At least for me, webchick's recent blog post had exactly the intended effect, i.e., that of a wake-up call: The module is already very usable, reasonably stable and even used on some production sites (that I know of) – but there is still no plan to finally create a stable release.
Of course, there are still a couple of issues that should be cleared before that, especially two current ones that will be major API changes:
- #1064884: Add support for indexing non-entities
- #1182614: Integrate with Facet API
As per webchick's orders, I've tagged them with "D7 stable release blocker".

Now the question to basically all stake-holders: What issues do you think also still need solving before we can make a stable 1.0 release? #955088: Provide (additional) access functionality? The numerous bugs in the database backend? The latter do make the database backend not really suitable for production use – but then again, withholding the whole release also isn't optimal, in my opinion.

The problem is also that there are of course several dependent modules which will all have to wait for this module's stable release to have a stable release themselves. (I could not find any other release blocker in any of my modules. If you do, please tag accordingly.) So completing a stable release for the Search API could effectively kick up webchick's counter by close to 10 modules.

Well, anyways, I'd love to hear some other opinions here, to hopefully make the right decision in the end. (Come to think of it: after three and a half years, today I created my first stable release ever. So, kinda inexperienced in that respect. ;) And maybe overcautious …)

Comments

klausi’s picture

I see no other blockers in my use cases, so just go ahead.

Shadlington’s picture

You've made a fantastic module that is pretty damn close to perfect. Tremendous work.

I'd echo klausi on this one - nothing else sticks out (though admittedly I've never really used the DB backend).

marcoka’s picture

awesome work on this! i think i will use search api in the future instead of the solr module.

Dave Reid’s picture

I added #955088: Provide (additional) access functionality as it's a pretty big security issue that needs to be solved prior to 1.0 since that's when the security team would need to get involved.

drunken monkey’s picture

After some consideration, also added #1231512: Use real Relationships instead of level magic in Views integration.

Oh, also – I guess while Entity API is in Beta, we can't really go stable, either, can we? After all, any Entity API update might trigger some major changes in the Search API …

realityloop’s picture

Status: Active » Fixed

just cleaning up

Status: Fixed » Closed (fixed)

Automatically closed -- issue fixed for 2 weeks with no activity.