#metabrainz

"Blocksafe wants to reduce gun violence with a hybrid IoT blockchain network"

it says here that "Compilations are also of primary type “album”. You have to filter these out manually if you don’t want them."

but i can't figure out how to do that

there doesn't seem to be a "compilation: true" field in any of the release items

(or anything like that)

compilation is a secondary release type

there should be a list of secondary release types, you just check if 'compilation' is one of them, iirc

this is my code http://paste.ofcode.org/BA6nRf45Nfgs6wQ4XKpWdM

running it produces (partially) this output: http://paste.ofcode.org/hePQkDvQM9dWC6cMizsEN4

it outputs out a bunch of releases, but that one is a compilation and not a proper album

and i don't see anything in the actual `release` object there, that indicates it's a compilation and not an album

i already specified album as my release type in `browse_releases()` (https://python-musicbrainzngs.readthedocs.io/en...)

alright, so the "type" you're looking at is actually the *release group* type

it's not stored on the release directly

you need to do the query with inc=release-groups

it's often good to look at the webservice xml, since the python library maps that fairly directly

the last release in that query shows the "Compilation" secondary-type in the associated release group

sorry, not sure if i'm following you correctly

are you saying that i need to use `musicbrainzngs.browse_release_groups()` instead of `musicbrainzngs.browse_releases()`?

no, a browse_releases is fine

but you need to add includes=["release-groups"] to pull in the info

like mb.browse_releases(artist=artist["id"], release_type="album", includes=["release-groups"])

that will turn into the webservice query: http://musicbrainz.org/ws/2/release?artist=db71...

which you'll note has a "<release-group>" inside each <release>

aha!

so there should then be a release group property on the release on python, with the type information

thank you very much :D

the list of linked information you can use with includes on a browse request is here: https://wiki.musicbrainz.org/Development/XML_We...

(and indeed, you should probably read that entire page to learn more about how the ws is structured)

yui, did you also ask in the forums? or was that somebody else?

CallerNo6: someone else

okay, cool, so they still need an answer :-)

I'd like to suggest a feature for musicbrainz in relation to use with Picard. Where is the right place to do that?

hi eharris. if you want to /discuss/ a suggestion, here or @ community.metabrainz.org is the place.

to actually make the suggestion, you'd create a ticket @ tickets.musicbrainz.org

more @ https://musicbrainz.org/doc/Bug_Tracker

Ok. So Picard uses my musicbrainz login. Shouldn't it be possible for musicbrainz to serve the data from my own edits to my instance of Picard even though they haven't officially been applied yet?

I'm having problems where I lose track of what I've done, because my entries in Picard don't show the edits I've already made.

so I end up with duplicate edits

hmm. that sounds like something that may have been requested already.

Unfortunately, the edit system isn't well-suited to that, though it would be possible.

For that matter, it might be nice if the same thing could be done through a "include pending edits" mode on the website as well, maybe with markup (colors?) to show the stuff that is being shown but isn't applied yet.

I don't see a relevant ticket. I'm sure other people would be interested in that capability, so by all means, create a ticket if you're up for it.

For a ticket to get attention from the devs, you probably need to 'champion' it in the forums too.

There are more good ideas than there are active devs, I'm afraid.

Well, this is just effectively doubling my workload, and I can't afford it. Since I'm making corrections as I go through my library, and Picard isn't reflecting what I've changed, and usually won't for a week.

pretty sure there was a ticket for that on our old trac site, even

it's very difficult to implement with our current edit system though

I'd really rather update musicbrainz and have Picard save the changes immediately to my library. If it's a question between updating it both in Picard for my own library AND in the website, I'm probably just going to have to stop updating the website.

Doing everything twice is too expensive.

"We're not going to fix this as it is. We're planning to move towards more things being auto-edits instead."

Is Picard's interface to it XML (I'd guess)

?

yes

it's not the interface that makes it difficult to implement, though

Hmm, in that case, is there some way I can get vetted for some level of "higher" access where my edits mostly get auto-applied?

aha. bitmap's search fu is powerful. (I assumed it would be a picard ticket)

It seems to me that having something more like how Wikipedia does it, where edits are immediately applied but can be easily revoked would solve this issue.

yeah, that's kind of been our plan for a while

The comment on that ticket that is no-fix is from 2014. I'm guessing I can't count on this being resolved either through my the proposed method reflecting the pending edits or going to auto-edits anytime soon then...

bitmap, do you mean that NES might still happen in some form?

Is there a "trusted editor" level of access what I might be able to apply for to have less wait time on my edits?

I think we'll be doing it incrementally with the current codebase

NES is dead for good

I'm still hoping to make BrainzStem a thing, but it's a long way away.

eharris: what kind of edits are you waiting on, mostly?

eharris, there is. http://musicbrainz.org/doc/Editor#Auto-editors . But there's not really an application process. It's more like, you get nominated if somebody notices that you're doing especially good work or something.

even AutoEditors can't apply merges and (some) removals automatically

Things like release group and release renames and merges, and adding of things like barcodes, recording label, catalog number, etc.

the main things I need are renames and merges. The catalog and other info I don't care about that much in my own library as long as the server has them, but I want the names and groupings right in my library.

are there really that many releases that need to be renamed?

No, there aren't a lot. But my library is big and I can't really afford to wait a week, or keep track of what I need to come back to and refresh. It's a lot of extra work, and impeding my workflow.

One of the biggest problems is that Picard doesn't give a good way to enter multi-disc sets, so I've been having to create each disk as its own release group (since the plugin supports that) and then come back and merge them.

but then I have to wait a week for the edit to take before I can "fix" my own library

hmm. I guess one thing you can do is copy the tracklist from the track-parser after picard opens the page, edit the previous release, and paste it there

and I think the plugin will submit multiple discs if the tags contain "discnumber" values

well, thanks for the info and tracking down that ticket. I hope the auto-edit stuff comes soon.

whenever I tried to use the plugin on more than one cluster, it refused to work

and since the names of the disks are different, they end up in different clusters

right, I guess they'd have to be in the same cluster and you'd need to do a bit of manual editing

but would take less time than merging afterward at least

I don't know. Merging is pretty easy than updating a bunch of individual tracks and numbers. It's just the waiting for the merge to take effect that is the killer

s/than/compared to/

merges are the hardest edit type to implement reverting for, so might be a while before we have that

I think using the track-parser is still a decent option though. wouldn't have to edit track or disc numbers, just copy and paste

Ok, thanks. I'll see if I can figure out how to make that work.

on the Tracklist tab, there's a "Track Parser" button under the medium where you can copy a textual representation of the tracklist

so in the release that was already added, you can click "Add Medium" at the bottom of the page and paste it in there

How do I get that out of Picard?

this is all on the website

I have my collection already mostly tagged. I want to pull the tags on my files in easily, without having to re-enter all the info.

I sometimes do a few edits to follow the style guidelines, but I don't re-enter everything