From 3766ab07d7061e946329443ed9a073ad3b41f457 Mon Sep 17 00:00:00 2001
From: Jarek Potiuk <jarek@potiuk.com>
Date: Tue, 15 Aug 2023 21:54:52 +0200
Subject: [PATCH] More complete fix for dev release scripts to filter commit
 for PR (#33418)

This is a more complete fix to #33411. This is also a follow up on
earlier implementation of #33261 that addressed checking if PRs
are merged. This one applies the same pattern to finding commit
but also improves it by checking if the (#NNNNNN) ends the subject
- so even if the PR is in the same form in the message, it will be
filtered out.

The previous "--reverse" quick fix in #33411 had potential of problem in
case there were releated PRs merged before the original PR (which is
quite posssible when you have a series of PRs referring to each other.
---
 dev/airflow-github | 24 ++++++++++++++----------
 1 file changed, 14 insertions(+), 10 deletions(-)

diff --git a/dev/airflow-github b/dev/airflow-github
index 69ca8ec53cc0..8109521e4ab6 100755
--- a/dev/airflow-github
+++ b/dev/airflow-github
@@ -83,27 +83,31 @@ def get_issue_type(issue):
 def get_commit_in_main_associated_with_pr(repo: git.Repo, issue: Issue) -> str | None:
     """For a PR, find the associated merged commit & return its SHA"""
     if issue.pull_request:
-        commit = repo.git.log("--reverse", f"--grep=#{issue.number}", "origin/main", "--format=%H")
-        if commit:
-            # We only want the oldest commit that referenced this PR number
-            return commit.splitlines()[0]
+        log_output = repo.git.log(f"--grep=(#{issue.number})$", "origin/main", "--format=%H %s")
+        if log_output:
+            for commit_line in log_output.splitlines():
+                # We only want the commit for the PR where squash-merge added (#PR) at the end of subject
+                if commit_line and commit_line.endswith(f"(#{issue.number})"):
+                    return commit_line.split(" ")[0]
+            return None
         else:
             pr: PullRequest = issue.as_pull_request()
             if pr.is_merged():
-                commit = pr.merge_commit_sha
-                return commit
+                return pr.merge_commit_sha
     return None
 
 
 def is_cherrypicked(repo: git.Repo, issue: Issue, previous_version: str | None = None) -> bool:
     """Check if a given issue is cherry-picked in the current branch or not"""
-    log_args = ["--format=%H", f"--grep=(#{issue.number})$"]
+    log_args = ["--format=%H %s", f"--grep=(#{issue.number})$"]
     if previous_version:
         log_args.append(previous_version + "..")
-    log = repo.git.log(*log_args)
+    log_output = repo.git.log(*log_args)
 
-    if log:
-        return True
+    for commit_line in log_output.splitlines():
+        # We only want the commit for the PR where squash-merge added (#PR) at the end of subject
+        if commit_line and commit_line.endswith(f"(#{issue.number})"):
+            return True
     return False