From 673721ec077645b585c9d15e60a6b7eab3c2362a Mon Sep 17 00:00:00 2001
From: Admin <killian.janod@alumni.univ-avignon.fr>
Date: Tue, 4 Jun 2013 17:08:26 +0200
Subject: [PATCH] ajout  phon il manque plus que le soap

---
 README.md             |  3 ++-
 processor/LiaTools.py | 18 ++++++++++++------
 webtagger.py          | 15 +++++++++++++--
 3 files changed, 27 insertions(+), 9 deletions(-)

diff --git a/README.md b/README.md
index 3d9bdf9..498e021 100644
--- a/README.md
+++ b/README.md
@@ -19,9 +19,10 @@ LIA's webTagger : Web api for the LIA POS TAGGER
 ## Requirement
 
 * [LIA\_TAGGER](http://pageperso.lif.univ-mrs.fr/~frederic.bechet/download.html)
+* [LIA\_PHON](http://pageperso.lif.univ-mrs.fr/~frederic.bechet/download.html)
 * Python > 2.5
 * Flask
-
+* nltk
 ## Advise 
 
 * Virtualenv
diff --git a/processor/LiaTools.py b/processor/LiaTools.py
index e9611d6..3de290b 100644
--- a/processor/LiaTools.py
+++ b/processor/LiaTools.py
@@ -3,6 +3,7 @@ import os
 from BaseProcessor import baseProcessor
 import nltk
 import re
+
 class Tagger(baseProcessor):
     """ a calling to lia_tagg class"""
     def clean(self,dirtyString):
@@ -16,11 +17,15 @@ class Tagger(baseProcessor):
     # This is used beceause lia_tagg deal with iso8859 only
     	return taggedString.decode('iso8859').encode("utf8")
     def lemm(self,cleanString):
- 	taggedString = self.taff(cleanString)       
-        return re.sub(r'<s> ',''," ".join([ x.split().pop(2) for x in taggedString.rstrip().split("\n")]))
+        print " cleannnnn " + cleanString
+ 	taggedString = self.tagg(cleanString)       
+        print "taggs full " + taggedString
+	sub = re.sub(r' </s>','',re.sub(r'<s> ',''," ".join([ x.rstrip().split().pop(2) for x in taggedString.rstrip().split("\n") if x])))
+        print " subbbbb" + sub 
+        return sub
     def isReady(self):
         os.environ["LIA_TAGG"]
-        return true
+        return True
 
 class Phoner(baseProcessor):
     """ a class which call the lia phoner """
@@ -36,10 +41,11 @@ class Phoner(baseProcessor):
     	return taggedString.decode('iso8859').encode("utf8")
     def isReady(self):
 	os.environ["LIA_PHON_REP"]
-        return true
+        return True
+
 class StopWord(baseProcessor):
     def isReady(self):
-        return true
+        return True
     def RemoveStopList(self,rowstring):
         """ Remove from set of word (splited String ) each words in the stoplist and join all of the other in a string """
-        return u" ".join(unicode(value) for value in list(set(test.split()) - set(nltk.corpus.stopwords.words("french"))))
+        return u" ".join(unicode(value) for value in list(set(rowstring.split()) - set(nltk.corpus.stopwords.words("french"))))
diff --git a/webtagger.py b/webtagger.py
index eb600ac..7535a1f 100644
--- a/webtagger.py
+++ b/webtagger.py
@@ -13,12 +13,23 @@ def docs():
 @app.route("/tagger",methods=['POST'])
 def cleaner():
     tagger = Tagger()
+    tagger.isReady()
+    phoner = Phoner()
+    phoner.isReady()
+    stoplist = StopWord()
+    stoplist.isReady()
     # Receive String from post parametre Raw text ( Json )
     dirtyString= request.json[u'string']
     # send the String throught LIA_TAGG script  thank's to pip
     # lia_clean split a word by line et markup the sentences
-    cleanString= tagger.clean(dirtyString)
-    taggedString= tagger.tagg(cleanString)
+    dirtyString = stoplist.RemoveStopList(dirtyString)
+    print " stop list " + dirtyString
+    lemm = tagger.lemm(tagger.clean(dirtyString))
+    print 'les lemm '+ lemm
+    dirtyString = dirtyString+" "+ lemm
+    cleanString= phoner.clean(dirtyString)
+    taggedString= phoner.phon(cleanString)
+    print taggedString
     return taggedString
 if __name__ == '__main__':
     app.debug = True
-- 
1.8.2.3