Hello all!
I think this patch is needed, also to adapt watchlist.py
to the new style, since all 'wikipedia' should become
replaced by 'pywikibot'. May be someone with write svn
access can commit this?
(please look at the attached file)
In the same process I wrote a completely new module
'botlist.py' which is an analogue to 'watchlist.py' not
completely finished, but works and caches all registred
bot from a site once a day. This gives a speed-up for
my bot. May be this can also be committed?
(please look at [1])
Thanks a lot and greetings!!
DrTrigon
[1]
https://fisheye.toolserver.org/browse/drtrigon/pywikipedia/dtbext/dtbext_botlist.py?r=HEAD
Index: watchlist.py
===================================================================
--- watchlist.py (Revision 8597)
+++ watchlist.py (Arbeitskopie)
@@ -21,28 +21,28 @@
__version__='$Id$'
-import wikipedia
import re, sys, pickle
import os.path
import time
+import wikipedia as pywikibot
cache = {}
def get(site = None):
if site is None:
- site = wikipedia.getSite()
+ site = pywikibot.getSite()
if site in cache:
# Use cached copy if it exists.
watchlist = cache[site]
else:
- fn = wikipedia.config.datafilepath('watchlists',
+ fn = pywikibot.config.datafilepath('watchlists',
'watchlist-%s-%s.dat' % (site.family.name, site.lang))
try:
# find out how old our saved dump is (in seconds)
file_age = time.time() - os.path.getmtime(fn)
# if it's older than 1 month, reload it
if file_age > 30 * 24 * 60 * 60:
- wikipedia.output(u'Copy of watchlist is one month old,
reloading')
+ pywikibot.output(u'Copy of watchlist is one month old,
reloading')
refresh(site)
except OSError:
# no saved watchlist exists yet, retrieve one
@@ -69,15 +69,15 @@
params = {
'action': 'query',
'list': 'watchlist',
- 'wllimit': wikipedia.config.special_page_limit,
+ 'wllimit': pywikibot.config.special_page_limit,
'wlprop': 'title',
}
- wikipedia.output(u'Retrieving watchlist for %s via API.' % repr(site))
- #wikipedia.put_throttle() # It actually is a get, but a heavy one.
+ pywikibot.output(u'Retrieving watchlist for %s via API.' % repr(site))
+ #pywikibot.put_throttle() # It actually is a get, but a heavy one.
watchlist = []
while True:
- data = wikipedia.query.GetData(params, site, sysop=sysop)
+ data = pywikibot.query.GetData(params, site, sysop=sysop)
if 'error' in data:
raise RuntimeError('ERROR: %s' % data)
watchlist.extend([w['title'] for w in data['query']['watchlist']])
@@ -90,10 +90,10 @@
# Save the watchlist to disk
# The file is stored in the watchlists subdir. Create if necessary.
if sysop:
- f = open(wikipedia.config.datafilepath('watchlists',
+ f = open(pywikibot.config.datafilepath('watchlists',
'watchlist-%s-%s-sysop.dat' % (site.family.name, site.lang)),
'w')
else:
- f = open(wikipedia.config.datafilepath('watchlists',
+ f = open(pywikibot.config.datafilepath('watchlists',
'watchlist-%s-%s.dat' % (site.family.name, site.lang)), 'w')
pickle.dump(watchlist, f)
f.close()
@@ -101,11 +101,11 @@
def _refreshOld(site, sysop=False):
# get watchlist special page's URL
path = site.watchlist_address()
- wikipedia.output(u'Retrieving watchlist for %s' % repr(site))
- #wikipedia.put_throttle() # It actually is a get, but a heavy one.
+ pywikibot.output(u'Retrieving watchlist for %s' % repr(site))
+ #pywikibot.put_throttle() # It actually is a get, but a heavy one.
watchlistHTML = site.getUrl(path, sysop=sysop)
- wikipedia.output(u'Parsing watchlist')
+ pywikibot.output(u'Parsing watchlist')
watchlist = []
for itemR in [re.compile(r'<li><input type="checkbox" name="id\[\]"
value="(.+?)" />'), re.compile(r'<li><input name="titles\[\]" type="checkbox"
value="(.+?)" />')]:
for m in itemR.finditer(watchlistHTML):
@@ -115,10 +115,10 @@
# Save the watchlist to disk
# The file is stored in the watchlists subdir. Create if necessary.
if sysop:
- f = open(wikipedia.config.datafilepath('watchlists',
+ f = open(pywikibot.config.datafilepath('watchlists',
'watchlist-%s-%s-sysop.dat' % (site.family.name, site.lang)),
'w')
else:
- f = open(wikipedia.config.datafilepath('watchlists',
+ f = open(pywikibot.config.datafilepath('watchlists',
'watchlist-%s-%s.dat' % (site.family.name, site.lang)), 'w')
pickle.dump(watchlist, f)
f.close()
@@ -126,17 +126,17 @@
def refresh_all(new = False, sysop=False):
if new:
import config
- wikipedia.output('Downloading All watchlists for your accounts in
user-config.py');
+ pywikibot.output('Downloading All watchlists for your accounts in
user-config.py');
for family in config.usernames:
for lang in config.usernames[ family ]:
- refresh(wikipedia.getSite( code = lang, fam = family ),
sysop=sysop )
+ refresh(pywikibot.getSite( code = lang, fam = family ),
sysop=sysop )
for family in config.sysopnames:
for lang in config.sysopnames[ family ]:
- refresh(wikipedia.getSite( code = lang, fam = family ),
sysop=sysop )
+ refresh(pywikibot.getSite( code = lang, fam = family ),
sysop=sysop )
else:
import dircache, time
- filenames =
dircache.listdir(wikipedia.config.datafilepath('watchlists'))
+ filenames =
dircache.listdir(pywikibot.config.datafilepath('watchlists'))
watchlist_filenameR = re.compile('watchlist-([a-z\-:]+).dat')
for filename in filenames:
match = watchlist_filenameR.match(filename)
@@ -144,13 +144,13 @@
arr = match.group(1).split('-')
family = arr[0]
lang = '-'.join(arr[1:])
- refresh(wikipedia.getSite(code = lang, fam = family))
+ refresh(pywikibot.getSite(code = lang, fam = family))
def main():
all = False
new = False
sysop = False
- for arg in wikipedia.handleArgs():
+ for arg in pywikibot.handleArgs():
if arg == '-all' or arg == '-update':
all = True
elif arg == '-new':
@@ -162,16 +162,16 @@
elif new:
refresh_all(new, sysop=sysop)
else:
- refresh(wikipedia.getSite(), sysop=sysop)
+ refresh(pywikibot.getSite(), sysop=sysop)
- watchlist = get(wikipedia.getSite())
- wikipedia.output(u'%i pages in the watchlist.' % len(watchlist))
+ watchlist = get(pywikibot.getSite())
+ pywikibot.output(u'%i pages in the watchlist.' % len(watchlist))
for pageName in watchlist:
- wikipedia.output( pageName, toStdout = True )
+ pywikibot.output( pageName, toStdout = True )
if __name__ == "__main__":
try:
main()
finally:
- wikipedia.stopme()
+ pywikibot.stopme()
_______________________________________________
Pywikipedia-l mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/pywikipedia-l
