User:YiFeiBot/wikiindex size.py: Difference between revisions

readable
No edit summary
(readable)
 
(9 intermediate revisions by the same user not shown)
Line 3: Line 3:
#!/usr/bin/python
#!/usr/bin/python
# -*- coding: utf-8  -*-
# -*- coding: utf-8  -*-
"""
This is not a complete bot; rather, it is a template from which simple
bots can be made. You can rename it to mybot.py, then edit it in
whatever way you want.
The following parameters are supported:
&params;
-summary:XYZ      Set the summary message text for the edit to XYZ, bypassing
                  the predefined message texts with original and replacements
                  inserted.
All other parameters will be regarded as part of the title of a single page,
and the bot will only work on that single page.
"""
#
# (C) Pywikipedia bot team, 2006-2011
#
# Distributed under the terms of the MIT license.
#
#
__version__ = '$Id: wikiindex-size.py 11155 2013-05-13 10:39:02Z xqt $'
__version__ = '$Id: NULL $'
#
#


Line 35: Line 15:
import sys
import sys
import urllib
import urllib
from datetime import datetime


# This is required for the text that is shown when you run this script
# This is required for the text that is shown when you run this script
Line 48: Line 29:
     # script (i.e. wikiindex-size.py in this case)
     # script (i.e. wikiindex-size.py in this case)


     def __init__(self, generator, summary):
     def __init__(self, generator, summary, debug=False):
         """
         """
         Constructor. Parameters:
         Constructor. Parameters:
Line 59: Line 40:
         self.generator = generator
         self.generator = generator
         # init constants
         # init constants
         self.site = pywikibot.getSite(code=pywikibot.default_code)
         self.site = pywikibot.getSite("en", "wikiindex")
         # Set the edit summary message
         # Set the edit summary message
         if summary:
         if summary:
Line 65: Line 46:
         else:
         else:
             self.summary = i18n.twtranslate(self.site, 'basic-changing')
             self.summary = i18n.twtranslate(self.site, 'basic-changing')
        self.debug = debug


     def run(self):
     def run(self):
Line 78: Line 60:
             return
             return


         ################################################################
         newtext = text
         # NOTE: Here you can modify the text in whatever way you want. #
         size_r = None
         ################################################################
         size_r = re.compile(ur"""(?P<all>\{\{\s*[Ss]ize\s*((\|\s*(?P<pages>pages|wiki[ _]pages)\s*=\s*(?P<pages_value>\d*)\s*[^\|\}]*\s*)|(\s*\|\s*(?P<pagesurl>statistics[ _]URL|wiki[ _]statistics[ _]URL)\s*=\s*(?P<pagesurl_value>https?://[^ \|\}\<]*)\s*[^\|\}]*\s*)|(\s*\|\s*(?P<wikifactor>wikiFactor)\s*=\s*(?P<wikifactor_value>\d*)\s*[^\|\}]*\s*)|(\s*\|\s*(?P<wikifactorurl>wikiFactor[ _]URL)\s*=\s*(?P<wikifactorurl_value>http://[^ \|\}\<]*)\s*[^\|\}]*\s*))+\s*\|?\s*\}\}(\s*\([Aa]s\s*of:?\s*(?P<day>\d+)\s*(?P<month>[A-Z][a-z]+)\s*(?P<year>\d+)\s*\)\s*(\<!--[ A-Za-z0-9/]+--\>)?)?)""")


        # If you find out that you do not want to edit this page, just return.
        # Example: This puts the text 'Test' at the beginning of the page.
        #text = 'Test ' + text
        newtext = text
       
        size_r = re.compile(ur"""(?im)(?P<all>\{\{\s*Size\s*((\s*\|\s*(?P<pages>pages|wiki[ _]pages)\s*=\s*(?P<pages_value>\d*)\s*[^\|\}]*\s*)|(\s*\|\s*(?P<pagesurl>statistics[ _]URL|wiki[ _]statistics[ _]URL)\s*=\s*(?P<pagesurl_value>https?://[^ \|\}\<]*)\s*[^\|\}]*\s*)|(\s*\|\s*(?P<wikifactor>wikiFactor)\s*=\s*(?P<wikifactor_value>\d*)\s*[^\|\}]*\s*)|(\s*\|\s*(?P<wikifactorurl>wikiFactor[ _]URL)\s*=\s*(?P<wikifactorurl_value>http://[^ \|\}\<]*)\s*[^\|\}]*\s*))+\s*\|?\s*\}\})\(\s*As[ _]of\s*(?P<day>\d*)\s*(?P<month>)\s*(?P<year>\d*)""")
         wtext = page.get()
         wtext = page.get()
         m = size_r.finditer(wtext)
         m = size_r.finditer(wtext)
       
         all = ""
         all = ""
         newvalues = ""
         newvalues = ""
Line 109: Line 84:
             year = i.group('year') and i.group('year').strip() or ''
             year = i.group('year') and i.group('year').strip() or ''
              
              
            if self.debug:
                pywikibot.output(u"text = " + text)
                pywikibot.output(u"all = " + all)
                pywikibot.output(u"pages = " + pages)
                pywikibot.output(u"pagesurl = " + pagesurl)
                pywikibot.output(u"wikifactor = " + wikifactor)
                pywikibot.output(u"wikifactorurl = " + wikifactorurl)
                pywikibot.output(u"pages_value = " + pages_value)
                pywikibot.output(u"pagesurl_value = " + pagesurl_value)
                pywikibot.output(u"wikifactor_value = " + wikifactor_value)
                pywikibot.output(u"wikifactorurl_value = " + wikifactorurl_value)
                pywikibot.output(u"day = " + day)
                pywikibot.output(u"month = " + month)
                pywikibot.output(u"year = " + year)
             #get new values
             #get new values
             n = re.findall(ur"(https?://[^\|\}\]]+\?action=raw|https?://[^\|\}\]]+:Statistics)", pagesurl_value)
             n = re.findall(ur"(https?://[^\|\}\]]+\?action=raw|https?://[^\|\}\]]+:Statistics)", pagesurl_value)
Line 125: Line 115:
                 if o:
                 if o:
                     if o[0] and int(pages_value) != int(o[0]):
                     if o[0] and int(pages_value) != int(o[0]):
                         summary = u"Robot: Updating size: %s -> %s" %    (pages_value, o[0])
                         self.summary = u"Robot: Updating size: %s -> %s" %    (pages_value, o[0])
                         pages_value = o[0]
                         pages_value = o[0]
                        newtime = True
                     else:
                     else:
                         break
                         break
Line 144: Line 135:
             print wikifactorurl, wikifactorurl_value"""
             print wikifactorurl, wikifactorurl_value"""
              
              
             newvalues = u"""{{Size <!--see Template:Size for full     detail-->
            if newtime:
                dt = datetime.date(datetime.utcnow())
               
                day = dt.strftime('%d')
                month = dt.strftime('%B')
                year = dt.strftime('%Y')
               
             newvalues = u"""{{Size <!--see Template:Size for full detail-->
| %s = %s <!--type the plain number of pages - NO thousands separators-->
| %s = %s <!--type the plain number of pages - NO thousands separators-->
| %s = %s <!--page count source (often a 'Statistics' page); if unknown type 'No'-->
| %s = %s <!--page count source (often a 'Statistics' page); if unknown type 'No'-->
| %s = %s <!--preferred; if unknown leave void; see: Category:wikiFactor for help-->
| %s = %s <!--preferred; if unknown leave void; see: Category:wikiFactor for help-->
| %s = %s <!--wF source (often 'PopularPages', 'Mostvisitedpages' or 'PageHits'); if unknown leave void-->
| %s = %s <!--wF source (often 'PopularPages', 'Mostvisitedpages' or 'PageHits'); if unknown leave void-->
}}""" % (pages and pages or 'pages', pages_value and pages_value or '', pagesurl and pagesurl or 'statistics URL', pagesurl_value and pagesurl_value or '', wikifactor and wikifactor or 'wikiFactor', wikifactor_value and wikifactor_value or '', wikifactorurl and wikifactorurl or 'wikiFactor URL', wikifactorurl_value and wikifactorurl_value or '')
}}(As of: %s %s %s)<!--manually add/amend date when stats are verified and/or updated-->""" % (
                pages or 'pages',
                pages_value or '',
                pagesurl or 'statistics URL',
                pagesurl_value or '',
                wikifactor or 'wikiFactor',
                wikifactor_value or '',
                wikifactorurl or 'wikiFactor URL',
                wikifactorurl_value or '',
                day or '',
                month or '',
                year or '')
             newtext = text.replace(all, newvalues)
             newtext = text.replace(all, newvalues)
          
          
Line 182: Line 191:
             pywikibot.showDiff(page.get(), text)
             pywikibot.showDiff(page.get(), text)
             pywikibot.output(u'Comment: %s' % comment)
             pywikibot.output(u'Comment: %s' % comment)
             choice = pywikibot.inputChoice(
             #choice = pywikibot.inputChoice(
                u'Do you want to accept these changes?',
            #    u'Do you want to accept these changes?',
                ['Yes', 'No'], ['y', 'N'], 'N')
            #    ['Yes', 'No'], ['y', 'N'], 'N')
             if choice == 'y':
             if True:
                 try:
                 try:
                     # Save the page
                     # Save the page
Line 205: Line 214:




class AutoWiSizeBot(WiSizeBot):
    # Intended for usage e.g. as cronjob without prompting the user.
    _REGEX_eol = re.compile(u'\n')
    def __init__(self):
        WiSizeBot.__init__(self, None, None)
    ## @since  10326
    #  @remarks needed by various bots
    def save(self, page, text, comment=None, **kwargs):
        pywikibot.output(u'\03{lightblue}Writing to wiki on %s...\03{default}'
                        % page.title(asLink=True))
        comment_output = comment or pywikibot.action
        pywikibot.output(u'\03{lightblue}Comment: %s\03{default}'
                        % comment_output)
        #pywikibot.showDiff(page.get(), text)
        for i in range(3):
            try:
                # Save the page
                page.put(text, comment=comment, **kwargs)
            except pywikibot.LockedPage:
                pywikibot.output(
                    u"\03{lightblue}Page %s is locked; skipping.\03{default}"
                    % page.title(asLink=True))
            except pywikibot.EditConflict:
                pywikibot.output(
                    u'\03{lightblue}Skipping %s because of edit '
                    u'conflict\03{default}' % (page.title()))
            except pywikibot.SpamfilterError, error:
                pywikibot.output(
                    u'\03{lightblue}Cannot change %s because of spam blacklist '
                    u'entry %s\03{default}' % (page.title(), error.url))
            else:
                return True
        return False
    ## @since  10326
    #  @remarks needed by various bots
    def append(self, page, text, comment=None, section=None, **kwargs):
        if section:
            pywikibot.output(
                u'\03{lightblue}Appending to wiki on %s in section '
                u'%s...\03{default}' % (page.title(asLink=True), section))
            for i in range(3):
                try:
                    # Append to page section
                    page.append(text, comment=comment, section=section,
                                **kwargs)
                except pywikibot.PageNotSaved, error:
                    pywikibot.output(
                        u'\03{lightblue}Cannot change %s because of '
                        u'%s\03{default}' % (page.title(), error))
                else:
                    return True
        else:
            content = self.load(page)    # 'None' if not existing page
            if not content:                # (create new page)
                content = u''
            content += u'\n\n'
            content += text
            return self.save(page, content, comment=comment, **kwargs)
    ## @since  10326
    #  @remarks needed by various bots
    def loadTemplates(self, page, template, default={}):
        """Get operating mode from page with template by searching the template.
          @param page: The user (page) for which the data should be retrieved.
          Returns a list of dict with the templates parameters found.
        """
        self._content = self.load(page)  # 'None' if not existing page
        templates = []
        if not self._content:
            return templates  # catch empty or not existing page
        for tmpl in pywikibot.extract_templates_and_params(self._content):
            if tmpl[0] == template:
                param_default = {}
                param_default.update(default)
                param_default.update(tmpl[1])
                templates.append(param_default)
        return templates
    ## @since  10326
    #  @remarks common interface to bot job queue on wiki
    def loadJobQueue(self, page, queue_security, reset=True):
        """Check if the data queue security is ok to execute the jobs,
          if so read the jobs and reset the queue.
          @param page: Wiki page containing job queue.
          @type  page: page
          @param queue_security: This string must match the last edit
                              comment, or else nothing is done.
          @type  queue_security: string
          Returns a list of jobs. This list may be empty.
        """
        try:
            actual = page.getVersionHistory(revCount=1)[0]
        except:
            pass
        secure = False
        for item in queue_security[0]:
            secure = secure or (actual[2] == item)
        secure = secure and (actual[3] == queue_security[1])
        if not secure:
            return []
        data = self._REGEX_eol.split(page.get())
        if reset:
            pywikibot.output(u'\03{lightblue}Job queue reset...\03{default}')
            pywikibot.setAction(u'reset job queue')
            page.put(u'', minorEdit=True)
        queue = []
        for line in data:
            queue.append(line[1:].strip())
        return queue




Line 350: Line 228:
     # summary message
     # summary message
     editSummary = ''
     editSummary = ''
    debug = False
     start = "!"
     start = "!"


Line 358: Line 237:
         elif arg.startswith('-start:'):
         elif arg.startswith('-start:'):
             start = arg[7:]
             start = arg[7:]
        elif arg == '-debug':
            debug = True
         else:
         else:
             pywikibot.output(u'Unknown argument: %s' % arg)
             pywikibot.output(u'Unknown argument: %s' % arg)


     cat = catlib.Category(pywikibot.getSite(), 'Category:MediaWiki')
     cat = catlib.Category(pywikibot.getSite("en", "wikiindex"), 'Category:MediaWiki')
     gen = pagegenerators.CategorizedPageGenerator(cat, start=start)
     gen = pagegenerators.CategorizedPageGenerator(cat, start=start)


Line 370: Line 251:
         # pages from the wiki simultaneously.
         # pages from the wiki simultaneously.
         gen = pagegenerators.PreloadingGenerator(gen)
         gen = pagegenerators.PreloadingGenerator(gen)
         bot = WiSizeBot(gen, editSummary)
         bot = WiSizeBot(gen, editSummary, debug)
         bot.run()
         bot.run()
     else:
     else:
Line 380: Line 261:
     finally:
     finally:
         pywikibot.stopme()
         pywikibot.stopme()
</pre>
</code>
checkuser, Administrators
2,635

edits