| name | calculate_github_community_metrics |
| description | Processes JSON data from GitHub Search results to compute specific repository health metrics including average merge time, top contributors, and bug resolution counts. |
When calculating metrics from the JSON output of a paginated GitHub Search API call, use the following logic patterns:
PR Metrics
- Merged Count: Filter items where
merged_at != null.
- Closed Count: Filter items where
state == "closed" and merged_at == null.
- Average Merge Days:
- Filter for merged PRs.
- Calculate the difference between
merged_at and created_at using ( (.merged_at | fromdateiso8601) - (.created_at | fromdateiso8601) ) / 86400.
- Use
add / length to find the average and round or multiplication/division logic for decimal precision.
- Top Contributor: Use
group_by(.user.login), sort by group length, and select the login of the last group.
Issue Metrics
- Bug Identification: Use a case-insensitive check for the substring "bug" within any of the issue's labels. Logic:
.labels[].name | ascii_downcase | contains("bug").
- Resolved Bugs: To accurately report bugs resolved during a month, query for issues closed in that period (
closed:YYYY-MM-DD..YYYY-MM-DD) and then filter the results for labels containing "bug".
JQ Integration Example
To compute average merge time to one decimal place:
jq '[.[] | select(.merged_at != null) | ((.merged_at | fromdateiso8601) - (.created_at | fromdateiso8601)) / 86400] | (add / length * 10 | round / 10)' input.json
To identify the user who opened the most PRs:
jq -r 'group_by(.user.login) | sort_by(length) | last | .[0].user.login' input.json
