Git Pre-Commit Hook for filter non UTF-8 encoding.

pre-commit

#!/usr/bin/env bash
#
# .git/hooks/pre-commit
# Prevent committing non–UTF‑8 files, but only for certain extensions

set -e          # exit on any error

# Define which file extensions to check (without the leading dot)
EXTENSIONS=(cs)

# Collect all added/copied/modified files in the index
files=$(git diff --cached --name-only --diff-filter=ACM)

# If no files to check, exit cleanly
[ -z "$files" ] && exit 0

# Helper: join array for grep
ext_pattern=$(printf '|%s' "${EXTENSIONS[@]}")
ext_pattern="(${ext_pattern:1})$"

# Array to hold non‑UTF‑8 files
non_utf8=()

for file in $files; do
  # Skip if it's not a regular file
  [ -f "$file" ] || continue

  # Skip files that don't match our extensions
  if [[ ! "$file" =~ \.${ext_pattern} ]]; then
    continue
  fi

  # Check the staged content for UTF‑8 validity
  if ! git show :"$file" | iconv -f utf-8 -t utf-8 >/dev/null 2>&1; then
    non_utf8+=( "$file" )
  fi
done

# If any non‑UTF‑8 files were found, list them and fail
if [ ${#non_utf8[@]} -ne 0 ]; then
  echo "✖ Error: the following files are not UTF-8 encoded. Please convert them before committing:" >&2
  for f in "${non_utf8[@]}"; do
    echo "  - $f" >&2
  done
  exit 1
fi

# All good
exit 0

Modify EXTENSIONS=(cs) for your file extensions.