Remove BOM from UTF8 files

find-bom.sh

#!/usr/bin/env bash
set -eu
# find-bom searches for nonbinary files starting with a BOM marker
# Path of BOM-encoded files is print to stdout
# Exit status is 0 if no BOM files has been found
# Exit status is 1 if at least one BOM file has been found

readonly TARGET_ROOT=${1:-$PWD}
echo "Searching for BOM in $TARGET_ROOT"

bomFiles=$(grep -rlI $'\xEF\xBB\xBF' "$TARGET_ROOT" --exclude-dir={.tox,.git,.venv,.hypothesis,.mypy_cache} || echo "")
if [ -z "$bomFiles" ]; then
    exit 0
fi
for f in $bomFiles; do 
    echo "$f"
done
exit 1

fix-bom.sh

#!/usr/bin/env bash
set -eu
# fix-bom search for BOM marker in non-binary files and removes them
# You can either call the script as-is, to run it in the current working folder, and all its children recursively
# Or you can pass a target folder as the command line argument
# 
# Sample use
#   ./fix-bom.sh
#   ./fix-bom.sh ./only/this/specific/tree/

readonly TARGET_ROOT=${1:-$PWD}
echo "Searching for BOM in $TARGET_ROOT"

bomFiles=$(grep -rlI $'\xEF\xBB\xBF' "$TARGET_ROOT" --exclude-dir={.tox,.git,.venv,.hypothesis,.mypy_cache} || echo "")
for f in $bomFiles; do
    echo "Removing BOM from $f"
    sed -i '1s/^\xEF\xBB\xBF//' "$f"
done