"Jan Dudík" <jan.dudik(a)gmail.com> wrote:
>in linked documentation i see
>>Make sure that the root of the rewrite package (the files you just
>>installed in the previous step) and the "pywikibot" directory in that
>>package are both in your PYTHONPATH environment value
>What is PYTHONPATH ?
http://docs.python.org/using/cmdline.html#envvar-PYTHONPATH
Hi, I was testing the rewrite redirect.py on eswiki and everything worked great, except for this <http://manuelt15.pastebin.com/Y4RN1M0d> after saving the last page the bot didn't stop, and after a while it tried to check for "Cambios recientes" (Recent changes) and then it stopped, can you give a check? Thanks
-- Manuel Taveras
_________________________________________________________________
Ahora no hace falta estar conectado para ver tus mails. Chequea tu Hotmail con Windows Live Mail, es fácil. Ver más.
http://www.microsoft.com/latam/windows/windowslive/products/mail.aspx
This is as more about the learning process than the actual task... but I
can't see what's wrong with my regex, and I'd appreciate any pointers.
I want to find cases where a header is not followed by a line break, e.g.
"== External links ==;http://www.americanbamboo.org/GeneralInfo.html
I tried various regex strings, including "([\r\n]==[^\n=]*==)([^=\r ])"
(allowing for the fact that excess spaces on the end aren't important) to
be replaced by "\\1\n\\2"
But I get many false matches. See http://www.pastie.org/913753 for details
of the good and the false matches.
In #regex on freenode I was told to try:
"s/((={2,}).*?\2)(?!\n)([^=]?)/\1\n\2/g"
but it gives the error: "bash: !\n: event not found".
I'm told that it works - http://pastie.org/913765 - but in my terminal I
can't get past the bash error.
Thanks!
--
Chris Watkins
Appropedia.org - Sharing knowledge to build rich, sustainable lives.
blogs.appropedia.orgcommunity.livejournal.com/appropedia
identi.ca/appropedia
twitter.com/appropedia
Hi, I plan on creating an enwiki mirror that will need to be kept
continually up to date. The plan is to download the data dump as a starting
point, and to grab every revision from enwiki as it is being made and insert
the revision into the mirror wiki's database. See
http://en.wikipedia.org/wiki/Wikipedia:Village_pump_%28technical%29#Real-ti…
more details.
Is there already a bot that can do something similar to this? I checked
http://meta.wikimedia.org/wiki/Pywikipediabot/Scripts and its subpage and
didn't see anything along those lines. Thanks, -Tisane
My bot's first test edits in my user space worked fine
(http://en.wikipedia.org/wiki/Special:Contributions/PotatoBot), but all
of a sudden it can't log in any more, nor can I manually log in using
login.py. I get the error message "Login failed. Wrong password or
CAPTCHA answer?". The password is fine, since I can log in on the
Wikipedia site. Changing the password doesn't help, nor does deleting
and re-creating user-config.py. I'm not aware that I changed any
settings. Can someone help? Thanks, Anypodetos
Hi all,
I've been struggling with the same problem Hannes Röst posted about
yesterday, that login.py (r8064) got stuck in the while True-loop on
line 170.
After reading MediaWiki bug 23076[1] closely, I noticed it referred to
a cookie, and debugging showed that the initial login request returns
a set-Cookie-header that is quietly ignored. From what I can tell the
if-statement in line 4979 of wikipedia.py (r8036) fails (because I
don't have any cookies at this point), and thus the cookies sent from
the server never get stored. Commenting out the line so
self.updateCookies(...) is always called fixes the login problem.
I'd appreciate if a developer could look into this and figure out the
proper solution to the problem as I'm unsure whether the if-statement
can safely be removed.
Regards,
Morten
References:
1: https://bugzilla.wikimedia.org/show_bug.cgi?id=23076