[RFC] Let CF users filter threads on attachments they made - Mailing list pgsql-www

From Jacob Champion
Subject [RFC] Let CF users filter threads on attachments they made
Date
Msg-id CAAWbhmg84OsO5VkaSjX4jokHy8mdpWpNKFgZJHHbb4mprXmtiQ@mail.gmail.com
Whole thread Raw
List pgsql-www
Hello,

I'm wading into commitfest app development, since I'm CFM'ing this
month, and there was an itch I wanted to scratch: the thread
attachment dialog almost never shows me the ones I'm interested in
without a few back-and-forth iterations through the search. I thought
it'd be nice to only show the threads that I directly posted
attachments to.

So I added a checkbox below the search field that does that, and
implemented author filtering in the pgarchives API. Screenshots and
patches (one each for pgarchives and pgcommitfest2) attached. WDYT?

Caveats:
- I think the new LIKE clause will probably perform terribly against a
large dataset. If you like this feature, I'll need assistance with
designing an appropriate index.
- Only one email is filtered (the "sender email" set in the commitfest profile).
- I've defaulted the checkbox to on, figuring that's probably what
most new users want. I'm relying on modern browser autocomplete to
keep the last checkbox setting you selected. Hopefully that's enough
to avoid annoying people who rely on the previous behavior.
- It's been several years since I did any frontend development. Please
be on the lookout for new reflections/injections, or whatever the kids
are using to mine their coins these days.

Some questions I have for the future:
- Should I cc pgsql-hackers on these sorts of threads?
- How do you all test with an actual production-style dataset? I've
cobbled together a test environment with multiple `./manage.py
runserver` invocations, and populated it with a couple entries, but I
figure there has to be a better way...?
- Have there been any prior discussions around changing the "new
patch" workflow so that it shows the thread selector by default? I
don't think I've ever used the "Specify thread msgid" field directly.
I know at least one new contributor got tripped up by that workflow
this commitfest, because they ended up emailing me to try to figure it
out.

Thanks!
--Jacob

Attachment

pgsql-www by date:

Previous
From: Joe Conway
Date:
Subject: Re: Wiki editor request
Next
From: Tom Lane
Date:
Subject: Weird mangling of a commit log entry in gitweb summary