Hello, I am currently developing a graphql endpoint for the musicbrainz api. I have the basic queries, i can search for annotations, areas, artists, etc.. However I don't know how to solve one problem. If i have for example a relations for one of the entity, i would like to batch a request and send it to the api to resolve those requested fields in
one request instead of 100. Is there a way to do so?
For example: An event has a relation to a place and an artist. How can I send a request to the api to get both of these entities with the data i would like to show? I have both ids, but I would like to get more data from them than the basic response.
right now we don't support what you're wanting, sorry. but we agree that that would be the blueprint for future versions of the API
scory
Okay, thank you for the quick answer.
ruaok
the only real way of doing what you want to do is to install a copy of MB on a VM and then query postgres directly, but that is much more work than calling an API. :(
scory
Yeah it is, and i only want to use what i can from the API. Is there a way to add a flag for the API to get a response without anything left out? For example, an entity has aliases, relations and tags but the don't always contain data so the response won't show the aliases. Is there a way to force the api to give those back as well? even if its
just a null value
ruaok
have you taken a look at the inc= arguments?
scory
I have, but in the example: https://musicbrainz.org/ws/2/event/?query=eid:b... it doesn't add anything. And those are used when you don't use the ?query tag if i understood it, and doesn't have an impact in this case. Or maybe it does and i use it in the wrong way.
I have known that you can search an entity with that, but that was still helpful in other way for me. Thank you.
The problem is if i have the 25 events for example, I can't really send 25 seperate requests for the API to get these details because the one request per second limit. Or if we talk about 100 it's more bothersome.
ruaok
I dont think it is realistic to make this endpoint using our API. With direct DB access, yes.
scory
It might be. Thank you both for the information.
scory has quit
reosarevok
ruaok: sorry, was distracted and I guess it's late now
[listenbrainz-server] dependabot-preview[bot] opened pull request #1267 (master…dependabot/pip/master/bleach-3.3.0): [Security] Bump bleach from 3.2.1 to 3.3.0 https://github.com/metabrainz/listenbrainz-serv...
[listenbrainz-server] dependabot-preview[bot] opened pull request #1268 (master…dependabot/pip/bleach-3.3.0): [Security] Bump bleach from 3.2.1 to 3.3.0 https://github.com/metabrainz/listenbrainz-serv...
[listenbrainz-server] dependabot-preview[bot] opened pull request #1269 (production…dependabot/pip/production/bleach-3.3.0): [Security] Bump bleach from 3.2.1 to 3.3.0 https://github.com/metabrainz/listenbrainz-serv...
v6lur joined the channel
v6lur_ joined the channel
v6lur has quit
dseomn has quit
dseomn joined the channel
dseomn has left the channel
dseomn joined the channel
Zhele joined the channel
zas
bitmap: ping
sumedh has quit
v6lur_ has quit
Gazooo7949440 has quit
Gazooo7949440 joined the channel
bitmap
zas: pong
zas
Hey
we had weird issues with sir-prod failing to connect to pg on floyd
I wonder why this is happening, perhaps too many connections at the same time or something like that, but I didn't find anything saying so
bitmap
hmm
zas
where are pgbouncer logs?
I noted some of postgresql logs are duplicated in various /var/log files, that's something to improve imho, prolly through proper syslog ng config inside container
bitmap
they're outputted to docker logs, at least
zas
no occurence of pgbouncer in last 1M lines
do I miss something?
bitmap
huh, not sure. there are some on pink though they are not labeled with "pgbouncer", just LOG stats
I don't think it's configured to log much beyond stats
but the postgres logs/graphs don't really show anything suspicious to me
zas
I checked firewall logs, no packet were blocked on pink or floyd