Closed GoogleCodeExporter closed 8 years ago
I would be against adding columns but somethimes we just have no choice.
I would also be against locking threads -- people are still going to talk in
other
ones, why be an ass and lock this one?
Original comment by andrey.o...@gmail.com
on 7 Mar 2009 at 8:39
That athens/robotechguy thread was terriblg, though. "Locking" threads by
hardcoding
parent numbers in cgi-board is hardly ideal.
And I was thinking if you didn't want to use a separate table for archive
posts. That
way we could add as many columns as we wanted.
Original comment by eksopl
on 7 Mar 2009 at 8:49
Can we at least make changes to be able to permasage threads?
Original comment by eksopl
on 7 Mar 2009 at 9:42
Really, permasage would be useful. But that means another column and another
index,
unless you want to do something ugly like using the email field to input the
permasage password (a true 4chan-like hack) or add an index to the delpass
column and
use a separate admin pass to permasage.
Original comment by eksopl
on 8 Mar 2009 at 5:06
I'll make it, I'll make it;
I'll just add another table, something like jp_comments, it will have fields
num,subnum, and then all these fancy columns we need for replies, and we'll
even able
to able to easy add new ones when we have to.
Anyway, even if we added a new col to main table, I don't think we'd have to add
index. This ``autosage'' field won't be in any where clause.
Original comment by andrey.o...@gmail.com
on 8 Mar 2009 at 6:24
It would, because you'd need to include something like and permasage = 0 in the
SQL
query that gets threads for the main page. Wait, it's more complex than that.
Threads
are not supposed to bump after they get permasaged, so you need to exclude all
posts
with the same parent after the post with the permasage.
But I agree that we should just add another table for comments so we can add
all the
internal posts columns we want, that's what I suggested in >>1.
Original comment by eksopl
on 8 Mar 2009 at 6:41
The changes I made to make View in Ghost Mode bearable (the xx_local tables)
makes it trivial to implement this now. It's just another column in xx_local.
I completely forgot about this discussion, though. I should have gone with
something a bit more like comment 5. Whoops.
Original comment by eksopl
on 9 Mar 2011 at 3:39
Ehhh, I don't think this really matters, now that I think about it.
Original comment by eksopl
on 20 Feb 2012 at 1:42
Original issue reported on code.google.com by
eksopl
on 7 Mar 2009 at 5:42