Clean coding standards results

Once this commit is deployed, the coding standards section results will look like this after they have been removed from the pift_ci_job_checkstyle table.

Closed issues won’t clear too much yet:

mysql> SELECT count(1), fis.field_issue_status_value IN (1,2,4,8,13,14,15,16) open FROM pift_ci_job_checkstyle jc INNER JOIN pift_ci_job j ON j.job_id = jc.job_id AND j.target_type = 'file' INNER JOIN field_data_field_issue_status fis ON fis.entity_id = j.issue_nid GROUP BY open;
+----------+------+
| count(1) | open |
+----------+------+
|      713 |    0 |
|   100316 |    1 |
+----------+------+

That includes Fixed issues as “open”. That is 65,106 rows which would be cleaned after their 2 weeks is up.

All non-current branch results

Rules I came up with:

• For custom one-off tests, keep for 30 days to match what is shown on project’s automated testing tab.
• For tests triggered on a schedule or commit, keep the most recent per-environment per-branch, and keep anything less than 24h old.

The results should be the same across environments, but they are all shown separately on the automated testing tab, so they are all expected to have results. I could see situations where having a little history (24h) would be useful for people.

This has been deployed to Drupal.org:

mysql> SELECT count(1) FROM pift_ci_job_checkstyle;
+----------+
| count(1) |
+----------+
|  7703181 |
+----------+
1 row in set (11.83 sec)

mysql> SELECT count(1) FROM pift_ci_job_checkstyle;
+----------+
| count(1) |
+----------+
|   617336 |
+----------+
1 row in set (0.30 sec)