first crack at psm flag

Author: AbeHandler

lib/docsplit.rb

@@ -18,18 +18,28 @@ module Docsplit
 
   DEPENDENCIES  = {:java => false, :gm => false, :pdftotext => false, :pdftk => false, :pdftailor => false, :tesseract => false}
 
+  osd = false
+
   # Check for all dependencies, and note their absence.
   dirs = ENV['PATH'].split(File::PATH_SEPARATOR)
   DEPENDENCIES.each_key do |dep|
     dirs.each do |dir|
       if File.executable?(File.join(dir, dep.to_s))
         DEPENDENCIES[dep] = true
+        if dep.to_s == 'tesseract'  #if the tesseract dependency is found
+          val = %x[ #{'tesseract --list-langs'} 2>&1 >/dev/null ]  #check for the osd plugin.
+          if val =~ /\bosd\b/     #osd will be listed in tesseract --listlangs
+            osd = true
+          end
+        end
         break
       end
     end
   end
 
-  # Raise an ExtractionFailed exception when the PDF is encrypted, or otherwise
+  DEPENDENCIES[:osd] = osd
+
+    # Raise an ExtractionFailed exception when the PDF is encrypted, or otherwise
   # broke.
   class ExtractionFailed < StandardError; end
 

lib/docsplit/command_line.rb

@@ -101,6 +101,9 @@ def parse_options
         opts.on('-r', '--rolling', 'generate images from each previous image') do |r|
           @options[:rolling] = true
         end
+        opts.on('-nod', '--no-orientation-detection', 'turn off automatic orientation detection in tesseract') do |r|
+          @options[:nod] = true
+        end
         opts.on_tail('-v', '--version', 'display docsplit version') do
           puts "Docsplit version #{Docsplit::VERSION}"
           exit

lib/docsplit/text_extractor.rb

@@ -74,7 +74,11 @@ def extract_from_ocr(pdf, pages)
         tiff = "#{tempdir}/#{@pdf_name}.tif"
         escaped_tiff = ESCAPE[tiff]
         run "MAGICK_TMPDIR=#{tempdir} OMP_NUM_THREADS=2 gm convert -despeckle #{MEMORY_ARGS} #{OCR_FLAGS} #{escaped_pdf} #{escaped_tiff} 2>&1"
-        run "tesseract #{escaped_tiff} #{base_path} -l #{@language} 2>&1"
+        psm = "-psm 1"
+        if @nod
+          psm = ""
+        end
+        run "tesseract #{escaped_tiff} #{base_path} -l #{@language} #{psm} 2>&1"
         clean_text(base_path + '.txt') if @clean_ocr
       end
     ensure