" email_events should be disabled
156if ($maintainer =~ /^NULL$/i) {
157    $email_events = "";
158}
159
160&main();
161
162if ($use_out) {
163    close OUT;
164
165    # if we've created a build log we'll copy it to the collection's etc
166    # directory
167    my ($final_etcdir);
168    if ($dontinstall) {
169    $final_etcdir = &util::filename_cat($collectdir, "etc", "build.log");
170    } else {
171    $final_etcdir = &util::filename_cat($ENV{'GSDLHOME'}, "collect", $collection, "etc", "build.log");
172    }
173   
174    &util::cp($outfile, $final_etcdir);
175}
176
177sub print_usage {
178    print STDOUT "\n";
179    print STDOUT "build: Builds a Greenstone collection (i.e. runs import.pl and buildcol.pl\n";
180    print STDOUT "       then copies the resulting indexes to the correct place).\n\n";
181    print STDOUT "  usage: $0 [options] collection-name\n\n";
182    print STDOUT "  options:\n";
183    print STDOUT "   -optionfile file        Get options from file, useful on systems where\n";
184    print STDOUT "                           long command lines may cause problems\n";
185    print STDOUT "   -indextype mg|mgpp|lucene \n";
186    print STDERR "                           Specify the type of indexer used in this collection\n";
187    print STDERR "                           If -append is used then -indextype is needed to \n";
188    print STDERR "                           determine how to run buildcol.pl as well as update\n";
189    print STDERR "                           'building' and 'index' according.\n";
190    print STDOUT "   -append                 Add new files to existing collection\n";
191    print STDOUT "   -manifest               Use manifest.xml file to determine which files to process.\n";
192    print STDOUT "   -remove_archives        Remove archives directory after successfully\n";
193    print STDOUT "                           building the collection.\n";
194    print STDOUT "   -remove_import          Remove import directory after successfully\n";
195    print STDOUT "                           importing the collection.\n";
196    print STDOUT "   -buildtype build|import If 'build' attempt to build directly\n";
197    print STDOUT "                           from archives directory (bypassing import\n";
198    print STDOUT "                           stage). Defaults to 'import'\n";
199    print STDOUT "   -maxdocs number         Maximum number of documents to build\n";
200    print STDOUT "   -download directory     Directory (or file) to get import documents from.\n";
201    print STDOUT "                           There may be multiple download directories and they\n";
202    print STDOUT "                           may be of type http://, ftp://, or file://\n";
203    print STDOUT "                           Note that any existing import directory will be\n";
204    print STDOUT "                           deleted to make way for the downloaded data if\n";
205    print STDOUT "                           a -download option is supplied\n";
206    print STDOUT "   -collectdir directory   Collection directory (defaults to " .
207    &util::filename_cat($ENV{'GSDLHOME'}). "collect for Greenstone2;\n";
208    print STDOUT"                            for Greenstone3 use -site option and then collectdir default will be\n";
209    print STDOUT "                            set to the collect folder within that site.)\n";
210    print STDOUT "   -site                   Specify the site within a Greenstone3 installation to use.\n";
211    print STDOUT "   -dontinstall            Only applicable if -collectdir is set to something\n";
212    print STDOUT "                           other than the default. -dontinstall will suppress\n";
213    print STDOUT "                           the default behaviour which is to install the\n";
214    print STDOUT "                           collection to the gsdl/collect directory once it has\n";
215    print STDOUT "                           been built.\n";
216    print STDOUT "   -save_archives          Create a copy of the existing archives directory\n";
217    print STDOUT "                           called archives.org\n";
218    print STDOUT "   -out                    Filename or handle to print output status to.\n";
219    print STDOUT "                           The default is STDERR\n";
220    print STDOUT "   -statsfile name         Filename or handle to print import statistics to.\n";
221    print STDOUT "                           The default is STDERR\n";
222    print STDOUT "   -make_writable          If set build will make the collection and any\n";
223    print STDOUT "                           temporary files it created globally writable after\n";
224    print STDOUT "                           it finishes\n";
225    print STDOUT "   -log_events             Log important events (collection built successfully\n";
226    print STDOUT "                           etc.) to event_log_file\n";
227    print STDOUT "   -event_log_file file    File to append important events to (defaults to\n";
228    print STDOUT "                           " . &util::filename_cat ($ENV{'GSDLHOME'}, "etc", "events.txt") . "\n";
229    print STDOUT "   -email_events addr      Comma separated list of email addresses to mail\n";
230    print STDOUT "                           details of important collection building events\n";
231    print STDOUT "   -mail_server server     The outgoing (SMTP) mail server to be used by\n";
232    print STDOUT "                           email_events. email_events will be disabled if\n";
233    print STDOUT "                           mail_server isn't set\n";
234    print STDOUT "   -event_header file      File containing a header to go on any event\n";
235    print STDOUT "                           messages. If not specified build will create a\n";
236    print STDOUT "                           generic header\n\n";
237    print STDOUT "  [Type \"build | more\" if this help text scrolled off your screen]";
238    print STDOUT "\n" unless $ENV{'GSDLOS'} =~ /^windows$/i;
239}
240
241sub main {
242
243    if ($save_archives && -d $archivedir) {
244    print $out "caching original archives to ${archivedir}.org\n";
245    &util::cp_r ($archivedir, "${archivedir}.org");
246    }
247
248    # do the download thing if we have any -download options
249    if (scalar (@download)) {
250    # remove any existing import data
251    if (&has_content ($importdir)) {
252        #print $out "build: WARNING: removing contents of $importdir\n";
253        #&util::rm_r ($importdir);
254    }
255   
256    foreach my $download_dir (@download) {
257
258        # remove any leading or trailing whitespace from filenames (just in case)
259        $download_dir =~ s/^\s+//;
260        $download_dir =~ s/\s+$//;
261       
262        if ($download_dir =~ /^(http|ftp):\/\//) {
263        # use wget to mirror http or ftp urls
264        # options used are:
265        #  -P = the directory to download documents to
266        #  -np = don't ascend to parent directories. this means that only documents
267        #        that live in the same directory or below on the same server as
268        #        the given url will be downloaded
269        #  -nv = not too verbose
270        #  -r = recursively mirror
271        #  -N = use time-stamping to see if an up-to-date local copy of each
272        #       file already exists. this may be useful if wget fails and
273        #       is restarted
274        #  -l inf = infinite recursion depth
275        #  -R "*\?*" = don't download cgi based urls
276        #  -o = the output file to write download status to (only used if the -out
277        #       option was given to build)
278
279        #Config{perlpath}, like $^X, is a special variable containing the full path to the current perl executable we are in
280        my $download_cmd = "\"$Config{perlpath}\" -S gsWget.pl -P \"$importdir\" -np -nv";
281        $download_cmd .= " -r -N -l inf -R \"*\\?*\"";
282        $download_cmd .= " -o \"$outfile.download\"" if $use_out;
283        $download_cmd .= " \"$download_dir\"";
284        system ($download_cmd);
285
286        # note that wget obeys the robot rules. this means that it will have
287        # downloaded a robots.txt file if one was present. since it's unlikely
288        # anyone really wants to include it in a collection we'll delete it.
289        # robots.txt shouldn't be more than two directories deep (I think it will
290        # always be exactly two deep but will look for it in the top directory too)
291        # so that's as far as we'll go looking for it.
292        if (opendir (DIR, $importdir)) {
293            my @files = readdir DIR;
294            closedir DIR;
295            foreach my $file (@files) {
296            next if $file =~ /^\.\.?$/;
297            if ($file =~ /^robots.txt$/i) {
298                &util::rm (&util::filename_cat ($importdir, $file));
299                last;
300            } else {
301                $file = &util::filename_cat ($importdir, $file);
302                if (-d $file) {
303                if (opendir (DIR, $file)) {
304                    my @files2 = readdir DIR;
305                    closedir DIR;
306                    foreach my $file2 (@files2) {
307                    if ($file2 =~ /^robots.txt$/i) {
308                        &util::rm (&util::filename_cat ($file, $file2));
309                        last;
310                    }
311                    }
312                }
313                }
314            }
315            }
316        }
317
318        # if using output directory append the file download output to it
319        &append_file ($out, "$outfile.download");
320       
321        } else {
322        # we assume anything not beginning with http:// or ftp://
323        # is a file or directory on the local file system.
324        $download_dir =~ s/^file:(\/\/)?//;
325        $download_dir =~ s/^\s+//; # may be whitespace between "file://" and the rest
326       
327        if (-e $download_dir) {
328            # copy download_dir and all it contains to the import directory
329            #Config{perlpath}, like $^X, is a special variable containing the full path to the current perl executable we are in
330            my $download_cmd = "\"$Config{perlpath}\" -S filecopy.pl";
331            $download_cmd .= " -collectdir \"$collectdir\"" if $collectdir =~ /\w/;
332            $download_cmd .= " -site \"$site\"" if $site =~ /\w/;
333            $download_cmd .= " -out \"$outfile.download\"" if $use_out;
334            $download_cmd .= " \"" . $download_dir . "\" " . $collection;
335
336            my $download_status = system ($download_cmd);
337            if ($download_status > 0)
338            {
339            die "Failed to execute: $download_cmd\n";
340            }
341
342
343            # if using output directory append the file download output to it
344            &append_file ($out, "$outfile.download");
345        } else {
346            print $out "WARNING: '$download_dir' does not exist\n";
347        }
348        }
349    }
350    }
351
352    `echo $importdir ; ls $importdir `;
353
354    my $col_cfg_file;
355    if ($gs_mode eq "gs3") {
356    $col_cfg_file = &util::filename_cat($etcdir, "collectionConfig.xml");
357    } else {
358    $col_cfg_file = &util::filename_cat($etcdir, "collect.cfg");
359    }
360
361    my $collect_cfg = &colcfg::read_collection_cfg ($col_cfg_file, $gs_mode);
362    # get the database type for this collection from its configuration file (may be undefined)
363    $infodbtype = $collect_cfg->{'infodbtype'} || &dbutil::get_default_infodb_type();
364
365    my $archiveinf_doc_file_path = &dbutil::get_infodb_file_path($infodbtype, "archiveinf-doc", $archivedir);
366    if (-e $archiveinf_doc_file_path) {
367    if (&has_content ($importdir)) {
368        if ($buildtype eq "build") {
369        &gsdl_build();
370        } else {
371        &gsdl_import();
372        &gsdl_build();
373        }
374    } else {
375        # there are archives but no import, build directly from archives
376        print $out "build: no import material was found, building directly\n";
377        print $out "       from archives\n";
378        &gsdl_build();
379    }
380    } else {
381    if (&has_content ($importdir)) {
382        if ($buildtype eq "build") {
383        print $out "build: can't build directly from archives as no\n";
384        print $out "       imported archives exist (did you forget to\n";
385        print $out "       move the contents of $collection/import to\n";
386        print $out "       collection/archives?)\n";
387        }
388        &gsdl_import();
389        if (&has_content ($archivedir, "^archiveinf-doc\..*\$")) {
390        &gsdl_build();
391        } else {
392        my $msg = "build: ERROR: The collection could not be built as no\n";
393        $msg .=   "       valid data was imported. Are at least some of\n";
394        $msg .=   "       the files you imported in a format that can be\n";
395        $msg .=   "       processed by the specified Greenstone plugins?\n";
396        print $out $msg;
397        &log_event ($msg);
398        &final_out (6) if $use_out;
399        die "\n";
400        }
401    } else {
402        # no import or archives
403        my $msg = "build: ERROR: The collection could not be built as it contains no data.\n";
404        print $out $msg;
405        &log_event ($msg);
406        &final_out (1) if $use_out;
407        die "\n";
408    }
409    }
410
411    if ($collectdir ne "" && !$dontinstall) {
412   
413    my $install_collectdir;
414    if (defined $ENV{'GSDL3HOME'})
415    {
416       
417        if ((defined $site) && ($site ne ""))
418        {
419        $install_collectdir = &util::filename_cat ($ENV{'GSDL3HOME'}, "sites", $site, "collect");
420        }
421        else
422        {
423        my $msg = "build: ERROR: Need to specify the site within the Greenstone3 installation.";
424        print $out $msg;
425        &log_event ($msg);
426        &final_out (6) if $use_out;
427        die "\n";
428        }
429    }
430    else
431    {
432        $install_collectdir = &util::filename_cat ($ENV{'GSDLHOME'}, "collect");
433    }
434
435    if (!&util::filenames_equal ($collectdir, $install_collectdir)) {
436       
437        # install collection to gsdl/collect
438        print $out "installing the $collection collection\n";
439        my $newdir = &util::filename_cat ($install_collectdir, $collection);
440        my $olddir = &util::filename_cat ($collectdir, $collection);
441        if (-d $newdir) {
442        my $msg = "build: Could not install collection as $newdir\n" .
443            "       already exists. Collection will remain at\n$olddir\n";
444
445        print $out $msg;
446        &log_event ($msg);
447        &final_out (4) if $use_out;
448        die "\n";
449        }
450        if (!&File::Copy::move ($olddir, $newdir)) {
451        my $msg = "build: Failed to install collection to $newdir\n" .
452            "       Collection will remain at $olddir\n";
453        print $out $msg;
454        &log_event ($msg);
455        &final_out (5) if $use_out;
456        die "\n";
457        }
458    }
459    }
460
461    &log_event ("The $collection collection was built successfully\n");
462    &final_out (0) if $use_out;
463}
464
465sub gsdl_import {
466
467    print $out "importing the $collection collection\n\n";
468
469    #Config{perlpath}, like $^X, is a special variable containing the full path to the current perl executable we are in
470    my $import_cmd = "\"$Config{perlpath}\" -S import.pl";
471    $import_cmd .= " -out \"$outfile.import\"" if $use_out;
472    if ($append) {
473    $import_cmd .= " -keepold";
474    if (not $manifest) {
475        # if we are appending, with no manifest, assume incremental
476        $import_cmd .= " -incremental";
477    }
478    } else {
479    $import_cmd .= " -removeold";
480    }
481
482    $import_cmd .= " -manifest manifest.xml" if ($manifest);
483    $import_cmd .= " -site \"$site\"" if $site =~ /\w/;
484    $import_cmd .= " -collectdir \"$collectdir\"" if $collectdir =~ /\w/;
485    $import_cmd .= " -statsfile \"$statsfile\"" if $statsfile =~ /\w/;
486    $import_cmd .= " $maxdocs $collection";
487
488    system ($import_cmd);
489    # if using output directory append the import output to it
490    &append_file ($out, "$outfile.import");
491
492    my $archiveinf_doc_file_path = &dbutil::get_infodb_file_path($infodbtype, "archiveinf-doc", $archivedir);
493
494    if (-e $archiveinf_doc_file_path) {
495    print $out "$collection collection imported successfully\n\n";
496    if ($remove_import) {
497        print $out "removing import directory ($importdir)\n";
498        &util::rm_r ($importdir);
499    }
500    } else {
501    print $out "$archiveinf_doc_file_path not found. archives contents:\n";
502    print $out `ls $archivedir`;
503
504    my $msg = "build: ERROR: import.pl failed\n";
505    print $out "\n$msg";
506    &log_event ($msg);
507    &final_out (2) if $use_out;
508    die "\n";
509    }
510}
511
512sub gsdl_build {
513
514    print $out "building the $collection collection\n\n";
515
516    #Config{perlpath}, like $^X, is a special variable containing the full path to the current perl executable we are in
517    my $build_cmd = "\"$Config{perlpath}\" -S buildcol.pl";
518
519    my $removeold = 1;
520    if ($append) {
521    if ($indextype eq "lucene") {
522        $build_cmd .= " -keepold -incremental -builddir $indexdir";
523        $removeold = 0;
524    }
525    else {
526        $build_cmd .= " -removeold";
527    }
528    }
529    else {
530    $build_cmd .= " -removeold";
531    }
532
533    $build_cmd .= " -out \"$outfile.build\"" if $use_out;
534    $build_cmd .= " -site \"$site\"" if $site =~ /\w/;
535    $build_cmd .= " -collectdir \"$collectdir\"" if $collectdir =~ /\w/;
536    $build_cmd .= " $maxdocs $collection";
537    system ($build_cmd);
538    # if using output directory append the buildcol output to it
539    &append_file ($out, "$outfile.build");
540
541    my @db_exts = ( ".ldb", ".bdb", ".gdb", ".db" );
542    my $build_ok = 0;
543    foreach my $db_ext (@db_exts) {
544    if ($removeold && (-e &util::filename_cat ($buildingdir, "text", "$collection$db_ext"))) {
545        $build_ok = 1;
546        last;
547    }
548    if (($removeold==0) && (-e &util::filename_cat ($indexdir, "text", "$collection$db_ext"))) {
549        $build_ok = 1;
550        last;
551    }
552    }
553
554    if ($build_ok) {
555    print $out "$collection collection built successfully\n\n";
556    if ($remove_archives) {
557        print $out "removing archives directory ($archivedir)\n";
558        &util::rm_r ($archivedir);
559    }
560    } else {
561    my $msg = "build: ERROR: buildcol.pl failed\n";
562    print $out "\n$msg";
563    &log_event ($msg);
564    &final_out (3) if $use_out;
565    die "\n";
566    }
567
568    if ($removeold) {
569    # replace old indexes with new ones
570    if (&has_content ($indexdir)) {
571        print $out "removing old indexes\n";
572        &util::rm_r ($indexdir);
573    }
574    rmdir ($indexdir) if -d $indexdir;
575    &File::Copy::move ($buildingdir, $indexdir);
576    }
577    else {
578    # Do nothing. We have built into index dir rather than building dir
579    }
580
581    # remove the cached archives
582    if ($save_archives && -d "${archivedir}.org") {
583    &util::rm_r ("${archivedir}.org");
584    }
585}
586
587# return 1 if $dir directory contains any files or sub-directories (other
588# than those specified in the $ignore regular expression)
589sub has_content {
590    my ($dir, $ignore) = @_;
591
592    if (!-d $dir) {return 0;}
593   
594    opendir (DIR, $dir) || return 0;
595    my @files = readdir DIR;
596    closedir DIR;
597   
598    foreach my $file (@files) {
599    if ($file !~ /^\.{1,2}$/) {
600        return 1 unless (defined $ignore && $file =~ /$ignore/);
601    }
602    }
603
604    return 0;
605}
606
607sub append_file {
608    my ($handle, $file) = @_;
609
610    open (FILE, $file) || return;
611    undef $/;
612    print $handle <FILE>;
613    $/ = "\n";
614    close FILE;
615    &util::rm ($file);
616}
617
618# creates a file called $outfile.final and writes an output code to it.
619# An output code of 0 specifies that there was no error
620sub final_out {
621    my ($exit_code) = @_;
622
623    if ($use_out && (!-e "$outfile.final")) {
624
625    if (open (FINAL, ">$outfile.final")) {
626        print FINAL $exit_code;
627        close FINAL;
628    }
629    }
630}
631
632sub log_event {
633    my ($msg) = @_;
634
635    return unless ($log_events || $email_events);
636
637    # get the event header
638    my $eheader = "[Build Event]\n";
639    $eheader .= "Date: " . scalar localtime() . "\n";
640    if ($event_header ne "" && open (HEADER, $event_header)) {
641    undef $/;
642    $eheader .= <HEADER>;
643    $/ = "\n";
644    close HEADER;
645    } else {
646    $eheader .= "Collection: $collection\n";
647    $eheader .= "GSDLHOME: $ENV{'GSDLHOME'}\n";
648    $eheader .= "Build Location: $collectdir\n";
649    }
650   
651    if ($log_events) {
652    my $fail = 0;
653    # append the event to the event log file
654    if ($event_log_file eq "" || !open (LOG, ">>$event_log_file")) {
655        # log file defaults to $GSDLHOME/etc/events.txt
656        $event_log_file = &util::filename_cat ($ENV{'GSDLHOME'}, "etc", "events.txt");
657        if (!open (LOG, ">>$event_log_file")) {
658        print $out "build: ERROR: Couldn't open event log file $event_log_file\n";
659        $fail = 1;
660        }
661    }
662    if (!$fail) {
663        print LOG $eheader;
664        print LOG $msg;
665        print LOG "\n";
666        close LOG;
667    }
668    }
669   
670    if ($email_events) {
671    # if mail_server isn't set email_events does nothing
672    if ($mail_server eq "") {
673        print $out "build: WARNING: mail_server was not set - email_events option was ignored\n";
674        return;
675    }
676   
677    my %mail = ('SMTP' => $mail_server,
678            'To' => $email_events,
679            'From' => $maintainer,
680            'Subject' => 'Greenstone Build Event'
681            );
682    $mail{'Message'} = $eheader . $msg;
683   
684    if (!sendmail %mail) {
685        print $out "build: ERROR sending mail to $email_events\n";
686        print $out "'$Mail::Sendmail::error'\n";
687    }
688    }
689}
690
691
692sub parse_args {
693    my ($argref) = @_;
694
695    if (!parsargv::parse($argref,
696             'optionfile/.*/', \$optionfile,
697             'indextype/^(mg|mgpp|lucene)$/mg', \$indextype,
698             'append', \$append,
699             'manifest', \$manifest,
700             'remove_archives', \$remove_archives,
701             'remove_import', \$remove_import,
702             'buildtype/^(build|import)$/import', \$buildtype,
703             'maxdocs/^\-?\d+/-1', \$maxdocs,
704             'download/.+', \@download,
705             'collectdir/.*/', \$collectdir,
706             'site/.*/', \$site,
707             'dontinstall', \$dontinstall,
708             'save_archives', \$save_archives,
709             'out/.*/STDERR', \$out,
710             'make_writable', \$make_writable,
711             'log_events', \$log_events,
712             'event_log_file/.*/', \$event_log_file,
713             'email_events/.*/', \$email_events,
714             'mail_server/.*/', \$mail_server,
715             'statsfile/.*/STDERR', \$statsfile,
716             'event_header/.*/', \$event_header)) {
717   
718    &print_usage();
719    die "\n";
720    }
721}
722
723
724END {
725
726    if ($make_writable) {
727    # chmod a+rw new collection
728    my $installed_collection = &util::filename_cat($ENV{'GSDLHOME'}, "collect", $collection);
729    &recursive_chmod($installed_collection);
730
731    # chmod a+rw anything we've left laying about in the tmp directory
732    if (($collectdir ne "") && 
733        (!&util::filenames_equal ($collectdir, &util::filename_cat($ENV{'GSDLHOME'}, "collect")))) {
734        &recursive_chmod($collectdir);
735    }
736    }
737
738    # this will produce a .final file if one doesn't exist yet - that
739    # should only happen if there's been an error somewhere in the perl
740    # code
741    &final_out(7);
742
743    sub recursive_chmod {
744    my ($dir) = @_;
745    return unless -d $dir;
746
747    chmod (0777, $dir);
748
749    opendir (DIR, $dir) || die;
750    my @files = readdir DIR;
751    closedir DIR;
752   
753    foreach my $file (@files) {
754        next if $file =~ /^\.\.?$/;
755        $file = &util::filename_cat($dir, $file);
756        if (-d $file) {
757        &recursive_chmod ($file);
758        } else {
759        chmod (0777, $file);
760        }
761    }
762    }
763}
Note: See TracBrowser for help on using the browser.