stashapp
[Feature] Multiple Users with Permissions
Is your feature request related to a problem? Please describe. I would like to have my videos at a safe place. So my idea is that I would create 2 accounts for myself; an admin and a user. User account credentials are saved in the browser thus it is easy to access the videos. Admin account is only used if organinizing if needed.
Describe the solution you'd like I would like to have the possiblity to have multiple accouns and to configure the permissions that an account can have. Especially in case of important features like Deleting videos.
I think multiple users with a permissions system could be pretty cool for Stash.
You could possibly see a wide array of uses if it's within the scope of the devs vision for Stash. Amateur/professionals could host their own install, allowing subscribers under a user role to see specific scenes/images/galleries. You would probably see some tube sites start using Stash to host content. Even on a local network, user permissions would be useful. My initial thoughts:
- Multiple users would necessitate an additional table for storing user data (name, pass, etc.). This table would evolve for feature requests (email validation and what have you).
- Permissions would range from viewing/creating/editing records (scenes, performers, etc.), managing users, and so on.
- Permissions could be set for individual media, collections/galleries, and for user roles.
Of course something of this depth would be much further down the road; but I think it's worth getting ideas written out in advance.
You would probably see some tube sites start using Stash to host content.
I think that, partially, using sqlite, and primarily, reading directly from block storage means that this will never scale past 25 users. It would be a huge rewrite to make stash "scalable"
I came here to see if this was possible, I just wanted to create permissions for my wife, and ideally let her have her own "like" count, and more importantly keep mine private and library settings. Would be nice to restrict content too.
This is something I'd like to add at some point. As others have pointed out, it will require a substantial amount of effort.
A multi-user environment would be great. We'll need a place in the settings where accounts can be managed. Plus a minimum of two assignable roles: admin and user. Admin is what we have now. Here are some requirements for a limited "user" role that is mostly read-only:
Access restrictions:
- No access to the "Settings" page (this includes browsing to the /settings URL manually)
- No access to the "Edit" tab on any page. (The "rating" option is here on some of them, and we probably want to let users rate things on their own. So ratings should be moved to another tab.)
- No access to the Tagger view or "Auto Tag" buttons
- No access to the "Operations" button on any page (allowing rescan, item deletion, etc)
- No access to the "Organized" button
- No access to the "Submit to Stash-Box" button
- The "File info" tab I am on the fence about... do users need to know those details?
- Also not sure if we want them creating markers
- The Help menu would be role-dependent; the "user" version should not reference the admin panel
Things that should be unique to each user account (rather than globally set):
- Filters - let them manage their own
- Homepage - let them set their own homepage
- Ratings
- O-counter
A single read-only user in addition to the admin user may also serve a lot of usecases, and would require significantly less effort to implement.
@byertsy leave 👍 or 👎 on the first post instead of +1 and other upvote comments.
Any plans to implement this? Doesn't have to be scalable initially. Few users for local use would be sufficient. Kind of like what jellyfin implemented.
Any plans to implement this? Doesn't have to be scalable initially. Few users for local use would be sufficient. Kind of like what jellyfin implemented.
This is in the roadmap (which is not currently published) and I've added it to the backlog to signify this. I have a WIP branch if someone wants to pick up the work, but I've got a number of higher priority things to work on before I can properly implement this.
