pygrep

#!/bin/sh

usage="Usage: $(basename $0) [-v|--verbose] [-q|--quiet] [-f|--filenames-only] [-i|--insensitive] EXPR [TARGET_BASE_DIR]: recursive grep in Python source files in order to search for the specified expression in the target sources, either from the TARGET_BASE_DIR directory, if specified, otherwise from the current directory.

  Options:
   -v or --verbose: be specifically verbose
   -q or --quiet: be specifically quiet, just listing matches
   -f or --filenames-only: display only filenames, not also the matched patterns, and if there are multiple matches in the same file, its filename will be output only once (implies quiet); useful for scripts
   -i or --insensitive: perform case-insensitive searches in the content of files, and also in the searched Python filenames

  Example: $(basename $0) -i 'ConfigParser' /tmp"


if [ "$1" = "-h" ] || [ "$1" = "--help" ]; then

	echo "  ${usage}"
	exit

fi


find="$(which find | grep -v ridiculously 2>/dev/null)"
grep="$(which grep | grep -v ridiculously 2>/dev/null)"

#echo "grep = ${grep}"
#echo "find = ${find}"


if [ ! -x "${grep}" ]; then
	echo "Error, no executable grep found." 1>&2
	exit 5
fi


if [ ! -x "${find}" ]; then
	echo "Error, no executable find found." 1>&2
	exit 6
fi


# Both are false by default:
verbose=1
quiet=1

insensitive_opt=""
filenames_only=1


# A useless file to ensure grep always has multiple files to scan.
# Done to force grep to display the filename (otherwise it will not)
# No need to create it, however.
#
force_multi_files="$HOME/.$(basename $0).to-delete.tmp"

name_opt="-name"

if [ $# -eq 0 ]; then
	echo "  Error, too few parameters.
${usage}" 1>&2
	exit 1
fi


if [ $# -ge 4 ]; then
	echo "  Error, too many parameters ($*).
${usage}" 1>&2
	exit 2
fi



# Read all known options:

token_eaten=0

while [ ${token_eaten} -eq 0 ]; do

	read_parameter="$1"
	#echo "read_parameter = ${read_parameter}"

	token_eaten=1

	if [ "${read_parameter}" = "--verbose" -o "${read_parameter}" = "-v" ]; then
		verbose=0
		quiet=1
		echo "Verbose mode activated."
		token_eaten=0
	fi


	if [ "${read_parameter}" = "--quiet" -o "${read_parameter}" = "-q" ]; then
		verbose=1
		quiet=0
		#echo "Quiet mode activated."
		token_eaten=0
	fi

	if [ "${read_parameter}" = "--filenames-only" -o "${read_parameter}" = "-f" ]; then

		verbose=1
		quiet=0
		filenames_only=0
		[ ${quiet} -eq 0 ] || echo "Only filenames will be displayed, not found patterns."
		token_eaten=0

	fi


	if [ "${read_parameter}" = "--insensitive" -o "${read_parameter}" = "-i" ];	then

		insensitive_opt="-i"
		name_opt="-iname"
		[ ${quiet} -eq 0 ] || echo "Case-insensitive mode activated."
		token_eaten=0

	fi

	if [ -n "${read_parameter}" ]; then
		shift
	fi

done

if [ -z "${read_parameter}" ]; then

	echo "  Error, too few parameters.
${usage}" 1>&2
	exit 3

fi


# full_base_dir is meant to be longer (too long to display before matches) yet
# more precise:
#
target_base_dir="$1"

if [ -z "${target_base_dir}" ]; then

	target_base_dir="."
	full_base_dir="$(realpath .)"
	#[ ${verbose} -eq 0 ] || echo "Proceeding from current directory ('${full_base_dir}')."
else
	full_base_dir=$(realpath "${target_base_dir}")
	#[ ${verbose} -eq 0 ] || echo "Proceeding from specified directory ('${full_base_dir}')."
	shift

fi

if [ ! -d "${target_base_dir}" ]; then

	echo "  Error, target base directory ('${target_base_dir}') does not exist." 1>&2

	exit 10

fi

#echo "target_base_dir = ${target_base_dir}"


searched_expr="${read_parameter}"
#echo "searched_expr = ${searched_expr}"


# Then check no parameter remains:

if [ $# -ge 1 ]; then

	echo "  Error, too many parameters (remaining: '$*').
${usage}" 1>&2
	exit 4

fi


base_grep_opt="--color"

# NEVER surround these single quotes by double ones:
filter_py='*.py'

# Unable to properly define a correctly-protected variable:
#filter_expr="${name_opt} ${filter_py}"

if [ ${verbose} -eq 0 ]; then

	#echo "Will search in following files from '${target_base_dir}' based on: ${find} . ${filter_expr}:"
	echo "Will search in following files from '${target_base_dir}' based on: ${find} . ${name_opt} ${filter_py}:"
	${find} "${target_base_dir}" ${name_opt} "${filter_py}"

fi



[ ${quiet} -eq 0 ] || echo "
  Recursive grep for expression <${searched_expr}> in all Python files from '${full_base_dir}' (i.e. '${target_base_dir}'):
	"

if [ ${filenames_only} -eq 1 ]; then

	# Most usual case:
	${find} "${target_base_dir}" ${name_opt} "${filter_py}" -exec ${grep} ${base_grep_opt} ${insensitive_opt} "${searched_expr}" '{}' ${force_multi_files} ';' 2>/dev/null

else

	${find} "${target_base_dir}" ${name_opt} "${filter_py}" -exec ${grep} ${base_grep_opt} ${insensitive_opt} "${searched_expr}" '{}' ${force_multi_files} ';' 2>/dev/null | sed 's|:.*$||1' | uniq

fi

[ ${quiet} -eq 0 ] || echo "
  End of search."