pos tagging fix

pickling of nltk corpuses in subprocesses fixed

workflows/nltoolkit/lib/tagging_common_parallel.py

@@ -95,7 +95,7 @@ def universal_sentence_tagger_hub(input_dict):
     print "evo nas!!!"
     #parallel for document in adc.documents:
     new_documents=pool.map(
-        partial(tag_document,
+        partial(sentance_tag_a_document,
                 tagger=tagger,
                 tagger_function=tagger_function,
                 args=args,

workflows/textflows.py

@@ -335,6 +335,12 @@ class NltkCorpus():
                 return getattr(self._corpus(),name)()
             return method
 
+    def __repr__(self):
+        return "NltkCorpus wrapper for "+self.corpus_name+" dataset"
+
+    def __getstate__(self):
+        return {'corpus_name': self.corpus_name,'_corpus_methods': self._corpus_methods}
+
 
 class NltkRegexpTokenizer():
     """ Wrapper for Nltk RegexTokenizer. Python's regular expressions are not picklable.