Some feeds, such as those from sourceforge.net, will sometimes show an error
page rather than a feed. In this case the feed cache used to be cleared for
that feed, causing all 'old' headlines to be flooded to the channel as soon
as the feed came back online.
This patch hopefully fixes that by only resetting the cache when the returned
page actually contains headlines.
Signed-off-by: James McCoy <jamessan@users.sourceforge.net>
This is better at avoiding repeats than just keeping the last fetch, since some feeds
shuffle items around (like google news search).
Conflicts:
plugins/RSS/config.py
When the feed has a specified encoding, we'll be dealing with unicode objects
in the response from feedparser.parse(). To avoid possible UnicodeErrors, we
need to encode() before handing the string off to other functions, so the
other functions are always dealing with bytestrings instead of bytestrings and
unicode objects. Mixing unicode and bytestrings will cause implicit
conversions of the unicode objects, which will most likely use the wrong
encoding.
Signed-off-by: James McCoy <jamessan@users.sourceforge.net>
Conflicts:
plugins/RSS/plugin.py
When the feed has a specified encoding, we'll be dealing with unicode objects
in the response from feedparser.parse(). To avoid possible UnicodeErrors, we
need to encode() before handing the string off to other functions, so the
other functions are always dealing with bytestrings instead of bytestrings and
unicode objects. Mixing unicode and bytestrings will cause implicit
conversions of the unicode objects, which will most likely use the wrong
encoding.
Signed-off-by: James McCoy <jamessan@users.sourceforge.net>
(cherry picked from commit 964c73f591)
Signed-off-by: Daniel Folkinshteyn <nanotube@users.sourceforge.net>
When the feed has a specified encoding, we'll be dealing with unicode objects
in the response from feedparser.parse(). To avoid possible UnicodeErrors, we
need to encode() before handing the string off to other functions, so the
other functions are always dealing with bytestrings instead of bytestrings and
unicode objects. Mixing unicode and bytestrings will cause implicit
conversions of the unicode objects, which will most likely use the wrong
encoding.
Signed-off-by: James McCoy <jamessan@users.sourceforge.net>
First: if an rss feed is called without the number of headlines argument,
we now have a default number of items it will output.
before, it tried to stuff the whole rss feed into the channel, which is quite
floody, if more than one 'mores' is set, or if oneToOne is false.
Second: when adding a new feed to announce, it again, tried to stuff the
whole rss feed into the channel, which ran into the same floody conditions
as above. now we have a default number of feeds to output when there's no
cached history.
