Changeset 1243 for trunk/gsdl

Timestamp:

2000-06-27T09:38:51+12:00 (24 years ago)

Author:

sjboddie

Message:

Caught HTMLPlug up with BasPlug. A few minor changes to some supporting
files (for new BasPlug options).

Location:

trunk/gsdl

Files:

• bin/script/import.pl (modified) (1 diff)
• perllib/mgbuilder.pm (modified) (1 diff)
• perllib/plugin.pm (modified) (2 diffs)
• perllib/plugins/HTMLPlug.pm (modified) (5 diffs)

trunk/gsdl/bin/script/import.pl

@@ -135 +135 @@
 
     # load all the plugins
-    $pluginfo = &plugin::load_plugins ($plugins);
+    $pluginfo = &plugin::load_plugins ($plugins, $verbosity);
     if (scalar(@$pluginfo) == 0) {
     print STDERR "No plugins were loaded.\n";

trunk/gsdl/perllib/mgbuilder.pm

@@ -99 +99 @@
     
     # load all the plugins
-    $self->{'pluginfo'} = &plugin::load_plugins ($plugins);
+    $self->{'pluginfo'} = &plugin::load_plugins ($plugins, $verbosity);
     if (scalar(@{$self->{'pluginfo'}}) == 0) {
     print STDERR "No plugins were loaded.\n";

trunk/gsdl/perllib/plugin.pm

@@ -29 +29 @@
 
 sub load_plugins {
-    my ($plugin_list) = @_;
+    my ($plugin_list, $verbosity) = @_;
     my @plugin_objects = ();
+
+    $verbosity = 2 unless defined $verbosity;
 
     foreach $pluginoptions (@$plugin_list) {
@@ -52 +54 @@
     die "$@" if $@;
     
+    # initialize plugin
+    $plugobj->init($verbosity);
+
     # add this object to the list
     push (@plugin_objects, $plugobj);

trunk/gsdl/perllib/plugins/HTMLPlug.pm

@@ -54 +54 @@
     print STDERR "\n  usage: plugin HTMLPlug [options]\n\n";
     print STDERR "  options:\n";
-    print STDERR "   -process_exp           A perl regular expression to match against filenames.\n";
-    print STDERR "                          Matching filenames will be processed by this plugin.\n";
-    print STDERR "                          Defaults to '(?i)\.html?\$' i.e. all documents ending in\n";
-    print STDERR "                          .htm or .html (case-insensitive).\n";
     print STDERR "   -nolinks               Don't make any attempt to trap links (setting this flag may\n";
     print STDERR "                          improve speed of building/importing but any relative links within\n";
     print STDERR "                          documents will be broken).\n";
-    print STDERR "   -block_exp             Files matching this regular expression will be blocked from\n";
-    print STDERR "                          being passed to any further plugins in the list. By default\n";
-    print STDERR "                          HTMLPlug blocks any files with .gif, .jpg, .jpeg, .png, .pdf,\n";
-    print STDERR "                          .rtf or .css file extensions.\n";
     print STDERR "   -keep_head             Don't remove headers from html files.\n";
     print STDERR "   -no_metadata           Don't attempt to extract any metadata from files.\n";
@@ -82 +74 @@
 sub new {
     my $class = shift (@_);
-    my $self = new BasPlug (@_);
+    my $self = new BasPlug ("HTMLPlug", @_);
 
     if (!parsargv::parse(\@_, 
-             q^process_exp/.*/(?i)\.html?$^, \$self->{'process_exp'},
              q^nolinks^, \$self->{'nolinks'},
-             q^block_exp/.*/(?i)\.(gif|jpe?g|png|pdf|rtf|css)$^, \$self->{'block_exp'},
              q^keep_head^, \$self->{'keep_head'},
              q^no_metadata^, \$self->{'no_metadata'},
@@ -95 +85 @@
              q^rename_assoc_files^, \$self->{'rename_assoc_files'},
              "allow_extra_options")) {
-
+    
     &print_usage();
     die "\n";
     }
-
+    
     $self->{'aux_files'} = {};
     $self->{'dir_num'} = 0;
     $self->{'file_num'} = 0;
-
+    
     return bless $self, $class;
 }
 
-sub is_recursive {
-    my $self = shift (@_);
-
-    return 0; # this is not a recursive plugin
-}
-
-# return number of files processed, undef if can't process
-# Note that $base_dir might be "" and that $file might 
-# include directories
-sub read {
-    my $self = shift (@_);
-    my ($pluginfo, $base_dir, $file, $metadata, $processor) = @_;
-
-    my $filename = &util::filename_cat($base_dir, $file);
-    return 0 if $filename =~ /$self->{'block_exp'}/;
-    if ($filename !~ /$self->{'process_exp'}/ || !-f $filename) {
-    return undef;
-    }
-    $file =~ s/^[\/\\]+//;
-
-    $self->{'verbosity'} = $processor->{'verbosity'};
+
+sub get_default_block_exp {
+    my $self = shift (@_);
+
+    return q^(?i)\.(gif|jpe?g|png|pdf|rtf|css)$^;
+}
+
+sub get_default_process_exp {
+    my $self = shift (@_);
+
+    return q^(?i)\.html?$^;
+}
+
+
+# do plugin specific processing of doc_obj
+sub process {
+    my $self = shift (@_);
+    my ($textref, $pluginfo, $base_dir, $file, $metadata, $doc_obj) = @_;
+
     print STDERR "HTMLPlug: processing $file\n" 
     if $self->{'verbosity'} > 1;
 
-    # create a new document
-    my $doc_obj = new doc ($file, "indexed_doc");
     my $cursection = $doc_obj->get_top_section();
-    
-    # read in HTML file ($text will be in utf8)
-    my $text = "";
-    $self->read_file ($filename, \$text);
-
-    if ($text !~ /\w/) {
-    print STDERR "HTMLPlug: ERROR: $file contains no text\n" if $self->{'verbosity'};
-    return 0;
-    }
-
-    $self->extra_metadata ($doc_obj, $cursection, $metadata);
-    $self->extract_metadata (\$text, $metadata, $doc_obj, $cursection) 
+
+    $self->extract_metadata ($textref, $metadata, $doc_obj, $cursection) 
     unless $self->{'no_metadata'};
 
@@ -158 +134 @@
     # remove header and footer
     if (!$self->{'keep_head'}) {
-    $text =~ s/^.*?<body[^>]*>//is;
-    $text =~ s/(<\/body[^>]*>|<\/html[^>]*>)//isg;
+    $$textref =~ s/^.*?<body[^>]*>//is;
+    $$textref =~ s/(<\/body[^>]*>|<\/html[^>]*>)//isg;
     }
 
@@ -166 +142 @@
 
     # usemap="./#index" not handled correctly => change to "#index"
-    $text =~ s/(<img[^>]*?usemap\s*=\s*\"?)([^\">\s]+)(\"?[^>]*>)/
+    $$textref =~ s/(<img[^>]*?usemap\s*=\s*\"?)([^\">\s]+)(\"?[^>]*>)/
         $self->replace_usemap_links($1, $2, $3)/isge;
 
-    $text =~ s/(<(?:a|area|frame|link)\s+[^>]*?(?:href|src)\s*=\s*\"?)([^\">\s]+)(\"?[^>]*>)/
+    $$textref =~ s/(<(?:a|area|frame|link)\s+[^>]*?(?:href|src)\s*=\s*\"?)([^\">\s]+)(\"?[^>]*>)/
         $self->replace_href_links ($1, $2, $3, $base_dir, $file, $doc_obj, $cursection)/isge;
     }
 
     # trap images
-    $text =~ s/(<img[^>]*?src\s*=\s*\"?)([^\">\s]+)(\"?[^>]*>)/
+    $$textref =~ s/(<img[^>]*?src\s*=\s*\"?)([^\">\s]+)(\"?[^>]*>)/
     $self->replace_images ($1, $2, $3, $base_dir, $file, $doc_obj, $cursection)/isge;
-
-    $doc_obj->add_utf8_text ($cursection, $text);
-
-    # add an OID
-    $doc_obj->set_OID();
-
-    # process the document
-    $processor->process($doc_obj);
-
-    return 1; # processed the file
 }