#musicbrainz-devel

jesus2099: not particularly, we know it won't happen any time soon anyway

hey guys i have null error and is there a way to fix it Thu Jun 13 05:15:24 2013 : psql:/home/musicbrainz/musicbrainz-server/admin/sql/InsertDefaultRows.sql:19: ERROR: null value in column "ha1" violates not-null constraint

Error during InsertDefaultRows.sql at ./admin/InitDb.pl line 108

nikki: because you don’t like it ?  ;P

I guess we haven't updated that sql file

because schema change

jesus2099: they way it currently looks it’s inconsistent with how MB currently highlights changes… but I don’t think this should be a high priority improvement

SultS: how inconsistent ? it’s like this in the release editor already

and this really looks like a problem when adding the release with a disc ID: http://musicbrainz.org/cdtoc/Pi_imdz8Y0MOHlZ9XU..., fixed in http://musicbrainz.org/edit/22568708

JonnyJD_: we can’t add release from TOC and reuse existing release (i dont know if that’s your problemà

jesus2099: sure we can add a release from disc ID

I am only doing that

im just runing carton exec ./admin/InitDb.pl -- --createdb --clean

me too

doesn't that still lose the disc id?

or is that only if you reuse an existing tracklist?

JonnyJD_: i add from TOC but doing so I have to forget any sorts of import from anything existing

Lewis: I don't think --clean has ever really worked

jesus2099: http://musicbrainz.org/edit/22568348 shows we use red for just editing an URL, not removing it… it would look on your image as yellow, if i understand correctly… so inconsistent

Lewis: what are you trying to do?

if you say take this or that release to build up your tracklist, it will forget your TOC

JonnyJD_: ↗

nikki: not that I know of. I put the disc in the drive, submit the ID and I can create a new release or attach the ID to another release

SultS: you’re linking to an edit page where e use a DIFF tool…

testing to make sure things will run when i put the big import in

SultS: my pic. is mostly what’s used on the release editor when you edit AR

it’s consistent with reled

JonnyJD_: http://tickets.musicbrainz.org/browse/MBS-2301 is what I'm thinking of

well, I use the "submit disc ID url" from libdiscid and it works

nikki: do you remember the « show edit change immediately while edit is pending (like PUID edits) » ticket ?

the number

nikki: well, that is a specific case when you have release duplicates. Not that you can't submit per disc ID in general.

JonnyJD_: no not you can’t submit, we should submit b TOC, it’s perfect

JonnyJD_: well, as I said "or is that only if you reuse an existing tracklist?" :P

but thanks for the link to the ticket, now I know what specifically isn't supposed to work. I normally don't have duplicates.

but when doing so, we must not use any funny stuff

when i need this i use the plain text tracklist copy/paste

instead

but then i have to map each recordings one by one… cucumbersome

jesus2099: not sure which ticket you mean

something like http://tickets.musicbrainz.org/browse/MBS-2310 but more general

maybe it does not exist

Project musicbrainz-server_beta build #489: FIXED in 12 min: http://ci.musicbrainz.org/job/musicbrainz-serve...

reosarevok: Fixing my idiocy

\o/

ruaok: ping

you're up next

yes

alastairp: :)

nikki: I'm not really sure how to resolve http://tickets.musicbrainz.org/browse/MBS-6275

I'm guessing that ticket is asking for exact release date diffs

rather than a general "from this set of release events to this set"

yes

I'd say match up based on country, where applicable, and then do a diff with the dates

yeah

I can't think of being able to do much more than that, so I'll see what I can do

so if it goes from UK 2013 + US 2012 and goes to UK 2013-05 + US 2013 + unknown 2014, I'd show UK/US (the countries themselves that is) as unchanged (as well as 2013), the unknown country and '05' as an addition, and the 2012-2013 as a change

probably can't do more than that though

Yea

am i here ?

hi ruaok, are you finishing that search build today ?

I'm at music hack day today and tomorrow.

I'm half afraid to do anything live since the wifi is a bit dodgy. :(

let me try it though

heh. logins are not even succeeding.

ocharles: do you have a branch for http://tickets.musicbrainz.org/browse/MBS-1378 from way-back-when when you did that screenshot?

um

(if you know, I guess :P)

doesn't look like it

looks like there's some mbs-1378 branches various places

I don't have anything locally, that is

yeah, understood

looks like there's at least something there, though it's from last august

that might be something that could be coerced into usability then :)

hopefully

do not we still have 9 tickets to fix for this release though

note*

though I guess only about 5 of those are directly actionable by anyone atm

date/country columns warp has a patch for I thought

country names in sidebar/edits I should do though, that's a good point

not sure if I can make reasonable headway on the others :/

you're no more or less capable than the rest of us :)

but anyway, not telling you what to do - just pointing out that we do have a responsibility to get these things fixed

I'm focusing on this fix version for now, so I'll hopefully get to them

yeah

wow, we scheduled within 12 months over a year ago?

so there's a comparable argument, but the country names thing should be a relatively easy fix so I should do that

I did say I expected exactly this to happen :)

but cool, sounds like it should be addressed anyway

we paid absolutely no attention to them, which is why it happened :P

no, we didn't even get past 3 months

so how could we get to them?

interesting. sacem has an API.

that's my take anyway

well, from what I saw we just set due dates and then just worked on whatever anyway

re tickets for this release http://tickets.musicbrainz.org/browse/MBS-4021 should be dropped as it cannot be done until MBS-5310 which is still unassigned

it's never really been clear how we pick tickets or what features we're using in jira or anything :(

nikki: not just you

ijabz: shall I bring MBS-5310 into this fix version then?

:)

man, this connection is terrible.

and my wired connection dongle is at the uni. :(

ok, roobarb is done.

oh, right I'll test it

ruaok: looks ok

good.

dora is done.

wanna take a look?

k

all looks ok, and website not failing :)

yay

so guess your done and just need ianmcorvidae to hotfix part 2 of his fix

yep, all done.

yes

pingu is currently freaking out.

not me personally, but ISEs being emailed to me

someone even got one on the main page :(

hmm. pingu is calming down.

I wonder if that error is something we get when a DB query times out and gets killed.

I think so

that is pingu cancelling whatever query it issued because the whole request has taken too long

hello!

"cancelled statement due to user request" is pingu saying "hay totoro, that query is taking way too long, please just give up"