ident.c: add split_ident_line() to parse formatted ident line

The commit formatting logic format_person_part() in pretty.c
implements the logic to split an author/committer ident line into
its parts, intermixed with logic to compute its output using these
piece it computes.

Separate the former out to a helper function split_ident_line() so
that other codepath can use the same logic, and rewrite the function
using the helper function.

Signed-off-by: Junio C Hamano <gitster@pobox.com>

3 files changed, 104 insertions, 44 deletions

diff --git a/cache.h b/cache.h
index 3a8e1258e7..7c42ecaff8 100644
--- a/cache.h
+++ b/cache.h
@@ -928,6 +928,22 @@ extern const char *fmt_name(const char *name, const char *email);
 extern const char *git_editor(void);
 extern const char *git_pager(int stdout_is_tty);
 
+struct ident_split {
+	const char *name_begin;
+	const char *name_end;
+	const char *mail_begin;
+	const char *mail_end;
+	const char *date_begin;
+	const char *date_end;
+	const char *tz_begin;
+	const char *tz_end;
+};
+/*
+ * Signals an success with 0, but time part of the result may be NULL
+ * if the input lacks timestamp and zone
+ */
+extern int split_ident_line(struct ident_split *, const char *, int);
+
 struct checkout {
 	const char *base_dir;
 	int base_dir_len;
diff --git a/ident.c b/ident.c
index f619619b82..87c697c2b0 100644
--- a/ident.c
+++ b/ident.c
@@ -220,6 +220,74 @@ static int copy(char *buf, size_t size, int offset, const char *src)
 	return offset;
 }
 
+/*
+ * Reverse of fmt_ident(); given an ident line, split the fields
+ * to allow the caller to parse it.
+ * Signal a success by returning 0, but date/tz fields of the result
+ * can still be NULL if the input line only has the name/email part
+ * (e.g. reading from a reflog entry).
+ */
+int split_ident_line(struct ident_split *split, const char *line, int len)
+{
+	const char *cp;
+	size_t span;
+	int status = -1;
+
+	memset(split, 0, sizeof(*split));
+
+	split->name_begin = line;
+	for (cp = line; *cp && cp < line + len; cp++)
+		if (*cp == '<') {
+			split->mail_begin = cp + 1;
+			break;
+		}
+	if (!split->mail_begin)
+		return status;
+
+	for (cp = split->mail_begin - 2; line < cp; cp--)
+		if (!isspace(*cp)) {
+			split->name_end = cp + 1;
+			break;
+		}
+	if (!split->name_end)
+		return status;
+
+	for (cp = split->mail_begin; cp < line + len; cp++)
+		if (*cp == '>') {
+			split->mail_end = cp;
+			break;
+		}
+	if (!split->mail_end)
+		return status;
+
+	for (cp = split->mail_end + 1; cp < line + len && isspace(*cp); cp++)
+		;
+	if (line + len <= cp)
+		goto person_only;
+	split->date_begin = cp;
+	span = strspn(cp, "0123456789");
+	if (!span)
+		goto person_only;
+	split->date_end = split->date_begin + span;
+	for (cp = split->date_end; cp < line + len && isspace(*cp); cp++)
+		;
+	if (line + len <= cp || (*cp != '+' && *cp != '-'))
+		goto person_only;
+	split->tz_begin = cp;
+	span = strspn(cp + 1, "0123456789");
+	if (!span)
+		goto person_only;
+	split->tz_end = split->tz_begin + 1 + span;
+	return 0;
+
+person_only:
+	split->date_begin = NULL;
+	split->date_end = NULL;
+	split->tz_begin = NULL;
+	split->tz_end = NULL;
+	return 0;
+}
+
 static const char *env_hint =
 "\n"
 "*** Please tell me who you are.\n"
diff --git a/pretty.c b/pretty.c
index 8688b8f2d4..f2dee308b8 100644
--- a/pretty.c
+++ b/pretty.c
@@ -531,41 +531,24 @@ static size_t format_person_part(struct strbuf *sb, char part,
 {
 	/* currently all placeholders have same length */
 	const int placeholder_len = 2;
-	int start, end, tz = 0;
+	int tz;
 	unsigned long date = 0;
-	char *ep;
-	const char *name_start, *name_end, *mail_start, *mail_end, *msg_end = msg+len;
 	char person_name[1024];
 	char person_mail[1024];
+	struct ident_split s;
+	const char *name_start, *name_end, *mail_start, *mail_end;
 
-	/* advance 'end' to point to email start delimiter */
-	for (end = 0; end < len && msg[end] != '<'; end++)
-		; /* do nothing */
-
-	/*
-	 * When end points at the '<' that we found, it should have
-	 * matching '>' later, which means 'end' must be strictly
-	 * below len - 1.
-	 */
-	if (end >= len - 2)
+	if (split_ident_line(&s, msg, len) < 0)
 		goto skip;
 
-	/* Seek for both name and email part */
-	name_start = msg;
-	name_end = msg+end;
-	while (name_end > name_start && isspace(*(name_end-1)))
-		name_end--;
-	mail_start = msg+end+1;
-	mail_end = mail_start;
-	while (mail_end < msg_end && *mail_end != '>')
-		mail_end++;
-	if (mail_end == msg_end)
-		goto skip;
-	end = mail_end-msg;
+	name_start = s.name_begin;
+	name_end = s.name_end;
+	mail_start = s.mail_begin;
+	mail_end = s.mail_end;
 
 	if (part == 'N' || part == 'E') { /* mailmap lookup */
-		strlcpy(person_name, name_start, name_end-name_start+1);
-		strlcpy(person_mail, mail_start, mail_end-mail_start+1);
+		strlcpy(person_name, name_start, name_end - name_start + 1);
+		strlcpy(person_mail, mail_start, mail_end - mail_start + 1);
 		mailmap_name(person_mail, sizeof(person_mail), person_name, sizeof(person_name));
 		name_start = person_name;
 		name_end = name_start + strlen(person_name);
@@ -581,28 +564,20 @@ static size_t format_person_part(struct strbuf *sb, char part,
 		return placeholder_len;
 	}
 
-	/* advance 'start' to point to date start delimiter */
-	for (start = end + 1; start < len && isspace(msg[start]); start++)
-		; /* do nothing */
-	if (start >= len)
-		goto skip;
-	date = strtoul(msg + start, &ep, 10);
-	if (msg + start == ep)
+	if (!s.date_begin)
 		goto skip;
 
+	date = strtoul(s.date_begin, NULL, 10);
+
 	if (part == 't') {	/* date, UNIX timestamp */
-		strbuf_add(sb, msg + start, ep - (msg + start));
+		strbuf_add(sb, s.date_begin, s.date_end - s.date_begin);
 		return placeholder_len;
 	}
 
 	/* parse tz */
-	for (start = ep - msg + 1; start < len && isspace(msg[start]); start++)
-		; /* do nothing */
-	if (start + 1 < len) {
-		tz = strtoul(msg + start + 1, NULL, 10);
-		if (msg[start] == '-')
-			tz = -tz;
-	}
+	tz = strtoul(s.tz_begin + 1, NULL, 10);
+	if (*s.tz_begin == '-')
+		tz = -tz;
 
 	switch (part) {
 	case 'd':	/* date */
@@ -621,8 +596,9 @@ static size_t format_person_part(struct strbuf *sb, char part,
 
 skip:
 	/*
-	 * bogus commit, 'sb' cannot be updated, but we still need to
-	 * compute a valid return value.
+	 * reading from either a bogus commit, or a reflog entry with
+	 * %gn, %ge, etc.; 'sb' cannot be updated, but we still need
+	 * to compute a valid return value.
 	 */
 	if (part == 'n' || part == 'e' || part == 't' || part == 'd'
 	    || part == 'D' || part == 'r' || part == 'i')