#!/bin/bash
#
# A hook script to check the commit log message to ensure that it has
# a well-formed commit summary and body, a valid Signed-off-by: line,
# and a Gerrit Change-Id: line (added automatically if missing).
#
# Called by git-commit with one argument, the name of the file
# that has the commit message.  The hook should exit with non-zero
# status after issuing an appropriate message if it wants to stop the
# commit.  The hook is allowed to edit the commit message file.
#
# Should be installed as .git/hooks/commit-msg.
#

init() {
        set -a
        readonly ORIGINAL="$1"
        readonly REVISED="$(mktemp "$ORIGINAL.XXXXXX")"
        readonly SAVE="$(basename $ORIGINAL).$(date +%Y%m%d.%H%M%S)"
        readonly SIGNOFF="Signed-off-by:"
        readonly CHANGEID="Change-Id:"
        readonly TESTPARAMS="Test-Parameters:"
        readonly INNOCUOUS=$(echo \
                        Acked-by \
                        Tested-by \
                        Reported-by \
                        Reviewed-by \
                        CC \
                | tr ' ' '|')
        readonly WIDTH_SUM=62
        readonly WIDTH_REG=70
        readonly JIRA_FMT_A="^[A-Z]\{2,5\}-[0-9]\{1,5\} [-a-z0-9]\{2,11\}: "
        readonly JIRA_FMT_B="^[A-Z]\{2,5\}-[0-9]\{1,5\} "

        # Identify a name followed by an email address.
        #
        readonly EMAILPAT=$'[ \t]*[^<> ]* [^<>]* <[^@ \t>]+@[a-zA-Z0-9.-]+\.[a-z]+>'

        HAS_ERROR=false
        HAS_SUMMARY=false
        HAS_LAST_BLANK=false
        HAS_BODY=false
        HAS_SIGNOFF=false
        HAS_CHANGEID=false

        IS_WRAPPING_UP=false

        LINE=""
        NUM=0
        set +a
}

# die: commit-msg fatal error: script error or empty input message
# All output redirected to stderr.
#
die() {
        echo "commit-msg fatal error:  $*"
        test -f "$REVISED" && rm -f "$REVISED"
        exit 1
} 1>&2

# Called when doing the final "wrap up" clause because we've found
# one of the tagged lines that belongs in the final section.
#
function ck_wrapup() {
        $IS_WRAPPING_UP && return

        $HAS_LAST_BLANK || error "blank line must preceed signoff section"
        $HAS_SUMMARY    || error "missing commit summary line."
        $HAS_BODY       || error "missing commit description."

        HAS_LAST_BLANK=false
        IS_WRAPPING_UP=true
}

function do_signoff() {
        ck_wrapup
        # Signed-off-by: First Last <email@host.domain>
        local txt=$(echo "${LINE#*: }" | grep -E "${EMAILPAT}")
        if (( ${#txt} == 0 )); then
                error "$SIGNOFF line requires name and email address"
        else
                HAS_SIGNOFF=true # require at least one
        fi
}

function do_changeid() {
        ck_wrapup
        $HAS_CHANGEID && error "multiple $CHANGEID lines are not allowed"

        # Change-Id: I1234567890123456789012345678901234567890
        # capital "I" plus 40 hex digits
        #
        local txt=$(echo "$LINE" | grep "^$CHANGEID I[0-9a-fA-F]\{40\}\$")
        (( ${#txt} > 0 )) ||
                error "has invalid $CHANGEID line for Gerrit tracking"

        HAS_CHANGEID=true
}

# All "innocuous" lines specify a person and email address
#
function do_innocuous() {
        ck_wrapup
        local txt=$(echo "${LINE#*: }" | grep -E "${EMAILPAT}")
        (( ${#txt} == 0 )) && error "invalid name and address"
}

function do_default_line() {
        $IS_WRAPPING_UP && {
                error "invalid signoff section line"
                return
        }
        if (( NUM == 1 )); then
                HAS_JIRA_COMPONENT=$(echo "$LINE" | grep "$JIRA_FMT_A")

                if (( ${#HAS_JIRA_COMPONENT} == 0 )); then
                        HAS_JIRA=$(echo "$LINE" | grep "$JIRA_FMT_B")
                        if (( ${#HAS_JIRA} > 0 )); then
                                error "has no component in summary."
                        else
                                error "missing JIRA ticket number."
                        fi
                elif (( ${#LINE} > WIDTH_SUM )); then
                        error "summary longer than $WIDTH_SUM columns."
                else
                        HAS_SUMMARY=true
                fi

        elif (( ${#LINE} > WIDTH_REG )); then
                error "has line longer than $WIDTH_REG columns."
        elif ! $HAS_BODY && ! $HAS_LAST_BLANK; then
                error "has no blank line after summary."
        else
                HAS_BODY=true
        fi
        HAS_LAST_BLANK=false
}

# Add a new unique Change-Id
#
new_changeid() {
        local NEWID=$({
                        git var GIT_AUTHOR_IDENT
                        git var GIT_COMMITTER_IDENT
                        git write-tree
                        git rev-parse HEAD 2>/dev/null
                        grep -v "^$SIGNOFF" "$ORIGINAL" | git stripspace -s
                } | git hash-object --stdin)
        (( ${#NEWID} > 0 )) ||
                die "git hash-object failed for $CHANGEID:"

        echo "$CHANGEID I$NEWID"
}

# A commit message error was encountered.
# All output redirected to stderr.
#
error() {
        (( ${#LINE} > 0 )) && echo "line $NUM: $LINE"
        echo "error: commit message $*" | fmt
        HAS_ERROR=true
} 1>&2

usage() {
        exec 1>&2
        cat <<- EOF

	See http://wiki.whamcloud.com/display/PUB/Commit+Comments
	for full details.  An example valid commit comment is:

	LU-nnn component: short description of change under 64 columns

	The "component:" should be a lower-case single-word subsystem of the
	Lustre code that best encompasses the change being made.  Examples of
	components include modules like: llite, lov, lmv, osc, mdc, ldlm, lnet,
	ptlrpc, mds, oss, osd, ldiskfs, libcfs, socklnd, o2iblnd; functional
	subsystems like: recovery, quota, grant; and auxilliary areas like:
	build, tests, docs.  This list is not exhaustive, but is a guideline.

	The commit comment should contain a detailed explanation of the change
	being made.  This can be as long as you'd like.  Please give details
	of what problem was solved (including error messages or problems that
	were seen), a good high-level description of how it was solved, and
	which parts of the code were changed (including important functions
	that were changed, if this is useful to understand the patch, and
	for easier searching).  Wrap lines at/under $WIDTH_REG columns.

	Finish the comment with a blank line and a blank-line-free
	sign off section:

	$SIGNOFF Your Real Name <your_email@domain.name>
	$CHANGEID Ixxxx(added automatically if missing)xxxx

	The "$CHANGEID" line should only be there when updating a previous
	commit/submission.  Copy the one from the original commit.

	The "sign off section" may also include several other tag lines:
	$(for T in $(tr '|' ' ' <<< "$INNOCUOUS"); do       \
	     echo "    $T: Some Person <email@domain.com>"; \
	  done)
	$TESTPARAMS optional additional test parameters
	{Organization}-bug-id: associated external change identifier
	EOF

        mv "$ORIGINAL" "$SAVE" &&
                echo "$0: saved original commit comment to $SAVE" 1>&2
}

init ${1+"$@"}
exec 3< "$ORIGINAL" 4> "$REVISED" || exit 1

while IFS= read -u3 LINE; do
        ((NUM += 1))
        case "$LINE" in
        $SIGNOFF* )   do_signoff   ;;
        $CHANGEID* )  do_changeid  ;;
        $TESTPARAMS* ) ck_wrapup   ;;

        "")
                HAS_LAST_BLANK=true
                $IS_WRAPPING_UP && continue
                ;;

        \#*)
                continue ## ignore and suppress comments
                ;;

        "diff --git a/"* )
                # Beginning of uncommented diffstat from "commit -v".  If
                # there are diff and index lines, skip the rest of the input:
                #   diff --git a/build/commit-msg b/build/commit-msg
                #   index 80a3442..acb4c50 100755
                #   deleted file mode 100644
                #   old mode 100644
                # If a "diff --git" line is not followed by one of these
                # lines, do the default line processing on both lines.
                #
                IFS= read -u3 INDEX || break
                ((NUM += 1))
                case "$INDEX" in
                "index "[0-9a-fA-F]*) break ;;
                "deleted file mode "*) break  ;;
                "old mode "*) break ;;
		"new file mode "*) break ;;
                esac
                LINE=${LINE}$'\n'${INDEX}
                do_default_line
                ;;

        *)
                if [[ "$LINE" =~ ^($INNOCUOUS): ]]; then
                        do_innocuous
                elif [[ "$LINE" =~ ^[A-Za-z0-9_-]+-bug-id: ]]; then
                        ck_wrapup
                else
                        # Allow arbitrary external bug identifiers for tracking.
                        # I can't seem to find a pattern for the "case" that
                        # checks for "*-bug-id", so this is checked here.
                        do_default_line
                fi
                ;;
        esac

        echo "$LINE" >&4
done

(( NUM <= 0 )) && die "empty commit message"

unset LINE
$HAS_SIGNOFF || error "missing valid $SIGNOFF: line."

if $HAS_ERROR; then
        exec 3<&- 4>&-
        usage
        rm "$REVISED"
        exit 1
fi

$HAS_CHANGEID || new_changeid >&4
exec 3<&- 4>&-

mv "$REVISED" "$ORIGINAL"

## Local Variables:
## Mode: shell-script
## sh-basic-offset:          8
## sh-indent-after-do:       8
## sh-indentation:           8
## sh-indent-for-case-label: 0
## sh-indent-for-case-alt:   8
## End: