root/gsdl/trunk/bin/script/replace_srcdoc_with_html.pl @ 15121

Revision 15121, 6.4 KB (checked in by ak19, 12 years ago)

New perl script replace_srcdoc_with_html.pl allows users to replace a source document in their collection with a Greenstone-generated html version of it when they rightclick in GLI's collection folder view. The script can also be run from the command line.

  • Property svn:executable set to *
Line 
1#!/usr/bin/perl -w
2
3
4BEGIN {
5    die "GSDLHOME not set\n" unless defined $ENV{'GSDLHOME'};
6    unshift (@INC, "$ENV{'GSDLHOME'}/perllib");
7    unshift (@INC, "$ENV{'GSDLHOME'}/perllib/plugins");
8}
9
10use strict;
11no strict 'subs'; # allow barewords (eg STDERR) as function arguments
12no strict 'refs'; # allow filehandles to be variables and vice versa
13
14use encodings;
15use printusage;
16use parse2;
17use FileHandle;
18
19my $arguments =
20    [
21      { 'name' => "language",
22    'desc' => "{scripts.language}",
23    'type' => "string",
24    'reqd' => "no",
25        'hiddengli' => "yes" },
26      { 'name' => "plugin",
27    'desc' => "{srcreplace.plugin}",
28    'type' => "string",
29    'reqd' => "yes",
30    'hiddengli' => "yes"},
31      { 'name' => "verbosity",
32    'desc' => "{import.verbosity}",
33    'type' => "int",
34    'range' => "0,",
35    'deft' => "1",
36    'reqd' => "no",
37    'modegli' => "4" },
38      # Do not remove the following option, it's a flag for generating the xml of the options
39      # It WILL be used!
40      { 'name' => "xml", # run with -xml, the output generated should be valid XML.  Used from GLI
41    'desc' => "",
42    'type' => "flag",
43    'reqd' => "no",
44    'hiddengli' => "yes" }
45      ];
46   
47my $options = { 'name' => "replace_srcdoc_with_html.pl",
48        'desc' => "{srcreplace.desc}",
49        'args' => $arguments };
50
51       
52sub main
53{
54    my ($language, $plugin, $verbosity);
55
56    my $xml = 0;
57
58    my $hashParsingResult = {};
59
60    # parse the options
61    my $intArgLeftinAfterParsing = parse2::parse(\@ARGV,$arguments,$hashParsingResult,"allow_extra_options");
62
63    # If parse returns -1 then something has gone wrong
64    if ($intArgLeftinAfterParsing == -1)
65    {
66    &PrintUsage::print_txt_usage($options, "{srcreplace.params}");
67    die "\n";
68    }
69
70    foreach my $strVariable (keys %$hashParsingResult)
71    {
72    eval "\$$strVariable = \$hashParsingResult->{\"\$strVariable\"}";
73    }
74
75    # If $language has been specified, load the appropriate resource bundle
76    # (Otherwise, the default resource bundle will be loaded automatically)
77    if ($language && $language =~ /\S/) {
78    &gsprintf::load_language_specific_resource_bundle($language);
79    }
80
81    if ($xml) {
82        &PrintUsage::print_xml_usage($options);
83    print "\n";
84    return;
85    }
86
87    # There should be one arg left after parsing (the filename)
88    # Or the user may have specified -h, in which case we output the usage
89    if($intArgLeftinAfterParsing != 1 || (@ARGV && $ARGV[0] =~ /^\-+h/))
90    {
91    &PrintUsage::print_txt_usage($options, "{srcreplace.params}");
92    die "\n";
93    }
94
95    # The filename of the document to be replaced is the first value
96    # that remains after the options have been parsed out
97    my $filename = $ARGV[0];
98    if (!defined $filename || $filename !~ /\w/) { 
99   
100    &PrintUsage::print_txt_usage($options, "{srcreplace.params}");
101    print STDERR "You need to specify a filename";
102    die "\n";
103    }
104    # check that file exists
105    if (!-e $filename) {
106    print STDERR "File $filename doesn't exist...\n";
107    die "\n";
108    }
109    # check required options
110    if (!defined $plugin || $plugin !~ /\w/) {
111    &PrintUsage::print_txt_usage($options, "{srcreplace.params}");
112    print STDERR "You need to specify a plugin";
113    die "\n";
114    }
115
116    # ConvertToPlug.pm's subclasses should be available here through GLI,
117    # but in cmdline version, these should be supplied
118    my $plugobj;
119    require "$plugin.pm";
120    eval ("\$plugobj = new $plugin()");
121    die "$@" if $@;
122
123    # ...and initialize it
124    $plugobj->init(1, "STDERR", "STDERR");
125   
126    # find the import directory, where we want to create it in. This is where the file
127    # passed as parameter by GLI is located.
128   
129    # derive tmp filename from input filename
130    my ($tailname, $import_dir, $suffix)
131    = &File::Basename::fileparse($filename, "\\.[^\\.]+\$");
132
133    # Use the plugin's tmp_area_convert_file function to avoid duplicating code.
134    # This method returns the name of the output file. In the case of Word docs,
135    # if converted with windows_scripting a "filename_files" folder might have been
136    # created for associated files. Same situation when using wvware with gsConvert.pl.
137    # (When old gsConvert.pl was used, wvware created no separate directory, instead files
138    # associated with the html generated would be at the same level in the tmp folder
139    # where the output file was created.) Now it's the same no matter whether wvware
140    # or windows_scripting did the conversion of the Word doc to html.
141    my $output_filename = $plugobj->tmp_area_convert_file("html", $filename);   
142    # my $output_filename = $plugobj->tmp_area_convert_file($output_ext, $input_filename, $textref);
143   
144    if(!defined $output_filename || (!-e $output_filename)) { # if no output html file was created, then error message
145    print STDERR "***replace_srcdoc_with_html.pl: no output file created for $filename ***\n";
146    die "No html file created for $filename. Replacement did not take place\n"; # Program NEEDS to die here,
147    # else the error that occurred is not transmitted to GLI and it thinks instead that execution was fine
148    #return 0; # error code 0 for false <- NO, needs to die, not return!
149    }
150    #else
151
152    # now, find out what to move:
153    # it may be a single file, or, if it is a word doc, it may also have an image folder
154    # which has the name "filename-without-extension_files"
155    my ($tmp_name, $tmp_dir, $ext) = &File::Basename::fileparse($output_filename, "\\.[^\\.]+\$");
156
157    if($import_dir ne $tmp_dir) { # output file is not in the same folder
158    my $assoc_folder = &util::filename_cat($tmp_dir, $tmp_name."_files");
159   
160    # use util to move it all to the import directory where the original file is located
161    #&util::mv($output_filename, &util::filename_cat($import_dir, $tmp_name.$ext)); # doesn't work
162    #&util::mv($output_filename, $import_dir); # doesn't work
163    &util::cp($output_filename, $import_dir);
164    &util::rm($output_filename);
165    if(defined $assoc_folder) { # move any associated folders too
166        print STDERR "****Folder for associated files is $assoc_folder\n";
167        #&util::mv($assoc_folder, $import_dir); # doesn't work for me
168        &util::cp_r($assoc_folder, $import_dir);
169        &util::rm_r($assoc_folder);
170    }
171    } # else output file was created in the same folder as the input file
172    # this happens for txt files processed by TEXTPlug
173    # in this case, there will be no image folder AND no moving of output html file is necessary
174
175    # Now we can remove the source doc permanently (there are no assocdirs for source doc)
176    &util::rm($filename);
177   
178}
179&main(@ARGV);
Note: See TracBrowser for help on using the browser.