Join GitHub today
GitHub is home to over 50 million developers working together to host and review code, manage projects, and build software together.
Sign upGitHub is where the world builds software
Millions of developers and companies build, ship, and maintain their software on GitHub — the largest and most advanced development platform in the world.
YouTube: -j --flat-playlist no longer gets titles #6699
Comments
|
I tried to debug this myself, but I got lost. I'm far from a Python expert, but I think it's harder than it ought to be. I documented my attempt in issue #6701. |
|
To try to fix my script, I tried to get I know that Thanks. |
|
By the way, this is a regression, so I'm not sure the "request" tag is accurate... |
|
@dstftw You might find the test I wrote useful, because it tests that the titles that are found match the known titles for certain videos. BTW, I appreciate your work on this project, and I know it's a very busy project with lots of people asking for fixes and submitting patches. Having said that, your "bedside manner" could use some work. Here was a regression that I reported twice, and after no response for a month on the second report, I fixed it myself, spending several hours figuring out how to work with the existing code. Then you abruptly instructed me to make certain changes. Being your project, I made them promptly, as requested. And I even wrote a comprehensive test to ensure that the regression wouldn't happen again. Then, three weeks passed with no response from any developers, the pull request waiting. Hey, people have lives, this is all volunteer work, that's fine. I'm busy too, that's why I didn't get around to working on it for a month. But then, with no comment, you merge your own code, completely disregarding all of my work. I understand that you wanted to refactor the code a bit, and that's fine. But it's rude to disregard someone's work in this way. The test I wrote was more comprehensive than the one you merged, yet you ignored it. And I even added missing docstrings and comments to make the code cleaner and easier to work with, but all of that was for naught. Well, I can handle being mildly snubbed. No big deal. The time wasn't entirely wasted, because I fixed the bug, which allowed my personal project to work again, and I learned some new things. And we're all human. But being treated this way doesn't encourage me to spend more time contributing in the future. And I'm just one random internet person. If this is a pattern, if this is how potential contributors are typically treated here, then it's a shame, because it's surely discouraging people who could help the project. Anyway, just some friendly advice from someone who's enjoyed using this project for several years. Thanks. |
|
I appreciate your efforts on fixing this issue and sorry for not commenting here - I'm working on several issues in row/parallel and don't always remember to comment everywhere. Here is rationale why I wasn't satisfied with solution provided. |
|
Thanks for your response. I hope I wasn't rude in what I said. I do understand that it was easier for you to do it than for me. Hey, if you have the time, I'm happy for you to do it yourself. I figured, as busy as you are, I would try to help reduce your workload. :) I do realize that if the playlists that are in the test change, the test will no longer pass. I chose a couple of videos that shouldn't be disappearing anytime soon, as they're part of a major channel that gets millions of hits every day. I thought it would be important to test that the titles were extracted correctly, matching known video titles, rather than simply testing that something was extracted. But you're the expert. :) I appreciate your offer, but it's not necessary to add me to the credits. I just want to see the project continue to do well. I know you get lots of bug reports and complaints, which tends to be a thankless job, but I'm glad you were able to get the time to fix it properly. Keep up the good work! :D |
Earlier this year I reported a similar issue, #4971. I guess YouTube has changed something on their end again. Hopefully it's still possible to get the video titles without fetching each video's page. It seems like it should be possible, because if I leave off
-j, I can see it fetching the playlist pages, and those include the video titles (at least, the HTML pages do), but I guessyoutube-dljust isn't parsing them out.If I leave off
--flat-playlist, I can get JSON with thefulltitleelement, but the JSON also includes everything, while all I want is the ID and title.Thanks for your help.