Change set verificationToken on User.confirm - #2426
Conversation
Use null instead of undefined
|
Can one of the admins verify this patch? To accept patch and trigger a build add comment ".ok\W+to\W+test." |
|
@slnode ok to test |
|
perhaps we should add a test case to prevent regression, not sure where we can add the test case tho, |
I'm not sure if I understand you correctly. Is this the same question you are asking in slack or it is different?
It is an expected behaviour see: https://github.com/strongloop/loopback-datasource-juggler/blob/058d9d46c09f249dd7aa14a09e3097aae8a7a21b/test/manipulation.test.js#L515-L526 |
|
Thanks @Amir-61 was trying to make sure the issue happens in all connectors, so the unit test could be added in loopback itself, if it was connector specific(which its not) maybe we would have needed to test on specific connector. @JonForest can you add a unit test to prevent regression, perhaps in this file Line 1147 in 42c83f6 |
|
related #2440 |
|
Will take a look at that unit test this weekend. |
|
@JonForest #2440 is basically a duplicate task, I would actually like to propose to close this and follow up there |
|
Works for me! As long as it gets fixed, all good. |
This addresses issue #2424
Simple change to setting the verificationToken to null, instead of undefined. undefined causes the field to not be updated via the Postgres adapter.
Note: the tests have a single failure for me at the moment, but they're also failing on master branch.