Closed SunburntRock89 closed 1 year ago
This breaks the current tests which expect an ID field to be present at all times, will try fix
This does have me ask the question of what the difference between id
and sid
is.
id
is the primary key but sid
seems to be used to reference the object in most places I've looked.
Also: Feel free to tell me I'm doing stuff wrong here, I've never actually made a proper PR before :)
Appreciate the contribution, @SunburntRock89 !
This does have me ask the question of what the difference between id and sid is. id is the primary key but sid seems to be used to reference the object in most places I've looked.
The idea was to not require the session id to be the same as the database record id... I'm not 100% sure that we had a great reason for retaining this flexibility, but maybe there are situations where you might want it. (Maybe all of your db records have integer ids, but you want your session ids to be prefixed with se-
when debugging in javascript or something like that...)
I think this PR not only enables/fixes support for MongoDB, but also makes the library more bulletproof, as its now less possible (I think?) for sid
and id
to ever get out of sync (i.e. situations where they aren't the same, AND dbRecordIdFunction()
fails to accurately produce id
given sid
).
Thanks again!
:tada: This PR is included in version 3.1.7 :tada:
The release is available on:
Your semantic-release bot :package::rocket:
This fixes issue #83 and should provide proper MongoDB support.