IGS / gEAR

The gEAR Portal was created as a data archive and viewer for gene expression data including microarrays, bulk RNA-Seq, single-cell RNA-Seq and more.
https://umgear.org
GNU Affero General Public License v3.0
13 stars 4 forks source link

Gene List Manager - How to handle "Lists shared with you" ownership #872

Open adkinsrs opened 4 weeks ago

adkinsrs commented 4 weeks ago

Currently for "ownership" filtering, there is an option to include/exclude gene lists that have been shared with you. Right now I just have this block on the server-side code set to "pass" because we have no current way to implement it. For the dataset explorer page, there is code to retrieve the share ID for a dataset if a user has been shared that dataset (via the dataset_shares table). Currently, no such equivalent table exists for gene lists.

So how to handle this?

@jorvis thoughts? I can make the code changes.

jorvis commented 4 weeks ago

To my knowledge the dataset_shares table isn't actually used at all. The only time shares are involved is if they're passed on the URL and they survive for that page load only. A solution to bring this back into use would be to insert into dataset_shares anytime a URL with a private share is accessed.

But since none of this is currently implemented anyway, we could drop dataset_shares and instead have something like "shares" which could generally store them for any object type.

shares

Certainly up for arguments.

adkinsrs commented 4 weeks ago

To my knowledge the dataset_shares table isn't actually used at all.

https://github.com/IGS/gEAR/blob/c23f910506aed171f1e73fb353535a2a016bc4f6/www/cgi/search_datasets.cgi#L120-L127. (plus line 54's function which queries the dataset_shares table)

adkinsrs commented 4 weeks ago

And as for updating the table to just be a generalized "shares", I have no arguments against it.

jorvis commented 4 weeks ago

Oh, absolutely. In the older interface there was a control page to formally share with a user, the user accepted, it inserted into the database, etc. But that was removed some time ago and there are probably some queries laying around which still reference it.