Repository navigation
Conversation
- analyzer_test.py: recall/precision comparison across ko.microsoft, ko.lucene, standard.lucene, keyword analyzers - synonym_test.py: synonym map effect (complements, not replaces, analyzer) - sample_data.json: 6 relevant + 4 distractor docs - README.md: customer diagnosis/remediation/verification playbook - deps pinned: azure-search-documents==12.0.0, azure-core==1.41.0, python-dotenv==1.2.2
📝 WalkthroughWalkthroughAdds a new ChangesAzure AI Search Korean Analyzer Comparison
Estimated code review effort: 3 (Moderate) | ~25 minutes Sequence Diagram(s)sequenceDiagram
participant Script as Test Script (analyzer_test.py / synonym_test.py)
participant IndexClient as SearchIndexClient
participant SearchClient as SearchClient
participant SampleData as sample_data.json
Script->>SampleData: load_data()
Script->>IndexClient: create_index() / create_synonym_map()
Script->>SearchClient: upload_docs()
Script->>SearchClient: search_hits() per query/analyzer/field
SearchClient-->>Script: matched document ids
Script->>Script: recall_report() computes recall and false positives
Compact metadata Poem 🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 2
🧹 Nitpick comments (2)
aisearch-v2/analyzer_test.py (1)
125-130: 📐 Maintainability & Code Quality | 🔵 Trivial | 💤 Low valueUnused loop variable
field_name.Only
analyzeris used in the loop body; static analysis correctly flags this.♻️ Proposed fix
- for field_name, analyzer in ANALYZERS.items(): + for _field_name, analyzer in ANALYZERS.items(): result = client.analyze_text(🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@aisearch-v2/analyzer_test.py` around lines 125 - 130, The loop in analyzer_test.py iterates over ANALYZERS.items() but does not use the field_name variable, so update the iteration in the test helper to avoid binding an unused name. Keep the logic in the analyze_text loop centered on ANALYZERS and analyzer, and either iterate only over the analyzer values or explicitly use both names if the field name is needed later.Source: Linters/SAST tools
aisearch-v2/synonym_test.py (1)
1-27: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick winInvalid escape sequences in docstring — use a raw string.
The Windows-path example (
.\.venv\Scripts\python.exe) inside this non-raw docstring triggers deprecation warnings for\.,\S,\pand is not forward-compatible with future Python releases that plan to make invalid escapes a hard error.♻️ Proposed fix
-""" +r""" Azure AI Search 동의어(Synonym Map) 효과 비교 테스트🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@aisearch-v2/synonym_test.py` around lines 1 - 27, The module docstring in synonym_test.py contains Windows backslash sequences in the execution example, which causes invalid escape warnings in a normal triple-quoted string. Update the top-level docstring to be a raw string (or otherwise escape the backslashes) so the path shown in the example is preserved without triggering invalid escape handling. Keep the change localized to the module docstring near the execution instructions.Source: Linters/SAST tools
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@aisearch-v2/README.md`:
- Line 186: The README section heading contains a broken replacement character
before the heading text, indicating an encoding issue; update the affected
headings in the README content so the title renders cleanly without the stray
glyph. Check the duplicated heading entries referenced by the Synonym Map
section and remove any corrupted character sequences while preserving the emoji
and Korean text.
- Around line 226-236: The README example for SynonymMap creation is using the
wrong shape for the synonyms value. Update the SynonymMap example to pass a list
of rule strings via the SynonymMap.synonyms field instead of building a
newline-joined string, and keep the example aligned with the SDK usage shown by
the SynonymMap and index_client.create_or_update_synonym_map symbols.
---
Nitpick comments:
In `@aisearch-v2/analyzer_test.py`:
- Around line 125-130: The loop in analyzer_test.py iterates over
ANALYZERS.items() but does not use the field_name variable, so update the
iteration in the test helper to avoid binding an unused name. Keep the logic in
the analyze_text loop centered on ANALYZERS and analyzer, and either iterate
only over the analyzer values or explicitly use both names if the field name is
needed later.
In `@aisearch-v2/synonym_test.py`:
- Around line 1-27: The module docstring in synonym_test.py contains Windows
backslash sequences in the execution example, which causes invalid escape
warnings in a normal triple-quoted string. Update the top-level docstring to be
a raw string (or otherwise escape the backslashes) so the path shown in the
example is preserved without triggering invalid escape handling. Keep the change
localized to the module docstring near the execution instructions.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro Plus
Run ID: a8a64f28-7679-47eb-a20f-52d14cd905eb
⛔ Files ignored due to path filters (2)
aisearch-v2/images/portal_ko_microsoft.pngis excluded by!**/*.pngaisearch-v2/images/portal_standard_lucene.pngis excluded by!**/*.png
📒 Files selected for processing (8)
aisearch-v2/.env.exampleaisearch-v2/.gitignoreaisearch-v2/README.mdaisearch-v2/analyzer_test.pyaisearch-v2/images/.gitkeepaisearch-v2/requirements.txtaisearch-v2/sample_data.jsonaisearch-v2/synonym_test.py
|
|
||
| --- | ||
|
|
||
| ## � 재현 3 — 동의어(Synonym Map)로 정확도 더 높이기 |
There was a problem hiding this comment.
📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick win
Broken/garbled characters in section headings.
Both headings contain a replacement character (�) before the emoji/text, indicating an encoding issue that will render as a broken glyph.
✏️ Proposed fix
-## � 재현 3 — 동의어(Synonym Map)로 정확도 더 높이기
+## 🧪 재현 3 — 동의어(Synonym Map)로 정확도 더 높이기-## �🛠️ 고객 적용 가이드 (진단 → 조치 → 검증)
+## 🛠️ 고객 적용 가이드 (진단 → 조치 → 검증)Also applies to: 250-250
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@aisearch-v2/README.md` at line 186, The README section heading contains a
broken replacement character before the heading text, indicating an encoding
issue; update the affected headings in the README content so the title renders
cleanly without the stray glyph. Check the duplicated heading entries referenced
by the Synonym Map section and remove any corrupted character sequences while
preserving the emoji and Korean text.
Comment on lines
+226
to
+236
| 동의어 맵 등록/필드 연결(SDK) 예시: | ||
| ```python | ||
| from azure.search.documents.indexes.models import SynonymMap, SearchableField, SearchFieldDataType | ||
|
|
||
| # 1) 동의어 맵 등록 | ||
| sm = SynonymMap(name="promo-synonyms", synonyms="\n".join([ | ||
| "온누리상품권, 온누리 상품권, 온누리", | ||
| "지역사랑상품권, 지역화폐", | ||
| ])) | ||
| index_client.create_or_update_synonym_map(sm) | ||
|
|
There was a problem hiding this comment.
🎯 Functional Correctness | 🟠 Major | ⚡ Quick win
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash
set -euo pipefail
# Locate the README and any related synonym-map examples or tests.
git ls-files 'aisearch-v2/README.md' 'aisearch-v2/**' | sed -n '1,200p'
printf '\n--- README excerpt ---\n'
sed -n '220,245p' aisearch-v2/README.md
printf '\n--- Search for SynonymMap usage ---\n'
rg -n "SynonymMap|synonyms=" aisearch-v2 -S
printf '\n--- Search for the specific warning / test mentioned in the review comment ---\n'
rg -n "character-by-character|join\\(self\\.synonyms\\)|synonym_test|broken synonym map" aisearch-v2 -SRepository: hellices/devguidesample
Length of output: 2406
🏁 Script executed:
#!/bin/bash
set -euo pipefail
printf '\n--- synonym_test.py excerpt ---\n'
sed -n '70,110p' aisearch-v2/synonym_test.py
printf '\n--- requirements.txt ---\n'
cat -n aisearch-v2/requirements.txt
printf '\n--- locate any SynonymMap docs or type hints in the repo ---\n'
rg -n "List\\[str\\]|SynonymMap\\(|synonym map|synonyms=" aisearch-v2 -SRepository: hellices/devguidesample
Length of output: 2081
🏁 Script executed:
#!/bin/bash
set -euo pipefail
# Read the Azure Search SDK version and, if available, inspect installed metadata without running repo code.
python3 - <<'PY'
from pathlib import Path
p = Path("aisearch-v2/requirements.txt")
for line in p.read_text(encoding="utf-8").splitlines():
if "azure-search-documents" in line:
print(line)
PYRepository: hellices/devguidesample
Length of output: 192
Use synonyms=[...] here
SynonymMap.synonyms should be a list of rule strings; "\n".join([...]) will serialize incorrectly and break the map. Use the list form instead.
Suggested fix
-sm = SynonymMap(name="promo-synonyms", synonyms="\n".join([
+sm = SynonymMap(name="promo-synonyms", synonyms=[
"온누리상품권, 온누리 상품권, 온누리",
"지역사랑상품권, 지역화폐",
-]))
+])📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
Suggested change
| 동의어 맵 등록/필드 연결(SDK) 예시: | |
| ```python | |
| from azure.search.documents.indexes.models import SynonymMap, SearchableField, SearchFieldDataType | |
| # 1) 동의어 맵 등록 | |
| sm = SynonymMap(name="promo-synonyms", synonyms="\n".join([ | |
| "온누리상품권, 온누리 상품권, 온누리", | |
| "지역사랑상품권, 지역화폐", | |
| ])) | |
| index_client.create_or_update_synonym_map(sm) | |
| 동의어 맵 등록/필드 연결(SDK) 예시: |
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@aisearch-v2/README.md` around lines 226 - 236, The README example for
SynonymMap creation is using the wrong shape for the synonyms value. Update the
SynonymMap example to pass a list of rule strings via the SynonymMap.synonyms
field instead of building a newline-joined string, and keep the example aligned
with the SDK usage shown by the SynonymMap and
index_client.create_or_update_synonym_map symbols.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
개요
Azure AI Search에서 필드 분석기(Analyzer) 선택이 한국어 검색 품질(recall/precision)에 미치는 영향을 실제 리소스로 재현·검증한 고객 가이드입니다.
ko.microsoft가 복합명사·조사 처리에서 가장 우수함을, 정답/오답(distractor) 데이터로 정량 검증합니다.포함 내용
analyzer_test.py— 4개 분석기(ko.microsoft,ko.lucene,standard.lucene,keyword) recall/precision 비교 (인덱스 생성 → 업로드 → Analyze 토큰 비교 → 검색 히트 → recall 리포트)synonym_test.py— 동의어 맵 효과 검증 (동의어는 분석기를 대체하지 않고 보완함을 실측)sample_data.json— 정답 6건 + distractor 4건 = 10건README.md— 진단 → 조치 → 검증 playbook + FAQrequirements.txt— 버전 고정:azure-search-documents==12.0.0,azure-core==1.41.0,python-dotenv==1.2.2.env.example,.gitignore검증 결과 (실제 실행, 두 스크립트 모두 EXIT=0)
msft+syn5/5,standard0/5, 문서에 없는 별칭컬처랜드0→3 회수ko.microsoft+ 동의어 맵 조합 권장 (analyzer 변경 시 재색인 필수)비고
.env(비밀키)와.venv/는.gitignore로 제외됨azure-search-documents12.0.0(major)의 breaking change는 본 스크립트가 사용하는 API에 영향 없음을 확인Summary by CodeRabbit
New Features
Documentation
Chores