Changeset 33299

Timestamp:

2019-07-04T19:10:03+12:00 (5 years ago)

Author:

ak19

Message:

1. Committing Dr Bainbridge's fix to remove duplicated heading titles in sectioned HTML produced from sectioned Word documents by StructuredHTMLPlugin. Committing this prior to debugging other issues. 2. Minor change to unicode.pm, improved description for function

Location:

main/trunk/greenstone2/perllib

Files:

• plugins/StructuredHTMLPlugin.pm (modified) (2 diffs)
• unicode.pm (modified) (1 diff)

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

@@ -159 +159 @@
     $body_text =~ s/(<p[^>]*><o:p>&nbsp;<\/o:p><\/p>)//isg;
     
-    # what was the following line for. effectively unused. do we need it??
+    # what was the following line for. effectively unused. do we need it??
     #$section_text .= "<!--\n<Section>\n-->\n";
     #my $top_section_tag = "<!--\n<Section>\n-->\n";
@@ -179 +179 @@
     foreach $hc ( @h_split )
     {
-    if ($hc =~ m/^([1-3])\s*.*?>(.*)$/s)
-    {
-        my $new_hnum = $1;
-        my $hc_after = $2;
-        
-        if ($hc_after =~ m/^(.*?)<\/h$new_hnum>/is)
-        {
-        my $h_text = $1;
-        $hc =~ s/^(\&nbsp\;)+/\&nbsp\;/g;
-        # boil HTML down to some interesting text
-        $h_text =~ s/^[1-3]>//;
-        $h_text =~ s/<\/?.*?>//sg;
-        $h_text =~ s/\s+/ /sg;
-        $h_text =~ s/^\s$//s;
-        $h_text =~ s/(&nbsp;)+\W*/&nbsp;/sg;
-        
-        if ($h_text =~ m/\w+/)
+        if ($hc =~ m/^([1-3])\s*.*?>(.*)$/s)
         {
-            if ($new_hnum > $hnum)
-            {
-            # increase section nesting
-            $hnum++;
-            while ($hnum < $new_hnum)
+            my $new_hnum = $1;
+            my $hc_after = $2;
+            my $filtered_hc = undef;
+            
+            if ($hc_after =~ m/^(.*?)<\/h$new_hnum>(.*)$/is)
             {
-                my $spacing = "  " x $hnum;
-                $section_text .= "<!--\n";
-                $section_text .= $spacing."<Section>\n";
-                $section_text .= "-->\n";
-                $hnum++;
+                my $h_text = $1;
+                $filtered_hc = $2; # This represents the remainder of $hc, after the (e.g.) <h2>xxxx</h2> element
+                
+                $hc =~ s/^(\&nbsp\;)+/\&nbsp\;/g;
+                # boil HTML down to some interesting text
+                $h_text =~ s/^[1-3]>//;
+                $h_text =~ s/<\/?.*?>//sg;
+                $h_text =~ s/\s+/ /sg;
+                $h_text =~ s/^\s$//s;
+                $h_text =~ s/(&nbsp;)+\W*/&nbsp;/sg;
+                
+                if ($h_text =~ m/\w+/)
+                {
+                    if ($new_hnum > $hnum)
+                    {
+                    # increase section nesting
+                    $hnum++;
+                    while ($hnum < $new_hnum)
+                    {
+                        my $spacing = "  " x $hnum;
+                        $section_text .= "<!--\n";
+                        $section_text .= $spacing."<Section>\n";
+                        $section_text .= "-->\n";
+                        $hnum++;
+                    }
+                    }
+                    else # ($new_hnum <= $hnum)
+                    {
+                    # descrease section nesting
+                    while ($hnum >= $new_hnum)
+                    {
+                        my $spacing = "  " x $hnum;
+                        $section_text .= "<!--\n";
+                        $section_text .= $spacing."</Section>\n";
+                        $section_text .= "-->\n";
+                        $hnum--;
+                    }
+                    $hnum++;
+                    }
+
+                    my $spacing = "  " x $hnum;
+                    $section_text .= "<!--\n";
+                    $section_text .= $spacing."<Section>\n";
+                    $section_text .= $spacing."  <Description>\n";
+                    $section_text .= $spacing."    <Metadata name=\"Title\">$h_text</Metadata>";
+                    $section_text .= $spacing."  </Description>\n";
+                    $section_text .= "-->\n";
+                    
+                    #print $outhandle $spacing."$h_text\n"
+                    #   if $self->{'verbosity'} > 2;
+                    
+                    $sectionh1++ if ($hnum==1);
+                }
             }
-            }
-            else # ($new_hnum <= $hnum)
-            {
-            # descrease section nesting
-            while ($hnum >= $new_hnum)
-            {
-                my $spacing = "  " x $hnum;
-                $section_text .= "<!--\n";
-                $section_text .= $spacing."</Section>\n";
-                $section_text .= "-->\n";
-                $hnum--;
+            else {
+###         print STDERR "***** hc = <h$hc\n\n";
             }
-            $hnum++;
-            }
-
-            my $spacing = "  " x $hnum;
-            $section_text .= "<!--\n";
-            $section_text .= $spacing."<Section>\n";
-            $section_text .= $spacing."  <Description>\n";
-            $section_text .= $spacing."    <Metadata name=\"Title\">$h_text</Metadata>";
-            $section_text .= $spacing."  </Description>\n";
-            $section_text .= "-->\n";
-            
-            #print $outhandle $spacing."$h_text\n"
-            #   if $self->{'verbosity'} > 2;
-            
-            $sectionh1++ if ($hnum==1);
+            
+            # This can probably be replaced by the first statement in the if-statement, because
+            # $filtered_hc should always be defined when it is assigned above (even if it resolves
+            # to be an empty string)
+            if (defined $filtered_hc) {
+                $section_text .= $filtered_hc;
+            }
+            else {
+                $section_text .= "<h$hc";
+            }
         }
-        }
-        else {
-###     print STDERR "***** hc = <h$hc\n\n";
-        }
-        $section_text .= "<h$hc";
-    }
-    else
-    {
-        $section_text .= "<h$hc";
-    }
+        else
+        {
+            $section_text .= "<h$hc";
+        }
     }
 

main/trunk/greenstone2/perllib/unicode.pm

@@ -765 +765 @@
 }
 
+# perl version of stringToHex
 # Useful method to print UTF8 (or other unicode) for debugging.
 # Characters that are easily displayed (that is, printable ASCII)