Utilisateur:Tavernier/cosmetic changes.py
(Redirigé depuis Utilisateur:J.M. Tavernier/cosmetic changes.py)
/*
*/ # -*- coding: utf-8 -*- """ This module can do slight modifications to a wiki page source code such that the code looks cleaner. The changes are not supposed to change the look of the rendered wiki page. WARNING: This module needs more testing! """ import wikipedia, pagegenerators import sys import re # Summary message when using this module as a stand-alone script msg_standalone = { 'de': u'Bot: Kosmetische Änderungen', 'en': u'Robot: Cosmetic changes', 'fr': u'Bot: Retouches cosmétiques', 'he': u'רובוט: שינויים קוסמטיים', 'pt': u'Bot: Mudanças triviais', } # Summary message that will be appended to the normal message when # cosmetic changes are made on the fly msg_append = { 'de': u'; kosmetische Änderungen', 'en': u'; cosmetic changes', 'fr': u'; retouches cosmétiques', 'he': u'; שינויים קוסמטיים', 'pt': u'; mudanças triviais', } deprecatedTemplates = { 'wikipedia': { 'de': [ u'Stub', ] } } class CosmeticChangesToolkit: def __init__(self, site, debug = False, acceptall = False): self.site = site self.debug = debug def change(self, text): """ Given a wiki source code text, returns the cleaned up version. """ oldText = text text = self.standardizeInterwiki(text) text = self.standardizeCategories(text) text = self.cleanUpLinks(text) text = self.cleanUpSectionHeaders(text) text = self.translateAndCapitalizeNamespaces(text) text = self.removeDeprecatedTemplates(text) text = self.resolveHtmlEntities(text) text = self.validXhtml(text) text = self.removeUselessSpaces(text) if self.debug: wikipedia.showDiff(oldText, text) return text def standardizeInterwiki(self, text): """ Makes sure that interwiki links are put to the correct position and into the right order. """ interwikiLinks = wikipedia.getLanguageLinks(text, insite = self.site) text = wikipedia.replaceLanguageLinks(text, interwikiLinks, site = self.site) return text def standardizeCategories(self, text): """ Makes sure that interwiki links are put to the correct position, but does not sort them. """ categories = wikipedia.getCategoryLinks(text, site = self.site) text = wikipedia.replaceCategoryLinks(text, categories, site = self.site) return text def translateAndCapitalizeNamespaces(self, text): """ Makes sure that localized namespace names are used. """ family = self.site.family for nsNumber in family.namespaces: thisNs = family.namespace(self.site.lang, nsNumber) defaultNs = family.namespace('_default', nsNumber) if thisNs != defaultNs: text = wikipedia.replaceExceptNowikiAndComments(text, r'\[\[\s*' + defaultNs + '\s*:(?P<nameAndLabel>.*?)\]\]', r'[[' + thisNs + ':\g<nameAndLabel>]]') if self.site.nocapitalize: for nsNumber in family.namespaces: thisNs = family.namespace(self.site.lang, nsNumber) lowerNs = thisNs[0].lower() + thisNs[1:] # this assumes that all NS names have length at least 2 text = wikipedia.replaceExceptNowikiAndComments(text, r'\[\[\s*' + lowerNs + '\s*:(?P<nameAndLabel>.*?)\]\]', r'[[' + thisNs + ':\g<nameAndLabel>]]') return text def cleanUpLinks(self, text): trailR = re.compile(self.site.linktrail()) # The regular expression which finds links. Results consist of four groups: # group title is the target page title, that is, everything before | or ]. # group section is the page section. It'll include the # to make life easier for us. # group label is the alternative link title, that's everything between | and ]. # group linktrail is the link trail, that's letters after ]] which are part of the word. # note that the definition of 'letter' varies from language to language. self.linkR = re.compile(r'\[\[(?P<titleWithSection>[^\]\|]+)(\|(?P<label>[^\]\|]*))?\]\](?P<linktrail>' + self.site.linktrail() + ')') curpos = 0 # This loop will run until we have finished the current page while True: m = self.linkR.search(text, pos = curpos) if not m: break # Make sure that next time around we will not find this same hit. curpos = m.start() + 1 titleWithSection = m.group('titleWithSection') if not self.site.isInterwikiLink(titleWithSection): # The link looks like this: # link_texttrailing_chars # We only work on namespace 0 because pipes and linktrails work # differently for images and categories. page = wikipedia.Page(self.site, titleWithSection) if page.namespace() == 0: # Replace underlines by spaces, also multiple underlines titleWithSection = re.sub('_+', ' ', titleWithSection) # Remove double spaces titleWithSection = re.sub(' +', ' ', titleWithSection) # Convert URL-encoded characters to unicode titleWithSection = wikipedia.url2unicode(titleWithSection, site = self.site) label = m.group('label') or titleWithSection trailingChars = m.group('linktrail') if trailingChars: label += trailingChars if titleWithSection == label: newLink = "%s" % titleWithSection # Check if we can create a link with trailing characters instead of a pipelink elif len(titleWithSection) <= len(label) and label[:len(titleWithSection)] == titleWithSection and re.sub(trailR, , label[len(titleWithSection):]) == : newLink = "%s%s" % (label[:len(titleWithSection)], label[len(titleWithSection):]) else: # Try to capitalize the first letter of the title. # Maybe this feature is not useful for languages that # don't capitalize nouns... #if not self.site.nocapitalize: if self.site.sitename() == 'wikipedia:de': titleWithSection = titleWithSection[0].upper() + titleWithSection[1:] newLink = "%s" % (titleWithSection, label) text = text[:m.start()] + newLink + text[m.end():] return text def resolveHtmlEntities(self, text): ignore = [ 38, # Ampersand (&) 60, # Less than (<) 62, # Less than (<) 160, # Non-breaking space ( ) - not supported by Firefox textareas ] text = wikipedia.html2unicode(text, ignore = ignore) return text def validXhtml(self, text): text = wikipedia.replaceExceptNowikiAndComments(text, r'
', r'
') return text def removeUselessSpaces(self, text): result = [] multipleSpacesR = re.compile(' +') spaceAtLineEndR = re.compile(' $') preR = re.compile('<pre', re.IGNORECASE) lines = text.split('\n') for line in lines: if len(line) > 0 and line[0] != ' ' and not preR.search(line): line = wikipedia.replaceExceptMathNowikiAndComments(line, multipleSpacesR, ' ') line = wikipedia.replaceExceptMathNowikiAndComments(line, spaceAtLineEndR, ) result.append(line) return '\n'.join(result) def cleanUpSectionHeaders(self, text): for level in range(1, 7): equals = '=' * level text = wikipedia.replaceExceptNowikiAndComments(text, r'\n' + equals + ' *(?P<title>[^=]+?) *' + equals + ' *\r\n', r'\n' + equals + ' \g<title> ' + equals + '\r\n') return text def removeDeprecatedTemplates(self, text): if deprecatedTemplates.has_key(self.site.family.name) and deprecatedTemplates[self.site.family.name].has_key(self.site.lang): for template in deprecatedTemplates[self.site.family.name][self.site.lang]: if not self.site.nocapitalize: template = '[' + template[0].upper() + template[0].lower() + ']' + template[1:] text = wikipedia.replaceExceptNowikiAndComments(text, r'\{\{([mM][sS][gG]:)?' + template + '(?P<parameters>\|[^}]+|)}}', ) return text class CosmeticChangesBot: def __init__(self, generator, acceptall = False): self.generator = generator self.acceptall = acceptall # Load default summary message. wikipedia.setAction(wikipedia.translate(wikipedia.getSite(), msg_standalone)) def run(self): for page in self.generator: try: ccToolkit = CosmeticChangesToolkit(page.site(), debug = True) changedText = ccToolkit.change(page.get()) if changedText != page.get(): if not self.acceptall: choice = wikipedia.inputChoice(u'Do you want to accept these changes?', ['Yes', 'No', 'All'], ['y', 'N', 'a'], 'N') if choice in ['a', 'A']: self.acceptall = True if self.acceptall or choice in ['y', 'Y']: page.put(changedText) except wikipedia.NoPage: print "Page %s does not exist?!" % page.aslink() except wikipedia.IsRedirectPage: print "Page %s is a redirect; skipping." % page.aslink() except wikipedia.LockedPage: print "Page %s is locked?!" % page.aslink() def main(): #page generator gen = None pageTitle = [] for arg in wikipedia.handleArgs(): if arg.startswith('-start:'): gen = pagegenerators.AllpagesPageGenerator(arg[7:]) elif arg.startswith('-ref:'): referredPage = wikipedia.Page(wikipedia.getSite(), arg[5:]) gen = pagegenerators.ReferringPageGenerator(referredPage) elif arg.startswith('-links:'): linkingPage = wikipedia.Page(wikipedia.getSite(), arg[7:]) gen = pagegenerators.LinkedPageGenerator(linkingPage) elif arg.startswith('-file:'): gen = pagegenerators.TextfilePageGenerator(arg[6:]) elif arg.startswith('-cat:'): cat = catlib.Category(wikipedia.getSite(), arg[5:]) gen = pagegenerators.CategorizedPageGenerator(cat) elif arg == '-always': acceptall = True else: pageTitle.append(arg) if pageTitle: page = wikipedia.Page(wikipedia.getSite(), ' '.join(pageTitle)) gen = iter([page]) if not gen: wikipedia.showHelp() else: preloadingGen = pagegenerators.PreloadingGenerator(gen) bot = CosmeticChangesBot(preloadingGen, acceptall) bot.run() if __name__ == "__main__": try: main() finally: wikipedia.stopme()