[Doc] add link checking for docs (backport #33090)

.github/workflows/ci-doc-checker.yml

@@ -0,0 +1,115 @@
+name: CI DOC Checker
+
+on:
+  pull_request_target:
+    types:
+      - opened
+      - synchronize
+    branches:
+      - main
+      - 'branch*'
+
+permissions:
+  issues: write
+  pull-requests: write
+
+jobs:
+  doc-checker:
+    runs-on: ubuntu-latest
+    name: DOC FILTER
+    if: >
+      !contains(github.event.pull_request.title, '(sync #') &&
+      !contains(github.event.pull_request.labels.*.name, 'sync') &&
+      (!startsWith(github.head_ref, github.base_ref) || !contains(github.head_ref, '-sync-'))
+    outputs:
+      output1: ${{ steps.doc-changes-info.outputs.doc }}
+    steps:
+      - uses: dorny/paths-filter@v2
+        id: changes
+        with:
+          filters: |
+            doc:
+            - 'docs/**'
+      - name: DOC CHECK INFO
+        id: doc-changes-info
+        run: |
+          echo "doc=${{ steps.changes.outputs.doc }}" >> $GITHUB_OUTPUT
+
+  add-doc-label:
+    needs: doc-checker
+    runs-on: ubuntu-latest
+    name: ADD DOC LABEL
+    if: ${{ needs.doc-checker.outputs.output1 == 'true' }}
+    steps:
+      - name: add document label
+        uses: actions-ecosystem/action-add-labels@v1
+        with:
+          github_token: ${{ secrets.PAT }}
+          labels: documentation
+
+  markdownlint:
+    runs-on: ubuntu-latest
+    needs: add-doc-label
+    env:
+      PR_NUMBER: ${{ github.event.number }}
+    steps:
+      - name: clean
+        run: |
+          rm -rf ${{ github.workspace }}
+          mkdir -p ${{ github.workspace }}
+      - name: BRANCH INFO
+        id: branch
+        run: |
+          echo ${{github.base_ref}}
+          echo "branch=${{github.base_ref}}" >> $GITHUB_OUTPUT
+
+      - uses: actions/checkout@v3
+        with:
+          fetch-depth: 0
+
+      - name: Checkout PR
+        run: |
+          BRANCH=${{steps.branch.outputs.branch}}
+          git config --global user.name "wanpengfei-git";
+          git config --global user.email "[email protected]";
+          git checkout $BRANCH;
+          git pull;
+          BRANCH_NAME="${BRANCH}-${PR_NUMBER}";
+          git fetch origin pull/${PR_NUMBER}/head:${BRANCH_NAME};
+          git checkout $BRANCH_NAME;
+          git checkout -b merge_pr;
+          git merge --squash --no-edit ${BRANCH} || (echo "::error::Merge conflict, please check." && exit -1);
+
+      - uses: articulate/[email protected]
+        with:
+          config: "docs/.markdownlint.yaml"
+          files: 'docs/**/*.md'
+          ignore: node_modules
+          version: 0.28.1
+      - name: docusaurus-mdx-checker
+        if: always()
+        run: |
+          npx docusaurus-mdx-checker -c docs
+      - name: link check
+        if: always()
+        uses: lycheeverse/[email protected]
+        with:
+          fail: true
+          args: >
+            --config docs/lychee.toml
+            --offline "docs/**/*.md"
+
+  behavior-unchange:
+    runs-on: ubuntu-latest
+    needs: add-doc-label
+    env:
+      PR_NUMBER: ${{ github.event.number }}
+      REPO: ${{ github.repository }}
+      GITHUB_TOKEN: ${{ secrets.PAT }}
+    steps:
+      - name: Set Body
+        run: |
+          body=$(gh pr view ${PR_NUMBER} -R ${REPO} --json body -q .body)
+          body=${body//"[x] Yes, this PR will result in a change in behavior."/"[ ] Yes, this PR will result in a change in behavior."}
+          body=${body//"[ ] No, this PR will not result in a change in behavior."/"[x] No, this PR will not result in a change in behavior."}
+          gh pr edit ${PR_NUMBER} -R ${REPO} -b "$body"

docs/lychee.toml

@@ -0,0 +1,104 @@
+#############################  Display  #############################
+
+# Verbose program output
+# Accepts log level: "error", "warn", "info", "debug", "trace"
+verbose = "info"
+
+# Show progress
+progress = false
+
+# Path to summary output file.
+# output = "report.md"
+
+#############################  Cache  ###############################
+
+# Enable link caching. This can be helpful to avoid checking the same links on
+# multiple runs.
+# If we start checking external links we should enable this, but at the
+# moment we are checking links to Markdown files only with arg --offline
+cache = false
+
+# Discard all cached requests older than this duration.
+# max_cache_age = "2d"
+
+#############################  Runtime  #############################
+
+# Number of threads to utilize.
+# Defaults to number of cores available to the system if omitted.
+# threads = 6
+
+# Maximum number of allowed redirects [default: 10]
+# max_redirects = 10
+
+# Maximum number of concurrent network requests [default: 128]
+max_concurrency = 30
+
+#############################  Requests  ############################
+
+# User agent to send with each request
+user_agent = "curl/7.83.1"
+
+# Website timeout from connect to response finished
+timeout = 10
+
+# Comma-separated list of accepted status codes for valid links.
+# Omit to accept all response types.
+#accept = "text/html"
+
+# Proceed for server connections considered insecure (invalid TLS)
+insecure = false
+
+# Comma-separated list of accepted status codes for valid links.
+accept = [200, 204, 301, 429]
+
+# Only test links with the given schemes (e.g. https).
+# Omit to check links with any scheme.
+scheme = [ "https" ]
+
+# When links are available using HTTPS, treat HTTP links as errors.
+require_https = true
+
+# Request method
+method = "get"
+
+# Custom request headers
+headers = []
+
+# Remap URI matching pattern to different URI.
+# remap = [ "https://example.com http://example.invalid" ]
+
+
+#############################  Exclusions  ##########################
+
+# Exclude URLs from checking (supports regex)
+
+ exclude = [
+   ]
+
+# Exclude these filesystem paths from getting checked.
+# We exclude StarRocks_intro as it generates the URLs
+# via a component that takes the path without the `.md`
+# extension, and lychee is looking for a valid path
+# including extension.
+exclude_path = ["./README.md", "./introduction/StarRocks_intro.md", "./assets/"]
+
+include = []
+
+# This prevents checking inside codeblocks
+include_verbatim = false
+
+# Exclude all private IPs from checking
+# Equivalent to setting `exclude_private`, `exclude_link_local`, and `exclude_loopback` to true
+exclude_all_private = true
+
+# # Exclude private IP address ranges from checking
+# exclude_private = false
+
+# # Exclude link-local IP address range from checking
+# exclude_link_local = false
+
+# # Exclude loopback IP address range and localhost from checking
+# exclude_loopback = false
+
+# Exclude all mail addresses from checking
+exclude_mail = true