simplify many codes.

[rec10/rec10-git.git] / rec10 / trunk / src / n_gram.py

diff --git a/rec10/trunk/src/n_gram.py b/rec10/trunk/src/n_gram.py
index e653d03..50bbd84 100644 (file)
--- a/rec10/trunk/src/n_gram.py
+++ b/rec10/trunk/src/n_gram.py
@@ -2,39 +2,41 @@
 # coding: UTF-8
 # Rec10 TS Recording Tools
 # Copyright (C) 2009 Yukikaze
+
 import zenhan
-def bigram(str1,str2):
-    str1=zenhan.toHankaku(str1)
-    str2=zenhan.toHankaku(str2)
-    str1=str1.replace(" ","")
-    str2=str2.replace(" ","")
+def bigram(str1, str2):
+    str1 = zenhan.toHankaku(str1)
+    str2 = zenhan.toHankaku(str2)
+    str1 = str1.replace(" ", "")
+    str2 = str2.replace(" ", "")
     """
     bigramによる単語の近さを計算します。
     """
-    gram=[]
-    if len(str1)<2:
+    gram = []
+    if len(str1) < 2:
         gram.append(str1)
     else:
-        tmp=str1[0]
+        tmp = str1[0]
         for x in str1[1:]:
-            gram.append(tmp+x)
-            tmp=x
-    point=0
+            gram.append(tmp + x)
+            tmp = x
+    point = 0
     for x in gram:
-        i=find_gram(x, str2)
-        if i>0:
-            i=90+10*i
+        i = find_gram(x, str2)
+        if i > 0:
+            i = 90 + 10 * i
         else:
-            i=0
-        point=point+i
+            i = 0
+        point = point + i
     return point
-def find_gram(gram,s):
+def find_gram(gram, s):
     """
     s中にあらわれるgramの数を調べる
     """
-    st=s
-    i=0
-    while st.find(gram)!=-1:
-       i=i+1
-       st=st[st.find(gram)+1:]
-    return i
\ No newline at end of file
+    st = s
+    i = 0
+    while st.find(gram) != -1:
+        i = i + 1
+    st = st[st.find(gram) + 1:]
+    return i
+""
\ No newline at end of file