From 1f3fed2182ef807ffaef99f539df652343c50f14 Mon Sep 17 00:00:00 2001
From: Roberto Rosario <roberto.rosario.gonzalez@gmail.com>
Date: Sun, 13 Jul 2014 11:32:41 -0400
Subject: [PATCH] Update the German OCR cleanup backend to be a class not a
 function

---
 mayan/apps/ocr/lang/deu.py | 67 ++++++++++++++++++++------------------
 1 file changed, 36 insertions(+), 31 deletions(-)

diff --git a/mayan/apps/ocr/lang/deu.py b/mayan/apps/ocr/lang/deu.py
index 16e4e359c7..df20573b09 100644
--- a/mayan/apps/ocr/lang/deu.py
+++ b/mayan/apps/ocr/lang/deu.py
@@ -1,43 +1,48 @@
 # -*- coding: utf-8 -*-
+from __future__ import absolute_import
+
 import re
 
+from . import BackendBase
 
-def check_word(word):
-    ALL_ALPHANUM = re.compile('([0-9a-zäöüß])', re.I)
-    NON_ALPHANUM = re.compile('([^0-9a-zäöüß])', re.I)
-    ALL_ALPHANUM = re.compile('([0-9a-z])', re.I)
-    NON_ALPHANUM = re.compile('([^0-9a-z])', re.I)
 
-    TOO_MANY_VOWELS = re.compile('[aäeioöuü]{4}', re.I)
-    TOO_MANY_CONSONANTS = re.compile('[bcdfghjklmnpqrstvwxyz]{4}', re.I)
-    ALL_ALPHA = re.compile('^[a-z]+$', re.I)
-    # SINGLE_LETTER_WORDS = re.compile('^$', re.I)
+class LanguageBackend(BackendBase):
+    def check_word(word):
+        ALL_ALPHANUM = re.compile('([0-9a-zäöüß])', re.I)
+        NON_ALPHANUM = re.compile('([^0-9a-zäöüß])', re.I)
+        ALL_ALPHANUM = re.compile('([0-9a-z])', re.I)
+        NON_ALPHANUM = re.compile('([^0-9a-z])', re.I)
 
-    #(L) If a string is longer than 40 characters, it is considered as garbage
-    # http://www.duden.de/sprachwissen/sprachratgeber/die-laengsten-woerter-im-dudenkorpus
-    # http://www.duden.de/sprachwissen/sprachratgeber/durchschnittliche-laenge-eines-deutschen-wortes
-    if len(word) > 40:
-        return None
+        TOO_MANY_VOWELS = re.compile('[aäeioöuü]{4}', re.I)
+        TOO_MANY_CONSONANTS = re.compile('[bcdfghjklmnpqrstvwxyz]{4}', re.I)
+        ALL_ALPHA = re.compile('^[a-z]+$', re.I)
+        # SINGLE_LETTER_WORDS = re.compile('^$', re.I)
 
-    #(A) If a string's ratio of alphanumeric characters to total
-    #characters is less than 50%, the string is garbage
-    if len(ALL_ALPHANUM.findall(word)) < len(word) / 2:
-        return None
+        #(L) If a string is longer than 40 characters, it is considered as garbage
+        # http://www.duden.de/sprachwissen/sprachratgeber/die-laengsten-woerter-im-dudenkorpus
+        # http://www.duden.de/sprachwissen/sprachratgeber/durchschnittliche-laenge-eines-deutschen-wortes
+        if len(word) > 40:
+            return None
 
-    #Remove word if all the letters in the word are non alphanumeric
-    if len(NON_ALPHANUM.findall(word)) == len(word):
-        return None
+        #(A) If a string's ratio of alphanumeric characters to total
+        #characters is less than 50%, the string is garbage
+        if len(ALL_ALPHANUM.findall(word)) < len(word) / 2:
+            return None
 
-    #Removed words with too many consecutie vowels
-    if TOO_MANY_VOWELS.findall(word):
-        return None
+        #Remove word if all the letters in the word are non alphanumeric
+        if len(NON_ALPHANUM.findall(word)) == len(word):
+            return None
 
-    #Removed words with too many consecutie consonants
-    if TOO_MANY_CONSONANTS.findall(word):
-        return None
+        #Removed words with too many consecutie vowels
+        if TOO_MANY_VOWELS.findall(word):
+            return None
 
-    # No single letter words in German
-    if len(word) == 1:
-        return None
+        #Removed words with too many consecutie consonants
+        if TOO_MANY_CONSONANTS.findall(word):
+            return None
 
-    return word
+        # No single letter words in German
+        if len(word) == 1:
+            return None
+
+        return word