Changeset 24971 for main/trunk

Timestamp:

2012-01-24T16:19:58+13:00 (12 years ago)

Author:

ak19

Message:

1. Introduced the util::filepath_to_url_format subroutine which will be used to convert filenames to URL style filenames to match the slashes used in the filename regex-es in extrameta keys used to index into extrameta data structures. 2. Fixed bug on windows where metadata.xml specifies filenames as regex with backslash in front of the file extension's period mark: DirectoryPlugin needed to unregex the filepath before calling fileparse on it, else the escaping backslash would interfere with perl's fileparse routine (only on windows, since backslash also represents a dirsep here). 3. Updated all those perl plugins where the new util::filepath_to_url_format needs to be called so that they use URL style filenames (thereafter regexed) to index into the extrameta data structures.

Location:

main/trunk/greenstone2/perllib

Files:

• plugins/DSpacePlugin.pm (modified) (2 diffs)
• plugins/DirectoryPlugin.pm (modified) (2 diffs)
• plugins/EmbeddedMetadataPlugin.pm (modified) (2 diffs)
• plugins/LOMPlugin.pm (modified) (2 diffs)
• plugins/MetadataCSVPlugin.pm (modified) (2 diffs)
• plugins/MetadataXMLPlugin.pm (modified) (1 diff)
• plugins/OAIPlugin.pm (modified) (2 diffs)
• util.pm (modified) (3 diffs)

main/trunk/greenstone2/perllib/plugins/DSpacePlugin.pm

@@ -50 +50 @@
 use ReadTextFile;
 use plugin;
+use util;
 use XMLParser;
 use strict;
@@ -249 +250 @@
     my @fname_list = map { "(".$_->{'file'}.")" } @$fnamemime_list;
     my $fname_re = join("|",@fname_list);
+    
+    # Indexing into the extrameta data structures requires the filename's style of slashes to be in URL format
+    # Then need to convert the filename to a regex, no longer to protect windows directory chars \, but for
+    # protecting special characters like brackets in the filepath such as "C:\Program Files (x86)\Greenstone".
+    $fname_re = &util::filepath_to_url_format($fname_re); # just in case there are slashes in there 
+    
     $fname_re =~ s/\./\\\./g;
 

main/trunk/greenstone2/perllib/plugins/DirectoryPlugin.pm

@@ -482 +482 @@
     # filter out any extrametakeys that mention subdirectories and store
     # for later use (i.e. when that sub-directory is being processed)
-    foreach my $ek (@extrametakeys) {
-    my ($subdir_re,$extrakey_dir) = &File::Basename::fileparse($ek);
-    $extrakey_dir = &util::unregex_filename($extrakey_dir);
-
-    my $dirsep_re = &util::get_re_dirsep();
-
-    my $ek_non_re = &util::unregex_filename($ek);
+    foreach my $ek (@extrametakeys) { # where each Extrametakey (which is a filename) is stored as a regex
+    my $ek_non_re = &util::unregex_filename($ek); # unregex it. This step is *also* required for the fileparse operation since the file suffix
+                                                  # will be specified as \.txt and we don't want fileparse to split it there making \.txt the "filename"
+    my ($subdir_re,$extrakey_dir) = &File::Basename::fileparse($ek_non_re);     
+    my $dirsep_re = &util::get_re_dirsep(); 
     if ($ek_non_re =~ m/$dirsep_re/) { # specifies at least one directory
         my $md = &extrametautil::getmetadata(\%extrametadata, $ek);
@@ -494 +492 @@
 
         my $subdir_extrametakeys = $self->{'subdir_extrametakeys'};
+        
+    # Indexing into the extrameta data structures requires the filename's style of slashes to be in URL format
+    # Then need to convert the filename to a regex, no longer to protect windows directory chars \, but for
+    # protecting special characters like brackets in the filepath such as "C:\Program Files (x86)\Greenstone".
+    $subdir_re = &util::filepath_to_url_format($subdir_re); # Possibly not necessary since subdir_re is the last segment of the filepath $ek #### 
+    $subdir_re = &util::filename_to_regex($subdir_re);      # Escape any special chars like brackets and . in subdir_re
         
         my $subdir_rec = { 're' => $subdir_re, 'md' => $md, 'mf' => $mf };

main/trunk/greenstone2/perllib/plugins/EmbeddedMetadataPlugin.pm

@@ -30 +30 @@
 use BasePlugin;
 use extrametautil;
+use util;
 
 use Encode;
@@ -325 +326 @@
     }
 
-    # Protect windows directory chars \
-    $file = &util::filename_to_regex($file); #### 
+    # Indexing into the extrameta data structures requires the filename's style of slashes to be in URL format
+    # Then need to convert the filename to a regex, no longer to protect windows directory chars \, but for
+    # protecting special characters like brackets in the filepath such as "C:\Program Files (x86)\Greenstone".
+    $file = &util::filepath_to_url_format($file);
+    $file = &util::filename_to_regex($file); 
     
     # Associate the metadata now

main/trunk/greenstone2/perllib/plugins/LOMPlugin.pm

@@ -36 +36 @@
 use MetadataPass;
 use MetadataRead;
+use util;
 use XMLParser;
 use Cwd;
@@ -166 +167 @@
     $file_re = $tail;
     }
+    
+    # Indexing into the extrameta data structures requires the filename's style of slashes to be in URL format
+    # Then need to convert the filename to a regex, no longer to protect windows directory chars \, but for
+    # protecting special characters like brackets in the filepath such as "C:\Program Files (x86)\Greenstone".
+    $file_re = &util::filepath_to_url_format($file_re);
     $file_re = &util::filename_to_regex($file_re);
     $self->{'lom_srcdoc'} = undef; # reset for next file to be processed

main/trunk/greenstone2/perllib/plugins/MetadataCSVPlugin.pm

@@ -36 +36 @@
 use extrametautil;
 use multiread;
+use util;
 
 use Encode;
@@ -205 +206 @@
 
     # Associate the metadata now
+    # Indexing into the extrameta data structures requires the filename's style of slashes to be in URL format
+    # Then need to convert the filename to a regex, no longer to protect windows directory chars \, but for
+    # protecting special characters like brackets in the filepath such as "C:\Program Files (x86)\Greenstone".
+    $csv_line_filename = &util::filepath_to_url_format($csv_line_filename);
     $csv_line_filename = &util::filename_to_regex($csv_line_filename);
 

main/trunk/greenstone2/perllib/plugins/MetadataXMLPlugin.pm

@@ -328 +328 @@
     if ($element eq "FileSet") {
     foreach my $target (@{$self->{'saved_targets'}}) {
+    
+        # Indexing into the extrameta data structures requires the filename's style of slashes to be in URL 
+        # format. But if metadata.xml was user-constructed, its FileName fields (stored in the saved_targets 
+        # map) could represent pathnames in Windows format using \ as file separator, instead of pathnames 
+        # being in URL format as Greenstone would've generated. Need to convert such filepaths to URL style 
+        # paths. Then need convert this to a regex to protect special characters like brackets in the 
+        # filepath, such as "C:\Program Files (x86)\Greenstone".
+        
+        $target = &util::unregex_filename($target); # filename is a regex, before converting it to URL form, unregex it
+        $target = &util::filepath_to_url_format($target); # convert to URL form
+        $target = &util::filename_to_regex($target); # regex it. Now we have regexed URL form of filename
+    
         my $file_metadata = &extrametautil::getmetadata($self->{'metadataref'}, $target);
         my $saved_metadata = $self->{'saved_metadata'};

main/trunk/greenstone2/perllib/plugins/OAIPlugin.pm

@@ -38 +38 @@
 use metadatautil;
 use MetadataRead;
+use util;
 
 # methods with identical signatures take precedence in the order given in the ISA list.
@@ -255 +256 @@
     # return all the metadata we have extracted to the caller.
     # Directory plug will pass it back in at read time, so we don't need to extract it again.
-    # extrametadata keys should be regular expressions
+    
+    # Extrametadata keys should be regular expressions
+    # Indexing into the extrameta data structures requires the filename's style of slashes to be in URL format
+    # Then need to convert the filename to a regex, no longer to protect windows directory chars \, but for
+    # protecting special characters like brackets in the filepath such as "C:\Program Files (x86)\Greenstone".
+    $filename_for_metadata = &util::filepath_to_url_format($filename_for_metadata);
     $filename_for_metadata = &util::filename_to_regex($filename_for_metadata);
+    
     &extrametautil::setmetadata($extrametadata, $filename_for_metadata, $new_metadata);
     &extrametautil::addmetakey($extrametakeys, $filename_for_metadata);

main/trunk/greenstone2/perllib/util.pm

@@ -837 +837 @@
     my $filename = shift (@_);
 
-    # need to put single backslash back to double so that regex works
+    # need to make single backslashes double so that regex works
     $filename =~ s/\\/\\\\/g; # if ($ENV{'GSDLOS'} =~ /^windows$/i);    
     
@@ -1067 +1067 @@
     # convert parameters only to / slashes if Windows
     
-    my $filename_urlformat = $filename;
-    my $within_dir_urlformat = $within_dir;
-    
-    if ($ENV{'GSDLOS'} =~ /^windows$/i) {
-        # Only need to worry about Windows, as Unix style directories already in url-format
-        # Convert Windows style \ => /
-        $filename_urlformat =~ s@\\@/@g;
-        $within_dir_urlformat =~ s@\\@/@g;
-    }
-    
+    my $filename_urlformat = &filepath_to_url_format($filename);
+    my $within_dir_urlformat = &filepath_to_url_format($within_dir);
+
     #if ($within_dir_urlformat !~ m/\/$/) {
         # make sure directory ends with a slash
@@ -1092 +1085 @@
     
     return $filename_urlformat;
+}
+
+# Convert parameter to use / slashes if Windows (if on Linux leave any \ as is,
+# since on Linux it doesn't represent a file separator but an escape char).
+sub filepath_to_url_format
+{
+    my ($filepath) = @_;
+    if ($ENV{'GSDLOS'} =~ /^windows$/i) {
+        # Only need to worry about Windows, as Unix style directories already in url-format
+        # Convert Windows style \ => /
+        $filepath =~ s@\\@/@g;      
+    }
+    return $filepath;
 }