fix: source_documents url error

.github/workflows/build-image.yml

@@ -4,48 +4,53 @@ on:
   push:
     branches:
       - main
-      - 'release/*'
+      - "release-*"
+    paths:
+      - '**.py'
+      - '**.yaml'
 
 jobs:
   build:
     runs-on: ubuntu-latest
 
     steps:
-    - name: Checkout repository
-      uses: actions/checkout@v2
-
-    - name: Set up docker buildx
-      id: buildx
-      uses: docker/setup-buildx-action@v2
-      with:
-        version: latest
-
-    - name: Docker login
-      env:
-        DOCKER_USERNAME: ${{ secrets.ALIYUN_REGISTRY_USERNAME }}
-        DOCKER_PASSWORD: ${{ secrets.ALIYUN_REGISTRY_PASSWORD }}
-      run: |
-        echo "${DOCKER_PASSWORD}" | docker login --username "${DOCKER_USERNAME}" --password-stdin registry.cn-shanghai.aliyuncs.com
-
-    - name: Extract branch name
-      id: extract_branch
-      run: echo "BRANCH_NAME=${GITHUB_REF#refs/heads/}" >> $GITHUB_ENV
-
-    - name: Extract Tag
-      id: extract_tag
-      run: echo ::set-output name=version::${GITHUB_REF/refs\/tags\//}
-
-    - name: Run buildx
-      run: |
-        tag=latest
-        if [[ "${{ steps.extract_branch.outputs.branch }}" == "main" ]];then 
-           tag=latest
-        elif [[ "${{ steps.extract_tag.outputs.version }}" == v* ]]; then
-          tag="${{ steps.extract_tag.outputs.version }}"
-        fi
-
-        echo extract_tag   ${{ steps.extract_tag.outputs.version }}
-        echo extract_branch  ${{ steps.extract_branch.outputs.branch }}
-        echo current tag is ${tag}
-
-        docker buildx build --platform linux/amd64,linux/arm64 --push -f deploy/docker/Dockerfile -t registry.cn-shanghai.aliyuncs.com/openhydra/aes-ai-tutor:${tag} .
+      - name: Checkout repository
+        uses: actions/checkout@v2
+
+      - name: Set up docker buildx
+        id: buildx
+        uses: docker/setup-buildx-action@v2
+        with:
+          version: latest
+
+      - name: Docker login
+        env:
+          DOCKER_USERNAME: ${{ secrets.ALIYUN_REGISTRY_USERNAME }}
+          DOCKER_PASSWORD: ${{ secrets.ALIYUN_REGISTRY_PASSWORD }}
+        run: |
+          echo "${DOCKER_PASSWORD}" | docker login --username "${DOCKER_USERNAME}" --password-stdin registry.cn-shanghai.aliyuncs.com
+
+      - name: Extract Tag
+        id: extract_tag
+        run: echo ::set-output name=version::${GITHUB_REF/refs\/tags\//}
+
+      - name: Extract branch name
+        shell: bash
+        run: echo "##[set-output name=branch;]$(echo ${GITHUB_REF#refs/heads/})"
+        id: extract_branch
+
+      - name: Run buildx
+        run: |
+          if [[ "${{ steps.extract_branch.outputs.branch }}" == "main" ]];then 
+             tag=latest
+          elif [[ "${{ steps.extract_tag.outputs.version }}" == v* ]]; then
+            tag="${{ steps.extract_tag.outputs.version }}"
+          else
+            tag="${{ steps.extract_branch.outputs.branch }}"
+          fi
+
+          echo extract_tag   ${{ steps.extract_tag.outputs.version }}
+          echo extract_branch  ${{ steps.extract_branch.outputs.branch }}
+          echo current tag is ${tag}
+
+          docker buildx build --platform linux/amd64,linux/arm64 --push -f deploy/docker/Dockerfile -t registry.cn-shanghai.aliyuncs.com/openhydra/aes-ai-tutor:${tag} .

deploy/install.yaml

@@ -18,7 +18,7 @@ spec:
       serviceAccountName: ai-education-studio-admin
       containers:
       - name: aes-ai-tutor-container
-        image:  registry.cn-shanghai.aliyuncs.com/openhydra/aes-ai-tutor:latest
+        image:  registry.cn-shanghai.aliyuncs.com/openhydra/aes-ai-tutor:release-2.0
         imagePullPolicy: IfNotPresent
         command: ["/bin/sh", "-c"]
         args: ["python cli.py init && python cli.py start --api"]

libs/aitutor/server/chat/kb_chat.py

@@ -101,9 +101,16 @@ async def knowledge_base_chat_iterator() -> AsyncIterable[str]:
                     file_name="",
                     metadata={},
                 )
-                source_documents = format_reference(
-                    kb_name, docs, api_address(is_public=True)
-                )
+                source_documents = []
+                for inum, doc in enumerate(docs):
+                    filename = doc.metadata.get("source")
+                    text = f"""出处 [{inum + 1}] [{filename}] \n\n{doc.page_content}\n\n"""
+                    source_documents.append(text)
+
+                if len(source_documents) == 0:  # 没有找到相关文档
+                    source_documents.append(
+                        f"""<span style='color:red'>未找到相关文档,该回答为大模型自身能力解答！</span>"""
+                    )
             elif mode == "temp_kb":
                 ok, msg = check_embed_model()
                 if not ok:
@@ -115,9 +122,16 @@ async def knowledge_base_chat_iterator() -> AsyncIterable[str]:
                     top_k=top_k,
                     score_threshold=score_threshold,
                 )
-                source_documents = format_reference(
-                    kb_name, docs, api_address(is_public=True)
-                )
+                source_documents = []
+                for inum, doc in enumerate(docs):
+                    filename = doc.metadata.get("source")
+                    text = f"""出处 [{inum + 1}] [{filename}] \n\n{doc.page_content}\n\n"""
+                    source_documents.append(text)
+
+                if len(source_documents) == 0:  # 没有找到相关文档
+                    source_documents.append(
+                        f"""<span style='color:red'>未找到相关文档,该回答为大模型自身能力解答！</span>"""
+                    )
             elif mode == "search_engine":
                 result = await run_in_threadpool(search_engine, query, top_k, kb_name)
                 docs = [x.dict() for x in result.get("docs", [])]