Closed saltydressage closed 7 years ago
Hm, maybe I can make a new field index and then hand parse the URLs going in and feed them into that index by themselves, would that work or am I talking out my ass? Trying it now.
I'd register with a custom search adapter:
https://github.com/etianen/django-watson/wiki/registering-models#tweaking-search-weighting
You can then customise how the fields you want to search are processed before being stored in the db. On Thu, 7 Apr 2016 at 19:20, Jeff S notifications@github.com wrote:
Hm, maybe I can make a new field index and then hand parse the URLs going in and feed them into that index by themselves, would that work or am I talking out my ass?
— You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub https://github.com/etianen/django-watson/issues/159#issuecomment-207035815
Thanks, I'll give this a better try. I poked at it a bit before but didn't seem to get anywhere. Good to know that's the expected direction so I'll put in more time.
I'm trying to figure out how to break up URLs so the individual components can be searched on individually. eg: http://www.cnn.com/test/big_news
I'd like that URL (and the content around it) to be matched on a search for "big_news".
The best I've seemed to find is this: http://stackoverflow.com/questions/14093230/word-separators-for-postgres-full-text-search-with-rails
Unfortunately my SQL-fu doesn't seem to be good enough right now to cajole this into working with Watson (taking into account different naming etc). I have it working in a standalone DB but Watson appears to be doing 'more' that gets in my way.
Are we able to get this added, or as a fallback (even betteR) can the author comment in this issue on how I can achieve this potentially with Watson?