Details
-
Type:
Improvement
-
Status: Done
-
Resolution: Done
-
Fix Version/s: None
-
Component/s: squash
-
Labels:None
-
Story Points:4.2
-
Epic Link:
-
Team:SQuaRE
Description
In DM-18343 we enabled markdown content in the Chronograf table cells which was required to display code changes reported by SQuaSH
Here we'll make the necessary changes in the SQuaSH API /code_changes endpoint to return that information and in the SQuaSH InfluxDB data model to store this information.
Attachments
Issue Links
- relates to
-
- To Do
Activity
Investigating the very poor performance of the /code_changes API endpoint. I suspect it comes from a UNION clause.
[2019-03-20 19:45:16,405] DEBUG in decorators: Time for get_previous: 10.1883s
|
[2019-03-20 19:49:32,288] DEBUG in decorators: Time for compute_code_changes: 10.9606s
|
[2019-03-20 19:49:33,718] DEBUG in decorators: Time for get: 267.5192s
|
Looks much better now:
[2019-03-20 22:27:42,744] DEBUG in decorators: Time for get_current: 0.1779s
|
[2019-03-20 22:27:43,006] DEBUG in decorators: Time for get_previous: 0.2615s
|
[2019-03-20 22:27:43,007] DEBUG in decorators: Time for compute_code_changes: 0.0007s
|
[2019-03-20 22:27:43,009] DEBUG in decorators: Time for get: 0.4428s
|
The squash-demo database in InfluxDB was recreated with two new fields added to the data model:
1. code_changes_counts a metric for the number of packages that changed in the current run wrt to the previous one.
2. code_changes a string that gives the Markdown content for the Chronograf table linking each package to the corresponding git commit.
Refactor the SQLAlchemy code to retrieve code changes for the jenkins environment given the ci_id.