Hey aerozol .. if possible could you please start making the designs for the album pages.. will be starting on those soon after I finish writing tests...
If you could get them to me in 3-4 days, that'd be great!
iconoclasthero has quit
pite has quit
Kladky has quit
ahvalmissaamine has quit
ahvalmissaamine joined the channel
Hobbyboy|BNC joined the channel
Hobbyboy has quit
lakkis has quit
genius3000 has quit
lakkis joined the channel
genius3000 joined the channel
Hobbyboy|BNC is now known as Hobbyboy
Kladky joined the channel
aerozol[m]
Hey pranav! Lucky you asked this week, I might be able to get them done in a few days. Can you please let me know what elements you want in the album pages (based on what’s in the website pages). Also, when can I review what you’ve done so far? Probably a good idea before starting another big one
pranav[m]
Sure
I had made a doc reg the same, am searching for it
aerozol[m]
I skimmed your forum post and missed this part until now - still accurate? Or you want to update
aerozol[m] uploaded an image: (23KiB) < https://matrix.chatbrainz.org/_matrix/media/v3/download/matrix.org/FYXertgnGwOInVigzUhNlCCT/image.png >
aerozol Pls go thru this, this has all that I require
aerozol[m]
Perfect thanks pranav , I thought you had made a doc, but I couldn’t find it in my bookmarks either!
Been a bit hectic lately so tried to have it for you ASAP while I have time pranav . Let me know if you spot any issues. A lot of the work I had done mocking up the album page: https://www.figma.com/design/tFLTQiq7QlA1oGRTJV...
Other input welcome too, see export below
aerozol[m] uploaded an image: (225KiB) < https://matrix.chatbrainz.org/_matrix/media/v3/download/matrix.org/EIyCSGhQmPActDoexoDUaMWk/LB%20app%20album%20page.png >
(Only real update from when I did the artist page draft is turning artist radio into its own section)
pranav[m]
Looks great aerozol
mayhem[m] joined the channel
mayhem[m]
<aerozol[m]> "ooh mayhem: https://www.reddit...."; <- Interesting, thanks. I just wish they would keep their superiority complex in check when someone provides them new toys. Thanks for calling our hard work janky.
d4rk-ph0enix has quit
d4rkie joined the channel
outsidecontext[m joined the channel
outsidecontext[m
as this is for navidrome it would be even simpler to use the subsonic integration directly, I think
monkey[m] joined the channel
monkey[m]
ansh: Hello! I have a question regarding the new BP, and removing items from queue. Please ping when available :)
outsidecontext[m
mayhem, lucifer: there has been some commit activity on the nmslib upstream project recently. Also the comment on https://github.com/nmslib/nmslib/pull/554 . Might be worth trying to upstream the Python compatibility changes from the MB repo.
ansh[m] joined the channel
ansh[m]
<monkey[m]> "ansh: Hello! I have a question..." <- Hi!
monkey[m]
Hi ansh, I think in the meantime I figured my issue out :)
The issue in question was for dispatching `REMOVE_TRACK_FROM_AMBIENT_QUEUE` after mapping listens, because the calculated `id` used to find items was completely different after mapping listens.
discordbrainz
<11muxxer.exe> Fellas, I'm currently not allowed to delete my ListenBrainz listens. Can anyone check into that? Tried like 5 times over the last 48 hours and not a single scrobble got deleted, I can only delete them individually.
monkey[m]
I resolved that by instead finding the item to deleted from the data used to dispatch SET_AMBIENT_QUEUE
A little bit of extra faff, but it works fine.
ansh[m]
So, are you now able to delete tracks?
monkey[m]
Yep, it's working fine now. I was mainly wondering if you had run into this sort of issue, and how you resolved it
ansh[m]
I discovered a new issue this morning while testing. [here](https://github.com/metabrainz/listenbrainz-server/pull/2827/files#diff-98b213be71f073ab070a99110a5e75b9a0c2f2ee05e59e7ebb0ab5fdd7a8f862R232)
the issue here is in `const { ambientQueue } = state;` If i try to log the ambientQueue here, it returns undefined. But if I try `const { ambientQueue: currAmbientQueue } = state;`, i'm able to console log currAmbientQueue.
ansh[m]: > <@anshgoyal:matrix.org> I discovered a new issue this morning while testing. [here](https://github.com/metabrainz/listenbrainz-server/pull/2827/files#diff-98b213be71f073ab070a99110a5e75b9a0c2f2ee05e59e7ebb0ab5fdd7a8f862R232)... (full message at <https://matrix.chatbrainz.org/_matrix/media/v3/download/chatbrainz.org/aHyovjYwvGGLwWYcVylRqPKB>)
ansh[m]
yep
monkey[m]
Hi muxxer.exe , can you tell me how you are removing listens? Is it via the API or the website?
ansh: Can you try something? Defining `const {queue, ambientQueue } = state;` before the switch block.
Does that change anything?
ansh[m]
sure! i'll try it out
lucifer[m] joined the channel
lucifer[m]
reosarevok: hi! will the oauth changes be released in prod today?
reosarevok[m] joined the channel
reosarevok[m]
bitmap, yvanzo: ^
discordbrainz
<11muxxer.exe> Website
<11muxxer.exe> Haven't touched the LBz API
ansh[m]
monkey: Strange, both of them are working now. I think we can ignore this.
monkey[m]
Hm
ansh[m]
lucifer: Hi! Regarding LB-1582. This ticket is about increasing the limit of artists in the map. As I checked, this stat is generated on demand using the entity stat for user, which has a limit of 1000 artists, and then we call the LAB API endpoint for fetching the country code.
So, in-order to remove the limit, how about we rewrite both the data fetching logic into one sql query (like how we do in YIM) and move it to labs? The stats will still be generated on demand
ansh: we can remove the limit but it will have to generated in spark only because it will take up too much resources to do it in labs for all the users.
if a small number of users use it then we could compute and cache it ideally but its unpredictable.
especially for all_time range, you don't want to aggregate reading from timescale at all.
ansh[m]
So, can we use on-demand generation with spark? Or do we have to compute the data regularly similar to how we do it for other stats?
yvanzo[m]
bitmap: Since `master` is matching `beta`, it is probably still time to merge your pull request to log response time.
lucifer[m]
currently we don't have on demand generation so it will have to be done like we do it for all other stats
yvanzo[m]
bitmap: Are there other pull requests that should be merged alike (bypassing the beta period) for the next release?
lucifer[m]
in few weeks we will probably have a second set of spark machines, can do on demand then.
yvanzo[m]
lucifer: The OAuth change is in beta at least. I won’t have time to update production today, maybe bitmap will update it, otherwise will do later on in the week.
@muxxer.exe I tested deleting a listen from my own dashboard, and that worked fine after I waited the requisite 1 hour. Can you please send me the page you are deleting listens from? I can try and see if there is an issue with said listens that are preventing the usual deletion workflow from running correctly
pite joined the channel
minimal joined the channel
bitmap[m]
yvanzo: I can do the MBS release today. I think the PR to log response times is the only one suitable for merging at this time
Currently, I have written unit tests for the user repository with 100% coverage
However, in order to write tests for the VM, I would need to mock social and listens repository as well.. is there a workaround or do I need to mock both of them and write tests for user VM?
Or maybe for now, I can implement the functions which r used in testing and the can test it with the mocks?
mayhem[m]
hiya! rimskii and lucifer : the deadline for rimskii 's project has been extended by two weeks.
Boomer joined the channel
BrainzGit
[listenbrainz-ios] 14gauravbhardwaj7 opened pull request #24 (03main…dashboard): Phase 2: Dashboard Components: Followers/Following and Stats section https://github.com/metabrainz/listenbrainz-ios/...