merge

Gemfile

@@ -8,7 +8,7 @@ platform :jruby do
   gem "rack"
   gem "tilt"
   gem "rufus-lru"
-  gem "tabula-extractor", '~>0.7.2', :require => "tabula"
+  gem "tabula-extractor", '~>0.7.2', :require => "tabula", :path => '../tabula-extractor'
 
   group :development do
     gem "rake"

Gemfile.lock

@@ -1,3 +1,9 @@
+PATH
+  remote: ../tabula-extractor
+  specs:
+    tabula-extractor (0.7.2-java)
+      trollop (~> 2.0)
+
 GEM
   remote: https://rubygems.org/
   specs:
@@ -27,6 +33,10 @@ DEPENDENCIES
   rack
   rake
   rufus-lru
+<<<<<<< Updated upstream
   tabula-extractor (~> 0.7.2)
+=======
+  tabula-extractor (~> 0.7.2)!
+>>>>>>> Stashed changes
   tilt
   warbler

webapp/static/js/pdf_view.js

@@ -323,6 +323,9 @@ Tabula.PDFView = Backbone.View.extend({
         return [coords.y1, coords.x1, coords.y2, coords.x2].join(',');
     },
 
+    debugTextChunks: function(image) {
+        return this._debugRectangularShapes(image, '/debug/' + this.PDF_ID + '/text_chunks');
+    },
 
     /* functions for the follow-you-around bar */
     total_selections: function(){

webapp/tabula_debug.rb

@@ -25,6 +25,32 @@ class TabulaDebug < Cuba
       }.to_json
     end
 
+    on ":file_id/text_chunks" do |file_id|
+      par = JSON.load(req.params['coords']).first
+      page = par['page']
+
+      pdf_path = File.join(TabulaSettings::DOCUMENTS_BASEPATH, file_id, 'document.pdf')
+      extractor = Tabula::Extraction::ObjectExtractor.new(pdf_path, [page])
+
+      text_elements = extractor.extract.next.get_text([par['y1'].to_f,
+                                                       par['x1'].to_f,
+                                                       par['y2'].to_f,
+                                                       par['x2'].to_f])
+
+      text_chunks = Tabula::TextElement.merge_words(text_elements)
+
+      puts text_chunks.inspect
+
+      res['Content-Type'] = 'application/json'
+      res.write text_chunks.map { |te|
+        { 'left' => te.left,
+          'top' => te.top,
+          'width' => te.width,
+          'height' => te.height,
+          'text' => te.text }
+      }.to_json
+    end
+
 
     on ":file_id/clipping_paths" do |file_id|
       par = JSON.load(req.params['coords']).first