1 | #!/usr/bin/perl -w
|
---|
2 |
|
---|
3 | ###########################################################################
|
---|
4 | #
|
---|
5 | # Enhance the DSpace media filter component with parallel processing
|
---|
6 | # capability. Currently only processes media of dc.type "Video" (which will
|
---|
7 | # be manually assigned to the *.TS files produced by ReplayMe!)
|
---|
8 | #
|
---|
9 | # Copyright (C) 2012 New Zealand Digital Library Project
|
---|
10 | #
|
---|
11 | # This program is free software; you can redistribute it and/or modify
|
---|
12 | # it under the terms of the GNU General Public License as published by
|
---|
13 | # the Free Software Foundation; either version 2 of the License, or
|
---|
14 | # (at your option) any later version.
|
---|
15 | #
|
---|
16 | # This program is distributed in the hope that it will be useful,
|
---|
17 | # but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
18 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
19 | # GNU General Public License for more details.
|
---|
20 | #
|
---|
21 | # You should have received a copy of the GNU General Public License
|
---|
22 | # along with this program; if not, write to the Free Software
|
---|
23 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
24 | #
|
---|
25 | ###########################################################################
|
---|
26 |
|
---|
27 | package parallel_dspace_filtermedia;
|
---|
28 |
|
---|
29 | # Pragma
|
---|
30 | use strict;
|
---|
31 |
|
---|
32 | BEGIN {
|
---|
33 | die "GSDLHOME not set\n" unless defined $ENV{'GSDLHOME'};
|
---|
34 | die "GSDLOS not set\n" unless defined $ENV{'GSDLOS'};
|
---|
35 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib");
|
---|
36 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib/cpan");
|
---|
37 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib/plugins");
|
---|
38 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib/plugouts");
|
---|
39 |
|
---|
40 | if (defined $ENV{'GSDLEXTS'}) {
|
---|
41 | my @extensions = split(/:/,$ENV{'GSDLEXTS'});
|
---|
42 | foreach my $e (@extensions) {
|
---|
43 | my $ext_prefix = "$ENV{'GSDLHOME'}/ext/$e";
|
---|
44 |
|
---|
45 | unshift (@INC, "$ext_prefix/perllib");
|
---|
46 | unshift (@INC, "$ext_prefix/perllib/cpan");
|
---|
47 | unshift (@INC, "$ext_prefix/perllib/plugins");
|
---|
48 | unshift (@INC, "$ext_prefix/perllib/plugouts");
|
---|
49 | }
|
---|
50 | }
|
---|
51 | }
|
---|
52 |
|
---|
53 | # Modules
|
---|
54 | use Time::HiRes qw( gettimeofday tv_interval );
|
---|
55 |
|
---|
56 | sub printUsage
|
---|
57 | {
|
---|
58 | my ($message) = @_;
|
---|
59 | if (defined $message)
|
---|
60 | {
|
---|
61 | print STDERR 'Error! ' . $message . "\n";
|
---|
62 | }
|
---|
63 | print STDERR 'Usage: parallel_dspace_filtermedia.pl -workers <int> -dspacehome <path> [-debug]' . "\n\n";
|
---|
64 | exit(0);
|
---|
65 | }
|
---|
66 |
|
---|
67 | sub main
|
---|
68 | {
|
---|
69 | my ($start_s, $start_u) = &gettimeofday();
|
---|
70 | print '[' . $start_s . '.' . $start_u . '] Parallel Media Filter started: ' . `date` . "\n";
|
---|
71 |
|
---|
72 | # 1. Initialization
|
---|
73 | my $dspace_home = '';
|
---|
74 | my $worker_count = 0;
|
---|
75 | my $debug = 0;
|
---|
76 | my $i;
|
---|
77 | my $argument;
|
---|
78 | for ($i = 0; $i < scalar(@ARGV); $i++)
|
---|
79 | {
|
---|
80 | $argument = $ARGV[$i];
|
---|
81 | if ('-dspacehome' eq $argument)
|
---|
82 | {
|
---|
83 | $i++;
|
---|
84 | $dspace_home = $ARGV[$i];
|
---|
85 | if (!-d $dspace_home)
|
---|
86 | {
|
---|
87 | &printUsage('DSpace home path given doesn\'t exist or isn\'t a directory');
|
---|
88 | }
|
---|
89 | }
|
---|
90 | elsif ('-workers' eq $argument)
|
---|
91 | {
|
---|
92 | $i++;
|
---|
93 | $worker_count = $ARGV[$i];
|
---|
94 | if ($worker_count !~ /^\d+$/)
|
---|
95 | {
|
---|
96 | &printUsage('Worker count must be an integer');
|
---|
97 | }
|
---|
98 | }
|
---|
99 | elsif ('-debug' eq $argument)
|
---|
100 | {
|
---|
101 | $debug = 1;
|
---|
102 | }
|
---|
103 | else
|
---|
104 | {
|
---|
105 | &printUsage('Unrecognized argument: ' . $argument);
|
---|
106 | }
|
---|
107 | }
|
---|
108 | if ('' eq $dspace_home)
|
---|
109 | {
|
---|
110 | &printUsage('Path to dspace home required!');
|
---|
111 | }
|
---|
112 |
|
---|
113 | # 2. Run metadata export to get the identifiers of video items in the
|
---|
114 | # collection
|
---|
115 | print ' * Retrieving the list of video item identifiers...' . "\n";
|
---|
116 | my @video_item_identifiers;
|
---|
117 | my $metadata_file_path = $dspace_home . '/log/metadata-' . time() . '.csv';
|
---|
118 | my $metadata_export_cmd = '"' . $dspace_home . '/bin/dspace" metadata-export -f "' . $metadata_file_path . '"';
|
---|
119 | if ($debug)
|
---|
120 | {
|
---|
121 | print '[DEBUG: ' . $metadata_export_cmd . "]\n";
|
---|
122 | }
|
---|
123 | `$metadata_export_cmd`;
|
---|
124 | open(FIN, "<:utf8", $metadata_file_path) || die('Error! Failed to open metadata file for reading: ' . $metadata_file_path);
|
---|
125 | my $line = '';
|
---|
126 | while ($line = <FIN>)
|
---|
127 | {
|
---|
128 | if ($line =~ /\"http:\/\/hdl.handle.net\/([^\"]+)\",.*\"Video\"/)
|
---|
129 | {
|
---|
130 | my $identifier = $1;
|
---|
131 | push(@video_item_identifiers, $identifier);
|
---|
132 | if ($debug)
|
---|
133 | {
|
---|
134 | print '[DEBUG: parsed identifier: ' . $identifier . "]\n";
|
---|
135 | }
|
---|
136 | }
|
---|
137 | }
|
---|
138 | close(FIN);
|
---|
139 | unlink($metadata_file_path);
|
---|
140 |
|
---|
141 | if ($worker_count > 0)
|
---|
142 | {
|
---|
143 | print " - Parallel processing video files...\n";
|
---|
144 | # 3. Write the identifiers to the file list
|
---|
145 | my $file_list_path = $dspace_home . '/log/filelist' . time() . '.txt';
|
---|
146 | open(FLOUT, ">:utf8", $file_list_path) or die('Error! Failed to open file for writing: ' . $file_list_path);
|
---|
147 | foreach my $video_item_identifier (@video_item_identifiers)
|
---|
148 | {
|
---|
149 | print FLOUT $video_item_identifier . "\n";
|
---|
150 | }
|
---|
151 | close(FLOUT);
|
---|
152 | # 4. Invoke mpidspacemediafilter (via mpirun) with the the filelist
|
---|
153 | # created above. MPIRun takes the number of worker threads to spawn
|
---|
154 |
|
---|
155 | # Determine if we've been provided a mpi.conf file to indicate the other
|
---|
156 | # machines (slave nodes) this parallizable process should run on
|
---|
157 | my $mpi_conf_path = $dspace_home . '/mpi.conf';
|
---|
158 | my $mpi_flags = '--show-progress --timestamp-output ';
|
---|
159 | if (-f $mpi_conf_path)
|
---|
160 | {
|
---|
161 | print STDERR " ***** CLUSTER MODE *****\n";
|
---|
162 | $mpi_flags .= '-machinefile "' . $mpi_conf_path . '" ';
|
---|
163 | #$mpi_flags .= '-nolocal ';
|
---|
164 | }
|
---|
165 | else
|
---|
166 | {
|
---|
167 | print STDERR " ***** SINGLE COMPUTER MODE *****\n";
|
---|
168 | }
|
---|
169 | # Excessive force! Ensure we bind to the correct network interface
|
---|
170 | $mpi_flags .= '--mca btl tcp,sm,self --mca btl_tcp_if_include eth0 ';
|
---|
171 |
|
---|
172 | my $mpi_cmd = 'mpirun ' . $mpi_flags . ' -np ' . ($worker_count + 1) . ' "' . $dspace_home . '/bin/mpidspacemediafilter" "' . $ENV{'GSDLHOME'} . '" "' . $dspace_home . '" "' . $file_list_path . '"';
|
---|
173 | if ($debug)
|
---|
174 | {
|
---|
175 | print '[DEBUG: ' . $mpi_cmd . "\n";
|
---|
176 | }
|
---|
177 | `$mpi_cmd`;
|
---|
178 | unlink($file_list_path);
|
---|
179 | }
|
---|
180 | # 5. With 0 worker threads we run a serial process calling media filter
|
---|
181 | # directly
|
---|
182 | else
|
---|
183 | {
|
---|
184 | print " - Serial processing video files...\n";
|
---|
185 | my $serial_mediafilter_cmd = '';
|
---|
186 | foreach my $video_item_identifier (@video_item_identifiers)
|
---|
187 | {
|
---|
188 | print ' - processing item: ' . $video_item_identifier . "\n";
|
---|
189 | $serial_mediafilter_cmd = '"' . $dspace_home . '/bin/dspace" filter-media -f -i "' . $video_item_identifier . '"';
|
---|
190 | if ($debug)
|
---|
191 | {
|
---|
192 | print '[DEBUG: ' . $serial_mediafilter_cmd . "]\n";
|
---|
193 | }
|
---|
194 | `$serial_mediafilter_cmd`;
|
---|
195 | }
|
---|
196 | }
|
---|
197 |
|
---|
198 | # 6. Complete!
|
---|
199 | my ($end_s, $end_u) = &gettimeofday();
|
---|
200 | print '[' . $end_s . '.' . $end_u . '] Parallel Media Filter Complete: ' . `date` . "\n";
|
---|
201 | print 'Elapsed Time: ' . sprintf('%0.6f', tv_interval([$start_s, $start_u], [$end_s, $end_u])) . " seconds\n";
|
---|
202 | }
|
---|
203 |
|
---|
204 | &main();
|
---|
205 |
|
---|
206 | exit;
|
---|
207 |
|
---|
208 | 1;
|
---|