#!/bin/bash
# gentoo-infra: infra/githooks.git:update-06-copyright
# Verify that GCO sign-off is present in commit messages
# Copyright 2018-2021 Michał Górny and others
# Distributed under the terms of the GNU General Public License v2 or later

# Disable filename expansion
set -f
# Force UTF-8
export LC_CTYPE=en_US.UTF-8
# Make == case-insensitive
shopt -s nocasematch

# --- Command line
refname=${1}
oldrev=${2}
newrev=${3}

# --- Safety check
if [[ -z ${GIT_DIR} ]]; then
	echo "Don't run this script from the command line." >&2
	echo " (if you want, you could supply GIT_DIR then run" >&2
	echo "  ${0} <ref> <oldrev> <newrev>)" >&2
	exit 1
fi

if [[ -z ${refname} || -z ${oldrev} || -z ${newrev} ]]; then
	echo "usage: ${0} <ref> <oldrev> <newrev>" >&2
	exit 1
fi

# Gentoo devs get extra realname checks
get_from_ldif() {
	local key=${1}
	local line

	while read -r line; do
		case ${line} in
			"${key}:: "*)
				# base64-encoded value
				base64 -d <<<"${line#*:: }"
				break
				;;
			"${key}: "*)
				echo "${line#*: }"
				break
				;;
		esac
	done
}

if [[ ${GL_USER} == *@gentoo.org ]]; then
	ldif=$(ldapsearch "uid=${GL_USER%@gentoo.org}" -D '' -Z -LLL \
		cn gecos -o ldif-wrap=no)
	cn_expected=$(get_from_ldif cn <<<"${ldif}")
	gecos_expected=$(get_from_ldif gecos <<<"${ldif}")

	if [[ -z ${cn_expected} ]]; then
		echo "Unable to get cn for ${GL_USER}, please report!" >&2
		exit 1
	fi
	if [[ -z ${gecos_expected} ]]; then
		echo "Unable to get gecos for ${GL_USER}, please report!" >&2
		exit 1
	fi
fi

ret=0

# special cases
zeros=0000000000000000000000000000000000000000
# branch removal
[[ ${newrev} == "${zeros}" ]] && exit 0
rev_list_arg="${oldrev}..${newrev}"
exec 10</dev/null
# new branch; check all commits that are reachable from ${newrev}
# but not reachable from any existing ref in refs/
if [[ ${oldrev} == "${zeros}" ]]; then
	rev_list_arg="${newrev}"
	exec 10< <(git rev-parse --not --exclude="${refname}" --all)
fi

while read -r commithash; do
	# verify that the commit message contains Signed-off-by
	signoff=no
	committer=$(git show -q --pretty=format:'%ce' "${commithash}")

	while read -r line; do
		if [[ ${line} == signed-off-by:* ]]; then
			# verify syntax first.  should be:
			# Signed-off-by: Real Name <email@address> [(maybe something)]
			if [[ ${line} != 'signed-off-by: '*' <'*@*.*'>'* ]]; then
				signoff=syntaxerr
				break
			fi

			# verify that DCO-1.1 is not used on licenses directory
			# (suggested by leio)
			if [[ ${line} == *DCO[-\ ]1* ]]; then
				while read -r -d $'\0' filename; do
					if [[ ${filename} == licenses/* ]]; then
						signoff=dcolicense
						break 2
					fi
				done < <(git diff -z --name-only "${commithash}^" "${commithash}")
			fi

			# if we already found the correct one, just verify syntax
			# of the rest
			[[ ${signoff} == ok ]] && continue

			# strip the key
			line=${line#*: }

			mail=${line#*<}
			mail=${mail%%>*}
			# different mail?  try other signoffs, maybe reject.
			if [[ ${mail} != "${committer}" ]]; then
				signoff=diffmail
				continue
			fi

			# is it the dev?  verify real name then.
			if [[ ${GL_USER} == *@gentoo.org && ${GL_USER} == "${committer}" ]]; then
				realname=${line%% <*}

				# require either CN or GECOS to match (to allow
				# for ASCII spelling)
				# also allow for a single comment after the name
				# e.g. for nickname, requested by jmbsvicetto
				if [[ ${realname} != "${cn_expected}" \
					&& ${realname} != "${gecos_expected}" \
					&& ${realname% (*} != "${cn_expected}" \
					&& ${realname% (*} != "${gecos_expected}" ]]
				then
					signoff=diffname
					break
				fi
			fi

			signoff=ok
		fi
	done < <(git show -q --pretty=format:'%b' "${commithash}")

	case ${signoff} in
		no)
			echo "${commithash}: missing Signed-off-by on commit"
			ret=1;;
		syntaxerr)
			echo "${commithash}: malformed Signed-off-by (should be: real name <email>)!"
			echo "  ${line}"
			ret=1;;
		dcolicense)
			echo "${commithash}: DCO-1.1 Signed-off-by used on license directory!"
			ret=1;;
		diffmail)
			echo "${commithash}: no Signed-off-by line matching committer's e-mail address found!"
			echo "  expected: ${committer}"
			echo "  last found: ${mail}"
			ret=1;;
		diffname)
			echo "${commithash}: name in Signed-off-by does not match realname in LDAP!"
			echo "  expected: ${cn_expected} (${gecos_expected})"
			echo "  last found: ${realname}"
			ret=1;;
	esac
done < <(git rev-list "${rev_list_arg}" --stdin <&10)

if [[ ${ret} == 1 ]]; then
	echo
	echo "Please make sure to read the copyright policy before adding Signed-off-by!"
	echo "  https://www.gentoo.org/glep/glep-0076.html"
fi

# --- Finished
exit "${ret}"