I am running the latest dev versions of all FeedAPI related modules including Feed API mapper and this module.
It works fine with urls that do not contain a symbols such as ? = & :. But for urls with a '?' there is some issue.

For example http://rss.cnn.com/rss/cnn_topstories.rss gives me a nice long list of elements in the 'Feed item' dropdown on the node/[id]/map page but http://www.curriki.org/xwiki/bin/view/Main/WebRss?xpage=rdf just gives me the standard author and tag options.

I have also tried this with other urls such as:
http://www.pustakalaya.org/list.php?cat=quick_filter&sort_by=searchKey0&...
http://activities.sugarlabs.org/en-US/sugar/browse/type:1/cat:all/format...

I am not sure which symbols do not work but ideally any symbols should work.

Comments

e2thex’s picture

e2thex commented
Assigned: Unassigned » e2thex

I do not see anything of the top of my head, (but might be a curl issue) I will take a look

chrowe’s picture

chrowe commented

Thanks,

A quick test of curl -C - -O http://www.pustakalaya.org/list.php?cat=quick_filter&sort_by=searchKey0&search_keys0=&tpl=3 seems to work on my server.

e2thex’s picture

e2thex commented

Ok I do not have any problem pulling

http://activities.sugarlabs.org/en-US/sugar/browse/type:1/cat:all/format...

which is an rss feed

but

http://www.curriki.org/xwiki/bin/view/Main/WebRss?xpage=rdf

is a rdf file which is FeedAPI Exhaustive Parser is not set up to parse (The main issue is that there is a channel node (like rss) but in rss the item nodes are children of the channel in in this file they are siblings (although this seems like the correct way to represent rdf))

http://www.pustakalaya.org/list.php?cat=quick_filter&sort_by=searchKey0&...

is also in a different format with no channel/feed node and the items right off the the base.

While it might be possible to add a hook to allow other to support these different file layouts, I think in general the module was meant to parse rss and atom formated files.

If anyone want to look at the code the relevant items happen on line 30 to 43. Where the code finds the items to parse.

I am thinking I could add a hook before the default channel finders so people could cut of the item in different ways. But it might better to try and find an rss feed for your info.