爲什麼我得到TypeError：需要一個整數？

Question

這是我在追蹤呼叫顯示錯誤代碼片段：

def categorize(title): 
    with conn: 
     cur= conn.cursor() 
     title_str= str(title) 
     title_words= re.split('; |, |\*|\n',title_str) 
     key_list= list(dictionary.keys()) 
     flag2= 1 
     for word in title_words: 
     title_letters= list(word) 
     flag1= 1 
     for key in key_list: 
      if key==title_letters[0]: 
       flag1= 0 
       break 

     if flag1== 0: 

     start=dictionary.get(title_letters[0]) 
     end= next_val(title_letters[0]) 

     for i in xrange (start,end): 
      if word==transfer_bag_of_words[i]: 
       flag2= 0 
       break 

     if flag2== 0: 
     cur.execute("select Id from articles where title=title") 
     row_id= cur.fetchone() 
     value= (row_id,'1') 
     s= str(value) 
     f.write(s) 
     f.write("\n") 
     break 

    return 


def next_val(text): 
    for i,v in enumerate(keyList): 
     if text=='t': 
     return len(transfer_bag_of_words) 
     elif v==text: 
     return dictionary[keyList[i+1]] 

這是回溯電話：

Traceback (most recent call last): 
File "categorize_words.py", line 93, in <module> 
query_database() 
File "categorize_words.py", line 45, in query_database 
categorize(row) 
File "categorize_words.py", line 67, in categorize 
for i in xrange (start,end): 
TypeError: an integer is required 

我沒有這裏給出的整個代碼。但我會解釋我正在嘗試做什麼。我試圖從sqlite數據庫中導入一個特定的字段，並檢查字段中的單個字是否與我的程序中已有的特定字符相匹配。我已經對這些單詞進行了簡單的分類，並且使用python字典爲它的索引分配了每一個新字母的開頭。我已經這樣做了，每當我檢查字詞袋中的字時，我就不必循環整個單詞。相反，我可以從這個詞的第一個字母的索引開始循環。

我檢查過，字典中get（）的返回類型是int，函數nextVal也應該返回一個int，因爲len（）和dictionary [keylist [i + 1]]都是int類型。

請幫忙。

編輯

這是我的全部代碼：

import sqlite3 as sql 
import re 

conn= sql.connect('football_corpus/corpus2.db') 

transfer_bag_of_words=['transfer','Transfer','transfers','Transfers','deal','signs','contract','rejects','bid','rumours','swap','moves', 
        'negotiation','negotiations','fee','subject','signings','agreement','personal','terms','pens','agent','in','for', 
        'joins','sell','buy','confirms','confirm','confirmed','signing','renew','joined','hunt','excited','move','sign', 
        'loan','loaned','loans','switch','complete','offer','offered','interest','price','tag','miss','signed','sniffing', 
        'remain','plug','pull','race','targeting','targets','target','eye','sale','clause','rejected', 
        'interested'] 

dictionary={} 
dictionary['a']=0; 
keyList=[] 
f= open('/home/surya/Twitter/corpus-builder/transfer.txt','w') 

def map_letter_to_pos(): 
    pos=0 
    transfer_bag_of_words.sort() 
    for word in transfer_bag_of_words: 
     flag=1 
     letters= list(word) 
     key_list= list(dictionary.keys()) 
     for key in key_list: 
     if key==letters[0]: 
      flag=0 
      break 

     if flag==1: 
     dictionary[letters[0]]=pos 
     pos+=1 
     else: 
     pos+=1 

    keyList= sorted(dictionary.keys()) 

def query_database(): 
    with conn: 
     cur= conn.cursor() 
     cur.execute("select title from articles") 
     row_titles= cur.fetchall() 

     for row in row_titles: 
     categorize(row) 

def categorize(title): 
    with conn: 
     cur= conn.cursor() 
     title_str= str(title) 
     title_words= re.split('; |, |\*|\n',title_str) 
     key_list= list(dictionary.keys()) 
     flag2= 1 
     for word in title_words: 
     title_letters= list(word) 
     flag1= 1 
     for key in key_list: 
      if key==title_letters[0]: 
       flag1= 0 
       break 

     if flag1== 0: 

     start=dictionary.get(title_letters[0]) 
     end= next_val(title_letters[0]) 

     for i in xrange (start,end): 
      if word==transfer_bag_of_words[i]: 
       flag2= 0 
       break 

     if flag2== 0: 
     cur.execute("select Id from articles where title=title") 
     row_id= cur.fetchone() 
     value= (row_id,'1') 
     s= str(value) 
     f.write(s) 
     f.write("\n") 
     break 

    return 


def next_val(text): 
    for i,v in enumerate(keyList): 
     if text=='t': 
     return len(transfer_bag_of_words) 
     elif v==text: 
     return dictionary[keyList[i+1]] 

if __name__=='__main__': 
    map_letter_to_pos() 
    query_database() 

這是下載鏈接到數據庫文件http://wikisend.com/download/702374/corpus2.db

Answer 1

map_letter_to_pos試圖修改全局變量密鑰列表沒有具體說明作爲全局變量，因此它只修改keyList的本地副本，然後丟棄它。這會導致next_val無法迭代，所以它永遠不會到達if elif，並返回None。

end = None 
range(start,end) # None is not an int