@0.0.8 , fixed bugs and improve load of languages stopwords files

setup.py

@@ -11,7 +11,7 @@
         'scipy',
         'networkx',
     ],
-    version = '0.0.7',
+    version = '0.0.8',
     description = 'A text summarization and keyword extraction package',
     author = 'Federico Barrios, Federico Lopez, Antonio Sanchez Pineda',
     author_email = '[email protected]',

textrank3/preprocessing/stopwords.py

@@ -1,6 +1,9 @@
 # encoding: UTF-8
 
 import json
+import os.path
+
+base_path = os.path.dirname(__file__)
 
 english = """
 all six eleven just less being indeed over both anyway detail four front already through yourselves fify
@@ -196,25 +199,27 @@
 yhtäällä yhtäältä yhtään yhä yksi yksin yksittäin yleensä ylemmäs yli ylös ympäri älköön älä
 """
 
-def plain_stopwords(json):
-    data = json.load(open(json))
+def plain_stopwords(json_filename):
+    json_file = os.path.abspath(os.path.join(base_path, 'languages', json_filename))
+    data = json.load(open(json_file))
     return ' '.join(data)
 
+
 LANGUAGES = {
-    "english": 'languages/stopwords-en',
-    "german": 'languages/stopwords-de',
-    "spanish": 'languages/stopwords-es',
-    "french": 'languages/stopwords-fr',
-    "italian": 'languages/stopwords-it',
-    "portuguese": 'languages/stopwords-pt',
-    "russian": 'languages/stopwords-ru',
-    "danish": 'languages/stopwords-da',
-    "dutch": 'languages/stopwords-nl',
-    "finnish": 'languages/stopwords-fi',
-    "norwegian": 'languages/stopwords-no',
-    "hungarian": 'languages/stopwords-hu',
-    "romanian": 'languages/stopwords-ro',
-    "swedish": 'languages/stopwords-sv',
+    "english": 'stopwords-en.json',
+    "german": 'stopwords-de.json',
+    "spanish": 'stopwords-es.json',
+    "french": 'stopwords-fr.json',
+    "italian": 'stopwords-it.json',
+    "portuguese": 'stopwords-pt.json',
+    "russian": 'stopwords-ru.json',
+    "danish": 'stopwords-da.json',
+    "dutch": 'stopwords-nl.json',
+    "finnish": 'stopwords-fi.json',
+    "norwegian": 'stopwords-no.json',
+    "hungarian": 'stopwords-hu.json',
+    "romanian": 'stopwords-ro.json',
+    "swedish": 'stopwords-sv.json',
 }
 
 

textrank3/preprocessing/textcleaner.py

@@ -37,8 +37,8 @@
 def set_stemmer_language(language):
     global STEMMER
     if not language in LANGUAGES:
-        raise ValueError("Valid languages are danish, dutch, english, finnish," +
-                 " french, german, hungarian, italian, norwegian, porter, portuguese," +
+        raise ValueError("Valid languages are: danish, dutch, english, finnish, " +
+                 "french, german, hungarian, italian, norwegian, porter, portuguese, " +
                  "romanian, russian, spanish, swedish")
     STEMMER = SnowballStemmer(language)