Revision: 7952 Author: xqt Date: 2010-02-23 11:33:54 +0000 (Tue, 23 Feb 2010)

Log Message: ----------- * Documentation for TEMPLATE_PAGE * enable a language code to work on * take month names from mediawiki if locale.nl_langinfo is not availlable * additional regexes needed for other languages * ignore time.tzset() if it's not availlable (bug #1844769)

Modified Paths: -------------- trunk/pywikipedia/archivebot.py

Modified: trunk/pywikipedia/archivebot.py =================================================================== --- trunk/pywikipedia/archivebot.py 2010-02-22 21:26:53 UTC (rev 7951) +++ trunk/pywikipedia/archivebot.py 2010-02-23 11:33:54 UTC (rev 7952) @@ -15,6 +15,44 @@ either basing on the thread's name or then name can contain a counter which will be incremented when the archive reaches a certain size.

+Trancluded template may contain the following parameters: + +{{TEMPLATE_PAGE +|archive = +|algo = +|counter = +|maxarchivesize = +|minthreadsleft = +|minthreadstoarchive = +|archiveheader = +|key = +}} + +Meanings of parameters are: + +archive Name of the page to which archived threads will be put. + Must be a subpage of the current page. Variables are + supported. +algo specifies the maximum age of a thread. Must be in the form + old(<delay>) where <delay> specifies the age in hours or + days like 24h or 5d. + Default ist old(24h) +counter The current value of a counter which could be assigned as + variable. Will be actualized by bot. Initial value is 1. +maxarchivesize The maximum archive size before incrementing the counter. + Value can be given with appending letter like K or M which + indicates KByte or MByte. Default value is 1000M. +minthreadsleft Minimum number of threads that should be left on a page. + Default value is 5. +minthreadstoarchive The minimum number of threads to archive at once. Default + value is 2. +archiveheader Content that will be put on new archive pages as the + header. This parameter supports the use of variables. + Default value is {{talkarchive}} +key A secret key that (if valid) allows archives to not be + subpages of the page being archived. + + Options: -h, --help show this help message and exit -f FILE, --file=FILE load list of pages from FILE @@ -26,6 +64,7 @@ -c PAGE, --calc=PAGE calculate key for PAGE and exit -l LOCALE, --locale=LOCALE switch to locale LOCALE + -L LANG, --lang=LANG set the language code to work on """ # # (C) Misza13, 2006-2007 @@ -182,12 +221,18 @@

def int2month(num): """Returns the locale's full name of month 'num' (1-12).""" - return locale.nl_langinfo(locale.MON_1+num-1).decode('utf-8') + if hasattr(locale, 'nl_langinfo'): + return locale.nl_langinfo(locale.MON_1+num-1).decode('utf-8') + Months = ['january', 'february', 'march', 'april', 'may', 'june', 'july', 'august', 'september', 'october', 'november', 'december'] + return Site.mediawiki_message(Months[num-1])

def int2month_short(num): """Returns the locale's abbreviated name of month 'num' (1-12).""" - return locale.nl_langinfo(locale.ABMON_1+num-1).replace('\xa0', '').decode('utf-8') + if hasattr(locale, 'nl_langinfo'): + return locale.nl_langinfo(locale.ABMON_1+num-1).replace('\xa0', '').decode('utf-8') + Months = ['jan', 'feb', 'mar', 'apr', 'may', 'jun', 'jul', 'aug', 'sep', 'oct', 'nov', 'dec'] + return Site.mediawiki_message(Months[num-1])

def txt2timestamp(txt, format): @@ -267,7 +312,11 @@ TM = re.search(r'(\d\d?). (\S+) (\d\d\d\d) kl.\W*(\d\d):(\d\d) (.*?)', line) #3. joulukuuta 2008 kello 16.26 (EET) if not TM: - TM = re.search(r'(\d\d?). (\S+) (\d\d\d\d) kello \W*(\d\d).(\d\d) (.*?)', line) + TM = re.search(r'(\d\d?). (\S+) (\d\d\d\d) kello \W*(\d\d).(\d\d) (.*?)', line) + if not TM: +# 14:23, 12. Jan. 2009 (UTC) + pat = re.compile(r'(\d\d):(\d\d), (\d\d?). (\S+).? (\d\d\d\d) (UTC)') + TM = pat.search(line) if TM: # wikipedia.output(TM) TIME = txt2timestamp(TM.group(0),"%d. %b %Y kl. %H:%M (%Z)") @@ -287,6 +336,8 @@ TIME = txt2timestamp(TM.group(0),"%H:%M, %b %d, %Y (%Z)") if not TIME: TIME = txt2timestamp(TM.group(0),"%d. %Bta %Y kello %H.%M (%Z)") + if not TIME: + TIME = txt2timestamp(TM.group(0),"%H:%M, %d. %b. %Y (%Z)") if TIME: self.timestamp = max(self.timestamp,time.mktime(TIME)) # wikipedia.output(u'Time to be parsed: %s' % TM.group(0)) @@ -522,6 +573,7 @@

def main(): + global Site, language from optparse import OptionParser parser = OptionParser(usage='usage: %prog [options] [LINKPAGE(s)]') parser.add_option('-f', '--file', dest='filename', @@ -538,6 +590,8 @@ help='calculate key for PAGE and exit', metavar='PAGE') parser.add_option('-l', '--locale', dest='locale', help='switch to locale LOCALE', metavar='LOCALE') + parser.add_option('-L', '--lang', dest='lang', + help='current language code', metavar='lang') parser.add_option('-T', '--timezone', dest='timezone', help='switch timezone to TIMEZONE', metavar='TIMEZONE') (options, args) = parser.parse_args() @@ -548,7 +602,8 @@ if options.timezone: os.environ['TZ'] = options.timezone #Or use the preset value - time.tzset() + if hasattr(time, 'tzset'): + time.tzset()

if options.calc: if not options.salt: @@ -569,6 +624,11 @@ else: force = False

+ if options.lang: + Site = wikipedia.getSite(options.lang) + language = Site.language() + if wikipedia.debug: print Site + for a in args: pagelist = [] if not options.filename and not options.pagename: @@ -595,10 +655,12 @@ Archiver.run() time.sleep(10) except: - wikipedia.output(u'Error occured while processing page [[%s]]' % pg.title()) + wikipedia.output(u'Error occured while processing page %s' % pg.aslink(True)) traceback.print_exc()

if __name__ == '__main__': - main() - wikipedia.stopme() + try: + main() + finally: + wikipedia.stopme()