Merge tag 'v0.5.15' into update/danswer-upstream

Author: sd109

.github/actions/custom-build-and-push/action.yml

@@ -0,0 +1,76 @@
+name: 'Build and Push Docker Image with Retry'
+description: 'Attempts to build and push a Docker image, with a retry on failure'
+inputs:
+  context:
+    description: 'Build context'
+    required: true
+  file:
+    description: 'Dockerfile location'
+    required: true
+  platforms:
+    description: 'Target platforms'
+    required: true
+  pull:
+    description: 'Always attempt to pull a newer version of the image'
+    required: false
+    default: 'true'
+  push:
+    description: 'Push the image to registry'
+    required: false
+    default: 'true'
+  load:
+    description: 'Load the image into Docker daemon'
+    required: false
+    default: 'true'
+  tags:
+    description: 'Image tags'
+    required: true
+  cache-from:
+    description: 'Cache sources'
+    required: false
+  cache-to:
+    description: 'Cache destinations'
+    required: false
+  retry-wait-time:
+    description: 'Time to wait before retry in seconds'
+    required: false
+    default: '5'
+
+runs:
+  using: "composite"
+  steps:
+    - name: Build and push Docker image (First Attempt)
+      id: buildx1
+      uses: docker/build-push-action@v5
+      continue-on-error: true
+      with:
+        context: ${{ inputs.context }}
+        file: ${{ inputs.file }}
+        platforms: ${{ inputs.platforms }}
+        pull: ${{ inputs.pull }}
+        push: ${{ inputs.push }}
+        load: ${{ inputs.load }}
+        tags: ${{ inputs.tags }}
+        cache-from: ${{ inputs.cache-from }}
+        cache-to: ${{ inputs.cache-to }}
+
+    - name: Wait to retry
+      if: steps.buildx1.outcome != 'success'
+      run: |
+        echo "First attempt failed. Waiting ${{ inputs.retry-wait-time }} seconds before retry..."
+        sleep ${{ inputs.retry-wait-time }}
+      shell: bash
+
+    - name: Build and push Docker image (Retry Attempt)
+      if: steps.buildx1.outcome != 'success'
+      uses: docker/build-push-action@v5
+      with:
+        context: ${{ inputs.context }}
+        file: ${{ inputs.file }}
+        platforms: ${{ inputs.platforms }}
+        pull: ${{ inputs.pull }}
+        push: ${{ inputs.push }}
+        load: ${{ inputs.load }}
+        tags: ${{ inputs.tags }}
+        cache-from: ${{ inputs.cache-from }}
+        cache-to: ${{ inputs.cache-to }}

.github/workflows/pr-helm-chart-testing.yml.disabled.txt

@@ -0,0 +1,67 @@
+# This workflow is intentionally disabled while we're still working on it
+# It's close to ready, but a race condition needs to be fixed with
+# API server and Vespa startup, and it needs to have a way to build/test against
+# local containers
+
+name: Helm - Lint and Test Charts
+
+on:
+  merge_group:
+  pull_request:
+    branches: [ main ]
+
+jobs:
+  lint-test:
+    runs-on: Amd64
+
+    # fetch-depth 0 is required for helm/chart-testing-action
+    steps:
+    - name: Checkout code
+      uses: actions/checkout@v3
+      with:
+        fetch-depth: 0
+        
+    - name: Set up Helm
+      uses: azure/[email protected]
+      with:
+        version: v3.14.4
+      
+    - name: Set up Python
+      uses: actions/setup-python@v4
+      with:
+        python-version: '3.11'
+        cache: 'pip'
+        cache-dependency-path: |
+          backend/requirements/default.txt
+          backend/requirements/dev.txt
+          backend/requirements/model_server.txt
+    - run: |
+        python -m pip install --upgrade pip
+        pip install -r backend/requirements/default.txt
+        pip install -r backend/requirements/dev.txt
+        pip install -r backend/requirements/model_server.txt
+
+    - name: Set up chart-testing
+      uses: helm/[email protected]
+
+    - name: Run chart-testing (list-changed)
+      id: list-changed
+      run: |
+        changed=$(ct list-changed --target-branch ${{ github.event.repository.default_branch }})
+        if [[ -n "$changed" ]]; then
+          echo "changed=true" >> "$GITHUB_OUTPUT"
+        fi
+
+    - name: Run chart-testing (lint)
+#       if: steps.list-changed.outputs.changed == 'true'
+      run: ct lint --all --config ct.yaml --target-branch ${{ github.event.repository.default_branch }}
+
+    - name: Create kind cluster
+#       if: steps.list-changed.outputs.changed == 'true'
+      uses: helm/[email protected]
+
+    - name: Run chart-testing (install)
+#       if: steps.list-changed.outputs.changed == 'true'
+      run: ct install --all --config ct.yaml
+#       run: ct install --target-branch ${{ github.event.repository.default_branch }}
+      

.github/workflows/pr-python-connector-tests.yml

@@ -0,0 +1,57 @@
+name: Connector Tests
+
+on:
+  pull_request:
+    branches: [main]
+  schedule:
+    # This cron expression runs the job daily at 16:00 UTC (9am PT)
+    - cron: "0 16 * * *"
+
+env:
+  # Confluence
+  CONFLUENCE_TEST_SPACE_URL: ${{ secrets.CONFLUENCE_TEST_SPACE_URL }}
+  CONFLUENCE_TEST_SPACE: ${{ secrets.CONFLUENCE_TEST_SPACE }}
+  CONFLUENCE_IS_CLOUD: ${{ secrets.CONFLUENCE_IS_CLOUD }}
+  CONFLUENCE_TEST_PAGE_ID: ${{ secrets.CONFLUENCE_TEST_PAGE_ID }}
+  CONFLUENCE_USER_NAME: ${{ secrets.CONFLUENCE_USER_NAME }}
+  CONFLUENCE_ACCESS_TOKEN: ${{ secrets.CONFLUENCE_ACCESS_TOKEN }}
+
+jobs:
+  connectors-check:
+    runs-on: ubuntu-latest
+
+    env:
+      PYTHONPATH: ./backend
+
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Set up Python
+        uses: actions/setup-python@v4
+        with:
+          python-version: "3.11"
+          cache: "pip"
+          cache-dependency-path: |
+            backend/requirements/default.txt
+            backend/requirements/dev.txt
+
+      - name: Install Dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install -r backend/requirements/default.txt
+          pip install -r backend/requirements/dev.txt
+
+      - name: Run Tests
+        shell: script -q -e -c "bash --noprofile --norc -eo pipefail {0}"
+        run: py.test -o junit_family=xunit2 -xv --ff backend/tests/daily/connectors
+
+      - name: Alert on Failure
+        if: failure() && github.event_name == 'schedule'
+        env:
+          SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
+        run: |
+          curl -X POST \
+            -H 'Content-type: application/json' \
+            --data '{"text":"Scheduled Connector Tests failed! Check the run at: https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}"}' \
+            $SLACK_WEBHOOK

.github/workflows/run-it.yml

@@ -28,60 +28,44 @@ jobs:
           username: ${{ secrets.DOCKER_USERNAME }}
           password: ${{ secrets.DOCKER_TOKEN }}
 
-      - name: Build Web Docker image
-        uses: docker/build-push-action@v5
-        with:
-          context: ./web
-          file: ./web/Dockerfile
-          platforms: linux/arm64
-          pull: true
-          push: true
-          load: true
-          tags: danswer/danswer-web-server:it
-          cache-from: type=registry,ref=danswer/danswer-web-server:it
-          cache-to: |
-            type=registry,ref=danswer/danswer-web-server:it,mode=max
-            type=inline
+      # NOTE: we don't need to build the Web Docker image since it's not used
+      # during the IT for now. We have a separate action to verify it builds 
+      # succesfully
+      - name: Pull Web Docker image
+        run: |
+          docker pull danswer/danswer-web-server:latest
+          docker tag danswer/danswer-web-server:latest danswer/danswer-web-server:it
 
       - name: Build Backend Docker image
-        uses: docker/build-push-action@v5
+        uses: ./.github/actions/custom-build-and-push
         with:
           context: ./backend
           file: ./backend/Dockerfile
           platforms: linux/arm64
-          pull: true
-          push: true
-          load: true
           tags: danswer/danswer-backend:it
           cache-from: type=registry,ref=danswer/danswer-backend:it
           cache-to: |
             type=registry,ref=danswer/danswer-backend:it,mode=max
             type=inline
 
       - name: Build Model Server Docker image
-        uses: docker/build-push-action@v5
+        uses: ./.github/actions/custom-build-and-push
         with:
           context: ./backend
           file: ./backend/Dockerfile.model_server
           platforms: linux/arm64
-          pull: true
-          push: true
-          load: true
           tags: danswer/danswer-model-server:it
           cache-from: type=registry,ref=danswer/danswer-model-server:it
           cache-to: |
             type=registry,ref=danswer/danswer-model-server:it,mode=max
             type=inline
 
       - name: Build integration test Docker image
-        uses: docker/build-push-action@v5
+        uses: ./.github/actions/custom-build-and-push
         with:
           context: ./backend
           file: ./backend/tests/integration/Dockerfile
           platforms: linux/arm64
-          pull: true
-          push: true
-          load: true
           tags: danswer/integration-test-runner:it
           cache-from: type=registry,ref=danswer/integration-test-runner:it
           cache-to: |
@@ -92,8 +76,11 @@ jobs:
         run: |
           cd deployment/docker_compose
           ENABLE_PAID_ENTERPRISE_EDITION_FEATURES=true \
+          AUTH_TYPE=basic \
+          REQUIRE_EMAIL_VERIFICATION=false \
+          DISABLE_TELEMETRY=true \
           IMAGE_TAG=it \
-          docker compose -f docker-compose.dev.yml -p danswer-stack up -d --build
+          docker compose -f docker-compose.dev.yml -p danswer-stack up -d
         id: start_docker
 
       - name: Wait for service to be ready
@@ -137,6 +124,7 @@ jobs:
             -e POSTGRES_PASSWORD=password \
             -e POSTGRES_DB=postgres \
             -e VESPA_HOST=index \
+            -e REDIS_HOST=cache \
             -e API_SERVER_HOST=api_server \
             -e OPENAI_API_KEY=${OPENAI_API_KEY} \
             danswer/integration-test-runner:it

.gitignore

@@ -4,6 +4,7 @@
 .mypy_cache
 .idea
 /deployment/data/nginx/app.conf
-.vscode/launch.json
+.vscode/
 *.sw?
 /backend/tests/regression/answer_quality/search_test_config.yaml
+**.tgz

.vscode/env_template.txt

@@ -1,5 +1,5 @@
-# Copy this file to .env at the base of the repo and fill in the <REPLACE THIS> values
-# This will help with development iteration speed and reduce repeat tasks for dev
+# Copy this file to .env in the .vscode folder
+# Fill in the <REPLACE THIS> values as needed, it is recommended to set the GEN_AI_API_KEY value to avoid having to set up an LLM in the UI
 # Also check out danswer/backend/scripts/restart_containers.sh for a script to restart the containers which Danswer relies on outside of VSCode/Cursor processes
 
 # For local dev, often user Authentication is not needed
@@ -15,7 +15,7 @@ LOG_LEVEL=debug
 
 # This passes top N results to LLM an additional time for reranking prior to answer generation
 # This step is quite heavy on token usage so we disable it for dev generally
-DISABLE_LLM_DOC_RELEVANCE=True
+DISABLE_LLM_DOC_RELEVANCE=False
 
 
 # Useful if you want to toggle auth on/off (google_oauth/OIDC specifically)
@@ -27,9 +27,9 @@ REQUIRE_EMAIL_VERIFICATION=False
 
 # Set these so if you wipe the DB, you don't end up having to go through the UI every time
 GEN_AI_API_KEY=<REPLACE THIS>
-# If answer quality isn't important for dev, use 3.5 turbo due to it being cheaper
-GEN_AI_MODEL_VERSION=gpt-3.5-turbo
-FAST_GEN_AI_MODEL_VERSION=gpt-3.5-turbo
+# If answer quality isn't important for dev, use gpt-4o-mini since it's cheaper
+GEN_AI_MODEL_VERSION=gpt-4o
+FAST_GEN_AI_MODEL_VERSION=gpt-4o
 
 # For Danswer Slack Bot, overrides the UI values so no need to set this up via UI every time
 # Only needed if using DanswerBot
@@ -38,7 +38,7 @@ FAST_GEN_AI_MODEL_VERSION=gpt-3.5-turbo
 
 
 # Python stuff
-PYTHONPATH=./backend
+PYTHONPATH=../backend
 PYTHONUNBUFFERED=1
 
 
@@ -49,4 +49,3 @@ BING_API_KEY=<REPLACE THIS>
 # Enable the full set of Danswer Enterprise Edition features
 # NOTE: DO NOT ENABLE THIS UNLESS YOU HAVE A PAID ENTERPRISE LICENSE (or if you are using this for local testing/development)
 ENABLE_PAID_ENTERPRISE_EDITION_FEATURES=False
-