Further fix to section aggregation. Gets some INASP journals working.

CrossRef · Apr 10, 2012 · 11ad669 · 11ad669
1 parent 1f56231
commit 11ad669
Show file tree

Hide file tree

Showing 3 changed files with 24 additions and 13 deletions.
diff --git a/lib/analysis/sections.rb b/lib/analysis/sections.rb
@@ -85,7 +85,7 @@ def self.include_in pdf
             not (column[:page] == region[:page] && Spatial.contains?(column, region))
           end
 
-          containers.first[:regions] << region unless containers.count.zero?
+          containers.first[:regions] << region unless containers.empty?
         end
 
         parser.after do
@@ -110,19 +110,22 @@ def self.include_in pdf
           merging_region = nil
 
           pages.each_pair do |page, columns|
-            columns.each do |c|
-              column = c[:column]
+            columns.each do |container|
+              column = container[:column]
 
-              c[:regions].each do |region|
+              container[:regions].each do |region|
                 if candidate? pdf, region, column
                   if !merging_region.nil? && match?(merging_region, region)
                     content = Spatial.merge_lines(merging_region, region, {})
+
                     merging_region.merge!(content)
 
                     merging_region[:components] << Spatial.get_dimensions(region)
                   elsif !merging_region.nil?
                     sections << merging_region
-                    merging_region = nil
+                    merging_region = region.merge({
+                      :components => [Spatial.get_dimensions(region)]
+                    })
                   else
                     merging_region = region.merge({
                       :components => [Spatial.get_dimensions(region)]

diff --git a/lib/font_metrics.rb b/lib/font_metrics.rb
@@ -42,7 +42,15 @@ def initialize font
         @ascent = font.ascent
         @descent = font.descent
         @bbox = font.bbox
-        @glyph_width_lookup = proc { |c| font.glyph_width c.codepoints.first }
+        @glyph_width_lookup = proc do |c|
+          begin
+            font.glyph_width c.codepoints.first
+          rescue TypeError => e
+            # It seems some fonts don't have a first char attribute in their
+            # descriptor and this causes problems for pdf-reader.
+            0
+          end
+        end
       end
 
       if not @bbox.nil?

diff --git a/lib/model/regions.rb b/lib/model/regions.rb
@@ -41,7 +41,7 @@ def self.append_line_spacing region
         height_taken = from_top + line[:height]
       end
     end
-    
+
     def self.include_in pdf
       pdf.spatials :regions, :paged => true, :depends_on => [:chunks] do |parser|
         chunks = []
@@ -51,7 +51,7 @@ def self.include_in pdf
           chunks = []
           regions = []
         end
-        
+
         parser.objects :chunks do |chunk|
           y = chunk[:y].floor
 
@@ -72,16 +72,16 @@ def self.include_in pdf
             chunk[:lines] = [Spatial.as_line(chunk)]
             chunk.delete :content
           end
-          
+
           compare_index = 1
           while chunks.count > compare_index
             b = chunks.first
             t = chunks[compare_index]
-              
+
             line_height = b[:line_height]
             line_slop = [line_height, t[:height]].min * pdf.settings[:line_slop]
             incident_y = (b[:y] + b[:height] + line_slop) >= t[:y]
-            
+
             if incident_y && incident(t, b)
               chunks[0] = Spatial.merge t, b, :lines => true
               chunks.delete_at compare_index
@@ -96,7 +96,7 @@ def self.include_in pdf
               compare_index = 1
             end
           end
-          
+
           regions << chunks.first unless chunks.first.nil?
 
           regions.each do |region|
@@ -112,6 +112,6 @@ def self.include_in pdf
         end
       end
     end
-    
+
   end
 end