fix(librarian): Filter for commits associated to a library (#2295)

Fixes: #2254

Changes:
- When parsing convention commits, the library association takes into
consideration the the commit description and tries to associate it with
the libraryID in the brackets (e.g. [LIBRARY-ID]). It is possible that
the brackets don't exist and in those cases, it will fall back to the
libraryID passed in (this will be based on the library's source root and
finds if the file modified is inside those source roots).
- Release Init will only update the `Changes` field in the library state
with the commits that are relevant. For example, if a nested commit has
6 nested commits for video and 4 for eventarc, then the changes will be
split accordingly.

---------

Signed-off-by: Lawrence Qiu <[email protected]>
Co-authored-by: Joe Wang <[email protected]>

Author: lqiu96

e2e_test.go

@@ -489,23 +489,23 @@ This pull request is generated with proto changes between
 
 BEGIN_COMMIT_OVERRIDE
 BEGIN_NESTED_COMMIT
-feat: [texttospeech] Support promptable voices by specifying a model name and a prompt
-feat: [texttospeech] Add enum value M4A to enum AudioEncoding
-docs: [texttospeech] A comment for method 'StreamingSynthesize' in service 'TextToSpeech' is changed
-docs: [texttospeech] A comment for enum value 'AUDIO_ENCODING_UNSPECIFIED' in enum 'AudioEncoding' is changed
-docs: [texttospeech] A comment for enum value 'OGG_OPUS' in enum 'AudioEncoding' is changed
-docs: [texttospeech] A comment for enum value 'PCM' in enum 'AudioEncoding' is changed
-docs: [texttospeech] A comment for field 'low_latency_journey_synthesis' in message '.google.cloud.texttospeech.v1beta1.AdvancedVoiceOptions' is changed
-docs: [texttospeech] A comment for enum value 'PHONETIC_ENCODING_IPA' in enum 'PhoneticEncoding' is changed
-docs: [texttospeech] A comment for enum value 'PHONETIC_ENCODING_X_SAMPA' in enum 'PhoneticEncoding' is changed
-docs: [texttospeech] A comment for field 'phrase' in message '.google.cloud.texttospeech.v1beta1.CustomPronunciationParams' is changed
-docs: [texttospeech] A comment for field 'pronunciations' in message '.google.cloud.texttospeech.v1beta1.CustomPronunciations' is changed
-docs: [texttospeech] A comment for message 'MultiSpeakerMarkup' is changed
-docs: [texttospeech] A comment for field 'custom_pronunciations' in message '.google.cloud.texttospeech.v1beta1.SynthesisInput' is changed
-docs: [texttospeech] A comment for field 'voice_clone' in message '.google.cloud.texttospeech.v1beta1.VoiceSelectionParams' is changed
-docs: [texttospeech] A comment for field 'speaking_rate' in message '.google.cloud.texttospeech.v1beta1.AudioConfig' is changed
-docs: [texttospeech] A comment for field 'audio_encoding' in message '.google.cloud.texttospeech.v1beta1.StreamingAudioConfig' is changed
-docs: [texttospeech] A comment for field 'text' in message '.google.cloud.texttospeech.v1beta1.StreamingSynthesisInput' is changed
+feat: [go-google-cloud-pubsub-v1] Support promptable voices by specifying a model name and a prompt
+feat: [go-google-cloud-pubsub-v1] Add enum value M4A to enum AudioEncoding
+docs: [go-google-cloud-pubsub-v1] A comment for method 'StreamingSynthesize' in service 'TextToSpeech' is changed
+docs: [go-google-cloud-pubsub-v1] A comment for enum value 'AUDIO_ENCODING_UNSPECIFIED' in enum 'AudioEncoding' is changed
+docs: [go-google-cloud-pubsub-v1] A comment for enum value 'OGG_OPUS' in enum 'AudioEncoding' is changed
+docs: [go-google-cloud-pubsub-v1] A comment for enum value 'PCM' in enum 'AudioEncoding' is changed
+docs: [go-google-cloud-pubsub-v1] A comment for field 'low_latency_journey_synthesis' in message '.google.cloud.texttospeech.v1beta1.AdvancedVoiceOptions' is changed
+docs: [go-google-cloud-pubsub-v1] A comment for enum value 'PHONETIC_ENCODING_IPA' in enum 'PhoneticEncoding' is changed
+docs: [go-google-cloud-pubsub-v1] A comment for enum value 'PHONETIC_ENCODING_X_SAMPA' in enum 'PhoneticEncoding' is changed
+docs: [go-google-cloud-pubsub-v1] A comment for field 'phrase' in message '.google.cloud.texttospeech.v1beta1.CustomPronunciationParams' is changed
+docs: [go-google-cloud-pubsub-v1] A comment for field 'pronunciations' in message '.google.cloud.texttospeech.v1beta1.CustomPronunciations' is changed
+docs: [go-google-cloud-pubsub-v1] A comment for message 'MultiSpeakerMarkup' is changed
+docs: [go-google-cloud-pubsub-v1] A comment for field 'custom_pronunciations' in message '.google.cloud.texttospeech.v1beta1.SynthesisInput' is changed
+docs: [go-google-cloud-pubsub-v1] A comment for field 'voice_clone' in message '.google.cloud.texttospeech.v1beta1.VoiceSelectionParams' is changed
+docs: [go-google-cloud-pubsub-v1] A comment for field 'speaking_rate' in message '.google.cloud.texttospeech.v1beta1.AudioConfig' is changed
+docs: [go-google-cloud-pubsub-v1] A comment for field 'audio_encoding' in message '.google.cloud.texttospeech.v1beta1.StreamingAudioConfig' is changed
+docs: [go-google-cloud-pubsub-v1] A comment for field 'text' in message '.google.cloud.texttospeech.v1beta1.StreamingSynthesisInput' is changed
 
 PiperOrigin-RevId: 799242210
 

internal/conventionalcommits/conventional_commits.go

@@ -42,6 +42,10 @@ var (
 	// e.g., "Reviewed-by: G. Gemini" or "BREAKING CHANGE: an API was changed".
 	footerRegex     = regexp.MustCompile(`^([A-Za-z-]+|` + breakingChangeKey + `):\s(.*)`)
 	sourceLinkRegex = regexp.MustCompile(`^\[googleapis\/googleapis@(?P<shortSHA>.*)\]\(https:\/\/github\.com\/googleapis\/googleapis\/commit\/(?P<sha>.*)\)$`)
+	// libraryIDRegex extracts the libraryID from the commit message in a generation PR.
+	// For a generation PR, each commit is expected to have the libraryID in brackets
+	// ('[]').
+	libraryIDRegex = regexp.MustCompile(`\[([^\]]+)\]`)
 )
 
 // ConventionalCommit represents a parsed conventional commit message.
@@ -80,6 +84,17 @@ type parsedHeader struct {
 	IsBreaking  bool
 }
 
+// extractLibraryID pulls the text between '[]' as the commit's libraryID
+// A non generation PR may not have the associated libraryID between [] and
+// will return an empty libraryID.
+func (header *parsedHeader) extractLibraryID() string {
+	matches := libraryIDRegex.FindStringSubmatch(header.Description)
+	if len(matches) == 0 {
+		return ""
+	}
+	return matches[1]
+}
+
 // commitPart holds the raw string of a commit message and whether it's nested.
 type commitPart struct {
 	message  string
@@ -212,6 +227,13 @@ func parseSimpleCommit(commitPart commitPart, commit *gitrepo.Commit, libraryID
 		}
 
 		subjects = append(subjects, []string{})
+		// If there is an association for the commit (i.e. the commit has '[LIBRARY_ID]' in the
+		// description), then use that libraryID. Otherwise, use the libraryID passed as the default.
+		headerLibraryID := header.extractLibraryID()
+		if headerLibraryID != "" {
+			libraryID = headerLibraryID
+		}
+
 		commits = append(commits, &ConventionalCommit{
 			Type:       header.Type,
 			Scope:      header.Scope,

internal/conventionalcommits/conventional_commits_test.go

@@ -35,7 +35,7 @@ func TestParseCommits(t *testing.T) {
 		wantErrPhrase string
 	}{
 		{
-			name:    "simple feat",
+			name:    "simple_commit_with_no_library_association",
 			message: "feat: add new feature",
 			want: []*ConventionalCommit{
 				{
@@ -51,7 +51,7 @@ func TestParseCommits(t *testing.T) {
 			},
 		},
 		{
-			name:    "feat with scope",
+			name:    "simple_commit_with_scope",
 			message: "feat(scope): add new feature",
 			want: []*ConventionalCommit{
 				{
@@ -68,7 +68,7 @@ func TestParseCommits(t *testing.T) {
 			},
 		},
 		{
-			name:    "feat with breaking change",
+			name:    "simple_commit_with_breaking_change",
 			message: "feat!: add new feature",
 			want: []*ConventionalCommit{
 				{
@@ -85,7 +85,7 @@ func TestParseCommits(t *testing.T) {
 			},
 		},
 		{
-			name:    "feat with single footer",
+			name:    "commit_with_single_footer",
 			message: "feat: add new feature\n\nCo-authored-by: John Doe <[email protected]>",
 			want: []*ConventionalCommit{
 				{
@@ -101,7 +101,7 @@ func TestParseCommits(t *testing.T) {
 			},
 		},
 		{
-			name:    "feat with multiple footers",
+			name:    "commit_with_multiple_footers",
 			message: "feat: add new feature\n\nCo-authored-by: John Doe <[email protected]>\nReviewed-by: Jane Smith <[email protected]>",
 			want: []*ConventionalCommit{
 				{
@@ -120,14 +120,14 @@ func TestParseCommits(t *testing.T) {
 			},
 		},
 		{
-			name:    "feat with multiple footers for generated changes",
+			name:    "commit_with_multiple_footers_for_generated_changes",
 			message: "feat: [library-name] add new feature\nThis is the body.\n...\n\nPiperOrigin-RevId: piper_cl_number\n\nSource-Link: [googleapis/googleapis@{source_commit_hash}](https://github.com/googleapis/googleapis/commit/abcdefg1234567)",
 			want: []*ConventionalCommit{
 				{
 					Type:       "feat",
 					Subject:    "[library-name] add new feature",
 					Body:       "This is the body.\n...",
-					LibraryID:  "example-id",
+					LibraryID:  "library-name",
 					IsNested:   false,
 					IsBreaking: false,
 					Footers: map[string]string{
@@ -141,7 +141,7 @@ func TestParseCommits(t *testing.T) {
 			},
 		},
 		{
-			name:    "feat with breaking change footer",
+			name:    "commit_with_breaking_change_footer",
 			message: "feat: add new feature\n\nBREAKING CHANGE: this is a breaking change",
 			want: []*ConventionalCommit{
 				{
@@ -159,7 +159,7 @@ func TestParseCommits(t *testing.T) {
 			},
 		},
 		{
-			name:    "feat with wrong breaking change footer",
+			name:    "commit_with_wrong_breaking_change_footer",
 			message: "feat: add new feature\n\nBreaking change: this is a breaking change",
 			want: []*ConventionalCommit{
 				{
@@ -177,7 +177,7 @@ func TestParseCommits(t *testing.T) {
 			},
 		},
 		{
-			name:    "feat with body and footers",
+			name:    "commit_with_body_and_footers",
 			message: "feat: add new feature\n\nThis is the body of the commit message.\nIt can span multiple lines.\n\nCo-authored-by: John Doe <[email protected]>",
 			want: []*ConventionalCommit{
 				{
@@ -194,7 +194,7 @@ func TestParseCommits(t *testing.T) {
 			},
 		},
 		{
-			name:    "feat with multi-line footer",
+			name:    "commit_with_multi-line_footer",
 			message: "feat: add new feature\n\nThis is the body.\n\nBREAKING CHANGE: this is a breaking change\nthat spans multiple lines.",
 			want: []*ConventionalCommit{
 				{
@@ -212,7 +212,7 @@ func TestParseCommits(t *testing.T) {
 			},
 		},
 		{
-			name: "commit override",
+			name: "commit_override",
 			message: `feat: original message
 
 BEGIN_COMMIT_OVERRIDE
@@ -238,13 +238,13 @@ END_COMMIT_OVERRIDE`,
 			},
 		},
 		{
-			name:    "invalid conventional commit",
+			name:    "invalid_conventional_commit",
 			message: "this is not a conventional commit",
 			wantErr: false,
 			want:    nil,
 		},
 		{
-			name:          "empty commit message",
+			name:          "empty_commit_message",
 			message:       "",
 			wantErr:       true,
 			wantErrPhrase: "empty commit",
@@ -303,7 +303,7 @@ END_NESTED_COMMIT
 			},
 		},
 		{
-			name: "commit with empty nested commit",
+			name: "commit_with_empty_nested_commit",
 			message: `feat(parser): main feature
 main commit body
 
@@ -326,7 +326,7 @@ END_NESTED_COMMIT
 			},
 		},
 		{
-			name: "commit override with nested commits",
+			name: "commit_override_with_nested_commits",
 			message: `feat: API regeneration main commit
 
 This pull request is generated with proto changes between
@@ -361,10 +361,10 @@ END_COMMIT_OVERRIDE
 					Type:       "feat",
 					Subject:    "[abc] nested commit 1",
 					Body:       "body of nested commit 1\n...",
-					LibraryID:  "example-id",
+					LibraryID:  "abc",
 					IsNested:   true,
 					Footers:    map[string]string{"PiperOrigin-RevId": "123456", "Source-Link": "fake-link"},
-					SHA:        sha.String(),
+					SHA:        sha.String(), // For each nested commit, the SHA should be the same (points to language repo's commit hash)
 					CommitHash: sha.String(),
 					When:       now,
 				},
@@ -373,16 +373,16 @@ END_COMMIT_OVERRIDE
 					Subject:    "[abc] nested commit 2",
 					IsNested:   true,
 					Body:       "body of nested commit 2\n...",
-					LibraryID:  "example-id",
+					LibraryID:  "abc",
 					Footers:    map[string]string{"PiperOrigin-RevId": "654321", "Source-Link": "fake-link"},
-					SHA:        sha.String(),
+					SHA:        sha.String(), // For each nested commit, the SHA should be the same (points to language repo's commit hash)
 					CommitHash: sha.String(),
 					When:       now,
 				},
 			},
 		},
 		{
-			name: "nest commit outside of override ignored",
+			name: "nest_commit_outside_of_override_ignored",
 			message: `feat: original message
 
 BEGIN_NESTED_COMMIT
@@ -411,7 +411,7 @@ END_NESTED_COMMIT`,
 			},
 		},
 		{
-			name: "parse multiple lines message inside nested commit, one line header",
+			name: "parse_multiple_lines_message_inside_nested_commit,_one_line_header",
 			message: `
 chore: Update generation configuration at Tue Aug 26 02:31:23 UTC 2025 (#11734)
 
@@ -436,7 +436,7 @@ END_COMMIT_OVERRIDE`,
 				{
 					Type:      "feat",
 					Subject:   "[texttospeech] Support promptable voices by specifying a model name and a prompt",
-					LibraryID: "example-id",
+					LibraryID: "texttospeech",
 					IsNested:  true,
 					Footers: map[string]string{
 						"PiperOrigin-RevId": "799242210",
@@ -449,7 +449,7 @@ END_COMMIT_OVERRIDE`,
 				{
 					Type:      "feat",
 					Subject:   "[texttospeech] Add enum value M4A to enum AudioEncoding",
-					LibraryID: "example-id",
+					LibraryID: "texttospeech",
 					IsNested:  true,
 					Footers: map[string]string{
 						"PiperOrigin-RevId": "799242210",
@@ -462,7 +462,7 @@ END_COMMIT_OVERRIDE`,
 				{
 					Type:      "docs",
 					Subject:   "[texttospeech] A comment for method 'StreamingSynthesize' in service 'TextToSpeech' is changed",
-					LibraryID: "example-id",
+					LibraryID: "texttospeech",
 					IsNested:  true,
 					Footers: map[string]string{
 						"PiperOrigin-RevId": "799242210",
@@ -499,7 +499,7 @@ END_COMMIT_OVERRIDE`,
 				{
 					Type:       "feat",
 					Subject:    "[texttospeech] Support promptable voices by specifying a model name and a prompt",
-					LibraryID:  "example-id",
+					LibraryID:  "texttospeech",
 					IsNested:   true,
 					Footers:    map[string]string{},
 					SHA:        sha.String(),
@@ -509,7 +509,7 @@ END_COMMIT_OVERRIDE`,
 				{
 					Type:       "feat",
 					Subject:    "[texttospeech] Add enum value M4A to enum AudioEncoding",
-					LibraryID:  "example-id",
+					LibraryID:  "texttospeech",
 					IsNested:   true,
 					Footers:    map[string]string{},
 					SHA:        sha.String(),
@@ -519,7 +519,54 @@ END_COMMIT_OVERRIDE`,
 				{
 					Type:       "docs",
 					Subject:    "[texttospeech] A comment for method 'StreamingSynthesize' in service 'TextToSpeech' is changed",
-					LibraryID:  "example-id",
+					LibraryID:  "texttospeech",
+					IsNested:   true,
+					Footers:    map[string]string{},
+					SHA:        sha.String(),
+					CommitHash: sha.String(),
+					When:       now,
+				},
+			},
+		},
+		{
+			name: "generation_commit_has_incorrect_libraryID_passed_in",
+			message: `
+chore: librarian generate pull request: 20250919T072957Z (#14501)
+
+This pull request is generated with proto changes between
+
+[googleapis/googleapis@f8776fe](googleapis/googleapis@f8776fe)
+(exclusive) and
+
+[googleapis/googleapis@36533b0](googleapis/googleapis@36533b0)
+(inclusive).
+
+BEGIN_COMMIT_OVERRIDE
+BEGIN_NESTED_COMMIT
+docs: [google-cloud-video-live-stream] Update requirements of resource ID fields to be more clear
+
+END_NESTED_COMMIT
+
+BEGIN_NESTED_COMMIT
+feat: [google-cloud-eventarc] add new fields to Eventarc resources
+
+END_NESTED_COMMIT
+END_COMMIT_OVERRIDE`,
+			want: []*ConventionalCommit{
+				{
+					Type:       "docs",
+					Subject:    "[google-cloud-video-live-stream] Update requirements of resource ID fields to be more clear",
+					LibraryID:  "google-cloud-video-live-stream",
+					IsNested:   true,
+					Footers:    map[string]string{},
+					SHA:        sha.String(),
+					CommitHash: sha.String(),
+					When:       now,
+				},
+				{
+					Type:       "feat",
+					Subject:    "[google-cloud-eventarc] add new fields to Eventarc resources",
+					LibraryID:  "google-cloud-eventarc",
 					IsNested:   true,
 					Footers:    map[string]string{},
 					SHA:        sha.String(),

internal/librarian/release_init.go

@@ -217,9 +217,22 @@ func (r *initRunner) processLibrary(library *config.LibraryState) error {
 	if err != nil {
 		return fmt.Errorf("failed to fetch conventional commits for library, %s: %w", library.ID, err)
 	}
+	// Filter specifically for commits relevant to a library
+	commits = filterCommitsByLibraryID(commits, library.ID)
 	return r.updateLibrary(library, commits)
 }
 
+// filterCommitsByLibraryID keeps the conventional commits that match a libaryID.
+func filterCommitsByLibraryID(commits []*conventionalcommits.ConventionalCommit, libraryID string) []*conventionalcommits.ConventionalCommit {
+	var filteredCommits []*conventionalcommits.ConventionalCommit
+	for _, commit := range commits {
+		if commit.LibraryID == libraryID {
+			filteredCommits = append(filteredCommits, commit)
+		}
+	}
+	return filteredCommits
+}
+
 // updateLibrary updates the library's state with the new release information:
 //
 // 1. Determines the library version's next version.