* David Cantrell <[EMAIL PROTECTED]> [2006-08-23 09:35]: > You probably want http://use.perl.org/~username/journal/rss
Actually, <http://use.perl.org/~username/journal/atom> – as well as <http://use.perl.org/search.pl?op=journals;content_type=atom>. The scraper pulls the latter and then iterates over the entries, pulling individual journal feeds to pluck out fulltext and transplant it, so I end up with the same feed that I get from <http://use.perl.org/search.pl?op=journals;content_type=atom>, only with fulltext included. It’s not necessary to poll every hour actually. I used to do that because originally I didn’t know about that hub feed and used a different feed to scrape: <http://use.perl.org/journal.pl?op=top;content_type=atom>. That one only lists journals, not individual entries, so you need to poll very frequently to avoid missing posts. Additionally, the next version of my scraper will have a cache so it can send the appropriate `If-Modified-Since` headers on requests, so that for feeds I’ve already seen the entire req/resp cycle takes all of 300 bytes or so instead of pulling down the whole 50KB shebang every time for no reason at all. (Yeah, I was really rude. I am embarrassed in retrospect.) Regards, -- Aristotle Pagaltzis // <http://plasmasturm.org/>
