Changeset 10168

Timestamp:

2005-06-24T12:16:01+12:00 (19 years ago)

Author:

kjdon

Message:

modified this to use a new xml format. it should work as before on the old format. now inherits from XMLPlug.

File:

• trunk/gsdl/perllib/plugins/PagedImgPlug.pm (modified) (11 diffs)

trunk/gsdl/perllib/plugins/PagedImgPlug.pm

@@ -34 +34 @@
 # document/book.
 #
+#There are two formats for the item files: a plain text format, and an xml 
+#format. You can use either format, and can have both formats in the same 
+#collection if you like. If you use the plain format, you must not start the 
+#file off with <PagedDocument>
+
+#### PLAIN FORMAT
 # The format of the xxx.item file is as follows:
 # The first lines contain any metadata for the whole document
@@ -60 +66 @@
 # should be rotated.
 # 
+
+#### XML FORMAT
+# The xml format looks like the following
+#<PagedDocument>
+#<Metadata name="Title">The Title of the entire document</Metadata>
+#<Page pagenum="1" imgfile="xxx.jpg" txtfile="yyy.jpg">
+#<Metadata name="Title">The Title of this page</Metadata>
+#</Page>
+#... more pages
+#</PagedDocument>
+#PagedDocument contains a list of Pages, Metadata and PageGroups. Any metadata 
+#that is not inside another tag will belong to the document.
+#Each Page has a pagenum (not used at the moment), an imgfile and/or a txtfile.
+#These are both optional - if neither is used, the section will have no content.
+#Pages can also have metadata associated with them.
+#PageGroups can be introduced at any point - they can contain Metadata and Pages and other PageGroups. They are used to introduce hierarchical structure into the document.
+#For example
+#<PagedDocument>
+#<PageGroup>
+#<Page>
+#<Page>
+#</PageGroup>
+#<Page>
+#</PagedDocument>
+#would generate a structure like
+#X
+#--X
+#  --X
+#  --X
+#--X
+#PageGroup tags can also have imgfile/textfile metadata if you like - this way they get some content themselves.
+
+#Currently the XML structure doesn't work very well with the paged document type, unless you use numerical Titles for each section.
+#There is still a bit of work to do on this format:
+#* enable other text file types, eg html, pdf etc
+#* make the document paging work properly
+#* add pagenum as Title unless a Title is present?
+
 # All the supplemetary image amd text files should be in the same folder as 
 # the .item file.
@@ -86 +130 @@
 # Additional metadata can be added into the .item files, alternatively you can 
 # use normal metadata.xml files, with the name of the xxx.item file as the 
-# FileName. 
+# FileName (only for document level metadata).
 
 package PagedImgPlug;
 
-use BasPlug;
+use XMLPlug;
 
 sub BEGIN {
-    @ISA = ('BasPlug');
+    @ISA = ('XMLPlug');
 }
 
@@ -175 +219 @@
         'args'     => $arguments };
 
-
 sub new {
     my ($class) = @_;
     my $plugin_name = shift (@_);
-    my $self = new BasPlug ("PagedImgPlug", @_);
+    $self = new XMLPlug ("PagedImgPlug", @_);
 
     my $option_list = $self->{'option_list'};
@@ -334 +377 @@
     } else {
     $doc_obj->add_metadata ($section, "srclink", 
-                "<a href=\"_httpcollection_/index/assoc/[parent:assocfilepath]/[Image]\">");
-    $doc_obj->add_metadata ($section, "srcicon", "<img src=\"_httpcollection_/index/assoc/[parent:assocfilepath]/[Image]\">");
+                "<a href=\"_httpcollection_/index/assoc/[parent(Top):assocfilepath]/[Image]\">");
+    $doc_obj->add_metadata ($section, "srcicon", "<img src=\"_httpcollection_/index/assoc/[parent(Top):assocfilepath]/[Image]\">");
 
     }
@@ -373 +416 @@
         $doc_obj->add_metadata ($section, "ThumbType", $thumbnailtype);
         $doc_obj->add_metadata ($section, "Thumb", $id."thumb.$thumbnailtype");
-        
-        $doc_obj->add_metadata ($section, "thumbicon", "<img src=\"_httpcollection_/index/assoc/[parent:assocfilepath]/[Thumb]\" width=[ThumbWidth] height=[ThumbHeight]>");
+        if ($top) {
+        $doc_obj->add_metadata ($section, "thumbicon", "<img src=\"_httpcollection_/index/assoc/[assocfilepath]/[Thumb]\" width=[ThumbWidth] height=[ThumbHeight]>");
+        } else {
+        $doc_obj->add_metadata ($section, "thumbicon", "<img src=\"_httpcollection_/index/assoc/[parent(Top):assocfilepath]/[Thumb]\" width=[ThumbWidth] height=[ThumbHeight]>");
+        }
     }
     
@@ -430 +476 @@
         $doc_obj->add_metadata ($section, "screenicon", "<img src=\"_httpcollection_/index/assoc/[assocfilepath]/[Screen]\" width=[ScreenWidth] height=[ScreenHeight]>");
         } else {
-        $doc_obj->add_metadata ($section, "screenicon", "<img src=\"_httpcollection_/index/assoc/{If}{[parent:assocfilepath],[parent:assocfilepath],[assocfilepath]}/[Screen]\" width=[ScreenWidth] height=[ScreenHeight]>");
+        $doc_obj->add_metadata ($section, "screenicon", "<img src=\"_httpcollection_/index/assoc/[parent(Top):assocfilepath]/[Screen]\" width=[ScreenWidth] height=[ScreenHeight]>");
 
         }
@@ -500 +546 @@
 
 sub read {
-    my $self = shift (@_);
+    $self = shift (@_);
     my ($pluginfo, $base_dir, $file, $metadata, $processor, $maxdocs, $total_count, $gli) = @_;
-
     my $outhandle   = $self->{'outhandle'};
     my $smart_block = $self->{'smart_block'};
@@ -523 +568 @@
     return 0; # blocked
     }
-
+    
     if ($filename !~ /$self->{'process_exp'}/ || !-f $filename) {
     return undef;
@@ -532 +577 @@
     print STDERR "<Processing n='$file' p='PagedImgPlug'>\n" if ($gli);
 
-    my ($dir);
-    ($dir, $file) = $filename =~ /^(.*?)([^\/\\]*)$/;
-
-    #process the .item file
-    my $doc_obj = $self->process_item($filename, $dir, $file, $processor);
-
+    # here we need to decide if we have an old text .item file, or a new xml 
+    # .item file - for now the test is if the first non-empty line is 
+    # <PagedDocument> then its xml
+    my $xml_version = 0;
+    open (ITEMFILE, $filename) || die "couldn't open $filename\n";
+    my $line = "";
+    my $num = 0;
+    $line = <ITEMFILE>;
+    while ($line !~ /\w/) {
+    $line = <ITEMFILE>;
+    }
+    chomp $line;
+    if ($line =~ /^<PagedDocument/) {
+    $xml_version = 1;
+    }
+    close ITEMFILE;
+    my $doc_obj;
+    if ($xml_version) {
+
+    $file =~ s/^[\/\\]+//; # $file often begins with / so we'll tidy it up
+    $self->{'file'} = $file;
+    $self->{'filename'} = $filename;
+    $self->{'processor'} = $processor;
+    $self->{'metadata'} = $metadata;
+    $self->{'gli'} = $gli;
+    eval {
+        $@ = "";
+        my $xslt = $self->{'xslt'};
+        if (defined $xslt && ($xslt ne "")) {
+        # perform xslt
+        my $transformed_xml = $self->apply_xslt($xslt,$filename);
+        
+        # feed transformed file (now in memory as string) into XML parser
+        #$self->{'parser'}->parse($transformed_xml);
+        $self->parse_string($transformed_xml);
+        }
+        else {
+        #$self->{'parser'}->parsefile($filename);
+        $self->parse_file($filename);
+        }
+    };
+    
+    if ($@) {
+        
+        # parsefile may either croak somewhere in XML::Parser (e.g. because
+        # the document is not well formed) or die somewhere in XMLPlug or a
+        # derived plugin (e.g. because we're attempting to process a
+        # document whose DOCTYPE is not meant for this plugin). For the
+        # first case we'll print a warning and continue, for the second
+        # we'll just continue quietly
+        
+        print STDERR "**** XML Parse Error is: $@\n";
+        
+        my ($msg) = $@ =~ /Carp::croak\(\'(.*?)\'\)/;
+        if (defined $msg) { 
+        my $outhandle = $self->{'outhandle'};
+        my $plugin_name = ref ($self);
+        print $outhandle "$plugin_name failed to process $file ($msg)\n";
+        }
+
+        # reset ourself for the next document
+        $self->{'section_level'}=0;
+        print STDERR "<ProcessingError n='$file'>\n" if ($gli);
+        return -1; # error during processing
+    }
+    $doc_obj = $self->{'doc_obj'};
+
+    } else {
+    my ($dir);
+    ($dir, $file) = $filename =~ /^(.*?)([^\/\\]*)$/;
+
+    #process the .item file
+    $doc_obj = $self->process_item($filename, $dir, $file, $processor);
+    
+    }
+    
     if ($self->{'cover_image'}) {
     $self->associate_cover_image($doc_obj, $filename);
@@ -578 +693 @@
 
     return 1;
+}
+
+sub xml_start_tag {
+    my $self = shift(@_);
+    my ($expat, $element) = @_;
+    $self->{'element'} = $element;
+    
+    my $doc_obj = $self->{'doc_obj'};
+    if ($element eq "PagedDocument") {
+    $self->{'current_section'} = $doc_obj->get_top_section();
+    } elsif ($element eq "PageGroup" || $element eq "Page") {
+    # create a new section as a child
+    $self->{'current_section'} = $doc_obj->insert_section($doc_obj->get_end_child($self->{'current_section'}));
+    $self->{'num_pages'}++;
+    # assign pagenum as  what??
+    my $pagenum = $_{'pagenum'}; #TODO!!
+    $doc_obj->set_utf8_metadata_element($self->{'current_section'}, 'PageNum', $pagenum);
+    my ($imgfile) = $_{'imgfile'};
+    if (defined $imgfile) {
+        $self->process_image($self->{'base_dir'}.$imgfile, $imgfile, $doc_obj, $self->{'current_section'});
+    }
+    my ($txtfile) = $_{'txtfile'};
+    if (defined($txtfile)) {
+        $self->process_text ($self->{'base_dir'}.$txtfile, $txtfile, $doc_obj, $self->{'current_section'});
+    } else {
+        # otherwise add in some dummy text 
+        $doc_obj->add_text($self->{'current_section'}, &gsprintf::lookup_string("{BasPlug.dummy_text}"));
+    }
+    } elsif ($element eq "Metadata") {
+    $self->{'metadata_name'} = $_{'name'};
+    }
+}
+
+sub xml_end_tag {
+    my $self = shift(@_);
+    my ($expat, $element) = @_;
+    
+    my $doc_obj = $self->{'doc_obj'};
+    if ($element eq "Page" || $element eq "PageGroup") {
+    # move the current section back to the parent
+    $self->{'current_section'} = $doc_obj->get_parent_section($self->{'current_section'});
+    } elsif ($element eq "Metadata") {
+    
+    $doc_obj->add_utf8_metadata ($self->{'current_section'}, $self->{'metadata_name'}, $self->{'metadata_value'});
+    $self->{'metadata_name'} = "";
+    $self->{'metadata_value'} = "";
+
+    }
+    # otherwise we ignore the end tag
+}
+
+
+sub xml_text {
+    my $self = shift(@_);
+    my ($expat) = @_; 
+
+    if ($self->{'element'} eq "Metadata") {
+    $self->{'metadata_value'} .= $_;
+    }
+}
+
+sub xml_doctype {
+}
+
+sub open_document {
+    my $self = shift(@_);
+    
+    # create a new document
+    $self->{'doc_obj'} = new doc ($self->{'filename'}, "indexed_doc");
+    my $doc_obj = $self->{'doc_obj'};
+    $doc_obj->set_OIDtype ($self->{'processor'}->{'OIDtype'});
+    my ($dir, $file) = $self->{'filename'} =~ /^(.*?)([^\/\\]*)$/;
+    $self->{'base_dir'} = $dir;
+    $self->{'num_pages'} = 0;
+    my $topsection = $doc_obj->get_top_section();
+    if ($self->{'doctype'} eq 'paged') {
+    # set the gsdlthistype metadata to Paged - this ensures this document will
+    # be treated as a Paged doc, even if Titles are not numeric
+    
+    $doc_obj->set_utf8_metadata_element ($topsection, "gsdlthistype", "Paged");
+    } else {
+    $doc_obj->set_utf8_metadata_element ($topsection, "gsdlthistype", "Hierarchy");
+    }
+
+    $doc_obj->add_metadata ($topsection, "Source", $file);
+    if ($self->{'headerpage'}) {
+    $doc_obj->add_text($topsection, &gsprintf::lookup_string("{BasPlug.dummy_text}"));
+    }
+
+}
+
+sub close_document {
+    my $self = shift(@_);
+    my $doc_obj = $self->{'doc_obj'};
+        
+    $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Plugin", "$self->{'plugin_type'}");
+    $doc_obj->add_metadata($doc_obj->get_top_section(), "FileFormat", "PagedImg");
+    
+    # add numpages metadata
+    $doc_obj->set_utf8_metadata_element ($doc_obj->get_top_section(), 'NumPages', $self->{'num_pages'});
+
+    # add an OID
+    $doc_obj->set_OID();
+    
 }