[GITEA] avoid superfluous synchronized pull_request run when opening a PR (#2236)

* Split TestPullRequest out of AddTestPullRequestTask
* Before scheduling the task, AddTestPullRequestTask stores the max
  index of the repository
* When the task runs, it does not take into account pull requests that
  have an index higher than the recorded max index

When AddTestPullRequestTask is called with isSync == true, it is the
direct consequence of a new commit being pushed. Forgejo knows nothing
of this new commit yet. If a PR is created later and its head
references the new commit, it will have an index that is higher and
must not be taken into account. It would be acting and triggering a
notification for a PR based on an event that happened before it
existed.

Refs: https://codeberg.org/forgejo/forgejo/issues/2009
Reviewed-on: https://codeberg.org/forgejo/forgejo/pulls/2236
Reviewed-by: Gusted <gusted@noreply.codeberg.org>
Co-authored-by: Earl Warren <contact@earl-warren.org>
Co-committed-by: Earl Warren <contact@earl-warren.org>
(cherry picked from commit b3be895a30b32bfae4acfa32db54406e1dd1dc21)

models/issues/issue_index.go

@@ -9,6 +9,14 @@ import (
 	"code.gitea.io/gitea/models/db"
 )
 
+func GetMaxIssueIndexForRepo(ctx context.Context, repoID int64) (int64, error) {
+	var max int64
+	if _, err := db.GetEngine(ctx).Select("MAX(`index`)").Table("issue").Where("repo_id=?", repoID).Get(&max); err != nil {
+		return 0, err
+	}
+	return max, nil
+}
+
 // RecalculateIssueIndexForRepo create issue_index for repo if not exist and
 // update it based on highest index of existing issues assigned to a repo
 func RecalculateIssueIndexForRepo(ctx context.Context, repoID int64) error {
@@ -18,8 +26,8 @@ func RecalculateIssueIndexForRepo(ctx context.Context, repoID int64) error {
 	}
 	defer committer.Close()
 
-	var max int64
-	if _, err = db.GetEngine(ctx).Select(" MAX(`index`)").Table("issue").Where("repo_id=?", repoID).Get(&max); err != nil {
+	max, err := GetMaxIssueIndexForRepo(ctx, repoID)
+	if err != nil {
 		return err
 	}
 

models/issues/issue_index_test.go

@@ -0,0 +1,38 @@
+// Copyright 2024 The Forgejo Authors
+// SPDX-License-Identifier: MIT
+
+package issues_test
+
+import (
+	"testing"
+
+	"code.gitea.io/gitea/models/db"
+	issues_model "code.gitea.io/gitea/models/issues"
+	repo_model "code.gitea.io/gitea/models/repo"
+	"code.gitea.io/gitea/models/unittest"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestGetMaxIssueIndexForRepo(t *testing.T) {
+	assert.NoError(t, unittest.PrepareTestDatabase())
+
+	repo := unittest.AssertExistsAndLoadBean(t, &repo_model.Repository{ID: 1})
+
+	maxPR, err := issues_model.GetMaxIssueIndexForRepo(db.DefaultContext, repo.ID)
+	assert.NoError(t, err)
+
+	issue := testCreateIssue(t, repo.ID, repo.OwnerID, "title1", "content1", false)
+	assert.Greater(t, issue.Index, maxPR)
+
+	maxPR, err = issues_model.GetMaxIssueIndexForRepo(db.DefaultContext, repo.ID)
+	assert.NoError(t, err)
+
+	pull := testCreateIssue(t, repo.ID, repo.OwnerID, "title2", "content2", true)
+	assert.Greater(t, pull.Index, maxPR)
+
+	maxPR, err = issues_model.GetMaxIssueIndexForRepo(db.DefaultContext, repo.ID)
+	assert.NoError(t, err)
+
+	assert.Equal(t, maxPR, pull.Index)
+}

models/issues/pull_list.go

@@ -50,6 +50,14 @@ func listPullRequestStatement(ctx context.Context, baseRepoID int64, opts *PullR
 	return sess, nil
 }
 
+func GetUnmergedPullRequestsByHeadInfoMax(ctx context.Context, repoID, maxIndex int64, branch string) ([]*PullRequest, error) {
+	prs := make([]*PullRequest, 0, 2)
+	sess := db.GetEngine(ctx).
+		Join("INNER", "issue", "issue.id = `pull_request`.issue_id").
+		Where("`pull_request`.head_repo_id = ? AND `pull_request`.head_branch = ? AND `pull_request`.has_merged = ? AND `issue`.is_closed = ? AND `pull_request`.flow = ? AND `issue`.`index` <= ?", repoID, branch, false, false, PullRequestFlowGithub, maxIndex)
+	return prs, sess.Find(&prs)
+}
+
 // GetUnmergedPullRequestsByHeadInfo returns all pull requests that are open and has not been merged
 func GetUnmergedPullRequestsByHeadInfo(ctx context.Context, repoID int64, branch string) ([]*PullRequest, error) {
 	prs := make([]*PullRequest, 0, 2)

models/issues/pull_test.go

@@ -4,6 +4,7 @@
 package issues_test
 
 import (
+	"fmt"
 	"testing"
 
 	"code.gitea.io/gitea/models/db"
@@ -158,6 +159,91 @@ func TestGetUnmergedPullRequestsByHeadInfo(t *testing.T) {
 	}
 }
 
+func TestGetUnmergedPullRequestsByHeadInfoMax(t *testing.T) {
+	assert.NoError(t, unittest.PrepareTestDatabase())
+
+	repoID := int64(1)
+	maxPR := int64(0)
+	prs, err := issues_model.GetUnmergedPullRequestsByHeadInfoMax(db.DefaultContext, repoID, maxPR, "branch2")
+	assert.NoError(t, err)
+	assert.Len(t, prs, 0)
+	maxPR, err = issues_model.GetMaxIssueIndexForRepo(db.DefaultContext, repoID)
+	assert.NoError(t, err)
+	prs, err = issues_model.GetUnmergedPullRequestsByHeadInfoMax(db.DefaultContext, repoID, maxPR, "branch2")
+	assert.NoError(t, err)
+	assert.Len(t, prs, 1)
+	for _, pr := range prs {
+		assert.Equal(t, int64(1), pr.HeadRepoID)
+		assert.Equal(t, "branch2", pr.HeadBranch)
+	}
+	pr := prs[0]
+
+	for _, testCase := range []struct {
+		table   string
+		field   string
+		id      int64
+		match   any
+		nomatch any
+	}{
+		{
+			table:   "issue",
+			field:   "is_closed",
+			id:      pr.IssueID,
+			match:   false,
+			nomatch: true,
+		},
+		{
+			table:   "pull_request",
+			field:   "flow",
+			id:      pr.ID,
+			match:   issues_model.PullRequestFlowGithub,
+			nomatch: issues_model.PullRequestFlowAGit,
+		},
+		{
+			table:   "pull_request",
+			field:   "head_repo_id",
+			id:      pr.ID,
+			match:   pr.HeadRepoID,
+			nomatch: 0,
+		},
+		{
+			table:   "pull_request",
+			field:   "head_branch",
+			id:      pr.ID,
+			match:   pr.HeadBranch,
+			nomatch: "something else",
+		},
+		{
+			table:   "pull_request",
+			field:   "has_merged",
+			id:      pr.ID,
+			match:   false,
+			nomatch: true,
+		},
+	} {
+		t.Run(testCase.field, func(t *testing.T) {
+			update := fmt.Sprintf("UPDATE `%s` SET `%s` = ? WHERE `id` = ?", testCase.table, testCase.field)
+
+			// expect no match
+			_, err = db.GetEngine(db.DefaultContext).Exec(update, testCase.nomatch, testCase.id)
+			assert.NoError(t, err)
+			prs, err = issues_model.GetUnmergedPullRequestsByHeadInfoMax(db.DefaultContext, repoID, maxPR, "branch2")
+			assert.NoError(t, err)
+			assert.Len(t, prs, 0)
+
+			// expect one match
+			_, err = db.GetEngine(db.DefaultContext).Exec(update, testCase.match, testCase.id)
+			assert.NoError(t, err)
+			prs, err = issues_model.GetUnmergedPullRequestsByHeadInfoMax(db.DefaultContext, repoID, maxPR, "branch2")
+			assert.NoError(t, err)
+			assert.Len(t, prs, 1)
+
+			// identical to the known PR
+			assert.Equal(t, pr.ID, prs[0].ID)
+		})
+	}
+}
+
 func TestGetUnmergedPullRequestsByBaseInfo(t *testing.T) {
 	assert.NoError(t, unittest.PrepareTestDatabase())
 	prs, err := issues_model.GetUnmergedPullRequestsByBaseInfo(db.DefaultContext, 1, "master")