From 2e75fdc6c4a9a14d6b0ec6fb3d96109ccdd880f8 Mon Sep 17 00:00:00 2001
From: killian <killian@VID-01-01.(none)>
Date: Wed, 10 Jul 2013 16:02:39 +0200
Subject: [PATCH] correction Ajouter les phonetisation des racine

---
 liaSoap.py         |  1 +
 processor/Orkis.py | 37 +++++++++++++++++++++++--------------
 2 files changed, 24 insertions(+), 14 deletions(-)

diff --git a/liaSoap.py b/liaSoap.py
index 8bfccc3..e4e784e 100644
--- a/liaSoap.py
+++ b/liaSoap.py
@@ -16,6 +16,7 @@ class getPhonService(ServiceBase):
     def get_phon(string):
         orkis=Orkis(string)
         orkis.getDico()
+	print(unicode(orkis))
         return unicode(orkis)
 
 application = Application([getPhonService],
diff --git a/processor/Orkis.py b/processor/Orkis.py
index eeea131..c1f588b 100644
--- a/processor/Orkis.py
+++ b/processor/Orkis.py
@@ -6,11 +6,11 @@ from LiaTools import *
 class Orkis(baseProcessor):
     """ Processor for Orkis """
     def __init__(self,dirtyString):
+	self.lem=u""
         self.tagger=Tagger()
         self.phoner=Phoner()
         self.dico ={}
         self.string=dirtyString
-	print self.string
     def isReady(self):
         self.phoner.isReady()
         self.tagger.isReady()
@@ -29,30 +29,39 @@ class Orkis(baseProcessor):
         stopword=StopWord()
         self.string=stopword.RemoveStopList(self.string)
     def insertLem(self):
-        self.cleanString=self.tagger.clean(self.string)
-        taggedString=self.tagger.tagg(self.cleanString)
+	self.lem=u""
+        self.cleanString=self.tagger.clean(self.string).rstrip()
+        taggedString=self.tagger.tagg(self.cleanString).rstrip()
         self.tableLem = taggedString.rstrip().split("\n")
         for line in taggedString.rstrip().split("\n"):
-	    if not re.match(r's>',line):
-            	table = line.rstrip().split(" ")
-            	if not table[0] in self.dico :
-                	self.dico[table[0]]=[set(),set()]
-            	self.dico[table[0]][0].add(table[2])
+            table = line.rstrip().split(" ")
+	    print("table2" + table[2])
+	    if not table[2].isspace():
+                if not table[0] in self.dico :
+                    self.dico[table[0]]=[set(),set()]
+                self.dico[table[0]][0].add(table[2])
+	        self.lem = self.lem +"\n"+ table[2]
     def insertPhon(self):
-        phonedString=self.phoner.phon(self.cleanString)
+	prephonedString=self.cleanString + self.lem.rstrip()
+        phonedString=self.phoner.phon(self.cleanString.rstrip() +self.lem.rstrip())
         self.tablephon= phonedString.rstrip().split("\n")
         for line in phonedString.rstrip().split("\n"):
 	    if not re.match(r's>',line):
                 table = line.rstrip().split(" ")
-                if table[0] in self.dico:
+                if table[0] in self.dico and not table[1].isspace() :
                     self.dico[table[0]][1].add(table[1])
+                elif table[0] not in self.dico and not table[1].isspace() :
+                    for mot,sets in self.dico.iteritems():
+		        if table[0] in sets[0]:
+                            self.dico[mot][1].add(table[1])
+        
     def getDico(self):
         self.clean()
         self.insertLem()
         self.insertPhon()
-        table=[]
+        self.table=[]
         for i,v in self.dico.iteritems():    
-            if not re.match(r"<s>",i):
+            if not re.match(r".s>",i):
                 list=[]          
                 list.append(i)
                 for indice in v[0]:
@@ -61,5 +70,5 @@ class Orkis(baseProcessor):
                     list.append(indice)
                 ligne= " ".join(list)     
 		
-                table.append(ligne) 
-        return "\n".join(table)
+                self.table.append(ligne) 
+        return "\n".join(self.table)
-- 
1.8.2.3