Analysis by GitHub issues closed from pull-request or not

[GeorgLink]

The goal of the analysis is to see whether GitHub issues are closed through pull-requests (either automatically or simply because an issue had a related, merged pull request) VERSUS issues are closed for other reasons (e.g., an idea dismissed without code change, request resolved).

The analysis may be expressed in total numbers and a ratio:

Week	Issues closed with PR	Issues closed without PR	Ratio with/total
1	12	88	12%
2	30	20	60%
3	5	15	25%

I don't think CHAOSS has documented this metric yet, so I started a conversation at https://github.com/chaoss/wg-evolution/issues/243

Is this supported? How would I do this in GrimoireLab?

[valeriocos]

The GitHub issue index contains part of this info. In a nutshell, if an issue contains PR data, the enriched item_type is pull request otherwise issue (see details at: https://github.com/chaoss/grimoirelab-elk/blob/c723df770bf8ff5f5bc5ee31551c77b73d89ac2c/grimoire_elk/enriched/github.py#L714). This should be enough to identify issues with PRs and issues without PRs.

What is missing from the issue index is whether the PRs were merged, this info is present in the GitHub pull request index (merged attribute: https://github.com/chaoss/grimoirelab-elk/blob/c723df770bf8ff5f5bc5ee31551c77b73d89ac2c/grimoire_elk/enriched/github.py#L600). However, I'm not sure whether the merged info is needed, since the goal is to measure talkative issue vs outcome-driven issues (so knowing if a PR was merged may be irrelevant).

[valeriocos]

@alpgarcia @sduenas feel free to jump into the conversation

[GeorgLink]

To clarify my request. Issue https://github.com/chaoss/wg-diversity-inclusion/issues/188 was closed automatically by merging pull request https://github.com/chaoss/wg-diversity-inclusion/pull/203

The analysis I want to see is the number of issues (like 188) that were linked to a merged pull request (like 203) compared to the rest of the issues