IF Archive experimental search feature

For oh about the past 25 years, when people mention the idea of searching ifarchive.org , I’ve said “just use Google with site:ifarchive.org.” (Maybe I didn’t say that in 2001. But shortly thereafter.)

Once IFDB launched, I said “just use Google or IFDB.”

This has never been completely satisfactory, though. IFDB only covers games, not interpreters or development tools or articles. And Google is messy these days.

I now have a trial of an IF Archive search form:

https://search.ifarchive.org/search

(The domain will eventually change to search.ifarchive.org but I haven’t done that yet.) (EDIT: I did it; edited link above.) (EDIT: In fact just go to search.ifarchive.org, that’s easier.)

The search engine behind this is a Python library called Whoosh. It’s rather old but it seems to work fine.

Please give it a spin and let me know what you think.

(I haven’t put the search form on the front page yet. I want to give it a bit of road testing first, which is what this post is about. Also, I don’t have any clear idea how much server load this will cause.)

14 Likes

work well, and is definitively fast (checked with sure-fire large-hits queries: zcode (421 hits), inform (1157 hits); the results came practically real-time) albeit perhaps having a “x results for page” option instead of the current fixed 10 result per page isn’t a bad idea for the easily predictable frequent queries (see my “sure-fires” above for an pair of examples…)

from one which has the Ifarch roadmap well-memorised and prefer the usage of ^F on all files by date (and has the patience of waiting the upload of the verbose version…) I can be 1000% sure that can be useful for non-veterans of the IF community.

Best regards from Italy,
dott. Piergiorgio.