feat(ci): add Vale style check workflow

.github/scripts/resolve-shared-content.sh

@@ -0,0 +1,45 @@
+#!/bin/bash
+# Resolve shared content files to their consuming product pages.
+#
+# Usage:
+#   echo "content/shared/foo.md" | ./resolve-shared-content.sh
+#   ./resolve-shared-content.sh < changed_files.txt
+#   ./resolve-shared-content.sh changed_files.txt
+#
+# For shared files (content/shared/*), finds all pages with matching
+# `source:` frontmatter and outputs those instead. Non-shared files
+# pass through unchanged.
+
+set -euo pipefail
+
+# Read input from file argument or stdin
+if [[ $# -gt 0 && -f "$1" ]]; then
+  INPUT=$(cat "$1")
+else
+  INPUT=$(cat)
+fi
+
+# Process each file
+while IFS= read -r file; do
+  [[ -z "$file" ]] && continue
+
+  if [[ "$file" == content/shared/* ]]; then
+    # Extract the shared path portion (e.g., /shared/influxdb3-cli/foo.md)
+    SHARED_PATH="${file#content}"
+
+    # Find all files that source this shared content
+    # The source frontmatter looks like: source: /shared/path/to/file.md
+    CONSUMERS=$(grep -rl "^source: ${SHARED_PATH}$" content/ 2>/dev/null | grep -v '^content/shared/' || true)
+
+    if [[ -n "$CONSUMERS" ]]; then
+      echo "$CONSUMERS"
+    else
+      # No consumers found - output the shared file itself
+      # (Vale can still lint it with default config)
+      echo "$file"
+    fi
+  else
+    # Non-shared file - pass through
+    echo "$file"
+  fi
+done <<< "$INPUT" | sort -u

.github/scripts/vale-check.sh

@@ -0,0 +1,113 @@
+#!/bin/bash
+# Run Vale on files with appropriate product-specific configs.
+#
+# Usage:
+#   ./vale-check.sh file1.md file2.md ...
+#   ./vale-check.sh < files.txt
+#   ./vale-check.sh --files files.txt
+#
+# Groups files by their Vale config, runs Vale once per group,
+# and outputs combined JSON results.
+#
+# Exit codes:
+#   0 - No errors (warnings/suggestions allowed)
+#   1 - One or more errors found
+
+set -euo pipefail
+
+# Parse arguments
+FILES=()
+while [[ $# -gt 0 ]]; do
+  case "$1" in
+    --files)
+      shift
+      while IFS= read -r f || [[ -n "$f" ]]; do
+        [[ -n "$f" ]] && FILES+=("$f")
+      done < "$1"
+      shift
+      ;;
+    *)
+      FILES+=("$1")
+      shift
+      ;;
+  esac
+done
+
+# Read from stdin if no files provided
+if [[ ${#FILES[@]} -eq 0 ]]; then
+  while IFS= read -r f || [[ -n "$f" ]]; do
+    [[ -n "$f" ]] && FILES+=("$f")
+  done
+fi
+
+if [[ ${#FILES[@]} -eq 0 ]]; then
+  echo "No files to check" >&2
+  exit 0
+fi
+
+# Map file to Vale config
+get_vale_config() {
+  local file="$1"
+  case "$file" in
+    content/influxdb3/cloud-dedicated/*)  echo "content/influxdb3/cloud-dedicated/.vale.ini" ;;
+    content/influxdb3/cloud-serverless/*) echo "content/influxdb3/cloud-serverless/.vale.ini" ;;
+    content/influxdb/v2/*)                echo "content/influxdb/v2/.vale.ini" ;;
+    content/*)                            echo ".vale.ini" ;;
+    *)                                    echo ".vale-instructions.ini" ;;
+  esac
+}
+
+# Group files by config
+declare -A CONFIG_GROUPS
+for file in "${FILES[@]}"; do
+  config=$(get_vale_config "$file")
+  CONFIG_GROUPS["$config"]+="$file"$'\n'
+done
+
+# Run Vale for each config group and collect results
+ALL_RESULTS='{"files": {}, "errors": 0, "warnings": 0, "suggestions": 0}'
+TOTAL_ERRORS=0
+
+for config in "${!CONFIG_GROUPS[@]}"; do
+  files="${CONFIG_GROUPS[$config]}"
+  # Remove trailing newline and convert to array
+  mapfile -t file_array <<< "${files%$'\n'}"
+
+  echo "Running Vale with config: $config (${#file_array[@]} files)" >&2
+
+  # Run Vale via Docker
+  RESULT=$(docker run --rm \
+    -v "$(pwd)":/workdir \
+    -w /workdir \
+    jdkato/vale:latest \
+    --config="$config" \
+    --output=JSON \
+    --minAlertLevel=suggestion \
+    "${file_array[@]}" 2>/dev/null || true)
+
+  if [[ -n "$RESULT" && "$RESULT" != "null" ]]; then
+    # Merge results - Vale outputs {filepath: [alerts...]}
+    # Count errors in this batch
+    BATCH_ERRORS=$(echo "$RESULT" | jq '[.[][].Severity] | map(select(. == "error")) | length')
+    BATCH_WARNINGS=$(echo "$RESULT" | jq '[.[][].Severity] | map(select(. == "warning")) | length')
+    BATCH_SUGGESTIONS=$(echo "$RESULT" | jq '[.[][].Severity] | map(select(. == "suggestion")) | length')
+
+    TOTAL_ERRORS=$((TOTAL_ERRORS + BATCH_ERRORS))
+
+    # Merge file results into ALL_RESULTS
+    ALL_RESULTS=$(echo "$ALL_RESULTS" | jq --argjson new "$RESULT" \
+      --argjson e "$BATCH_ERRORS" \
+      --argjson w "$BATCH_WARNINGS" \
+      --argjson s "$BATCH_SUGGESTIONS" \
+      '.files += $new | .errors += $e | .warnings += $w | .suggestions += $s')
+  fi
+done
+
+# Output combined results
+echo "$ALL_RESULTS"
+
+# Exit with error if any errors found
+if [[ $TOTAL_ERRORS -gt 0 ]]; then
+  exit 1
+fi
+exit 0