-
Notifications
You must be signed in to change notification settings - Fork 89
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add support for migrating secret scanning resolution comments #1241
base: main
Are you sure you want to change the base?
Add support for migrating secret scanning resolution comments #1241
Conversation
Unit Test Results811 tests 811 ✅ 22s ⏱️ Results for commit cc8aa0c. ♻️ This comment has been updated with latest results. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for the contribution. Had a few questions but looks good overall.
{ | ||
state, | ||
resolution, | ||
resolution_comment = resolutionComment ?? string.Empty |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Why is it being set to empty string? If nothing is passed it makes sense for the default to be null
.
resolution_comment = resolutionComment ?? string.Empty | |
resolution_comment = resolutionComment |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hello, thanks for the comments.
Added it that way to align with what is done for the dismissedComment for code scanning alerts (cf. https://github.com/github/gh-gei/blob/main/src/Octoshift/Services/GithubApi.cs#L933 )
var payload = state == "open"
? (new { state })
: (object)(new
{
state,
dismissed_reason = dismissedReason,
dismissed_comment = dismissedComment ?? string.Empty
});
await _client.PatchAsync(url, payload);
Your suggestion makes sense though, so would you prefer me to do that in a separate pull request to align both changes, or change both it in this pull request?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
👋 ok got it. So I guess as you said the other one needs to be changed too. I think it's better to do that one in a separate PR.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Understood. I have opened #1248 and added your suggestions there. Thanks again!
@@ -2782,6 +2789,30 @@ public async Task UpdateSecretScanningAlert_Calls_The_Right_Endpoint_With_Payloa | |||
_githubClientMock.Verify(m => m.PatchAsync(url, It.Is<object>(x => x.ToJson() == payload.ToJson()), null)); | |||
} | |||
|
|||
|
|||
[Fact] | |||
public async Task UpdateSecretScanningAlert_Replaces_Null_Resolution_Comment_With_Empty_String() |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Based on my earlier comment we probably won't need this test.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Agreed.
And if we align with code scanning, we probably don't need that test either: UpdateCodeScanningAlert_Replaces_Null_Dismissed_Comment_With_Empty_String
As per previous comment, happy to do a separate pull request to align both code scanning and secret scanning, or do the changes in this one.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I have opened #1248 and added your suggestions there. Thanks again!
Happy to work on it next.
} | ||
|
||
[Fact] | ||
public async Task One_Secret_Updated_With_No_Resolution_Comment() |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Not sure if this adds value, seems like a bit of an overkill IMO.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What would be your preferred option:
- remove completely that test. This would align with the current tests for code scanning where a dismissedComment is always set (so we are not testing the use case where no comments were set)
- rework both code scanning and secret scanning tests to "squeeze" one alert with no comment (to exercise the code, but not as a separate test case).
- keep that test separate and add a similar overkill test in code scanning 😄
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
So I am leaning towards option 1, since the way I think about this is a scenario where no comments are set still means we're setting it but to a null
value and we have test coverage that shows we indeed use the passed value and pass it along.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks again for all the comments. I have now removed that test case.
Added support for migrating secret scanning resolution comments
ThirdPartyNotices.txt
(if applicable)Fixes #1240