You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When computing the oldestCommit value for BranchAge Bot Action, we currently calculate by comparing the created_at timestamp to the timestamp of the time we receive the Merge Request Hook event from GitLab.
However, I have observed cases where the created_at and committed_date values (along with the actual id of the commit) change while the authored_date stays the same for the same commit. Further investigation is required to understand under which circumstances this change takes place.
I propose that we update BranchAge to calculate oldestCommit value based on authored_date.
See below for a real example (with user data masked) pulled from the logs today:
The commit when the MR event first got sent to GRDBot
{
"name": "BranchAge",
"statusCode": 200,
"action": {
"goodGitPractice": false,
"mrNote": ":loudspeaker: This merge request has a pretty old commit. You should try and merge more frequently to keep your commits on branches fresh. [#BranchAgeAnalysis](https:\/\/github.com\/Cigna\/GitRDoneBot#2-branch-age)"
},
"computedValues": {
"oldestCommit": {
"id": "a1234",
"short_id": "a12",
"created_at": "2020-12-11T20:07:07.000Z",
"parent_ids": [],
"title": "does some stuff",
"message": "does some stuff",
"author_name": "River Tam",
"author_email": "RiverTam@fake.com",
"authored_date": "2020-12-08T20:18:48.000Z",
"committer_name": "River Tam",
"committer_email": "RiverTam@fake.com",
"committed_date": "2020-12-11T20:07:07.000Z"
}
}
The same commit when the MR sent an update event 10 minutes later
{
"name": "BranchAge",
"statusCode": 200,
"action": {
"goodGitPractice": true,
"mrNote": ":star: It\u2019s great that you\u2019re committing and merging code frequently - the commits on this branch aren\u2019t old or stale. Good job! [#BranchAgeAnalysis](https:\/\/github.com\/Cigna\/GitRDoneBot#2-branch-age)"
},
"computedValues": {
"oldestCommit": {
"id": "b1234",
"short_id": "b12",
"created_at": "2021-01-11T19:01:50.000Z",
"parent_ids": [],
"title": "does some stuff",
"message": "does some stuff",
"author_name": "River Tam",
"author_email": "RiverTam@fake.com",
"authored_date": "2020-12-08T20:18:48.000Z",
"committer_name": "River Tam",
"committer_email": "RiverTam@fake.com",
"committed_date": "2021-01-11T19:01:50.000Z"
}
}
}
The text was updated successfully, but these errors were encountered:
When computing the
oldestCommit
value forBranchAge
Bot Action, we currently calculate by comparing thecreated_at
timestamp to the timestamp of the time we receive the Merge Request Hook event from GitLab.However, I have observed cases where the
created_at
andcommitted_date
values (along with the actualid
of the commit) change while theauthored_date
stays the same for the same commit. Further investigation is required to understand under which circumstances this change takes place.I propose that we update
BranchAge
to calculateoldestCommit
value based onauthored_date
.See below for a real example (with user data masked) pulled from the logs today:
The commit when the MR event first got sent to GRDBot
The same commit when the MR sent an update event 10 minutes later
The text was updated successfully, but these errors were encountered: