starred/ytmusicapi

Fork 0

mirror of https://github.com/sigma67/ytmusicapi.git synced 2026-04-25 15:26:01 +03:00

[GH-ISSUE #520] script to setup test env #378

New issue

Closed

opened 2026-02-27 23:00:28 +03:00 by kerem · 6 comments

kerem commented

2026-02-27 23:00:28 +03:00

Owner

Originally created by @jcbirdwell on GitHub (Jan 14, 2024).
Original GitHub issue: https://github.com/sigma67/ytmusicapi/issues/520

Is your feature request related to a problem? Please describe.
Austin mentioned it in their PR #517 and I wholeheartedly agree. I've probably spent about as much time setting up/fixing my test environment as I have actually contributing code and the setup almost dissuaded me from requesting the pull... It's a solution that I considered implementing during the tail end of my oauth overhaul when a bunch of errors seemed to stem from discrepancies in test.cfg setup, but didn't as it was already too long.

Describe the solution you'd like
CLI python script similar to setup that walks you through adding oauth and browser auth (copy paste etc but writes the values both to files in the test dir by default and adds inputs to tests.cfg), creating the brand accounts (under their correct auth accounts), followed by a simple script for building out a suitable and standardized brand account library to run the tests against ie saving tracks, albums, creating playlists etc.

Describe alternatives you've considered
DevOps is by no means my specialty so I'm open to suggestions in terms of fancy pipeline/pdm etc solutions, but I think even a partial implementation of the above spec would be well worth it.
Further documentation of the test setup could alleviate some of the pain, but I'd consider it alone to be inadequate.

Originally created by @jcbirdwell on GitHub (Jan 14, 2024). Original GitHub issue: https://github.com/sigma67/ytmusicapi/issues/520 **Is your feature request related to a problem? Please describe.** Austin mentioned it in their PR #517 and I wholeheartedly agree. I've probably spent about as much time setting up/fixing my test environment as I have actually contributing code and the setup almost dissuaded me from requesting the pull... It's a solution that I considered implementing during the tail end of my oauth overhaul when a bunch of errors seemed to stem from discrepancies in test.cfg setup, but didn't as it was already too long. **Describe the solution you'd like** CLI python script similar to setup that walks you through adding oauth and browser auth (copy paste etc but writes the values both to files in the test dir by default and adds inputs to tests.cfg), creating the brand accounts (under their correct auth accounts), followed by a simple script for building out a suitable and standardized brand account library to run the tests against ie saving tracks, albums, creating playlists etc. **Describe alternatives you've considered** DevOps is by no means my specialty so I'm open to suggestions in terms of fancy pipeline/pdm etc solutions, but I think even a partial implementation of the above spec would be well worth it. Further documentation of the test setup could alleviate some of the pain, but I'd consider it alone to be inadequate.

kerem

2026-02-27 23:00:28 +03:00

closed this issue
added the
documentation
label

kerem commented

2026-02-27 23:00:29 +03:00

Author

Owner

@sigma67 commented on GitHub (Jan 14, 2024):

I agree and I'm aware of the issue. Not sure if creating the brand account is worth automating, I think it's not super trivial to do.

But filling the test.cfg could probably be done in an automated manner. Before automating should probably attempt to simplify that file though (fewer entries where not needed), to avoid too much effort.

@sigma67 commented on GitHub (Jan 14, 2024): I agree and I'm aware of the issue. Not sure if creating the brand account is worth automating, I think it's not super trivial to do. But filling the `test.cfg` could probably be done in an automated manner. Before automating should probably attempt to simplify that file though (fewer entries where not needed), to avoid too much effort.

kerem commented

2026-02-27 23:00:29 +03:00

Author

Owner

@jcbirdwell commented on GitHub (Jan 14, 2024):

Not sure if creating the brand account is worth automating, I think it's not super trivial to do.

Maybe not fully automated, but jumping the user to the correct links via webbrowser with some instruction might streamline the process.

But filling the test.cfg could probably be done in an automated manner. Before automating should probably attempt to simplify that file though

Automation of the brand account library contents would drastically simplify the test.cfg by removing the need for any of the limits, uploads, albums, playlists, and queries sections. Items to be added/kept in the brand account could be externalized to a json file.

(fewer entries where not needed), to avoid too much effort.

I would say lowering some of the current limit values would aid the creation process, like 300 library artists may be a bit excessive for an account only used for testing.

@jcbirdwell commented on GitHub (Jan 14, 2024): > Not sure if creating the brand account is worth automating, I think it's not super trivial to do. Maybe not fully automated, but jumping the user to the correct links via webbrowser with some instruction might streamline the process. > But filling the `test.cfg` could probably be done in an automated manner. Before automating should probably attempt to simplify that file though Automation of the brand account library contents would drastically simplify the test.cfg by removing the need for any of the limits, uploads, albums, playlists, and queries sections. Items to be added/kept in the brand account could be externalized to a json file. > (fewer entries where not needed), to avoid too much effort. I would say lowering some of the current limit values would aid the creation process, like 300 library artists may be a bit excessive for an account only used for testing.

kerem commented

2026-02-27 23:00:29 +03:00

Author

Owner

@sigma67 commented on GitHub (Jan 14, 2024):

I see what you're getting at now with automatically populating a new brand account 👍 Great idea.

I think we could also narrow down the different accounts that are currently used to just three:

main account for uploads - brand accounts can't do uploads (using browser auth)
brand account for content manipulation (using oauth)
empty account for empty content tests

@sigma67 commented on GitHub (Jan 14, 2024): I see what you're getting at now with automatically populating a new brand account 👍 Great idea. I think we could also narrow down the different accounts that are currently used to just three: - main account for uploads - brand accounts can't do uploads (using browser auth) - brand account for content manipulation (using oauth) - empty account for empty content tests

kerem commented

2026-02-27 23:00:29 +03:00

Author

Owner

@sigma67 commented on GitHub (Feb 4, 2024):

@jcbirdwell I added a script, feel free to review

@sigma67 commented on GitHub (Feb 4, 2024): @jcbirdwell I added a script, feel free to review

kerem commented

2026-02-27 23:00:29 +03:00

Author

Owner

@sigma67 commented on GitHub (Feb 27, 2024):

To be fair I don't think the whole brand account for testing thing is realistic anymore. At least not without a change to YouTube's policy.

They have already banned my testing account due to their "spam" policy

@sigma67 commented on GitHub (Feb 27, 2024): To be fair I don't think the whole brand account for testing thing is realistic anymore. At least not without a change to YouTube's policy. They have already banned my testing account due to their "spam" policy

kerem commented

2026-02-27 23:00:29 +03:00

Author

Owner

@sigma67 commented on GitHub (Mar 1, 2024):

After appealing they unblocked it again - so it seems they're not actively opposed, it's just automatically caught.

@sigma67 commented on GitHub (Mar 1, 2024): After appealing they unblocked it again - so it seems they're not actively opposed, it's just automatically caught.