Python:CorrectTyop.py

# DRAFT
# TO BE TESTED
# ROBOT REQUESTED BY MEMBERS OF ZH:WIKIPEDIA
# c.f.:
#http://zh.wikipedia.org/w/index.php?title=Wikipedia:%E6%9C%BA%E5%99%A8%E4%BA%BA/%E6%8F%90%E8%AE%AE&variant=zh-tw
# 自動把「膺品」改成「贗品」的機器人
 
 
#SET-UP
pageNameList=[u'數學',
              u'物理',
              u'生物',
              u'音樂',] # LIST YOUR PAGES HERE
 
 
# LIST OF CONVERSIONS 
typoList= [ (ur'膺品',ur'贗品'),
            (ur'修茸',ur'修葺'),
            (ur'謚號',ur'諡號'),
          ]
 
#IMPORTS
import re
import wikipedia
 
# COMPILE THE REGEX LIST
typoXList=[]
for i,j in typoList:
  typoXList.append((re.compile(i), flags=re.U),j) # ADD THE PAIR (REGEX,REPLACEMENT) TO THE END OF THE LIST
 
# GET THE SITE OBJECT
site = wikipedia.getSite()
 
# MAIN LOOP
for pageName in pageNameList:
  print '-------------------------------------------------------\n'
  print '[[' + pageName + ']]:'
  page = wikipedia.Page( site, pageName) # CREATE PAGE OBJECT
  text=page.get()               # GET THE TEXT OF THE PAGE
  typoFound = False             #SET FLAG
  for r,s in typoXList:
    text, count= r.subn(s,text)        # REPLACE ANY OCCURANCE OF r IN text BY s
    if count > 0: 
       print count, 'typo found, converted to', s
       typoFound = True
  if typoFound:
    page.put(text,'機修錯別字:[[botwiki:Python:CorrectTyop.py]]')
    print('page [[' + pageName + ']] modified. \n')
 
wikipedia.stopme()