Skip to content

⚡ Bolt: [성능 개선] label_section 다중 패턴 텍스트 스캔 복잡도 최적화 (O(L*N) -> O(N)) #1302

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
seonghobae wants to merge 5 commits into from
Closed

⚡ Bolt: [성능 개선] label_section 다중 패턴 텍스트 스캔 복잡도 최적화 (O(L*N) -> O(N)) #1302

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
4f667b5
⚡ Bolt: [성능 개선] label_section 다중 패턴 텍스트 스캔 복잡도 최적화 (O(L*N) -> O(N))
seonghobae Aug 24, 2026
84202e1
Merge branch 'main' into bolt-optimize-label-scan-7937440534787671718
seonghobae Aug 25, 2026
e1de793
perf(ci): avoid repeated label rescans
seonghobae Aug 25, 2026
9262da9
test(normalizer): keep coverage assertions in original test
seonghobae Aug 25, 2026
6250a5c
⚡ Bolt: [성능 개선] label_section 다중 패턴 텍스트 스캔 복잡도 최적화 (O(L*N) -> O(N))
seonghobae Aug 25, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 3 additions & 0 deletions .jules/bolt.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,3 +47,6 @@
## 2026-08-09 - [대용량 로그 스캔 시 정규표현식 실행 전 O(N) 서브스트링 검증 선행]
**Learning:** `classify_testthat_failure`에서 테스트 실패 내역이 없는 2MB 로그 파일을 대상으로 정규표현식을 실행하면 약 20ms가 소요되지만, 단순 문자열 검색은 약 1ms만 소요됩니다. 문자열 존재 여부가 정규표현식 매칭의 전제 조건일 때, 콜드 패스(Cold Path)에서 순서 최적화는 매우 큰 성능 차이를 만듭니다.
**Action:** 대용량 텍스트 입력(CI 로그 등)에서 복잡한 정규표현식을 파싱하기 전에 항상 빠른 O(N) 문자열 존재 여부 확인을 먼저 수행하십시오.
## 2024-11-28 - [성능 개선] 다중 라벨 텍스트 스캔 정규표현식 최적화
**Learning:** `scripts/ci/opencode_review_normalize_output.py`의 `label_section` 함수는 다음 섹션의 시작 위치를 찾기 위해 *모든* 검증 라벨에 대해 전체 남은 텍스트를 재스캔하는 리스트 컴프리헨션을 사용하여 `O(L * N)` 복잡도를 가졌으며, 이는 패턴이 많은 대용량 텍스트 로그에서 심각한 성능 병목이었습니다.
**Action:** 다중 하위 문자열 패턴에 대한 순차적 검색을 단일 컴파일된 정규표현식(`ANY_LABEL_PATTERN = re.compile("|".join(...))`)으로 결합하고 `.search(text, start)`를 사용하여 다음 매칭 섹션 마커로 O(N) 시간 내에 이동하도록 변경했습니다.
21 changes: 12 additions & 9 deletions scripts/ci/opencode_review_normalize_output.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -137,6 +137,8 @@
label: re.compile(re.escape(label)) for label in APPROVAL_VERIFICATION_LABELS
}

ANY_LABEL_PATTERN = re.compile("|".join(re.escape(label) for label in APPROVAL_VERIFICATION_LABELS))

@devin-ai-integration devin-ai-integration Bot Aug 25, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

📝 Info: Combined-regex correctness is fragile to future label additions

Python alternation is leftmost-then-first-alternative, not longest-match. ANY_LABEL_PATTERN (opencode_review_normalize_output.py) stays correct only because no label is a prefix of another. Adding a label that shares a start position with an existing one could produce a shorter, wrong match depending on tuple order.

Open in Devin Review

Was this helpful? React with 👍 or 👎 to provide feedback.


SOURCE_LIKE_CHANGED_FILE_EXTENSIONS = frozenset(
{
".bash",
Expand Down Expand Up @@ -979,15 +981,16 @@ def label_starts(candidate: str) -> list[int]:
if not starts:
return ""
Comment thread
seonghobae marked this conversation as resolved.
start = starts[-1] + len(label)
next_starts = [
candidate_start
for candidate in APPROVAL_VERIFICATION_LABELS
if candidate != label
for candidate_start in label_starts(candidate)
if candidate_start >= start
]
end = min(next_starts) if next_starts else len(text)
return text[start:end]

match = ANY_LABEL_PATTERN.search(text, start)
while match:
idx = match.start()
matched_label = match.group(0)
if matched_label != label and idx in label_starts(matched_label):
return text[start:idx]
match = ANY_LABEL_PATTERN.search(text, match.end())
Comment thread
coderabbitai[bot] marked this conversation as resolved.
Outdated

return text[start:]

@devin-ai-integration devin-ai-integration Bot Aug 24, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

📝 Info: label_section refactor is behaviorally equivalent

The rewrite preserves the original semantics. ANY_LABEL_PATTERN lists coverage: before docstring coverage:, but finditer is non-overlapping and left-anchored, so at the d of docstring coverage: it matches the whole label, not the inner coverage:; no label is a textual prefix of another, so alternation order causes no mismatch. starts[-1] picks the last target occurrence, and ascending match order makes the first later differing candidate equal to the old min(next_starts). The docstring skip at line 968 is now near-dead but harmless.

Open in Devin Review

Was this helpful? React with 👍 or 👎 to provide feedback.



def coverage_section_is_valid(section: str) -> bool:
Expand Down
Loading