inxomnyaa
commented
5 years ago I can confirm this, this was the main reason i reported to the owner of the repo.
I have not found time to analyze it yet, i will add it to my #todo list, and debug it asap.
I'm reporting the following bugs in case you weren't aware of it/them: 1) it's probably by design or limitations that the downloading queue gets cleared when the page is refreshed.
2) when downloading larger files (>30 MB) and a queue develops cause it can't download them as fast as requests get added, SCDL seems to have one of the following issues with requests that haven't yet started to download: 2a) SCDL stalls/fails to start downloading the next request and the downloads stay pending. "SCDL currently downloading #" in which the # doesn't decrease, tho will increase when more downloads are requested. The only fix is to refresh the page, which clears the queue (bug #1), then request the downloads again, usually one-by-one otherwise we are back to bug #2. 2b) SCDL queue seems to overwrite the source URL connected to the different requests, such SCDL downloads the same file, yet renames them with the intended meta data.
Example: If you add the following files in quick succession to be downloaded: Hello.mp3 (50 MB), My.mp3 (60 MB), Crazy.mp3 (70 MB), World.mp3 (80 MB) You may get:
Hope this helps and you're able to reproduce these. I didn't do extensive testing so it's possible there are more related bugs. I did do the above tests enough that they were reproduced and the above conclusions where reached. The difference whether the first file downloaded correctly or not was if a few second delay was given before adding the additional files, hence it already started downloading/caching.