->data->{old} and ->{new} get filled with hashes, not objects. or is that not true for all edits?
2012-01-31 03139, 2012
djce1 joined the channel
2012-01-31 03116, 2012
bitmap_ joined the channel
2012-01-31 03109, 2012
ocharles
yea, i guess that's the only place that has to stay as hashes
2012-01-31 03111, 2012
ocharles
as those are partial
2012-01-31 03118, 2012
warp
why is partial a problem there? our objects can handle that, can't they?
2012-01-31 03153, 2012
mueslix joined the channel
2012-01-31 03103, 2012
ocharles
warp: imo, a sane object model matches the schema
2012-01-31 03113, 2012
ocharles
ie, every artist must have a "gender_id" - even if it's Nothing
2012-01-31 03118, 2012
muesli joined the channel
2012-01-31 03121, 2012
ocharles
but in our edits, sometimes we just don't know that, because we don't store it
2012-01-31 03126, 2012
tjaalton joined the channel
2012-01-31 03128, 2012
ocharles
so it could be 1, it could be undef, but we don't know
2012-01-31 03105, 2012
warp
ocharles: I thought we could (in Moose) distinguish between properties which haven't been set at all and properties which aren't defined. (so the different between exists and defined basically).
2012-01-31 03112, 2012
ocharles
true
2012-01-31 03122, 2012
ocharles
but still, it's a partial object
2012-01-31 03128, 2012
ocharles
which is a potential bug down the line
2012-01-31 03113, 2012
warp
in general we only load into the object that stuff which is needed. so it being a partial object doesn't strike me as a problem in itself.
2012-01-31 03108, 2012
warp
ofcourse we nowadays want ->{old} to not be partial, so we know what the object looked at the time the edit was entered. perhaps we should just put full objects in ->{new} as well?
2012-01-31 03150, 2012
ocharles
well this leads us to the new edit systme
2012-01-31 03102, 2012
ocharles
which for me is to just not serialize objects into the database, but have a complete history everywhere
2012-01-31 03104, 2012
warp
yay! new edit system!
2012-01-31 03114, 2012
warp
ocharles: right.
2012-01-31 03106, 2012
warp
ocharles: so, for now I should just keep putting hashes into ->data->{old} and ->data->{new}, but try not use them anywhere else?
reosarevok: I spent the minimum amount of time to move my User:kuno stuff into Development and add a few other links, but I'm going back to actual development now.
2012-01-31 03157, 2012
ruaok
also 2: data is quite interesting. the data we have so far for UMG is pretty clean and looks useful.
2012-01-31 03108, 2012
ruaok
not sure we can get something deployed today, but we have stuff that is working.
2012-01-31 03125, 2012
reosarevok
warp: agreed, cool - I'll need to study today + tomorrow, can dedicate time to it after that
2012-01-31 03140, 2012
warp
reosarevok: great, let me know if you need help.
2012-01-31 03149, 2012
reosarevok
I will
2012-01-31 03152, 2012
ruaok
reosarevok: would some last.fm stickers do instead of a pony?
Might make sense to flesh the specific years a bit
2012-01-31 03139, 2012
Leftmost joined the channel
2012-01-31 03105, 2012
ruaok joined the channel
2012-01-31 03134, 2012
ruaok
<BANG>
2012-01-31 03135, 2012
ruaok
meeting time.
2012-01-31 03139, 2012
ruaok takes a sip of beer
2012-01-31 03139, 2012
warp
hello!
2012-01-31 03147, 2012
nikkiphone
hi
2012-01-31 03113, 2012
ruaok
warp, you go.
2012-01-31 03125, 2012
ruaok
we're still working out some setup issues.
2012-01-31 03130, 2012
warp
heh
2012-01-31 03152, 2012
ruaok
(with the laptops, not the beer, mind you)
2012-01-31 03130, 2012
warp
there is nothing particularly noteworthy I worked on last week.
2012-01-31 03150, 2012
warp
we did the release, so that involved some testing and fixing test failures before the release.
2012-01-31 03136, 2012
warp
I spent friday looking at paul's /ws/2/all search
2012-01-31 03105, 2012
warp
trying to come up with an interface for that, and doing some experiments in getting similar results from the direct search
2012-01-31 03145, 2012
warp
(which was a bit pointless, but I only learned that afterward when someone explained what the dismax=true argument does to the results from the search server)
2012-01-31 03119, 2012
warp
everything else is just small tickets and other tasks.
2012-01-31 03133, 2012
warp
one thing to note though
2012-01-31 03149, 2012
warp
the "Unassigned, Within 3 months" list of issues is completely empty now.
2012-01-31 03100, 2012
ocharles
\o/
2012-01-31 03105, 2012
ocharles
well
2012-01-31 03106, 2012
ruaok
nice
2012-01-31 03109, 2012
ocharles
you should check "unassigned" :)
2012-01-31 03116, 2012
ocharles
wait, that is unassigned
2012-01-31 03117, 2012
ruaok
time to triage more bugs.
2012-01-31 03118, 2012
warp
not all are fixed, but the remaining ones are assigned to folks and are hopefully being worked on.
