Add disallowed words pre-commit check (#20)

# Motivation
<!-- Why is this change necessary? -->

# Content
<!-- Please include a summary of the change -->
# Testing
<!-- How was the change tested? -->
# Please check the following before marking your PR as ready for review

- [ ] I have added tests for my changes
- [ ] I have updated the documentation or added new documentation as
needed
- [ ] I have read and agree to the [Contributor License
Agreement](../CLA.md)

---------

Co-authored-by: Edward Li <[email protected]>

Author: Edward-Codegen

.gitattributes

@@ -3,3 +3,4 @@ tests/codemod/repos/extra/*.json filter=lfs diff=lfs merge=lfs -text
 **/expected_diff.patch filter=lfs diff=lfs merge=lfs -text
 tests/codemod/repos/repos.json filter=lfs diff=lfs merge=lfs -text
 tests/verified_codemods/** filter=lfs diff=lfs merge=lfs -text
+.github/disallowed-words.txt filter=lfs diff=lfs merge=lfs -text

.github/disallowed-words.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac0b7be00533f882fb174c498c27372fa78d3ec2d65c0e885475bf92b43b7303
+size 337

.pre-commit-config.yaml

@@ -79,6 +79,13 @@ repos:
         pass_filenames: false
         always_run: true
         entry: bash -c "uv run --frozen --all-extras --dev deptry src codegen-git/src --ignore DEP001"
+  - repo: "local"
+    hooks:
+      - id: disallowed-words-check
+        name: Check for disallowed words
+        entry: scripts/disallowed-words-check.sh
+        language: script
+        files: ''  # Check all files
   - repo: https://github.com/renovatebot/pre-commit-hooks
     rev: 39.120.3
     hooks:

scripts/disallowed-words-check.sh

@@ -0,0 +1,71 @@
+#!/usr/bin/env bash
+
+DISALLOWED_WORDS_FILE=".github/disallowed-words.txt"
+
+# 1) If the file doesn't exist, fail the commit.
+if [[ ! -f "$DISALLOWED_WORDS_FILE" ]]; then
+  echo "ERROR: $DISALLOWED_WORDS_FILE not found."
+  echo "Cannot proceed with disallowed word checks."
+  exit 1
+fi
+
+# 2) If Git LFS isn't installed, fail the commit.
+if ! command -v git-lfs &>/dev/null; then
+  echo "ERROR: Git LFS not installed or not in PATH."
+  echo "Cannot proceed with disallowed word checks."
+  exit 1
+fi
+
+# 3) If the file is still an LFS pointer (not synced), fail the commit.
+if grep -q "https://git-lfs.github.com/spec/v1" "$DISALLOWED_WORDS_FILE"; then
+  echo "ERROR: $DISALLOWED_WORDS_FILE is an LFS pointer but not synced."
+  echo "Cannot proceed with disallowed word checks."
+  exit 1
+fi
+
+# 4) Read the disallowed words (one per line).
+DISALLOWED_WORDS="$(grep -v '^[[:space:]]*$' "$DISALLOWED_WORDS_FILE")"
+if [[ -z "$DISALLOWED_WORDS" ]]; then
+  echo "ERROR: No disallowed words found in $DISALLOWED_WORDS_FILE."
+  echo "Cannot proceed with disallowed word checks."
+  exit 1
+fi
+
+# Build a single regex WITHOUT word boundaries.
+# NOTE: This is intentionally strict. For example, banning "cat" also flags "catastrophic".
+# Will tweak and change this to a more lenient regex later.
+DISALLOWED_REGEX="($(echo "$DISALLOWED_WORDS" | paste -s -d '|' -))"
+
+# 5) Find staged files that are Added (A) or Modified (M).
+FILES_TO_CHECK=$(git diff --cached --name-status | egrep '^[AM]' | cut -f2)
+# FILES_TO_CHECK=$(git ls-files)  # Uncomment this to check ALL files - CAUTION: SLOW!
+
+FAILED=0
+
+# 6) For each file:
+#    - Check the filename itself.
+#    - Check the file contents (if it exists).
+for FILE in $FILES_TO_CHECK; do
+  FILENAME_MATCHES=$(echo "$FILE" | grep -i -E -o "$DISALLOWED_REGEX")
+  if [[ -n "$FILENAME_MATCHES" ]]; then
+    echo "ERROR: Filename '$FILE' contains these disallowed words:"
+    echo "$FILENAME_MATCHES"
+    FAILED=1
+  fi
+
+  if [[ -f "$FILE" ]]; then
+    CONTENT_MATCHES=$(grep -I -i -E -o "$DISALLOWED_REGEX" "$FILE" | sort -u)
+    if [[ -n "$CONTENT_MATCHES" ]]; then
+      echo "ERROR: File '$FILE' contains these disallowed words:"
+      echo "$CONTENT_MATCHES"
+      FAILED=1
+    fi
+  fi
+done
+
+# 7) Block commit if any violations were found.
+if [[ $FAILED -eq 1 ]]; then
+  exit 1
+fi
+
+exit 0