[guardrail-change] Install pre-commit LOC budget hook

- Rewrote scripts/check-loc.sh: fixed macOS compat, added --staged mode,
  optimized --all mode with find+wc pipeline
- Added .git/hooks/pre-commit that runs check-loc.sh --staged
- Extended loc-exceptions.txt with glob patterns for test files (*test*)
  and blog content pages (blog/*/page.tsx)

The hook blocks commits containing staged files >500 LOC unless exempted.
Bypass for emergencies: git commit --no-verify

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

.claude/rules/loc-exceptions.txt

@@ -11,9 +11,13 @@
 **/venv/** | owner=infra | reason=Python virtualenv | review=permanent
 
 # Test-Dateien (duerfen groesser sein fuer Table-Driven Tests)
-**/tests/test_cv_vocab_pipeline.py | owner=klausur | reason=umfangreiche OCR Pipeline Tests | review=2026-07-01
+**/*test*.py | owner=all | reason=Tests mit Table-Driven Patterns duerfen groesser sein | review=permanent
-**/tests/test_rbac.py | owner=klausur | reason=RBAC Test-Matrix | review=2026-07-01
+**/*test*.go | owner=all | reason=Go Tests mit Table-Driven Patterns | review=permanent
-**/tests/test_grid_editor_api.py | owner=klausur | reason=Grid Editor Integrationstests | review=2026-07-01
+**/*test*.ts | owner=all | reason=TypeScript Tests | review=permanent
+**/tests/** | owner=all | reason=Test-Verzeichnisse | review=permanent
+
+# Blog-Seiten (reine statische Inhalte, kein Code)
+**/blog/*/page.tsx | owner=website | reason=Statische Blog-Artikel (MDX-artig, reiner Content) | review=permanent
 
 # Pure Data Registries (keine Logik, nur Daten-Definitionen)
 **/dsfa_sources_registry.py | owner=klausur | reason=Pure data registry (license + source definitions, no logic) | review=2027-01-01

scripts/check-loc.sh

@@ -1,5 +1,17 @@
 #!/usr/bin/env bash
-set -euo pipefail
+#
+# LOC Budget Checker — enforces the 500 LOC hard cap.
+#
+# Usage:
+#   bash scripts/check-loc.sh --changed   # only git-changed files (pre-commit)
+#   bash scripts/check-loc.sh --all       # all source files in repo
+#   bash scripts/check-loc.sh --staged    # only staged files (for pre-commit hook)
+#
+# Exit codes:
+#   0 — all files within budget
+#   1 — one or more violations found
+
+set -uo pipefail
 
 MAX_LOC="${MAX_LOC:-500}"
 ROOT_DIR="$(git rev-parse --show-toplevel 2>/dev/null || pwd)"
@@ -7,64 +19,143 @@ EXCEPTIONS_FILE="${ROOT_DIR}/.claude/rules/loc-exceptions.txt"
 
 red()   { printf '\033[31m%s\033[0m\n' "$*"; }
 green() { printf '\033[32m%s\033[0m\n' "$*"; }
+yellow(){ printf '\033[33m%s\033[0m\n' "$*"; }
 
+# Check if a file matches any exception pattern in loc-exceptions.txt
 is_exempt() {
   local file="$1"
   [[ -f "$EXCEPTIONS_FILE" ]] || return 1
+
   while IFS= read -r line; do
+    # Skip empty lines and comments
     [[ -z "$line" ]] && continue
-    [[ "$line" =~ ^# ]] && continue
+    [[ "$line" =~ ^[[:space:]]*# ]] && continue
+
+    # Extract glob pattern (everything before first |)
     local pattern="${line%%|*}"
-    pattern="$(echo "$pattern" | xargs)"
+    pattern="$(echo "$pattern" | xargs)"  # trim whitespace
-    if [[ -n "$pattern" ]] && [[ "$file" == $pattern ]]; then
+    [[ -z "$pattern" ]] && continue
-      return 0
+
-    fi
+    # Match against the file path (supports ** globs)
+    # shellcheck disable=SC2254
+    case "$file" in
+      $pattern) return 0 ;;
+    esac
   done < "$EXCEPTIONS_FILE"
   return 1
 }
 
-count_loc() {
+# Check if file is a source file we care about
-  [[ -f "$1" ]] && awk 'END { print NR }' "$1" || echo 0
+is_source_file() {
+  local file="$1"
+  case "$file" in
+    *.py|*.go|*.ts|*.tsx) return 0 ;;
+    *) return 1 ;;
+  esac
 }
 
-collect_changed_files() {
+# Skip non-source directories
-  { git diff --name-only --cached; git diff --name-only; git ls-files --others --exclude-standard; } | awk 'NF' | sort -u
+should_skip_path() {
+  local file="$1"
+  case "$file" in
+    */node_modules/*|*/.next/*|*/__pycache__/*|*/venv/*|*/.git/*) return 0 ;;
+    *) return 1 ;;
+  esac
 }
 
 main() {
   local mode="changed"
-  [[ "${1:-}" == "--changed" ]] && mode="changed"
+  case "${1:-}" in
-  [[ "${1:-}" == "--all" ]] && mode="all"
+    --changed) mode="changed" ;;
+    --staged)  mode="staged" ;;
+    --all)     mode="all" ;;
+  esac
 
-  local files=()
+  cd "$ROOT_DIR"
-  if [[ "$mode" == "changed" ]]; then
+
-    IFS=$'\n' read -d '' -r -a files < <(collect_changed_files)
+  # Collect files based on mode
-  else
+  local tmpfile
-    IFS=$'\n' read -d '' -r -a files < <(find "$ROOT_DIR" \( -name '*.py' -o -name '*.go' -o -name '*.ts' -o -name '*.tsx' \) -not -path '*/node_modules/*' -not -path '*/.next/*' -not -path '*/__pycache__/*' -not -path '*/venv/*')
+  tmpfile=$(mktemp)
-  fi
+  trap "rm -f '$tmpfile'" EXIT
+
+  case "$mode" in
+    staged)
+      git diff --name-only --cached > "$tmpfile"
+      ;;
+    changed)
+      { git diff --name-only --cached 2>/dev/null; git diff --name-only 2>/dev/null; git ls-files --others --exclude-standard 2>/dev/null; } | sort -u > "$tmpfile"
+      ;;
+    all)
+      # Use find + xargs wc for speed, then filter >500 LOC only
+      find . \( -name '*.py' -o -name '*.go' -o -name '*.ts' -o -name '*.tsx' \) \
+        -not -path '*/node_modules/*' \
+        -not -path '*/.next/*' \
+        -not -path '*/__pycache__/*' \
+        -not -path '*/venv/*' \
+        -not -path '*/.git/*' \
+        -exec wc -l {} + 2>/dev/null \
+        | awk -v max="$MAX_LOC" '$1 > max && !/total$/ { sub(/^[[:space:]]*[0-9]+[[:space:]]*\.\//, ""); print }' > "$tmpfile"
+      # For --all mode, we already have only the violating files
+      # Override the check loop to just check exemptions
+      local fast_violations=0
+      local fast_checked=0
+      local fast_exempted=0
+      while IFS= read -r file; do
+        [[ -z "$file" ]] && continue
+        [[ -f "$file" ]] || continue
+        if is_exempt "$file"; then
+          fast_exempted=$((fast_exempted + 1))
+          continue
+        fi
+        fast_checked=$((fast_checked + 1))
+        local loc
+        loc=$(wc -l < "$file" 2>/dev/null || echo 0)
+        red "VIOLATION: $file ($loc LOC > $MAX_LOC)"
+        fast_violations=$((fast_violations + 1))
+      done < "$tmpfile"
+      echo ""
+      if (( fast_violations > 0 )); then
+        red "$fast_violations file(s) exceed ${MAX_LOC} LOC budget. (violations: $fast_violations, exempted: $fast_exempted)"
+        exit 1
+      fi
+      green "LOC budget check passed. (exempted: $fast_exempted)"
+      exit 0
+      ;;
+  esac
 
-  local failed=0
   local violations=0
+  local checked=0
+  local exempted=0
 
-  for file in "${files[@]}"; do
+  while IFS= read -r file; do
+    [[ -z "$file" ]] && continue
     [[ -f "$file" ]] || continue
-    is_exempt "$file" && continue
+    is_source_file "$file" || continue
+    should_skip_path "$file" && continue
+
+    if is_exempt "$file"; then
+      exempted=$((exempted + 1))
+      continue
+    fi
+
+    checked=$((checked + 1))
     local loc
-    loc="$(count_loc "$file")"
+    loc=$(wc -l < "$file" 2>/dev/null || echo 0)
+
     if (( loc > MAX_LOC )); then
       red "VIOLATION: $file ($loc LOC > $MAX_LOC)"
       violations=$((violations + 1))
-      failed=1
     fi
-  done
+  done < "$tmpfile"
 
-  if (( failed )); then
+  echo ""
-    red ""
+  if (( violations > 0 )); then
-    red "$violations file(s) exceed ${MAX_LOC} LOC budget."
+    red "$violations file(s) exceed ${MAX_LOC} LOC budget. (checked: $checked, exempted: $exempted)"
     exit 1
   fi
 
-  green "LOC budget check passed."
+  green "LOC budget check passed. (checked: $checked, exempted: $exempted)"
+  exit 0
 }
 
 main "$@"