m.amanullah7: look at the other files in the `listenbrainz/db` directory. we are not using orm in most cases.
plain sql queries.
mamanullah7[m]
lucifer:Sure let me check i'll try not to use then!
lucifer[m]
holycow23: let's rename time_bracket to hour and also the data type from string to integer.
holycow23[m]
cool
will do that
what's the next step though?
lucifer[m]
then update the api to read from couchdb.
and the frontend to render the chart using the api data.
holycow23[m]
Just the last step of StatsDB to Webserver is left right over [here](https://listenbrainz.readthedocs.io/en/latest/developers/spark-architecture.html)
lucifer[m]
that is the api serving the stats after reading it from couchdb.
holycow23[m]
Yeah but that's the only step left right mainly?
lucifer[m]
yes
holycow23[m]
Also I was trying to import listenbrainz.db.stats as db_stats in the terminal but gives an import error of psycopg2 and while installing it gives an error
lucifer[m]
try pip install psycopg2-binary
holycow23[m]
which I feel needs certain sudo installs should I go ahead with them?
Did that but still the same import issue
lucifer[m]
i checked and its an issue with the pip installation on wolf. this worked, python3.12 -m pip install psycopg2-binary --break-system-packages.
holycow23[m]
psycopg2-binary installed without any error for me
but the import of import listenbrainz.db.stats as db_stats still gives an error
lucifer[m]
what's the error?
holycow23[m]
ModuleNotFoundError: No module named 'psycopg2'
lucifer[m]
okay, there's an issue with the python installed on wolf. but anyway. you should create a virtual environment and use python in that.
python3.12 -m venv venv
.venv/bin/activate
pip install -r requirements.txt
should work fine then
* . venv/bin/activate
pite has quit
holycow23[m]
inside the environment do I start the shell using the same docker exec -it listenbrainzspark-request_consumer-1 pyspark --driver-memory 16g?
lucifer[m]
if you use docker then the environment won't be used
it only matters if you are running commands outside docker
pite joined the channel
holycow23[m]
yes that's what I assumed
Also the pip install gave an error
uWSGI gave an error
lucifer[m]
what's the error?
holycow23[m]
ERROR: Could not build wheels for uWSGI, which is required to install pyproject.toml-based projects
holycow23: can you please update your wolf setup to expose ports only on 127.0.0.1 ?
that should be the default in docker compose files iirc.
holycow23[m]
- "127.0.0.1:9000:9000"
- "127.0.0.1:9870:9870"
this is what I have
lucifer[m]
in the other docker-compose.yml
for listenbrainz web/websockets etc.
holycow23[m]
Everything is on 127.0.0.1
lucifer[m]
hmm
holycow23: i see the docker file was missing the default host for some containers, and the default is 0.0.0.0 so the issue. i have updated the docker compose file on your setup. just stop and start the containers when ready
m.amanullah7: create a unique index on user_id and funkwhale_server_id.
mamanullah7[m]
okay!
lucifer: Thanks it woking now!
lucifer: last time i tried to upload one song on fw but it failed to upload! is there any way to add songs? i wants to test fw in brainzplayer! should i push changes can u look into?
lucifer[m]
m.amanullah7: try again now and if it doesn't work, share the error you see.
HemangMishra[m] has quit
mamanullah7[m]
i'm not able to upload songs on fw ?
* lucifer: i'm not
lucifer[m]
how are you trying to upload the songs?
mamanullah7[m]
lucifer: ive downloaded locally one song and in settings there is an option to upload songs, i'm uploading there
lucifer[m]
mayhem: is the wolf nginx + le setup running from your account?
want to increase the upload size limit on it.
m.amanullah7: its failing because currently the max upload size allowed is likely 1 MB? if you have a small audio file or maybe crop an existing one. that could work until i can update the size on wolf.
mamanullah7[m]
lucifer: let me try!
zas[m] has quit
lucifer: i tried to upload the audio less than 1MB it showing its uploaded but i cant able to find that songs anywhere?
lucifer[m]
i can't see it on my end if you uploaded it with privacy set to you. can you try uploading it again with privacy set to instance?
lucifer[m] uploaded an image: (14KiB) < https://matrix.chatbrainz.org/_matrix/media/v3/download/chatbrainz.org/eabwRwrDunfCpJijQnqgjxEB/image.png >
this setting on the top left of the upload modal
mamanullah7[m]
lucifer: uplaoded one song can u able to see
lucifer[m]
nope.
i'll try to upload something and debug this when the upload is fixed.
* the upload issue is fixed.
q3lont has quit
q3lont joined the channel
mamanullah7[m]
sure!
lusciouslover has quit
lusciouslover joined the channel
RandomMushroom12 has quit
holycow23[m]
lucifer: I just pulled down the docker and re ran the stats so for some reason the genre stats have gone away and giving this error when generating any other type of stat
```request_consumer-1 | listenbrainz_spark.exceptions.PathNotFoundException: Path not found: /release_metadata_cache
request_consumer-1 | [PATH_NOT_FOUND] Path does not exist: hdfs://namenode:9000/release_metadata_cache.```
* lucifer: I just pulled down the docker and re ran the stats so for some reason the genre stats have gone away and giving this error when generating any other type of stat... (full message at <https://matrix.chatbrainz.org/_matrix/media/v3/...>)
mayhem[m]
<lucifer[m]> "mayhem: is the wolf nginx + le..." <- look in /home/production
recqd[m] joined the channel
recqd[m]
is there anyway to submit edits like changing a country, and voting yes/no or abstain on edits via the API?
