Python:I6.py

#!/usr/bin/env python
# -*- coding: utf-8 -*-
import os, sys, re, traceback
sys.path.append(os.environ['HOME']+'/pywikipedia')
import wikipedia, catlib, pagegenerators
 
 
def bufline(ch=u'*', clr=10, L=40):
    wikipedia.output(ch*L,colors=[clr for i in range(L)])
 
 
def colorize(text):
    colors = [None for i in range(len(text))]
 
    RX = re.compile(r'{{[^}]*}}',re.DOTALL)
    for match in RX.finditer(text):
        for i in range(match.start(),match.end()):
            colors[i] = 14
 
    wikipedia.output(text, colors=colors)
 
 
def halt_brackets(txt, lvl, open, close):
    txt = list(txt)
    ret = u''
    while True:
        ch = txt.pop(0)
        ret += ch
        if ch==open:
            lvl += 1
        elif ch==close:
            lvl -= 1
        if len(txt)==0 or lvl==0:
            return ret, ''.join(txt)
 
 
def delink(pg, image):
    wikipedia.output(u'Delinking from [[%s]]...' % (pg.title()))
    old_txt = pg.get()
    new_txt = u''
 
    R = r'^(?P<pre>.*?)(?P<img>(?:Image:)?%s)(?P<post>.*)$' % re.sub('[ _]','[ _]',image.titleWithoutNamespace().replace('.',r'\.'))
    #print R
    rx = re.compile(R,re.IGNORECASE)
 
    for L in old_txt.split('\n'):
        M = rx.search(L)
        newL = ''
        if M:
            #print M.groups()
            pre = M.group('pre')
            img = M.group('img')
            post = M.group('post')
            if pre == '':
                wikipedia.output(u'Image inside a <gallery>')
                newL = u'<!--Removed deleted image: ' + img + post + u'-->'
            elif re.search(r'{{ *(?:audio|listen) *\| *(?: *filename *= *)?$',pre):
                wikipedia.output(u'Media file inside {{audio}} or {{listen}}')
                rest1, rest2 = halt_brackets(post,2,'{','}')
                newL = re.sub(r'{{ *((?:audio|listen) *(?:\| *filename *= *|\|))$',r'<!--Removed deleted file: {{\1',pre) + img + rest1 + '-->' + rest2
            elif re.search(r'\[\[ *$',pre):
                wikipedia.output(u'Regular image inside brackets')
                rest1, rest2 = halt_brackets(post,2,'[',']')
                newL = re.sub(r'\[\[ *$','',pre) + u'<!--Removed deleted image: [[' + img + rest1 + '-->' + rest2
            elif re.search('= *$',pre):
                wikipedia.output(u'Image inside infobox, no brackets')
                newL = pre + u'<!--Removed deleted image: ' + img + u'-->' + post
            elif re.search(r'\[\[[^\]\[]*$',pre):
                pass #We caught a suffix, eg. [[Image:spam eggs.jpg]] while searching for [[Image:eggs.jpg]]
            else:
                return False #It's there, but we can't remove it
        if newL:
            new_txt += newL + '\n'
        else:
            new_txt += L + '\n'
    if old_txt != new_txt:
        wikipedia.showDiff(old_txt,new_txt)
        pg.put(
                new_txt,
                comment = u'Removed image (%s) deleted per [[WP:CSD#I6]]' % image.titleWithoutNamespace(),
                minorEdit = True,
                )
        return True
    else:
        return False
 
 
def delink_and_delete(img):
    wikipedia.output(u'Delinking image...')
 
    for pg in img.usingPages():
        if pg.namespace()==0:
            rm = delink(pg,img)
            if not rm:
                wikipedia.output(u'Failed to delink. Aborting.')
                return
 
    reason = u'Image missing fair use rationale ([[WP:CSD#I6]])'
    if img.usingPages():
        reason += u' (removed from: %s)' % u', '.join([u'[[%s]]' % pg.title() for pg in img.usingPages()])
    wikipedia.output(u'Deleting image...')
    img.delete(
            reason = reason,
            prompt = False,
            )
    talk = img.toggleTalkPage()
    if talk.exists():
        wikipedia.output(u'Deleting talk page...')
        talk.delete(
                reason = u'Talk page of a deleted image [[WP:CSD#G8]]',
                prompt = False,
                )
 
 
def main(day, startAt):
    Site = wikipedia.getSite()
    cat = catlib.Category(Site,'Category:Images with no fair use rationale as of ' + day)
 
    for pg in cat.articles():
        if startAt and pg.title() < 'Image:'+startAt:
            continue
        try:
            img = wikipedia.ImagePage(Site,pg.title())
            bufline(ch=u'=', clr=11, L=70)
            wikipedia.output(u'[[%s]]\n\tImage page description follows (%d bytes):' % (img.title(),len(img.get())))
            bufline(ch=u'=', clr=12, L=70)
            colorize(img.get())
            bufline(ch=u'=', clr=12, L=70)
            wikipedia.output(u'Used in: ' + ', '.join(['[[%s]]' % p.title() for p in img.usingPages()]))
            bufline(ch=u'=', clr=12, L=70)
 
            ch = wikipedia.inputChoice('Delete image?',['y','n'],['y','n'])
 
            if ch=='y':
                delink_and_delete(img)
 
        except wikipedia.Error:
            traceback.print_exc()
 
 
if __name__ == '__main__':
    try:
        day = sys.argv[1]
        try:
            startAt = sys.argv[2]
        except:
            startAt = None
        main(day, startAt)
    finally:
        wikipedia.stopme()