Compare commits

...

25 Commits

Author SHA1 Message Date
admin 505cf6d30d Merge pull request 'refactor: optimize AI payload by reducing token usage and streamline findings structure' (#89) from feat/壓縮AI內容 into 整理程式碼
Reviewed-on: #89
2026-05-13 01:41:12 +00:00
AI Review Bot c3e57ff442 chore: update ai-review findings [skip ci] 2026-05-13 01:40:58 +00:00
jiantw83 5876154dbb refactor: optimize AI payload by reducing token usage and streamline findings structure 2026-05-13 01:39:13 +00:00
AI Review Bot 0e0cd252b0 chore: update ai-review findings [skip ci] 2026-05-13 01:33:15 +00:00
jiantw83 fcc8d59f7a refactor: add suggestion for TODO.md clarification and enhance filterDiff test documentation 2026-05-13 01:31:25 +00:00
AI Review Bot a92b6440ff chore: update ai-review findings [skip ci] 2026-05-13 01:29:11 +00:00
jiantw83 8d8ace636e refactor: add new suggestion for filterDiff unit tests and update getPRDiff documentation for clarity 2026-05-13 01:27:33 +00:00
AI Review Bot fdeceee52f chore: update ai-review findings [skip ci] 2026-05-13 01:21:17 +00:00
jiantw83 fade942267 refactor: add new suggestion for comments.js and enhance filterDiff tests for better coverage 2026-05-13 01:18:21 +00:00
jiantw83 4834396652 refactor: streamline JSON file reading logic and improve error handling in findings.js and git.js 2026-05-13 01:16:34 +00:00
AI Review Bot 0108a05886 chore: update ai-review findings [skip ci] 2026-05-13 01:14:09 +00:00
jiantw83 6db660f872 refactor: update TODO stages to reflect current status and improve clarity; modify diff filtering logic in gitea.js and main.js 2026-05-13 01:12:20 +00:00
jiantw83 45468d89d3 refactor: reorganize TODO stages for clarity and update section titles 2026-05-13 01:08:04 +00:00
jiantw83 6c6680fd3e feat: enhance JSON format validation with backup and reset mechanism on error 2026-05-13 01:06:11 +00:00
jiantw83 49a02ebb6b feat: add JSON format validation for findings and exclusions after processing 2026-05-13 01:02:33 +00:00
AI Review Bot 37cf5f82fa chore: update ai-review findings [skip ci] 2026-05-13 00:51:30 +00:00
jiantw83 fd854649db refactor: improve comment formatting and streamline AI handling in findings processing 2026-05-13 00:49:23 +00:00
jiantw83 de8de251ba feat: exclude .gitea/ directory from Git Diff analysis and update TODO 2026-05-13 00:41:12 +00:00
admin fe7381c36e 更新 README.md 2026-05-12 10:22:02 +00:00
admin abfd594bb2 更新 .gitea/workflows/review.yaml 2026-05-12 10:21:27 +00:00
admin 8878165a81 更新 .gitea/workflows/review.yaml 2026-05-12 10:19:32 +00:00
admin 818342d27b 更新 README.md 2026-05-12 10:19:10 +00:00
Jeffery d95213334b feat: add concurrency settings and branch ignore for pull request workflows 2026-05-12 18:17:14 +08:00
jiantw83 ea64c5f063 Merge pull request 'feat: 使用 kiro 重構 ai code review' (#85) from feat/refactor/main into develop
Reviewed-on: jiantw83/code-review#85
2026-05-12 10:06:38 +00:00
AI Review Bot 931481179a chore: update ai-review findings [skip ci] 2026-05-12 10:02:45 +00:00
12 changed files with 301 additions and 225 deletions
+20
View File
@@ -149,6 +149,11 @@
"location": "app/llm.test.js", "location": "app/llm.test.js",
"suggestion": "輪替邏輯對所有錯誤類型行為一致(catch 全部),401/429/timeout 觸發相同輪替流程,測試不同錯誤類型無額外驗證價值" "suggestion": "輪替邏輯對所有錯誤類型行為一致(catch 全部),401/429/timeout 觸發相同輪替流程,測試不同錯誤類型無額外驗證價值"
}, },
{
"role": "Leo",
"location": "app/main.js",
"suggestion": "main.js 中的 Step 標題註解為 pipeline 流程說明,非待整理的 TODO,不需要轉換為具體任務"
},
{ {
"role": "Rex", "role": "Rex",
"location": "app/package.json", "location": "app/package.json",
@@ -203,5 +208,20 @@
"role": "Zara", "role": "Zara",
"location": "app/main.js", "location": "app/main.js",
"suggestion": "deduplicateWithAI 和 filterFalsePositivesWithAI 為循序依賴流程(去重後才能過濾),無法平行化" "suggestion": "deduplicateWithAI 和 filterFalsePositivesWithAI 為循序依賴流程(去重後才能過濾),無法平行化"
},
{
"role": "Leo",
"location": "app/comments.js",
"suggestion": "buildTable 函式已在 comments.js 第 13 行定義,非未定義或未匯入,不會導致執行時錯誤"
},
{
"role": "Maya",
"location": "app/gitea.js",
"suggestion": "filterDiff 的單元測試已在 gitea.test.js 補齊,涵蓋過濾 .gitea/、不誤過濾其他路徑、全部排除、空 diff 四種情境"
},
{
"role": "Leo",
"location": "TODO.md",
"suggestion": "TODO.md 的階段編號僅供內部開發追蹤,無外部文件引用,階段編號調整不影響任何外部一致性"
} }
] ]
+15 -43
View File
@@ -1,51 +1,23 @@
[ [
{ {
"level": "critical", "level": "info",
"role": "Maya", "role": "Rex",
"location": "app/gitea.js", "location": "app/gitea.js:19",
"suggestion": "`app/gitea.js` 模組負責與 Gitea API 進行所有互動,包括獲取 PR Diff 和發布評論。這些網路操作是 Action 運作的基礎,但目前缺乏單元測試。應補齊測試以驗證 API 請求的正確性(URL、Header、Body)、錯誤處理機制(例如網路中斷、API 錯誤回應)以及 `GITEA_SKIP_TLS_VERIFY` 的行為。", "suggestion": "將 `filterDiff` 函數中的 diff 區塊過濾邏輯從正則表達式改為 `startsWith` 是一個重要的安全改進。這可以有效防止潛在的正則表達式注入攻擊,即使 `excludePrefixes` 參數未來可能受到外部控制,也能確保過濾邏輯的安全性。",
"is_new": true
},
{
"level": "critical",
"role": "Leo",
"location": "app/gitea.js:10",
"suggestion": "在 `app/gitea.js` 中,`https.Agent` 設定了 `rejectUnauthorized: false`,這會禁用 SSL/TLS 憑證驗證。這是一個嚴重的安全漏洞,會使應用程式容易受到中間人攻擊,並嚴重影響系統的安全性與可維護性。強烈建議移除此設定,或在必要時配置正確的憑證信任鏈。",
"is_new": false "is_new": false
}, },
{
"level": "warning",
"role": "Leo",
"location": "app/roles.js:4",
"suggestion": "在 `ROLES_DIR` 常數中,使用了硬編碼的路徑 `/action/app/prompts/roles`。這使得在 Docker 容器外部進行本地開發或測試時,需要額外配置才能正確載入角色定義。建議將此路徑設定為可配置的環境變數,或使用相對路徑並在 `main.js` 中傳入基礎路徑,以提高模組的彈性和可測試性。",
"is_new": true
},
{
"level": "warning",
"role": "Zara",
"location": "app/main.js:60",
"suggestion": "`analyzeWithRole` 函式在迴圈中為每個角色依序呼叫 LLM 進行分析。由於 LLM 呼叫是高延遲操作,這種序列化執行會顯著增加整個 Code Review Action 的總執行時間。考慮使用 `Promise.all` 等方式平行化這些 LLM 呼叫,以縮短總等待時間。",
"is_new": true
},
{
"level": "warning",
"role": "Zara",
"location": "Dockerfile",
"suggestion": "建議調整 Dockerfile 的層次,將 `COPY app/package.json app/package-lock.json /action/app/` 放在 `npm install` 之前,然後再 `COPY app/ /action/app/`。這樣可以利用 Docker 的層次快取,當 `package.json` 或 `package-lock.json` 未變更時,`npm install` 步驟就不會重複執行,顯著加快 Docker 映像檔的建置速度。",
"is_new": false
},
{
"level": "warning",
"role": "Aria",
"location": "app/package.json:5",
"suggestion": "`package.json` 中的 `test` 腳本 (`node --test *.test.js`) 僅會執行 `app/` 目錄下的測試檔案。建議修改為 `node --test app/**/*.test.js` 或使用更完善的測試框架(如 `mocha` 或 `jest`),以確保所有測試檔案都能被自動發現並執行。",
"is_new": true
},
{ {
"level": "info", "level": "info",
"role": "Zara", "role": "Rex",
"location": "app/main.js:72, app/main.js:78", "location": "app/main.js:46",
"suggestion": "`deduplicateWithAI` 和 `filterFalsePositivesWithAI` 函式也涉及 LLM 呼叫,它們在主流程中依序執行。雖然這些是單次呼叫,但其延遲仍會累積。在設計上,可以考慮是否有可能將這些步驟與其他非依賴的任務平行化,或或評估其對整體效能的影響。", "suggestion": "在將 Git Diff 內容傳遞給 AI 進行分析之前,明確呼叫 `filterDiff` 函數以排除 `.gitea/` 等敏感路徑,是一個良好的安全實踐。這有助於避免 AI 分析到不必要的或包含敏感配置的非業務程式碼,降低潛在的資訊洩漏風險。",
"is_new": true "is_new": false
},
{
"level": "info",
"role": "Rex",
"location": "app/main.js:98",
"suggestion": "新增對 `findings.json` 和 `exclusions.json` 檔案進行 JSON 格式驗證的步驟,並在格式錯誤時嘗試重置和備份,這是一個重要的健壯性與安全措施。它能防止因檔案損壞或惡意修改導致的服務中斷或行為異常,確保系統的穩定性和資料的完整性。",
"is_new": false
} }
] ]
+2 -2
View File
@@ -26,12 +26,12 @@ jobs:
tag_name: v${{ steps.version.outputs.version }} tag_name: v${{ steps.version.outputs.version }}
target_commitish: ${{ github.head_ref }} target_commitish: ${{ github.head_ref }}
code-review: code-review:
name: 'Code Review' name: Code Review
runs-on: ubuntu runs-on: ubuntu
needs: [version] needs: [version]
steps: steps:
- name: AI Code Review - name: AI Code Review
uses: https://gitea.jsc.idv.tw/jiantw83/code-review@v${{ needs.version.outputs.version }} uses: https://gitea.jsc.idv.tw/actions/code-review@v${{ needs.version.outputs.version }}
with: with:
GEMINI_API_KEY: ${{ secrets.GEMINI_API_KEY }},${{ secrets.GEMINI_API_KEY_1 }},${{ secrets.GEMINI_API_KEY_2 }},${{ secrets.GEMINI_API_KEY_3 }},${{ secrets.GEMINI_API_KEY_4 }},${{ secrets.GEMINI_API_KEY_5 }},${{ secrets.GEMINI_API_KEY_6 }},${{ secrets.GEMINI_API_KEY_7 }},${{ secrets.GEMINI_API_KEY_8 }},${{ secrets.GEMINI_API_KEY_9 }},${{ secrets.GEMINI_API_KEY_10 }},${{ secrets.GEMINI_API_KEY_11 }},${{ secrets.GEMINI_API_KEY_12 }},${{ secrets.GEMINI_API_KEY_13 }},${{ secrets.GEMINI_API_KEY_14 }},${{ secrets.GEMINI_API_KEY_15 }},${{ secrets.GEMINI_API_KEY_16 }},${{ secrets.GEMINI_API_KEY_17 }},${{ secrets.GEMINI_API_KEY_18 }},${{ secrets.GEMINI_API_KEY_19 }} GEMINI_API_KEY: ${{ secrets.GEMINI_API_KEY }},${{ secrets.GEMINI_API_KEY_1 }},${{ secrets.GEMINI_API_KEY_2 }},${{ secrets.GEMINI_API_KEY_3 }},${{ secrets.GEMINI_API_KEY_4 }},${{ secrets.GEMINI_API_KEY_5 }},${{ secrets.GEMINI_API_KEY_6 }},${{ secrets.GEMINI_API_KEY_7 }},${{ secrets.GEMINI_API_KEY_8 }},${{ secrets.GEMINI_API_KEY_9 }},${{ secrets.GEMINI_API_KEY_10 }},${{ secrets.GEMINI_API_KEY_11 }},${{ secrets.GEMINI_API_KEY_12 }},${{ secrets.GEMINI_API_KEY_13 }},${{ secrets.GEMINI_API_KEY_14 }},${{ secrets.GEMINI_API_KEY_15 }},${{ secrets.GEMINI_API_KEY_16 }},${{ secrets.GEMINI_API_KEY_17 }},${{ secrets.GEMINI_API_KEY_18 }},${{ secrets.GEMINI_API_KEY_19 }}
GEMINI_BASE_URL: https://generativelanguage.googleapis.com/v1beta GEMINI_BASE_URL: https://generativelanguage.googleapis.com/v1beta
+47 -13
View File
@@ -22,6 +22,9 @@
4. 盡量將應用程式放在 ./app,修改 entrypoint.sh 與 Dockerfile 讓程式可以正常運行 4. 盡量將應用程式放在 ./app,修改 entrypoint.sh 與 Dockerfile 讓程式可以正常運行
5. 將提示詞放到 ./app/prompts 內供程式讀取 5. 將提示詞放到 ./app/prompts 內供程式讀取
6. API Key 支援逗號分隔傳入多個,隨機順序各嘗試一次,全部失敗則 exit 1 6. API Key 支援逗號分隔傳入多個,隨機順序各嘗試一次,全部失敗則 exit 1
7. 讀取 Git Diff 時排除 `.gitea/` 資料夾內的所有檔案,避免 AI 分析 workflow 設定等非業務程式碼
8. 階段五完成後驗證 `findings.json``exclusions.json` 是否為合法 JSON 格式,格式錯誤時先嘗試重置為空陣列並備份原檔,修正失敗才 exit 1
9. 傳給 AI 的 findings 只保留必要欄位(level、role、location、suggestion),排除 `is_new` 等內部欄位;system prompt 精簡為指令核心;exclusions hint 只傳 location 與 suggestion,減少 token 用量
# 使用說明 # 使用說明
@@ -29,19 +32,26 @@
2.`.gitea/workflows` 資料夾中建立 `ai-review.yaml' 2.`.gitea/workflows` 資料夾中建立 `ai-review.yaml'
3.`ai-review.yaml` 中填入以下內容(選擇一個使用) 3.`ai-review.yaml` 中填入以下內容(選擇一個使用)
> **權限說明**:此 Action 需要 `contents: write`(寫入 findings.json)、`pull-requests: write`(發佈 PR comment)、`issues: write`(發佈 issue comment)三項權限,為正常運作所必要,無法縮減。
### 1. OpenAI ### 1. OpenAI
```yaml ```yaml
name: AI name: AI
concurrency:
group: ${{ github.workflow }}-${{ github.head_ref }}
cancel-in-progress: true
on: on:
pull_request: pull_request:
branches-ignore:
- master
types: [opened, synchronize] types: [opened, synchronize]
jobs: jobs:
code-review: code-review:
name: 'Code Review' name: Code Review
runs-on: ubuntu runs-on: ubuntu
steps: steps:
- name: AI Code Review - name: AI Code Review
uses: https://gitea.jsc.idv.tw/jiantw83/code-review@${{ vars.ACTION_CODE_REVIEW_VERSION }} uses: https://gitea.jsc.idv.tw/actions/code-review@${{ vars.ACTION_CODE_REVIEW_VERSION }}
with: with:
OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }} # 支援逗號分隔多個 Key OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }} # 支援逗號分隔多個 Key
OPENAI_BASE_URL: https://api.openai.com/v1 OPENAI_BASE_URL: https://api.openai.com/v1
@@ -55,16 +65,21 @@ jobs:
### 2. OpenRouter ### 2. OpenRouter
```yaml ```yaml
name: AI name: AI
concurrency:
group: ${{ github.workflow }}-${{ github.head_ref }}
cancel-in-progress: true
on: on:
pull_request: pull_request:
branches-ignore:
- master
types: [opened, synchronize] types: [opened, synchronize]
jobs: jobs:
code-review: code-review:
name: 'Code Review' name: Code Review
runs-on: ubuntu runs-on: ubuntu
steps: steps:
- name: AI Code Review - name: AI Code Review
uses: https://gitea.jsc.idv.tw/jiantw83/code-review@${{ vars.ACTION_CODE_REVIEW_VERSION }} uses: https://gitea.jsc.idv.tw/actions/code-review@${{ vars.ACTION_CODE_REVIEW_VERSION }}
with: with:
OPENAI_API_KEY: ${{ secrets.OPENROUTER_API_KEY }},${{ secrets.OPENROUTER_API_KEY_1 }} OPENAI_API_KEY: ${{ secrets.OPENROUTER_API_KEY }},${{ secrets.OPENROUTER_API_KEY_1 }}
OPENAI_BASE_URL: https://openrouter.ai/api/v1 OPENAI_BASE_URL: https://openrouter.ai/api/v1
@@ -78,16 +93,21 @@ jobs:
### 3. Anthropic Claude ### 3. Anthropic Claude
```yaml ```yaml
name: AI name: AI
concurrency:
group: ${{ github.workflow }}-${{ github.head_ref }}
cancel-in-progress: true
on: on:
pull_request: pull_request:
branches-ignore:
- master
types: [opened, synchronize] types: [opened, synchronize]
jobs: jobs:
code-review: code-review:
name: 'Code Review' name: Code Review
runs-on: ubuntu runs-on: ubuntu
steps: steps:
- name: AI Code Review - name: AI Code Review
uses: https://gitea.jsc.idv.tw/jiantw83/code-review@${{ vars.ACTION_CODE_REVIEW_VERSION }} uses: https://gitea.jsc.idv.tw/actions/code-review@${{ vars.ACTION_CODE_REVIEW_VERSION }}
with: with:
CLAUDE_API_KEY: ${{ secrets.CLAUDE_API_KEY }} # 支援逗號分隔多個 Key CLAUDE_API_KEY: ${{ secrets.CLAUDE_API_KEY }} # 支援逗號分隔多個 Key
CLAUDE_BASE_URL: https://api.anthropic.com/v1 CLAUDE_BASE_URL: https://api.anthropic.com/v1
@@ -100,16 +120,21 @@ jobs:
### 4. Google Gemini ### 4. Google Gemini
```yaml ```yaml
name: AI name: AI
concurrency:
group: ${{ github.workflow }}-${{ github.head_ref }}
cancel-in-progress: true
on: on:
pull_request: pull_request:
branches-ignore:
- master
types: [opened, synchronize] types: [opened, synchronize]
jobs: jobs:
code-review: code-review:
name: 'Code Review' name: Code Review
runs-on: ubuntu runs-on: ubuntu
steps: steps:
- name: AI Code Review - name: AI Code Review
uses: https://gitea.jsc.idv.tw/jiantw83/code-review@${{ vars.ACTION_CODE_REVIEW_VERSION }} uses: https://gitea.jsc.idv.tw/actions/code-review@${{ vars.ACTION_CODE_REVIEW_VERSION }}
with: with:
GEMINI_API_KEY: ${{ secrets.GEMINI_API_KEY }},${{ secrets.GEMINI_API_KEY_1 }},${{ secrets.GEMINI_API_KEY_2 }},${{ secrets.GEMINI_API_KEY_3 }},${{ secrets.GEMINI_API_KEY_4 }},${{ secrets.GEMINI_API_KEY_5 }},${{ secrets.GEMINI_API_KEY_6 }},${{ secrets.GEMINI_API_KEY_7 }},${{ secrets.GEMINI_API_KEY_8 }},${{ secrets.GEMINI_API_KEY_9 }},${{ secrets.GEMINI_API_KEY_10 }},${{ secrets.GEMINI_API_KEY_11 }},${{ secrets.GEMINI_API_KEY_12 }},${{ secrets.GEMINI_API_KEY_13 }},${{ secrets.GEMINI_API_KEY_14 }},${{ secrets.GEMINI_API_KEY_15 }},${{ secrets.GEMINI_API_KEY_16 }},${{ secrets.GEMINI_API_KEY_17 }},${{ secrets.GEMINI_API_KEY_18 }},${{ secrets.GEMINI_API_KEY_19 }} GEMINI_API_KEY: ${{ secrets.GEMINI_API_KEY }},${{ secrets.GEMINI_API_KEY_1 }},${{ secrets.GEMINI_API_KEY_2 }},${{ secrets.GEMINI_API_KEY_3 }},${{ secrets.GEMINI_API_KEY_4 }},${{ secrets.GEMINI_API_KEY_5 }},${{ secrets.GEMINI_API_KEY_6 }},${{ secrets.GEMINI_API_KEY_7 }},${{ secrets.GEMINI_API_KEY_8 }},${{ secrets.GEMINI_API_KEY_9 }},${{ secrets.GEMINI_API_KEY_10 }},${{ secrets.GEMINI_API_KEY_11 }},${{ secrets.GEMINI_API_KEY_12 }},${{ secrets.GEMINI_API_KEY_13 }},${{ secrets.GEMINI_API_KEY_14 }},${{ secrets.GEMINI_API_KEY_15 }},${{ secrets.GEMINI_API_KEY_16 }},${{ secrets.GEMINI_API_KEY_17 }},${{ secrets.GEMINI_API_KEY_18 }},${{ secrets.GEMINI_API_KEY_19 }}
GEMINI_BASE_URL: https://generativelanguage.googleapis.com/v1beta GEMINI_BASE_URL: https://generativelanguage.googleapis.com/v1beta
@@ -123,16 +148,21 @@ jobs:
### 5. Amazon Q ### 5. Amazon Q
```yaml ```yaml
name: AI name: AI
concurrency:
group: ${{ github.workflow }}-${{ github.head_ref }}
cancel-in-progress: true
on: on:
pull_request: pull_request:
branches-ignore:
- master
types: [opened, synchronize] types: [opened, synchronize]
jobs: jobs:
code-review: code-review:
name: 'Code Review' name: Code Review
runs-on: ubuntu runs-on: ubuntu
steps: steps:
- name: AI Code Review - name: AI Code Review
uses: https://gitea.jsc.idv.tw/jiantw83/code-review@${{ vars.ACTION_CODE_REVIEW_VERSION }} uses: https://gitea.jsc.idv.tw/actions/code-review@${{ vars.ACTION_CODE_REVIEW_VERSION }}
with: with:
AMAZONQ_API_KEY: ${{ secrets.AMAZONQ_API_KEY }} # 支援逗號分隔多個 Key AMAZONQ_API_KEY: ${{ secrets.AMAZONQ_API_KEY }} # 支援逗號分隔多個 Key
AMAZONQ_BASE_URL: https://q.api.aws AMAZONQ_BASE_URL: https://q.api.aws
@@ -146,22 +176,26 @@ jobs:
```yaml ```yaml
name: AI name: AI
concurrency:
group: ${{ github.workflow }}-${{ github.head_ref }}
cancel-in-progress: true
on: on:
pull_request: pull_request:
branches-ignore:
- master
types: [opened, synchronize] types: [opened, synchronize]
jobs: jobs:
code-review: code-review:
name: 'Code Review' name: Code Review
runs-on: ubuntu runs-on: ubuntu
steps: steps:
- name: AI Code Review - name: AI Code Review
uses: https://gitea.jsc.idv.tw/jiantw83/code-review@${{ vars.ACTION_CODE_REVIEW_VERSION }} uses: https://gitea.jsc.idv.tw/actions/code-review@${{ vars.ACTION_CODE_REVIEW_VERSION }}
with: with:
OLLAMA_BASE_URL: https://ollama.jsc.idv.me/v1 OLLAMA_BASE_URL: https://ollama.jsc.idv.me/v1
OLLAMA_MODEL: ${{ vars.OLLAMA_MODEL }} OLLAMA_MODEL: ${{ vars.OLLAMA_MODEL }}
permissions: permissions:
contents: write contents: write
pull-requests: write pull-requests: write
issues: write issues: write
``` ```
+29 -18
View File
@@ -3,43 +3,54 @@
## 階段一:基本流程串接 ## 階段一:基本流程串接
- 目標:確保 action 可以被觸發,pipeline 各步驟依序執行,log 出每個主要階段的進入與完成。 - 目標:確保 action 可以被觸發,pipeline 各步驟依序執行,log 出每個主要階段的進入與完成。
- 驗收:log 中能看到每個階段(如「Step1: pipeline start」、「Step2: findings merge」等)明確訊息,且流程能走完(即使還沒產生 findings)。 - 驗收:log 中能看到每個階段(如「Step1: pipeline start」、「Step2: findings merge」等)明確訊息,且流程能走完(即使還沒產生 findings)。
- 完成 - 未驗收
## 階段二:Findings 產生與合併 ## 階段二:Git Diff 排除 .gitea/ 資料夾
- 目標:讀取 Git Diff 時排除 `.gitea/` 資料夾內的所有檔案,避免 AI 分析 workflow 設定等非業務程式碼。
- 驗收:PR 中有 `.gitea/` 路徑的變更時,diff 內容不包含該路徑的區塊,AI 分析結果不含 `.gitea/` 相關問題。
- 未驗收
## 階段三:Findings 產生與合併
- 目標:各角色(style/security/performance/maintainability/testing)能產生 findings,並正確合併新舊 findings。 - 目標:各角色(style/security/performance/maintainability/testing)能產生 findings,並正確合併新舊 findings。
- 驗收:log 中能看到每個角色 findings 數量、合併後 findings 統計,並有「Step3: merged findings total=...」等訊息。 - 驗收:log 中能看到每個角色 findings 數量、合併後 findings 統計,並有「Step3: merged findings total=...」等訊息。
- 完成 - 未驗收
## 階段AI 去重與角色確認 ## 階段AI 去重與角色確認
- 目標:嘗試呼叫 LLM 進行 findings 去重與角色確認,API 額度不足時要有降級處理 log。 - 目標:嘗試呼叫 LLM 進行 findings 去重與角色確認,API 額度不足時要有降級處理 log。
- 驗收:log 中能看到 deduplication/resolution confirmation 成功或失敗(如 402),降級時有「保留所有問題」等明確訊息。 - 驗收:log 中能看到 deduplication/resolution confirmation 成功或失敗(如 402),降級時有「保留所有問題」等明確訊息。
- 完成 - 未驗收
## 階段AI 排除問題過濾 ## 階段AI 排除問題過濾
- 目標:讀取排除問題檔案(`.gitea/ai-review/exclusions.json`)進行規則過濾,並呼叫 AI 判斷剩餘問題是否為誤報或不適用,兩層過濾後產生最終問題清單。 - 目標:讀取排除問題檔案(`.gitea/ai-review/exclusions.json`)進行規則過濾,並呼叫 AI 判斷剩餘問題是否為誤報或不適用,兩層過濾後產生最終問題清單。
- 驗收:log 中能看到排除問題檔案讀取成功或不存在的訊息、規則過濾數量變化,以及「AI 誤報過濾: N -> M 筆」或降級訊息。 - 驗收:log 中能看到排除問題檔案讀取成功或不存在的訊息、規則過濾數量變化,以及「AI 誤報過濾: N -> M 筆」或降級訊息。
- 完成 - 未驗收
## 階段findings 寫入與 comment 發布 ## 階段findings 寫入與 comment 發布
- 目標:`.gitea/ai-review/findings.json` 正確寫入,comment 發布順序正確(舊問題→非嚴重→嚴重),每步有 log。 - 目標:`.gitea/ai-review/findings.json` 正確寫入,comment 發布順序正確(舊問題→非嚴重→嚴重),每步有 log。
- 驗收:log 中能看到 `.gitea/ai-review/findings.json` 寫入、comment sync 的詳細訊息與順序。 - 驗收:log 中能看到 `.gitea/ai-review/findings.json` 寫入、comment sync 的詳細訊息與順序。
- 完成 - 未驗收
## 階段六:記憶區 commit/push 與錯誤處理 ## 階段七:階段六後驗證 JSON 格式
- 目標:階段六完成後驗證 `findings.json``exclusions.json` 是否為合法 JSON 格式,格式錯誤時先嘗試重置為空陣列並備份原檔,修正失敗才 exit 1。
- 驗收:log 中能看到兩個檔案的驗證結果(成功或失敗),格式錯誤時有「嘗試修正」訊息與備份路徑,修正失敗時 workflow 狀態為失敗。
- 未驗收
## 階段八:記憶區 commit/push 與錯誤處理
- 目標:記憶區能成功 commit/push,錯誤時有明確 log,流程結束有總結訊息。 - 目標:記憶區能成功 commit/push,錯誤時有明確 log,流程結束有總結訊息。
- 驗收:log 有「persisted findings」、「commit=...」、「push=...」等訊息,錯誤時有「Runner failed: ...」等明確錯誤說明。 - 驗收:log 有「persisted findings」、「commit=...」、「push=...」等訊息,錯誤時有「Runner failed: ...」等明確錯誤說明。
- 完成 - 未驗收
## 階段:阻擋嚴重問題 PR(第 8 點) ## 階段:阻擋嚴重問題 PR(第 8 點)
- 目標:如果 PR 問題表格中有嚴重(critical)問題,workflow 需直接 exit 1,不讓流程成功。 - 目標:如果 PR 問題表格中有嚴重(critical)問題,workflow 需直接 exit 1,不讓流程成功。
- 驗收:log 中能看到「critical 問題存在,workflow 結束(exit 1)」等明確訊息,且 workflow 狀態為失敗。 - 驗收:log 中能看到「critical 問題存在,workflow 結束(exit 1)」等明確訊息,且 workflow 狀態為失敗。
- 完成 - 未驗收
## 階段API Key 輪替 ## 階段API Key 輪替
- 目標:所有平台的 API Key 支援逗號分隔傳入多個,隨機順序各嘗試一次,單一 Key 失敗時自動換下一個,全部失敗則 exit 1。 - 目標:所有平台的 API Key 支援逗號分隔傳入多個,隨機順序各嘗試一次,單一 Key 失敗時自動換下一個,全部失敗則 exit 1。
- 驗收:log 中能看到「key[N/M] 失敗」等訊息,換 key 後繼續執行;傳入單一 Key 時行為與原本相同;全部 Key 失敗時 log「所有 API Key 均失敗,終止流程」且 workflow 狀態為失敗。 - 驗收:log 中能看到「key[N/M] 失敗」等訊息,換 key 後繼續執行;傳入單一 Key 時行為與原本相同;全部 Key 失敗時 log「所有 API Key 均失敗,終止流程」且 workflow 狀態為失敗。
- 完成 - 未驗收
--- ## 階段十一:壓縮 AI 傳入內容減少 token 用量
- 目標:傳給 AI 的 findings 只保留必要欄位(level、role、location、suggestion);system prompt 精簡為指令核心;exclusions hint 只傳 location 與 suggestionAI 回傳後補回原始完整欄位(含 is_new)。
所有階段驗收通過 - 驗收:AI 呼叫的 payload 不含 is_new 等內部欄位,去重與誤報過濾後的 findings 仍保有完整欄位供後續流程使用
- 未驗收
+1 -1
View File
@@ -63,7 +63,7 @@ export async function postNewCriticalComments(findings) {
return; return;
} }
for (const f of criticals) { for (const f of criticals) {
const body = `## 🚨 嚴重問題\n\n| 審查員 | 位置 | 建議 |\n|--------|------|------|\n| ${f.role} | ${f.location} | ${f.suggestion} |`; const body = `## 🚨 嚴重問題\n\n${buildTable([f])}`;
await postComment(body); await postComment(body);
console.log(` ✅ 嚴重問題 comment 發布: [${f.role}] ${f.location}`); console.log(` ✅ 嚴重問題 comment 發布: [${f.role}] ${f.location}`);
} }
+45 -64
View File
@@ -11,7 +11,6 @@ const LEVELS = ['critical', 'warning', 'info'];
export async function analyzeWithRole(role, diff) { export async function analyzeWithRole(role, diff) {
console.log(` [${role.name}] 開始分析...`); console.log(` [${role.name}] 開始分析...`);
const findings = await chatJSON(role.system_prompt, `以下是 Git Diff 內容:\n\n${diff}`); const findings = await chatJSON(role.system_prompt, `以下是 Git Diff 內容:\n\n${diff}`);
// 確保每筆都有必要欄位,並標記為新問題
const valid = findings.filter(f => f.level && f.role && f.location && f.suggestion) const valid = findings.filter(f => f.level && f.role && f.location && f.suggestion)
.map(f => ({ ...f, is_new: true })); .map(f => ({ ...f, is_new: true }));
console.log(` [${role.name}] 找到 ${valid.length} 個問題`); console.log(` [${role.name}] 找到 ${valid.length} 個問題`);
@@ -19,28 +18,33 @@ export async function analyzeWithRole(role, diff) {
} }
/** /**
* 讀取舊 findings(從 workspace 的 FINDINGS_PATH * 讀取 JSON 陣列檔案,失敗或不存在時回傳空陣列
*/ */
export function loadOldFindings(workspace) { function readJSONArray(fullPath, label) {
const fullPath = path.join(workspace, FINDINGS_PATH);
if (!fs.existsSync(fullPath)) { if (!fs.existsSync(fullPath)) {
console.log(' 舊 findings 檔案不存在,視為空'); console.log(` ${label}檔案不存在,視為空`);
return []; return [];
} }
try { try {
const data = JSON.parse(fs.readFileSync(fullPath, 'utf8')); const data = JSON.parse(fs.readFileSync(fullPath, 'utf8'));
const old = (Array.isArray(data) ? data : []).map(f => ({ ...f, is_new: false })); return Array.isArray(data) ? data : [];
console.log(` 讀取舊 findings: ${old.length}`);
return old;
} catch (e) { } catch (e) {
console.log(` ⚠️ 讀取舊 findings 失敗: ${e.message},視為空`); console.log(` ⚠️ 讀取${label}失敗: ${e.message},視為空`);
return []; return [];
} }
} }
/**
* 讀取舊 findings(從 workspace 的 FINDINGS_PATH
*/
export function loadOldFindings(workspace) {
const old = readJSONArray(path.join(workspace, FINDINGS_PATH), '舊 findings ').map(f => ({ ...f, is_new: false }));
console.log(` 讀取舊 findings: ${old.length}`);
return old;
}
/** /**
* 合併新舊 findings,以 (role + location + suggestion前50字) 為 key 去除重複 * 合併新舊 findings,以 (role + location + suggestion前50字) 為 key 去除重複
* 舊問題保留,新問題若與舊問題重複則捨棄
*/ */
export function mergeFindings(oldFindings, newFindings) { export function mergeFindings(oldFindings, newFindings) {
const key = f => `${f.role}|${f.location}|${String(f.suggestion).slice(0, 50)}`; const key = f => `${f.role}|${f.location}|${String(f.suggestion).slice(0, 50)}`;
@@ -63,56 +67,49 @@ export function sortByLevel(findings) {
} }
/** /**
* 呼叫 LLM 進行語意去重,回傳去重後的 findings * AI 呼叫失敗時的統一降級處理
* 失敗時降級回傳原始 findings */
function fallback(label, findings, e) {
const status = e.response?.status;
const reason = (status === 402 || status === 429) ? `${status} 額度/限流` : e.message;
console.log(` ⚠️ ${label}失敗(${reason}),降級:保留所有問題`);
return findings;
}
/** 只保留 AI 需要的欄位,減少 token 用量 */
function toAIPayload(findings) {
return findings.map(({ level, role, location, suggestion }) => ({ level, role, location, suggestion }));
}
/**
* 呼叫 LLM 進行語意去重,失敗時降級回傳原始 findings
*/ */
export async function deduplicateWithAI(findings) { export async function deduplicateWithAI(findings) {
if (findings.length === 0) return findings; if (findings.length === 0) return findings;
const systemPrompt = `你是一位程式碼審查問題去重專家。 const systemPrompt = `移除語意重複的程式碼審查問題(JSON 陣列)。保留等級較高者(critical > warning > info)。只回傳去重後的 JSON 陣列。`;
給你一份問題清單(JSON 陣列),請移除語意重複的問題(即使描述文字不同,但指的是同一個問題)。
保留等級較高的版本,優先保留 critical > warning > info。
只回傳去重後的 JSON 陣列,不要有其他文字。`;
const userContent = `以下是問題清單,請去除語意重複的項目:\n\n${JSON.stringify(findings, null, 2)}`;
try { try {
const result = await chatJSON(systemPrompt, userContent); const result = await chatJSON(systemPrompt, JSON.stringify(toAIPayload(findings)));
if (Array.isArray(result) && result.length > 0) { if (Array.isArray(result) && result.length > 0) {
console.log(` AI 去重: ${findings.length} -> ${result.length}`); console.log(` AI 去重: ${findings.length} -> ${result.length}`);
return result; // 以 location+suggestion 為 key,將原始 findings 的完整欄位(含 is_new)補回
const origMap = new Map(findings.map(f => [`${f.location}|${String(f.suggestion).slice(0, 50)}`, f]));
return result.map(r => origMap.get(`${r.location}|${String(r.suggestion).slice(0, 50)}`) ?? r);
} }
throw new Error('AI 回傳空陣列'); throw new Error('AI 回傳空陣列');
} catch (e) { } catch (e) {
const status = e.response?.status; return fallback('AI 去重', findings, e);
if (status === 402 || status === 429) {
console.log(` ⚠️ AI 去重失敗(${status} 額度/限流),降級:保留所有問題`);
} else {
console.log(` ⚠️ AI 去重失敗(${e.message}),降級:保留所有問題`);
}
return findings;
} }
} }
/** /**
* 讀取排除問題檔案(從 workspace 的 EXCLUSIONS_PATH * 讀取排除問題檔案(從 workspace 的 EXCLUSIONS_PATH
* 格式:[{ role, location, suggestion }],欄位可部分省略,省略表示萬用
*/ */
export function loadExclusions(workspace) { export function loadExclusions(workspace) {
const fullPath = path.join(workspace, EXCLUSIONS_PATH); const exclusions = readJSONArray(path.join(workspace, EXCLUSIONS_PATH), '排除問題');
if (!fs.existsSync(fullPath)) {
console.log(' 排除問題檔案不存在,跳過過濾');
return [];
}
try {
const data = JSON.parse(fs.readFileSync(fullPath, 'utf8'));
const exclusions = Array.isArray(data) ? data : [];
console.log(` 讀取排除問題: ${exclusions.length}`); console.log(` 讀取排除問題: ${exclusions.length}`);
return exclusions; return exclusions;
} catch (e) {
console.log(` ⚠️ 讀取排除問題失敗: ${e.message},跳過過濾`);
return [];
}
} }
/** /**
@@ -125,49 +122,33 @@ export function applyExclusions(findings, exclusions) {
const filtered = findings.filter(f => !exclusions.some(ex => { const filtered = findings.filter(f => !exclusions.some(ex => {
const fPath = String(f.location).split(':')[0]; const fPath = String(f.location).split(':')[0];
const exPath = ex.location ? String(ex.location).split(':')[0] : null; const exPath = ex.location ? String(ex.location).split(':')[0] : null;
return (!exPath || fPath === exPath) && return (!exPath || fPath === exPath) && (!ex.role || ex.role === f.role);
(!ex.role || ex.role === f.role);
})); }));
console.log(` 排除過濾: ${before} -> ${filtered.length} 筆(排除 ${before - filtered.length} 筆)`); console.log(` 排除過濾: ${before} -> ${filtered.length} 筆(排除 ${before - filtered.length} 筆)`);
return filtered; return filtered;
} }
/** /**
* 呼叫 AI 判斷哪些問題是誤報或不需處理,回傳需保留的 findings * 呼叫 AI 判斷哪些問題是誤報或不需處理,失敗時降級回傳原始 findings
* exclusions 為已知誤報清單,供 AI 參考判斷
* 失敗時降級回傳原始 findings
*/ */
export async function filterFalsePositivesWithAI(findings, exclusions = []) { export async function filterFalsePositivesWithAI(findings, exclusions = []) {
if (findings.length === 0) return findings; if (findings.length === 0) return findings;
const exclusionHint = exclusions.length > 0 const exclusionHint = exclusions.length > 0
? `\n\n以下是已知誤報或不需處理的問題清單(供參考,相同檔案路徑且語意相近的問題應一併排除):\n${JSON.stringify(exclusions, null, 2)}` ? `\n已知誤報(相同路徑且語意相近一併排除):\n${JSON.stringify(exclusions.map(({ location, suggestion }) => ({ location, suggestion })))}`
: ''; : '';
const systemPrompt = `你是一位資深程式碼審查專家,負責判斷審查問題是否為誤報或不需處理。 const systemPrompt = `判斷以下程式碼審查問題是否為誤報或不適用(如已正確使用 secrets、CI/CD 必要權限等),移除後只回傳需保留的 JSON 陣列。${exclusionHint}`;
給你一份問題清單(JSON 陣列),每筆包含 level、role、location、suggestion。
請移除以下類型的問題:
1. 誤報:問題描述與實際程式碼不符(例如:程式碼已正確使用環境變數或 secrets,卻被標記為硬編碼敏感資料)
2. 不適用:問題在此專案情境下不需處理(例如:CI/CD action 本來就需要透過環境變數傳遞 token)
3. 與已知誤報清單語意相近的問題(檔案路徑相同且建議內容相似)
只回傳需要保留的問題 JSON 陣列,不要有其他文字。${exclusionHint}`;
const userContent = `請判斷以下問題清單,移除誤報或不需處理的問題:\n\n${JSON.stringify(findings, null, 2)}`;
try { try {
const result = await chatJSON(systemPrompt, userContent); const result = await chatJSON(systemPrompt, JSON.stringify(toAIPayload(findings)));
if (Array.isArray(result) && result.length > 0) { if (Array.isArray(result) && result.length > 0) {
console.log(` AI 誤報過濾: ${findings.length} -> ${result.length}`); console.log(` AI 誤報過濾: ${findings.length} -> ${result.length}`);
return result; const origMap = new Map(findings.map(f => [`${f.location}|${String(f.suggestion).slice(0, 50)}`, f]));
return result.map(r => origMap.get(`${r.location}|${String(r.suggestion).slice(0, 50)}`) ?? r);
} }
throw new Error('AI 回傳空陣列或非陣列'); throw new Error('AI 回傳空陣列或非陣列');
} catch (e) { } catch (e) {
const status = e.response?.status; return fallback('AI 誤報過濾', findings, e);
if (status === 402 || status === 429) {
console.log(` ⚠️ AI 誤報過濾失敗(${status} 額度/限流),降級:保留所有問題`);
} else {
console.log(` ⚠️ AI 誤報過濾失敗(${e.message}),降級:保留所有問題`);
}
return findings;
} }
} }
+18 -29
View File
@@ -3,6 +3,8 @@ import fs from 'fs';
import path from 'path'; import path from 'path';
import { GITEA_SERVER_URL, GITEA_REPOSITORY, GITEA_TOKEN, PR_HEAD_BRANCH, FINDINGS_PATH } from './config.js'; import { GITEA_SERVER_URL, GITEA_REPOSITORY, GITEA_TOKEN, PR_HEAD_BRANCH, FINDINGS_PATH } from './config.js';
const remoteUrl = `${GITEA_SERVER_URL.replace(/\/$/, '')}/${GITEA_REPOSITORY}.git`;
function makeRunner(spawn) { function makeRunner(spawn) {
return function run(args, cwd, env) { return function run(args, cwd, env) {
const opts = { cwd, encoding: 'utf8' }; const opts = { cwd, encoding: 'utf8' };
@@ -14,20 +16,25 @@ function makeRunner(spawn) {
}; };
} }
function withAskpass(workspace, fn) {
const askpassScript = path.join(workspace, '.git-askpass.sh');
fs.writeFileSync(askpassScript, '#!/bin/sh\necho "$GIT_TOKEN"\n', { mode: 0o700 });
const credEnv = { ...process.env, GIT_ASKPASS: askpassScript, GIT_USERNAME: 'x-token', GIT_TOKEN: GITEA_TOKEN };
try {
return fn(credEnv);
} finally {
try { fs.unlinkSync(askpassScript); } catch {}
}
}
/** /**
* Clone PR head branch to workspace/repo (idempotent) * Clone PR head branch to workspace/repo (idempotent)
*/ */
export function cloneRepo(workspace, _spawnSync = spawnSync) { export function cloneRepo(workspace, _spawnSync = spawnSync) {
const run = makeRunner(_spawnSync); const run = makeRunner(_spawnSync);
const baseUrl = GITEA_SERVER_URL.replace(/\/$/, '');
const remoteUrl = `${baseUrl}/${GITEA_REPOSITORY}.git`;
const repoDir = path.join(workspace, 'repo'); const repoDir = path.join(workspace, 'repo');
const askpassScript = path.join(workspace, '.git-askpass.sh'); return withAskpass(workspace, credEnv => {
fs.writeFileSync(askpassScript, '#!/bin/sh\necho "$GIT_TOKEN"\n', { mode: 0o700 });
const credEnv = { ...process.env, GIT_ASKPASS: askpassScript, GIT_USERNAME: 'x-token', GIT_TOKEN: GITEA_TOKEN };
try {
if (!fs.existsSync(repoDir)) { if (!fs.existsSync(repoDir)) {
run(['clone', '--depth=1', '--branch', PR_HEAD_BRANCH, remoteUrl, repoDir], workspace, credEnv); run(['clone', '--depth=1', '--branch', PR_HEAD_BRANCH, remoteUrl, repoDir], workspace, credEnv);
console.log(` ✅ repo cloned to ${repoDir}`); console.log(` ✅ repo cloned to ${repoDir}`);
@@ -36,37 +43,20 @@ export function cloneRepo(workspace, _spawnSync = spawnSync) {
run(['checkout', PR_HEAD_BRANCH], repoDir); run(['checkout', PR_HEAD_BRANCH], repoDir);
console.log(` ✅ repo already exists, fetched latest`); console.log(` ✅ repo already exists, fetched latest`);
} }
} finally {
try { fs.unlinkSync(askpassScript); } catch {}
}
return repoDir; return repoDir;
});
} }
export async function commitAndPush(workspace, _spawnSync = spawnSync) { export async function commitAndPush(workspace, _spawnSync = spawnSync) {
const run = makeRunner(_spawnSync); const run = makeRunner(_spawnSync);
const baseUrl = GITEA_SERVER_URL.replace(/\/$/, '');
const remoteUrl = `${baseUrl}/${GITEA_REPOSITORY}.git`;
const repoDir = path.join(workspace, 'repo');
// Write a temporary askpass script that reads the token from an env var,
// so the token value never appears in the script file itself
const askpassScript = path.join(workspace, '.git-askpass.sh');
fs.writeFileSync(askpassScript, '#!/bin/sh\necho "$GIT_TOKEN"\n', { mode: 0o700 });
const credEnv = { ...process.env, GIT_ASKPASS: askpassScript, GIT_USERNAME: 'x-token', GIT_TOKEN: GITEA_TOKEN };
try { try {
if (!fs.existsSync(repoDir)) { const repoDir = cloneRepo(workspace, _spawnSync);
run(['clone', '--depth=1', '--branch', PR_HEAD_BRANCH, remoteUrl, repoDir], workspace, credEnv);
}
await withAskpass(workspace, async credEnv => {
run(['config', 'user.email', 'ai-review[bot]@gitea'], repoDir); run(['config', 'user.email', 'ai-review[bot]@gitea'], repoDir);
run(['config', 'user.name', 'AI Review Bot'], repoDir); run(['config', 'user.name', 'AI Review Bot'], repoDir);
run(['fetch', 'origin', PR_HEAD_BRANCH], repoDir, credEnv);
run(['checkout', PR_HEAD_BRANCH], repoDir);
// 將 findings.json 從 workspace 複製到 clone 的 repo
const srcFindings = path.join(workspace, FINDINGS_PATH); const srcFindings = path.join(workspace, FINDINGS_PATH);
const destFindings = path.join(repoDir, FINDINGS_PATH); const destFindings = path.join(repoDir, FINDINGS_PATH);
fs.mkdirSync(path.dirname(destFindings), { recursive: true }); fs.mkdirSync(path.dirname(destFindings), { recursive: true });
@@ -84,9 +74,8 @@ export async function commitAndPush(workspace, _spawnSync = spawnSync) {
const commitHash = out.match(/\[.+ ([a-f0-9]+)\]/)?.[1] || 'unknown'; const commitHash = out.match(/\[.+ ([a-f0-9]+)\]/)?.[1] || 'unknown';
run(['push', remoteUrl, PR_HEAD_BRANCH], repoDir, credEnv); run(['push', remoteUrl, PR_HEAD_BRANCH], repoDir, credEnv);
console.log(` ✅ persisted findings commit=${commitHash} push=${PR_HEAD_BRANCH}`); console.log(` ✅ persisted findings commit=${commitHash} push=${PR_HEAD_BRANCH}`);
});
} catch (e) { } catch (e) {
console.log(` ⚠️ Runner failed: commit/push 失敗: ${e.message}`); console.log(` ⚠️ Runner failed: commit/push 失敗: ${e.message}`);
} finally {
try { fs.unlinkSync(askpassScript); } catch {}
} }
} }
+14
View File
@@ -6,11 +6,25 @@ const httpsAgent = GITEA_SKIP_TLS_VERIFY ? new https.Agent({ rejectUnauthorized:
const headers = () => ({ Authorization: `token ${GITEA_TOKEN}`, 'Content-Type': 'application/json' }); const headers = () => ({ Authorization: `token ${GITEA_TOKEN}`, 'Content-Type': 'application/json' });
const api = (path) => `${GITEA_SERVER_URL.replace(/\/$/, '')}/api/v1${path}`; const api = (path) => `${GITEA_SERVER_URL.replace(/\/$/, '')}/api/v1${path}`;
/**
* 取得 PR 的原始 Git Diff 內容。
* 注意:回傳值未經路徑過濾,呼叫端須使用 filterDiff 排除敏感路徑(如 .gitea/)後再傳給 AI。
*/
export async function getPRDiff() { export async function getPRDiff() {
const resp = await axios.get(api(`/repos/${GITEA_REPOSITORY}/pulls/${PR_NUMBER}.diff`), { headers: headers(), timeout: 60000, httpsAgent }); const resp = await axios.get(api(`/repos/${GITEA_REPOSITORY}/pulls/${PR_NUMBER}.diff`), { headers: headers(), timeout: 60000, httpsAgent });
return resp.data; return resp.data;
} }
/**
* 過濾 diff 內容,移除路徑符合 excludePrefixes 的區塊。
* 每個區塊以 "diff --git a/<prefix>" 開頭判斷,使用 startsWith 精確比對前綴。
*/
export function filterDiff(diff, excludePrefixes) {
return diff.split(/(?=^diff --git )/m)
.filter(block => !excludePrefixes.some(p => block.startsWith(`diff --git a/${p}`)))
.join('');
}
export async function postComment(body) { export async function postComment(body) {
const resp = await axios.post(api(`/repos/${GITEA_REPOSITORY}/issues/${PR_NUMBER}/comments`), { body }, { headers: headers(), timeout: 30000, httpsAgent }); const resp = await axios.post(api(`/repos/${GITEA_REPOSITORY}/issues/${PR_NUMBER}/comments`), { body }, { headers: headers(), timeout: 30000, httpsAgent });
return resp.data; return resp.data;
+30 -5
View File
@@ -2,13 +2,10 @@ import { describe, it, afterEach, mock } from 'node:test';
import assert from 'node:assert/strict'; import assert from 'node:assert/strict';
import axios from 'axios'; import axios from 'axios';
// gitea.js reads env vars at module load time (ESM cache), so we test
// the actual values baked in at import time and verify behavior via axios mocks.
afterEach(() => mock.restoreAll()); afterEach(() => mock.restoreAll());
describe('gitea', async () => { describe('gitea', async () => {
const { getPRDiff, postComment } = await import('./gitea.js'); const { getPRDiff, filterDiff, postComment } = await import('./gitea.js');
it('getPRDiff calls Gitea diff API with Authorization header', async () => { it('getPRDiff calls Gitea diff API with Authorization header', async () => {
let capturedUrl, capturedOpts; let capturedUrl, capturedOpts;
@@ -48,7 +45,6 @@ describe('gitea', async () => {
return { data: '' }; return { data: '' };
}); });
await getPRDiff(); await getPRDiff();
// httpsAgent is undefined when GITEA_SKIP_TLS_VERIFY !== 'true'
assert.equal(capturedOpts.httpsAgent, undefined); assert.equal(capturedOpts.httpsAgent, undefined);
}); });
@@ -62,3 +58,32 @@ describe('gitea', async () => {
await assert.rejects(() => postComment('test'), /api error/); await assert.rejects(() => postComment('test'), /api error/);
}); });
}); });
describe('filterDiff', async () => {
const { filterDiff } = await import('./gitea.js');
const block = (file) => `diff --git a/${file} b/${file}\n--- a/${file}\n+++ b/${file}\n@@ -1 +1 @@\n-old\n+new\n`;
it('filters out .gitea/ blocks', () => {
const diff = block('.gitea/workflows/review.yaml') + block('src/index.js');
const result = filterDiff(diff, ['.gitea/']);
assert.ok(!result.includes('.gitea/'));
assert.ok(result.includes('src/index.js'));
});
it('does not filter non-.gitea/ blocks', () => {
const diff = block('src/index.js') + block('README.md');
const result = filterDiff(diff, ['.gitea/']);
assert.equal(result, diff);
});
it('returns empty string when all blocks are excluded', () => {
const diff = block('.gitea/workflows/review.yaml') + block('.gitea/ai-review/findings.json');
const result = filterDiff(diff, ['.gitea/']);
assert.equal(result, '');
});
it('returns empty string for empty diff', () => {
assert.equal(filterDiff('', ['.gitea/']), '');
});
});
+3 -4
View File
@@ -29,12 +29,11 @@ export async function chat(systemPrompt, userContent) {
} }
export async function chatJSON(systemPrompt, userContent) { export async function chatJSON(systemPrompt, userContent) {
const text = await chat(systemPrompt, userContent);
try { try {
let text = await chat(systemPrompt, userContent); return JSON.parse(text.trim().replace(/^```[^\n]*\n?/, '').replace(/```$/, '').trim());
text = text.trim().replace(/^```[^\n]*\n?/, '').replace(/```$/, '').trim();
return JSON.parse(text);
} catch (e) { } catch (e) {
console.log(` [LLM] 解析失敗: ${e.message}`); console.log(` [LLM] JSON 解析失敗: ${e.message}`);
return []; return [];
} }
} }
+58 -27
View File
@@ -1,6 +1,8 @@
import { GITEA_REPOSITORY, PR_NUMBER, PR_HEAD_BRANCH, PR_BASE_BRANCH, getLLMConfig } from './config.js'; import fs from 'fs';
import path from 'path';
import { GITEA_REPOSITORY, PR_NUMBER, PR_HEAD_BRANCH, PR_BASE_BRANCH, getLLMConfig, FINDINGS_PATH, EXCLUSIONS_PATH } from './config.js';
import { loadRoles, getRoleIntro } from './roles.js'; import { loadRoles, getRoleIntro } from './roles.js';
import { getPRDiff, postComment } from './gitea.js'; import { getPRDiff, filterDiff, postComment } from './gitea.js';
import { analyzeWithRole, loadOldFindings, mergeFindings, sortByLevel, deduplicateWithAI, loadExclusions, applyExclusions, filterFalsePositivesWithAI } from './findings.js'; import { analyzeWithRole, loadOldFindings, mergeFindings, sortByLevel, deduplicateWithAI, loadExclusions, applyExclusions, filterFalsePositivesWithAI } from './findings.js';
import { saveFindings, postOldFindingsComment, postNewNonCriticalComment, postNewCriticalComments } from './comments.js'; import { saveFindings, postOldFindingsComment, postNewNonCriticalComment, postNewCriticalComments } from './comments.js';
import { cloneRepo, commitAndPush } from './git.js'; import { cloneRepo, commitAndPush } from './git.js';
@@ -13,7 +15,6 @@ async function main() {
console.log(` repo=${GITEA_REPOSITORY} PR=#${PR_NUMBER}`); console.log(` repo=${GITEA_REPOSITORY} PR=#${PR_NUMBER}`);
console.log(` ${PR_HEAD_BRANCH} -> ${PR_BASE_BRANCH}`); console.log(` ${PR_HEAD_BRANCH} -> ${PR_BASE_BRANCH}`);
// 偵測 LLM
const { provider, baseURL, model } = getLLMConfig(); const { provider, baseURL, model } = getLLMConfig();
if (!provider) { if (!provider) {
console.error('❌ 未設定任何 LLM API Key,請檢查 action inputs'); console.error('❌ 未設定任何 LLM API Key,請檢查 action inputs');
@@ -21,11 +22,9 @@ async function main() {
} }
console.log(` LLM: provider=${provider} model=${model} base_url=${baseURL}`); console.log(` LLM: provider=${provider} model=${model} base_url=${baseURL}`);
// 載入角色
const roles = loadRoles(); const roles = loadRoles();
console.log(` 已載入 ${roles.length} 個角色: [${roles.map(r => r.name).join(', ')}]`); console.log(` 已載入 ${roles.length} 個角色: [${roles.map(r => r.name).join(', ')}]`);
// 取得 PR diff
let diff; let diff;
try { try {
diff = await getPRDiff(); diff = await getPRDiff();
@@ -40,7 +39,6 @@ async function main() {
process.exit(0); process.exit(0);
} }
// 發布角色介紹 comment
try { try {
const intro = getRoleIntro(roles) + `\n\n> 🔍 服務:${provider} 模型:${model}`; const intro = getRoleIntro(roles) + `\n\n> 🔍 服務:${provider} 模型:${model}`;
await postComment(intro); await postComment(intro);
@@ -48,10 +46,19 @@ async function main() {
} catch (e) { } catch (e) {
console.log(` ⚠️ comment 發布失敗(繼續執行): ${e.message}`); console.log(` ⚠️ comment 發布失敗(繼續執行): ${e.message}`);
} }
console.log(' Step1 完成');
// Step2: 各角色分析 diff 產生新 findings // Step2: 排除 .gitea/ 資料夾內的所有檔案
console.log('\n📊 Step2: Findings 產生'); console.log('\n🗂️ Step2: Git Diff 過濾');
diff = filterDiff(diff, ['.gitea/']);
console.log(` 排除 .gitea/ 後 diff 長度: ${diff.length} 字元`);
if (!diff.trim()) {
console.log(' ⚠️ 過濾後 diff 為空,無需審查');
process.exit(0);
}
// Step3: 各角色分析 diff 產生新 findings
console.log('\n📊 Step3: Findings 產生');
const results = await Promise.allSettled(roles.map(role => analyzeWithRole(role, diff))); const results = await Promise.allSettled(roles.map(role => analyzeWithRole(role, diff)));
const newFindings = []; const newFindings = [];
for (let i = 0; i < results.length; i++) { for (let i = 0; i < results.length; i++) {
@@ -61,10 +68,10 @@ async function main() {
console.log(` ⚠️ [${roles[i].name}] 分析失敗(跳過): ${results[i].reason?.message}`); console.log(` ⚠️ [${roles[i].name}] 分析失敗(跳過): ${results[i].reason?.message}`);
} }
} }
console.log(` Step2 完成: 新 findings 總計 ${newFindings.length}`); console.log(` Step3 完成: 新 findings 總計 ${newFindings.length}`);
// Step3: 讀取舊 findings,合併去重(含 AI 語意去重) // Step4: 讀取舊 findings,合併去重(含 AI 語意去重)
console.log('\n🔀 Step3: Findings 合併'); console.log('\n🔀 Step4: Findings 合併');
// Clone repo 以讀取舊 findings 與排除清單 // Clone repo 以讀取舊 findings 與排除清單
let repoDir; let repoDir;
try { try {
@@ -74,39 +81,64 @@ async function main() {
} }
const oldFindings = loadOldFindings(repoDir || WORKSPACE); const oldFindings = loadOldFindings(repoDir || WORKSPACE);
const mergedFindings = mergeFindings(oldFindings, newFindings); const mergedFindings = mergeFindings(oldFindings, newFindings);
console.log(` Step3 merged findings total=${mergedFindings.length}`); console.log(` Step4 merged findings total=${mergedFindings.length}`);
console.log('\n🤖 Step3b: AI 語意去重'); console.log('\n🤖 Step4b: AI 語意去重');
const deduped = await deduplicateWithAI(mergedFindings); const deduped = await deduplicateWithAI(mergedFindings);
const sorted = sortByLevel(deduped); const sorted = sortByLevel(deduped);
console.log(` Step3b dedup findings total=${sorted.length} (critical=${sorted.filter(f=>f.level==='critical').length} warning=${sorted.filter(f=>f.level==='warning').length} info=${sorted.filter(f=>f.level==='info').length})`); console.log(` Step4b dedup findings total=${sorted.length} (critical=${sorted.filter(f=>f.level==='critical').length} warning=${sorted.filter(f=>f.level==='warning').length} info=${sorted.filter(f=>f.level==='info').length})`);
// Step4: 讀取排除問題檔案,過濾 PR 問題表格,並請 AI 判斷誤報 // Step5: 讀取排除問題檔案,過濾 PR 問題表格,並請 AI 判斷誤報
console.log('\n🚫 Step4: AI 排除問題過濾'); console.log('\n🚫 Step5: AI 排除問題過濾');
// 輸入至 findings 用於 AI 誤報過濾,exclusions 同時作為已知誤報參考
const exclusions = loadExclusions(repoDir || WORKSPACE); const exclusions = loadExclusions(repoDir || WORKSPACE);
const ruleFiltered = applyExclusions(sorted, exclusions); const ruleFiltered = applyExclusions(sorted, exclusions);
const filtered = await filterFalsePositivesWithAI(ruleFiltered, exclusions); const filtered = await filterFalsePositivesWithAI(ruleFiltered, exclusions);
console.log(` Step4 完成: findings total=${filtered.length}`); console.log(` Step5 完成: findings total=${filtered.length}`);
// Step5: 寫入 findings.json,依序發布 comment // Step6: 寫入 findings.json,依序發布 comment
console.log('\n📝 Step5: Findings 寫入與 Comment 發布'); console.log('\n📝 Step6: Findings 寫入與 Comment 發布');
saveFindings(WORKSPACE, filtered); saveFindings(WORKSPACE, filtered);
try { try {
await postOldFindingsComment(filtered); await postOldFindingsComment(filtered);
await postNewNonCriticalComment(filtered); await postNewNonCriticalComment(filtered);
await postNewCriticalComments(filtered); await postNewCriticalComments(filtered);
console.log(' Step5 完成'); console.log(' Step6 完成');
} catch (e) { } catch (e) {
console.log(` ⚠️ comment 發布失敗(繼續執行): ${e.message}`); console.log(` ⚠️ comment 發布失敗(繼續執行): ${e.message}`);
} }
// Step6: commit/push findings.json 到來源分支 // Step7: 驗證 findings.json 與 exclusions.json 為合法 JSON
console.log('\n💾 Step6: 記憶區 Commit/Push'); console.log('\n🔎 Step7: JSON 格式驗證');
for (const relPath of [FINDINGS_PATH, EXCLUSIONS_PATH]) {
const fullPath = path.join(repoDir || WORKSPACE, relPath);
if (!fs.existsSync(fullPath)) {
console.log(` ⚠️ ${relPath} 不存在,跳過驗證`);
continue;
}
try {
JSON.parse(fs.readFileSync(fullPath, 'utf8'));
console.log(`${relPath} JSON 格式正確`);
} catch (e) {
console.error(`${relPath} JSON 格式錯誤: ${e.message},嘗試修正...`);
try {
const backupPath = fullPath + '.bak';
fs.copyFileSync(fullPath, backupPath);
fs.writeFileSync(fullPath, '[]\n', 'utf8');
console.log(`${relPath} 已重置為空陣列(原檔備份至 ${relPath}.bak`);
} catch (repairErr) {
console.error(`${relPath} 修正失敗: ${repairErr.message}`);
process.exit(1);
}
}
}
// Step8: commit/push findings.json 到來源分支
console.log('\n💾 Step8: 記憶區 Commit/Push');
await commitAndPush(WORKSPACE); await commitAndPush(WORKSPACE);
// Step7: 有 critical 問題則 exit 1 // Step9: 有 critical 問題則 exit 1
console.log('\n🚦 Step7: 嚴重問題檢查'); console.log('\n🚦 Step9: 嚴重問題檢查');
const criticalCount = filtered.filter(f => f.level === 'critical').length; const criticalCount = filtered.filter(f => f.level === 'critical').length;
if (criticalCount > 0) { if (criticalCount > 0) {
console.log(` ❌ 發現 ${criticalCount} 個嚴重問題,workflow 結束(exit 1`); console.log(` ❌ 發現 ${criticalCount} 個嚴重問題,workflow 結束(exit 1`);
@@ -114,7 +146,6 @@ async function main() {
process.exit(1); process.exit(1);
} }
console.log(' ✅ 無嚴重問題'); console.log(' ✅ 無嚴重問題');
console.log('\n✅ Pipeline 完成'); console.log('\n✅ Pipeline 完成');
console.log('='.repeat(60)); console.log('='.repeat(60));
} }