September 21, 2006
Leveraging searchfeedr with pageSearch
I took a day off today fro a bit of guilt free playing with some of the beta invite apps I've got backlogged, but as ever I got sidetracked and just had to scratch an itch that came up last night. The itch started when I realised that the feeds from pages like delicious, wink, technorati etc. that I can generate from user page URLs, and use to feed searchfeedr, are actually producing info in a machine readable way that can be obtained 'by force' - by scraping the links from the page directly, an approach I've used before in my search links on this page tools.
Anyway, searching back over the OUseful blog for those tools, I also turned up a post to the demo pagelinks2opml generator. Which of course got me thinking...if this were a pagelinks2json tool, that scraped links off a page given it's url, and made them available via a JSON feed, I could feed them as limits directly into searchffedr.
Like this: pageSearch - (searchfeedr edition).
To see how it works - sort of - here's another image served up directly from Gliffy:
This image was created by copying an image I'd already drawn on Gliffy, and just making one or two minor changes.
The pageSearch tool, too, only required very minor changes to the searchfeedr script to get it to work (one line, in fact...)
Posted by ajh59 at September 21, 2006 03:09 PM