Is there a way to configure the OPDS feed at all?

[SamKook]

I've been trying stump and it looks great so far with only minor issues, but I just realized that the OPDS v1.2 feed is using some name instead of the series for the folder name and, worst of all, it's adding that name at the beginning of the book filename which completely messes up the sorting in my reader since I rely on the filename.

I've been looking in the settings of the UI, but I don't even see any mention of OPDS anywhere so I was hoping I missed it or that it can be tweaked in a config file somewhere.

[kenny-ish]

There is no OPDS setting for this, in the UI or in a config file. The only OPDS-related server option is ENABLE_OPDS_PROGRESSION, and the per-library settings have nothing about feed naming. What you are seeing is metadata with a fallback: a series feed is titled with the name from a series.json in that folder when one exists, otherwise the folder name (v1_2.rs#L538-L540), and each book entry is titled with the metadata title (ComicInfo.xml <Title>, the EPUB title, or a title written by the metadata fetching integration), falling back to the file name only when no title is stored (entry.rs#L286-L291).

Stump itself does not prepend anything to the file. The acquisition link ends with the real file name and the download response sets Content-Disposition from the path on disk (serve_media.rs#L48-L57), so the prefix is your reader building names out of the feed or entry titles. The feed order itself is already by file name.

So the lever is the metadata, not Stump's config. If the series shows a name you did not choose, look for a series.json in that folder or a metadata fetch that filled the series title; the same goes for the book titles in ComicInfo.xml. Once no title is stored, the feed falls back to folder and file names. Switching to the v2.0 feed will not change anything, it uses the same fallback. The open request for OPDS ordering options is #653, if you want to add your use case there.

[SamKook]

I see. Comparing the feed with what I was using before(jellyfin OPDS plugin), it seems the issue is that librera(the reader I use) makes use of the authors field for the folder it saves the files in locally and adds it to the beginning of every files, but the jellyfin plugin simply used the folder name for the authors entry field(which suited me since that was the series name instead) since it didn't read metadata at all.

So what I need would be a way to control what gets inserted into the author field when creating the opds entry, which is kind of hacky or hope librera adds a configuration for how it saves files from OPDS, which is unlikely to happen any time soon since I asked about this last year and got no reply(also got no reply on a bug report that got fixed a while back but is still open so I don't think they pay attention to the git page all that much).

Thanks for the answer, at least I know what's happening now. I really like the musician reading mode(auto scrolls with a speed that can be adjusted on the fly) librera has so I guess I'll have to live with it.