add robust upstream md->mdx transformer tooling

.github/workflows/pull-from-bazel-build.yml

@@ -40,6 +40,15 @@ jobs:
         with:
           bazelisk-cache: true
           repository-cache: true
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: '20'
+
+      - name: Install mdx transformer dependencies
+        working-directory: tools/mdx-transform
+        run: npm ci
 
       - name: Build reference documentation
         working-directory: upstream
@@ -133,4 +142,4 @@ jobs:
             echo "Changes committed and pushed successfully"
           else
             echo "No changes detected, skipping commit"
-          fi
+          fi

copy-upstream-docs.sh

@@ -22,38 +22,7 @@ LOCAL_FILES="
 
 # Files that are not valid MDX syntax
 # This output pasted from a CI job - we should burn it down to zero
-BROKEN_FILES="
-community/roadmaps-configurability.mdx
-concepts/build-files.mdx
-concepts/dependencies.mdx
-concepts/labels.mdx
-configure/integrate-cpp.mdx
-contribute/docs-style-guide.mdx
-contribute/search.mdx
-docs/cc-toolchain-config-reference.mdx
-docs/user-manual.mdx
-extending/config.mdx
-external/mod-command.mdx
-external/registry.mdx
-external/migration_tool.mdx
-query/language.mdx
-query/quickstart.mdx
-reference/be/functions.mdx
-reference/be/platforms-and-toolchains.mdx
-reference/command-line-reference.mdx
-reference/flag-cheatsheet.mdx
-reference/test-encyclopedia.mdx
-remote/dynamic.mdx
-rules/lib/globals/bzl.mdx
-rules/lib/repo/cache.mdx
-rules/lib/repo/git.mdx
-rules/lib/repo/http.mdx
-rules/lib/repo/local.mdx
-rules/lib/repo/utils.mdx
-rules/lib/globals/module.mdx
-rules/windows.mdx
-run/build.mdx
-"
+BROKEN_FILES=""
 
 # Verify that at least one source exists
 if [ ! -d "$UPSTREAM_SITE" ] && [ ! -d "$REFERENCE_DOCS" ]; then
@@ -91,7 +60,7 @@ transform_docs() {
 
     # Transform and copy the file
     echo "Transforming and copying $source_file to $DEST_DIR/$target_file"
-    awk -f transform-docs.awk "$source_file" > "$DEST_DIR/$target_file"
+    node tools/mdx-transform/transform.mjs "$source_file" "$DEST_DIR/$target_file"
     done
 }