[PATCH v2 0/7] Allow git log to use mailmap file

[PATCH v2 0/7] Allow git log to use mailmap file

[Antoine Pelisse]

Implement the feature suggested by Rich Mindwinter and Junio C Hamano
(and following his advices)



Allows git show/log commands to map author and committer
names and emails using the mailmap file.

Updates related to this second series:
  - All tests are successful after each patch
  - Use split_ident_line in shortlog.c
  - Documentation has been added to git-log.txt
  - Test has been added to check that we use the file
  - Lots of improvements in the way strbufs are used
  - New interface to map_user()
  - Bunch of other fixes

git glog --committer/--author is still not looking for mailmap user
names.

Antoine Pelisse (7):
  Use split_ident_line to parse author and committer
  mailmap: Remove buffer length limit in map_user
  mailmap: Simplify map_user() interface
  mailmap: Add mailmap structure to rev_info and pp
  pretty: Use mailmap to display username and email
  log: Add --use-mailmap option
  test: Add test for --use-mailmap option

 Documentation/git-log.txt |    5 ++
 builtin/blame.c           |  189 +++++++++++++++++++++------------------------
 builtin/log.c             |    9 ++-
 builtin/shortlog.c        |   53 ++++---------
 commit.h                  |    1 +
 log-tree.c                |    1 +
 mailmap.c                 |   56 ++++++--------
 mailmap.h                 |    4 +-
 pretty.c                  |  111 ++++++++++++++------------
 revision.h                |    1 +
 t/t4203-mailmap.sh        |   14 ++++
 11 files changed, 219 insertions(+), 225 deletions(-)

--
1.7.9.5

[PATCH v2 1/7] Use split_ident_line to parse author and committer

[Antoine Pelisse]

Currently blame.c::get_acline, pretty.c::pp_user_info() and
shortlog.c::insert_one_record are parsing author name, email, time and
tz themselves.

Use ident.c::split_ident_line for better code reuse.

Signed-off-by: Antoine Pelisse <apelisse@gmail.com>
---
 builtin/blame.c    |   59 +++++++++++++++++++---------------------------------
 builtin/shortlog.c |   36 +++++++++-----------------------
 pretty.c           |   35 ++++++++++++++++++-------------
 3 files changed, 52 insertions(+), 78 deletions(-)

diff --git a/builtin/blame.c b/builtin/blame.c
index cfae569..dd4aff9 100644
--- a/builtin/blame.c
+++ b/builtin/blame.c
@@ -1343,8 +1343,9 @@ static void get_ac_line(const char *inbuf, const char *what,
 			int mail_len, char *mail,
 			unsigned long *time, const char **tz)
 {
-	int len, tzlen, maillen;
-	char *tmp, *endp, *timepos, *mailpos;
+	struct ident_split ident;
+	int len, tzlen, maillen, namelen;
+	char *tmp, *endp, *mailpos;
 
 	tmp = strstr(inbuf, what);
 	if (!tmp)
@@ -1355,7 +1356,10 @@ static void get_ac_line(const char *inbuf, const char *what,
 		len = strlen(tmp);
 	else
 		len = endp - tmp;
-	if (person_len <= len) {
+	if (person_len <= len)
+		goto error_out;
+
+	if (split_ident_line(&ident, tmp, len)) {
 	error_out:
 		/* Ugh */
 		*tz = "(unknown)";
@@ -1364,47 +1368,26 @@ static void get_ac_line(const char *inbuf, const char *what,
 		*time = 0;
 		return;
 	}
-	memcpy(person, tmp, len);
 
-	tmp = person;
-	tmp += len;
-	*tmp = 0;
-	while (person < tmp && *tmp != ' ')
-		tmp--;
-	if (tmp <= person)
-		goto error_out;
-	*tz = tmp+1;
-	tzlen = (person+len)-(tmp+1);
+	namelen = ident.name_end - ident.name_begin;
+	memcpy(person, ident.name_begin, namelen);
+	person[namelen] = 0;
 
-	*tmp = 0;
-	while (person < tmp && *tmp != ' ')
-		tmp--;
-	if (tmp <= person)
-		goto error_out;
-	*time = strtoul(tmp, NULL, 10);
-	timepos = tmp;
+	maillen = ident.mail_end - ident.mail_begin + 2;
+	memcpy(mail, ident.mail_begin - 1, maillen);
+	mail[maillen] = 0;
 
-	*tmp = 0;
-	while (person < tmp && !(*tmp == ' ' && tmp[1] == '<'))
-		tmp--;
-	if (tmp <= person)
-		return;
-	mailpos = tmp + 1;
-	*tmp = 0;
-	maillen = timepos - tmp;
-	memcpy(mail, mailpos, maillen);
+	*time = strtoul(ident.date_begin, NULL, 10);
 
-	if (!mailmap.nr)
-		return;
+	tzlen = ident.tz_end - ident.tz_begin;
 
-	/*
-	 * mailmap expansion may make the name longer.
-	 * make room by pushing stuff down.
-	 */
-	tmp = person + person_len - (tzlen + 1);
-	memmove(tmp, *tz, tzlen);
+	/* Place tz at the end of person */
+	*tz = tmp = person + person_len - (tzlen + 1);
+	memcpy(tmp, ident.tz_begin, tzlen);
 	tmp[tzlen] = 0;
-	*tz = tmp;
+
+	if (!mailmap.nr)
+		return;
 
 	/*
 	 * Now, convert both name and e-mail using mailmap
diff --git a/builtin/shortlog.c b/builtin/shortlog.c
index b316cf3..03c6cd7 100644
--- a/builtin/shortlog.c
+++ b/builtin/shortlog.c
@@ -40,40 +40,24 @@ static void insert_one_record(struct shortlog *log,
 	char emailbuf[1024];
 	size_t len;
 	const char *eol;
-	const char *boemail, *eoemail;
 	struct strbuf subject = STRBUF_INIT;
+	struct ident_split ident;
 
-	boemail = strchr(author, '<');
-	if (!boemail)
-		return;
-	eoemail = strchr(boemail, '>');
-	if (!eoemail)
+	if (split_ident_line(&ident, author, strlen(author)))
 		return;
 
 	/* copy author name to namebuf, to support matching on both name and email */
-	memcpy(namebuf, author, boemail - author);
-	len = boemail - author;
-	while (len > 0 && isspace(namebuf[len-1]))
-		len--;
+	len = ident.name_end - ident.name_begin;
+	memcpy(namebuf, ident.name_begin, len);
 	namebuf[len] = 0;
 
 	/* copy email name to emailbuf, to allow email replacement as well */
-	memcpy(emailbuf, boemail+1, eoemail - boemail);
-	emailbuf[eoemail - boemail - 1] = 0;
-
-	if (!map_user(&log->mailmap, emailbuf, sizeof(emailbuf), namebuf, sizeof(namebuf))) {
-		while (author < boemail && isspace(*author))
-			author++;
-		for (len = 0;
-		     len < sizeof(namebuf) - 1 && author + len < boemail;
-		     len++)
-			namebuf[len] = author[len];
-		while (0 < len && isspace(namebuf[len-1]))
-			len--;
-		namebuf[len] = '\0';
-	}
-	else
-		len = strlen(namebuf);
+	len = ident.mail_end - ident.mail_begin;
+	memcpy(emailbuf, ident.mail_begin, len);
+	emailbuf[len] = 0;
+
+	map_user(&log->mailmap, emailbuf, sizeof(emailbuf), namebuf, sizeof(namebuf));
+	len = strlen(namebuf);
 
 	if (log->email) {
 		size_t room = sizeof(namebuf) - len - 1;
diff --git a/pretty.c b/pretty.c
index 5bdc2e7..350d1df 100644
--- a/pretty.c
+++ b/pretty.c
@@ -387,29 +387,36 @@ void pp_user_info(const struct pretty_print_context *pp,
 		  const char *what, struct strbuf *sb,
 		  const char *line, const char *encoding)
 {
+	struct ident_split ident;
+	int linelen, namelen;
+	char *line_end, *date;
 	int max_length = 78; /* per rfc2822 */
-	char *date;
-	int namelen;
 	unsigned long time;
 	int tz;
 
 	if (pp->fmt == CMIT_FMT_ONELINE)
 		return;
-	date = strchr(line, '>');
-	if (!date)
+
+	line_end = strchr(line, '\n');
+	if (!line_end) {
+		line_end = strchr(line, '\0');
+		if (!line_end)
+			return;
+	}
+
+	linelen = ++line_end - line;
+	if (split_ident_line(&ident, line, linelen))
 		return;
-	namelen = ++date - line;
-	time = strtoul(date, &date, 10);
+
+	namelen = ident.mail_end - ident.name_begin + 1;
+	time = strtoul(ident.date_begin, &date, 10);
 	tz = strtol(date, NULL, 10);
 
 	if (pp->fmt == CMIT_FMT_EMAIL) {
-		char *name_tail = strchr(line, '<');
 		int display_name_length;
-		if (!name_tail)
-			return;
-		while (line < name_tail && isspace(name_tail[-1]))
-			name_tail--;
-		display_name_length = name_tail - line;
+
+		display_name_length = ident.name_end - ident.name_begin;
+
 		strbuf_addstr(sb, "From: ");
 		if (needs_rfc2047_encoding(line, display_name_length, RFC2047_ADDRESS)) {
 			add_rfc2047(sb, line, display_name_length,
@@ -427,10 +434,10 @@ void pp_user_info(const struct pretty_print_context *pp,
 		}
 		if (namelen - display_name_length + last_line_length(sb) > max_length) {
 			strbuf_addch(sb, '\n');
-			if (!isspace(name_tail[0]))
+			if (!isspace(ident.name_end[0]))
 				strbuf_addch(sb, ' ');
 		}
-		strbuf_add(sb, name_tail, namelen - display_name_length);
+		strbuf_add(sb, ident.name_end, namelen - display_name_length);
 		strbuf_addch(sb, '\n');
 	} else {
 		strbuf_addf(sb, "%s: %.*s%.*s\n", what,
-- 
1.7.9.5

[PATCH v2 2/7] mailmap: Remove buffer length limit in map_user

[Antoine Pelisse]

Remove the hard limit set by mail buffer in map_user and
use the strbuf API to replace it.

Signed-off-by: Antoine Pelisse <apelisse@gmail.com>
---
 mailmap.c |   17 ++++++-----------
 1 file changed, 6 insertions(+), 11 deletions(-)

diff --git a/mailmap.c b/mailmap.c
index ea4b471..3bc6491 100644
--- a/mailmap.c
+++ b/mailmap.c
@@ -193,7 +193,7 @@ int map_user(struct string_list *map,
 	char *end_of_email;
 	struct string_list_item *item;
 	struct mailmap_entry *me;
-	char buf[1024], *mailbuf;
+	struct strbuf mailbuf = STRBUF_INIT;
 	int i;
 
 	/* figure out space requirement for email */
@@ -204,18 +204,14 @@ int map_user(struct string_list *map,
 		if (!end_of_email)
 			return 0;
 	}
-	if (end_of_email - email + 1 < sizeof(buf))
-		mailbuf = buf;
-	else
-		mailbuf = xmalloc(end_of_email - email + 1);
 
 	/* downcase the email address */
+	strbuf_grow(&mailbuf, end_of_email - email);
 	for (i = 0; i < end_of_email - email; i++)
-		mailbuf[i] = tolower(email[i]);
-	mailbuf[i] = 0;
+		strbuf_addch(&mailbuf, tolower(email[i]));
 
-	debug_mm("map_user: map '%s' <%s>\n", name, mailbuf);
-	item = string_list_lookup(map, mailbuf);
+	debug_mm("map_user: map '%s' <%s>\n", name, mailbuf.buf);
+	item = string_list_lookup(map, mailbuf.buf);
 	if (item != NULL) {
 		me = (struct mailmap_entry *)item->util;
 		if (me->namemap.nr) {
@@ -226,8 +222,7 @@ int map_user(struct string_list *map,
 				item = subitem;
 		}
 	}
-	if (mailbuf != buf)
-		free(mailbuf);
+	strbuf_release(&mailbuf);
 	if (item != NULL) {
 		struct mailmap_info *mi = (struct mailmap_info *)item->util;
 		if (mi->name == NULL && (mi->email == NULL || maxlen_email == 0)) {
-- 
1.7.9.5

[PATCH v2 3/7] mailmap: Simplify map_user() interface

[Antoine Pelisse]

mailmap.c::map_user() is simplified to take two strbuf parameters
instead of statically allocated buffers.

The buffers are still modified in-place, when mapping is required.

It actually makes things easier for callers:
 - Caller must prepare the call by copying author name and email in
 strbuf (instead of copying in static buffer)
 - Caller can easily manipulate the buffer afterward.
 - Caller doesn't have to call strlen() to calculate the size of the
 new string.

Signed-off-by: Antoine Pelisse <apelisse@gmail.com>
---
 builtin/blame.c    |  164 +++++++++++++++++++++++++++-------------------------
 builtin/shortlog.c |   31 ++++------
 mailmap.c          |   51 ++++++++--------
 mailmap.h          |    4 +-
 pretty.c           |   40 ++++++-------
 5 files changed, 139 insertions(+), 151 deletions(-)

diff --git a/builtin/blame.c b/builtin/blame.c
index dd4aff9..8586850 100644
--- a/builtin/blame.c
+++ b/builtin/blame.c
@@ -1321,31 +1321,30 @@ static void pass_blame(struct scoreboard *sb, struct origin *origin, int opt)
  * Information on commits, used for output.
  */
 struct commit_info {
-	const char *author;
-	const char *author_mail;
+	struct strbuf author;
+	struct strbuf author_mail;
 	unsigned long author_time;
-	const char *author_tz;
+	struct strbuf author_tz;
 
 	/* filled only when asked for details */
-	const char *committer;
-	const char *committer_mail;
+	struct strbuf committer;
+	struct strbuf committer_mail;
 	unsigned long committer_time;
-	const char *committer_tz;
+	struct strbuf committer_tz;
 
-	const char *summary;
+	struct strbuf summary;
 };
 
 /*
  * Parse author/committer line in the commit object buffer
  */
 static void get_ac_line(const char *inbuf, const char *what,
-			int person_len, char *person,
-			int mail_len, char *mail,
-			unsigned long *time, const char **tz)
+			struct strbuf *name, struct strbuf *mail,
+			unsigned long *time, struct strbuf *tz)
 {
 	struct ident_split ident;
-	int len, tzlen, maillen, namelen;
-	char *tmp, *endp, *mailpos;
+	int len;
+	char *tmp, *endp;
 
 	tmp = strstr(inbuf, what);
 	if (!tmp)
@@ -1356,51 +1355,61 @@ static void get_ac_line(const char *inbuf, const char *what,
 		len = strlen(tmp);
 	else
 		len = endp - tmp;
-	if (person_len <= len)
-		goto error_out;
 
 	if (split_ident_line(&ident, tmp, len)) {
 	error_out:
 		/* Ugh */
-		*tz = "(unknown)";
-		strcpy(person, *tz);
-		strcpy(mail, *tz);
+		tmp = "(unknown)";
+		strbuf_addstr(name, tmp);
+		strbuf_addstr(mail, tmp);
+		strbuf_addstr(tz, tmp);
 		*time = 0;
 		return;
 	}
 
-	namelen = ident.name_end - ident.name_begin;
-	memcpy(person, ident.name_begin, namelen);
-	person[namelen] = 0;
+	len = ident.name_end - ident.name_begin;
+	strbuf_add(name, ident.name_begin, len);
 
-	maillen = ident.mail_end - ident.mail_begin + 2;
-	memcpy(mail, ident.mail_begin - 1, maillen);
-	mail[maillen] = 0;
+	len = ident.mail_end - ident.mail_begin;
+	strbuf_add(mail, ident.mail_begin, len);
 
 	*time = strtoul(ident.date_begin, NULL, 10);
 
-	tzlen = ident.tz_end - ident.tz_begin;
+	len = ident.tz_end - ident.tz_begin;
+	strbuf_add(tz, ident.tz_begin, len);
 
-	/* Place tz at the end of person */
-	*tz = tmp = person + person_len - (tzlen + 1);
-	memcpy(tmp, ident.tz_begin, tzlen);
-	tmp[tzlen] = 0;
-
-	if (!mailmap.nr)
-		return;
-
-	/*
+        /*
 	 * Now, convert both name and e-mail using mailmap
 	 */
-	if (map_user(&mailmap, mail+1, mail_len-1, person, tmp-person-1)) {
-		/* Add a trailing '>' to email, since map_user returns plain emails
-		   Note: It already has '<', since we replace from mail+1 */
-		mailpos = memchr(mail, '\0', mail_len);
-		if (mailpos && mailpos-mail < mail_len - 1) {
-			*mailpos = '>';
-			*(mailpos+1) = '\0';
-		}
-	}
+	if (mailmap.nr)
+		map_user(&mailmap, mail, name);
+
+	strbuf_insert(mail, 0, "<", 1);
+	strbuf_addch(mail, '>');
+}
+
+static void commit_info_init(struct commit_info *ci)
+{
+
+	strbuf_init(&ci->author, 0);
+	strbuf_init(&ci->author_mail, 0);
+	strbuf_init(&ci->author_tz, 0);
+	strbuf_init(&ci->committer, 0);
+	strbuf_init(&ci->committer_mail, 0);
+	strbuf_init(&ci->committer_tz, 0);
+	strbuf_init(&ci->summary, 0);
+}
+
+static void commit_info_destroy(struct commit_info *ci)
+{
+
+	strbuf_release(&ci->author);
+	strbuf_release(&ci->author_mail);
+	strbuf_release(&ci->author_tz);
+	strbuf_release(&ci->committer);
+	strbuf_release(&ci->committer_mail);
+	strbuf_release(&ci->committer_tz);
+	strbuf_release(&ci->summary);
 }
 
 static void get_commit_info(struct commit *commit,
@@ -1410,11 +1419,8 @@ static void get_commit_info(struct commit *commit,
 	int len;
 	const char *subject, *encoding;
 	char *reencoded, *message;
-	static char author_name[1024];
-	static char author_mail[1024];
-	static char committer_name[1024];
-	static char committer_mail[1024];
-	static char summary_buf[1024];
+
+	commit_info_init(ret);
 
 	/*
 	 * We've operated without save_commit_buffer, so
@@ -1432,11 +1438,7 @@ static void get_commit_info(struct commit *commit,
 	encoding = get_log_output_encoding();
 	reencoded = logmsg_reencode(commit, encoding);
 	message   = reencoded ? reencoded : commit->buffer;
-	ret->author = author_name;
-	ret->author_mail = author_mail;
-	get_ac_line(message, "\nauthor ",
-		    sizeof(author_name), author_name,
-		    sizeof(author_mail), author_mail,
+	get_ac_line(message, "\nauthor ", &ret->author, &ret->author_mail,
 		    &ret->author_time, &ret->author_tz);
 
 	if (!detailed) {
@@ -1444,21 +1446,16 @@ static void get_commit_info(struct commit *commit,
 		return;
 	}
 
-	ret->committer = committer_name;
-	ret->committer_mail = committer_mail;
-	get_ac_line(message, "\ncommitter ",
-		    sizeof(committer_name), committer_name,
-		    sizeof(committer_mail), committer_mail,
-		    &ret->committer_time, &ret->committer_tz);
+	get_ac_line(message, "\ncommitter ", &ret->committer,
+		    &ret->committer_mail, &ret->committer_time,
+		    &ret->committer_tz);
 
-	ret->summary = summary_buf;
 	len = find_commit_subject(message, &subject);
-	if (len && len < sizeof(summary_buf)) {
-		memcpy(summary_buf, subject, len);
-		summary_buf[len] = 0;
-	} else {
-		sprintf(summary_buf, "(%s)", sha1_to_hex(commit->object.sha1));
-	}
+	if (len)
+		strbuf_add(&ret->summary, subject, len);
+	else
+		strbuf_addf(&ret->summary, "(%s)", sha1_to_hex(commit->object.sha1));
+
 	free(reencoded);
 }
 
@@ -1487,15 +1484,15 @@ static int emit_one_suspect_detail(struct origin *suspect, int repeat)
 
 	suspect->commit->object.flags |= METAINFO_SHOWN;
 	get_commit_info(suspect->commit, &ci, 1);
-	printf("author %s\n", ci.author);
-	printf("author-mail %s\n", ci.author_mail);
+	printf("author %s\n", ci.author.buf);
+	printf("author-mail %s\n", ci.author_mail.buf);
 	printf("author-time %lu\n", ci.author_time);
-	printf("author-tz %s\n", ci.author_tz);
-	printf("committer %s\n", ci.committer);
-	printf("committer-mail %s\n", ci.committer_mail);
+	printf("author-tz %s\n", ci.author_tz.buf);
+	printf("committer %s\n", ci.committer.buf);
+	printf("committer-mail %s\n", ci.committer_mail.buf);
 	printf("committer-time %lu\n", ci.committer_time);
-	printf("committer-tz %s\n", ci.committer_tz);
-	printf("summary %s\n", ci.summary);
+	printf("committer-tz %s\n", ci.committer_tz.buf);
+	printf("summary %s\n", ci.summary.buf);
 	if (suspect->commit->object.flags & UNINTERESTING)
 		printf("boundary\n");
 	if (suspect->previous) {
@@ -1503,6 +1500,9 @@ static int emit_one_suspect_detail(struct origin *suspect, int repeat)
 		printf("previous %s ", sha1_to_hex(prev->commit->object.sha1));
 		write_name_quoted(prev->path, stdout, '\n');
 	}
+
+	commit_info_destroy(&ci);
+
 	return 1;
 }
 
@@ -1689,11 +1689,11 @@ static void emit_other(struct scoreboard *sb, struct blame_entry *ent, int opt)
 		if (opt & OUTPUT_ANNOTATE_COMPAT) {
 			const char *name;
 			if (opt & OUTPUT_SHOW_EMAIL)
-				name = ci.author_mail;
+				name = ci.author_mail.buf;
 			else
-				name = ci.author;
+				name = ci.author.buf;
 			printf("\t(%10s\t%10s\t%d)", name,
-			       format_time(ci.author_time, ci.author_tz,
+			       format_time(ci.author_time, ci.author_tz.buf,
 					   show_raw_time),
 			       ent->lno + 1 + cnt);
 		} else {
@@ -1712,14 +1712,14 @@ static void emit_other(struct scoreboard *sb, struct blame_entry *ent, int opt)
 				const char *name;
 				int pad;
 				if (opt & OUTPUT_SHOW_EMAIL)
-					name = ci.author_mail;
+					name = ci.author_mail.buf;
 				else
-					name = ci.author;
+					name = ci.author.buf;
 				pad = longest_author - utf8_strwidth(name);
 				printf(" (%s%*s %10s",
 				       name, pad, "",
 				       format_time(ci.author_time,
-						   ci.author_tz,
+						   ci.author_tz.buf,
 						   show_raw_time));
 			}
 			printf(" %*d) ",
@@ -1734,6 +1734,8 @@ static void emit_other(struct scoreboard *sb, struct blame_entry *ent, int opt)
 
 	if (sb->final_buf_size && cp[-1] != '\n')
 		putchar('\n');
+
+	commit_info_destroy(&ci);
 }
 
 static void output(struct scoreboard *sb, int option)
@@ -1858,9 +1860,9 @@ static void find_alignment(struct scoreboard *sb, int *option)
 			suspect->commit->object.flags |= METAINFO_SHOWN;
 			get_commit_info(suspect->commit, &ci, 1);
 			if (*option & OUTPUT_SHOW_EMAIL)
-				num = utf8_strwidth(ci.author_mail);
+				num = utf8_strwidth(ci.author_mail.buf);
 			else
-				num = utf8_strwidth(ci.author);
+				num = utf8_strwidth(ci.author.buf);
 			if (longest_author < num)
 				longest_author = num;
 		}
@@ -1872,6 +1874,8 @@ static void find_alignment(struct scoreboard *sb, int *option)
 			longest_dst_lines = num;
 		if (largest_score < ent_score(sb, e))
 			largest_score = ent_score(sb, e);
+
+		commit_info_destroy(&ci);
 	}
 	max_orig_digits = decimal_width(longest_src_lines);
 	max_digits = decimal_width(longest_dst_lines);
diff --git a/builtin/shortlog.c b/builtin/shortlog.c
index 03c6cd7..20e8599 100644
--- a/builtin/shortlog.c
+++ b/builtin/shortlog.c
@@ -36,39 +36,30 @@ static void insert_one_record(struct shortlog *log,
 	const char *dot3 = log->common_repo_prefix;
 	char *buffer, *p;
 	struct string_list_item *item;
-	char namebuf[1024];
-	char emailbuf[1024];
-	size_t len;
 	const char *eol;
 	struct strbuf subject = STRBUF_INIT;
+	struct strbuf name = STRBUF_INIT;
+	struct strbuf mail = STRBUF_INIT;
 	struct ident_split ident;
 
 	if (split_ident_line(&ident, author, strlen(author)))
 		return;
 
-	/* copy author name to namebuf, to support matching on both name and email */
-	len = ident.name_end - ident.name_begin;
-	memcpy(namebuf, ident.name_begin, len);
-	namebuf[len] = 0;
+	strbuf_add(&mail, ident.mail_begin, ident.mail_end - ident.mail_begin);
+	strbuf_add(&name, ident.name_begin, ident.name_end - ident.name_begin);
 
-	/* copy email name to emailbuf, to allow email replacement as well */
-	len = ident.mail_end - ident.mail_begin;
-	memcpy(emailbuf, ident.mail_begin, len);
-	emailbuf[len] = 0;
+	map_user(&log->mailmap, &mail, &name);
 
-	map_user(&log->mailmap, emailbuf, sizeof(emailbuf), namebuf, sizeof(namebuf));
-	len = strlen(namebuf);
+	if (log->email)
+		strbuf_addf(&name, " <%s>", mail.buf);
 
-	if (log->email) {
-		size_t room = sizeof(namebuf) - len - 1;
-		int maillen = strlen(emailbuf);
-		snprintf(namebuf + len, room, " <%.*s>", maillen, emailbuf);
-	}
-
-	item = string_list_insert(&log->list, namebuf);
+	item = string_list_insert(&log->list, name.buf);
 	if (item->util == NULL)
 		item->util = xcalloc(1, sizeof(struct string_list));
 
+	strbuf_release(&name);
+	strbuf_release(&mail);
+
 	/* Skip any leading whitespace, including any blank lines. */
 	while (*oneline && isspace(*oneline))
 		oneline++;
diff --git a/mailmap.c b/mailmap.c
index 3bc6491..d5bffc6 100644
--- a/mailmap.c
+++ b/mailmap.c
@@ -187,57 +187,52 @@ void clear_mailmap(struct string_list *map)
 	debug_mm("mailmap: cleared\n");
 }
 
-int map_user(struct string_list *map,
-	     char *email, int maxlen_email, char *name, int maxlen_name)
+int map_user(struct string_list *map, struct strbuf *mail, struct strbuf *name)
 {
-	char *end_of_email;
+
 	struct string_list_item *item;
 	struct mailmap_entry *me;
-	struct strbuf mailbuf = STRBUF_INIT;
+	struct strbuf lowermail = STRBUF_INIT;
 	int i;
 
-	/* figure out space requirement for email */
-	end_of_email = strchr(email, '>');
-	if (!end_of_email) {
-		/* email passed in might not be wrapped in <>, but end with a \0 */
-		end_of_email = memchr(email, '\0', maxlen_email);
-		if (!end_of_email)
-			return 0;
-	}
-
 	/* downcase the email address */
-	strbuf_grow(&mailbuf, end_of_email - email);
-	for (i = 0; i < end_of_email - email; i++)
-		strbuf_addch(&mailbuf, tolower(email[i]));
+	strbuf_grow(&lowermail, mail->len);
+	for (i = 0; i < mail->len; i++)
+		strbuf_addch(&lowermail, tolower(mail->buf[i]));
 
-	debug_mm("map_user: map '%s' <%s>\n", name, mailbuf.buf);
-	item = string_list_lookup(map, mailbuf.buf);
+	debug_mm("map_user: map '%s' <%s>\n", name->buf, lowermail.buf);
+	item = string_list_lookup(map, lowermail.buf);
 	if (item != NULL) {
 		me = (struct mailmap_entry *)item->util;
 		if (me->namemap.nr) {
 			/* The item has multiple items, so we'll look up on name too */
 			/* If the name is not found, we choose the simple entry      */
-			struct string_list_item *subitem = string_list_lookup(&me->namemap, name);
+			struct string_list_item *subitem = string_list_lookup(&me->namemap, name->buf);
 			if (subitem)
 				item = subitem;
 		}
 	}
-	strbuf_release(&mailbuf);
+	strbuf_release(&lowermail);
 	if (item != NULL) {
 		struct mailmap_info *mi = (struct mailmap_info *)item->util;
-		if (mi->name == NULL && (mi->email == NULL || maxlen_email == 0)) {
+		if (mi->name == NULL && mi->email == NULL) {
 			debug_mm("map_user:  -- (no simple mapping)\n");
 			return 0;
 		}
-		if (maxlen_email && mi->email)
-			strlcpy(email, mi->email, maxlen_email);
-		else
-			*end_of_email = '\0';
-		if (maxlen_name && mi->name)
-			strlcpy(name, mi->name, maxlen_name);
-		debug_mm("map_user:  to '%s' <%s>\n", name, mi->email ? mi->email : "");
+		if (mi->email) {
+			strbuf_reset(mail);
+			strbuf_addstr(mail, mi->email);
+		}
+
+		if (mi->name) {
+			strbuf_reset(name);
+			strbuf_addstr(name, mi->name);
+		}
+
+		debug_mm("map_user:  to '%s' <%s>\n", name->buf, mi->email ? mi->email : "");
 		return 1;
 	}
+
 	debug_mm("map_user:  --\n");
 	return 0;
 }
diff --git a/mailmap.h b/mailmap.h
index d5c3664..88dca51 100644
--- a/mailmap.h
+++ b/mailmap.h
@@ -4,7 +4,5 @@
 int read_mailmap(struct string_list *map, char **repo_abbrev);
 void clear_mailmap(struct string_list *map);
 
-int map_user(struct string_list *mailmap,
-	     char *email, int maxlen_email, char *name, int maxlen_name);
-
+int map_user(struct string_list *map, struct strbuf *name, struct strbuf *mail);
 #endif
diff --git a/pretty.c b/pretty.c
index 350d1df..d05c675 100644
--- a/pretty.c
+++ b/pretty.c
@@ -593,14 +593,14 @@ char *logmsg_reencode(const struct commit *commit,
 	return out;
 }
 
-static int mailmap_name(char *email, int email_len, char *name, int name_len)
+static int mailmap_name(struct strbuf *mail, struct strbuf *name)
 {
 	static struct string_list *mail_map;
 	if (!mail_map) {
 		mail_map = xcalloc(1, sizeof(*mail_map));
 		read_mailmap(mail_map, NULL);
 	}
-	return mail_map->nr && map_user(mail_map, email, email_len, name, name_len);
+	return mail_map->nr && map_user(mail_map, mail, name);
 }
 
 static size_t format_person_part(struct strbuf *sb, char part,
@@ -610,39 +610,39 @@ static size_t format_person_part(struct strbuf *sb, char part,
 	const int placeholder_len = 2;
 	int tz;
 	unsigned long date = 0;
-	char person_name[1024];
-	char person_mail[1024];
+	struct strbuf person_name = STRBUF_INIT;
+	struct strbuf person_mail = STRBUF_INIT;
 	struct ident_split s;
-	const char *name_start, *name_end, *mail_start, *mail_end;
 
 	if (split_ident_line(&s, msg, len) < 0)
 		goto skip;
 
-	name_start = s.name_begin;
-	name_end = s.name_end;
-	mail_start = s.mail_begin;
-	mail_end = s.mail_end;
+	strbuf_add(&person_name, s.name_begin, (int)(s.name_end - s.name_begin));
+	strbuf_add(&person_mail, s.mail_begin, (int)(s.mail_end - s.mail_begin));
 
 	if (part == 'N' || part == 'E') { /* mailmap lookup */
-		snprintf(person_name, sizeof(person_name), "%.*s",
-			 (int)(name_end - name_start), name_start);
-		snprintf(person_mail, sizeof(person_mail), "%.*s",
-			 (int)(mail_end - mail_start), mail_start);
-		mailmap_name(person_mail, sizeof(person_mail), person_name, sizeof(person_name));
-		name_start = person_name;
-		name_end = name_start + strlen(person_name);
-		mail_start = person_mail;
-		mail_end = mail_start +  strlen(person_mail);
+		mailmap_name(&person_mail, &person_name);
 	}
 	if (part == 'n' || part == 'N') {	/* name */
-		strbuf_add(sb, name_start, name_end-name_start);
+		strbuf_addbuf(sb, &person_name);
+
+		strbuf_release(&person_name);
+		strbuf_release(&person_mail);
+
 		return placeholder_len;
 	}
 	if (part == 'e' || part == 'E') {	/* email */
-		strbuf_add(sb, mail_start, mail_end-mail_start);
+		strbuf_addbuf(sb, &person_mail);
+
+		strbuf_release(&person_name);
+		strbuf_release(&person_mail);
+
 		return placeholder_len;
 	}
 
+	strbuf_release(&person_name);
+	strbuf_release(&person_mail);
+
 	if (!s.date_begin)
 		goto skip;
 
-- 
1.7.9.5

[PATCH v2 4/7] mailmap: Add mailmap structure to rev_info and pp

[Antoine Pelisse]

the mailmap string_list structure filled with mailmap
information is passed along from rev_info to pretty_print_context
to provide mailmap information to pretty print each commits
with the correct username and email.

Signed-off-by: Antoine Pelisse <apelisse@gmail.com>
---
 commit.h   |    1 +
 log-tree.c |    1 +
 revision.h |    1 +
 3 files changed, 3 insertions(+)

diff --git a/commit.h b/commit.h
index b6ad8f3..7f8f987 100644
--- a/commit.h
+++ b/commit.h
@@ -89,6 +89,7 @@ struct pretty_print_context {
 	char *notes_message;
 	struct reflog_walk_info *reflog_info;
 	const char *output_encoding;
+	struct string_list *mailmap;
 };
 
 struct userformat_want {
diff --git a/log-tree.c b/log-tree.c
index 4f86def..92254fd 100644
--- a/log-tree.c
+++ b/log-tree.c
@@ -671,6 +671,7 @@ void show_log(struct rev_info *opt)
 	ctx.preserve_subject = opt->preserve_subject;
 	ctx.reflog_info = opt->reflog_info;
 	ctx.fmt = opt->commit_format;
+	ctx.mailmap = opt->mailmap;
 	pretty_print_commit(&ctx, commit, &msgbuf);
 
 	if (opt->add_signoff)
diff --git a/revision.h b/revision.h
index 059bfff..83a79f5 100644
--- a/revision.h
+++ b/revision.h
@@ -143,6 +143,7 @@ struct rev_info {
 	const char	*subject_prefix;
 	int		no_inline;
 	int		show_log_size;
+	struct string_list *mailmap;
 
 	/* Filter by commit log message */
 	struct grep_opt	grep_filter;
-- 
1.7.9.5

[PATCH v2 5/7] pretty: Use mailmap to display username and email

[Antoine Pelisse]

Use the mailmap information to display the correct
username and email address in all log commands.

Signed-off-by: Antoine Pelisse <apelisse@gmail.com>
---
 pretty.c |   48 ++++++++++++++++++++++++++++--------------------
 1 file changed, 28 insertions(+), 20 deletions(-)

diff --git a/pretty.c b/pretty.c
index d05c675..fbf87b7 100644
--- a/pretty.c
+++ b/pretty.c
@@ -387,6 +387,8 @@ void pp_user_info(const struct pretty_print_context *pp,
 		  const char *what, struct strbuf *sb,
 		  const char *line, const char *encoding)
 {
+	struct strbuf name;
+	struct strbuf mail;
 	struct ident_split ident;
 	int linelen, namelen;
 	char *line_end, *date;
@@ -408,42 +410,48 @@ void pp_user_info(const struct pretty_print_context *pp,
 	if (split_ident_line(&ident, line, linelen))
 		return;
 
-	namelen = ident.mail_end - ident.name_begin + 1;
+	strbuf_init(&mail, 0);
+	strbuf_init(&name, 0);
+
+	strbuf_add(&mail, ident.mail_begin, ident.mail_end - ident.mail_begin);
+	strbuf_add(&name, ident.name_begin, ident.name_end - ident.name_begin);
+
+	if (pp->mailmap)
+		map_user(pp->mailmap, &mail, &name);
+
+	namelen = name.len + mail.len + 3; /* ' ' + '<' + '>' */
 	time = strtoul(ident.date_begin, &date, 10);
 	tz = strtol(date, NULL, 10);
 
 	if (pp->fmt == CMIT_FMT_EMAIL) {
-		int display_name_length;
-
-		display_name_length = ident.name_end - ident.name_begin;
-
 		strbuf_addstr(sb, "From: ");
-		if (needs_rfc2047_encoding(line, display_name_length, RFC2047_ADDRESS)) {
-			add_rfc2047(sb, line, display_name_length,
-						encoding, RFC2047_ADDRESS);
+		if (needs_rfc2047_encoding(name.buf, name.len, RFC2047_ADDRESS)) {
+			add_rfc2047(sb, name.buf, name.len,
+				    encoding, RFC2047_ADDRESS);
 			max_length = 76; /* per rfc2047 */
-		} else if (needs_rfc822_quoting(line, display_name_length)) {
+		} else if (needs_rfc822_quoting(name.buf, name.len)) {
 			struct strbuf quoted = STRBUF_INIT;
-			add_rfc822_quoted(&quoted, line, display_name_length);
+			add_rfc822_quoted(&quoted, name.buf, name.len);
 			strbuf_add_wrapped_bytes(sb, quoted.buf, quoted.len,
 							-6, 1, max_length);
 			strbuf_release(&quoted);
 		} else {
-			strbuf_add_wrapped_bytes(sb, line, display_name_length,
-							-6, 1, max_length);
+			strbuf_add_wrapped_bytes(sb, name.buf, name.len,
+						 -6, 1, max_length);
 		}
-		if (namelen - display_name_length + last_line_length(sb) > max_length) {
+		if (namelen - name.len + last_line_length(sb) > max_length)
 			strbuf_addch(sb, '\n');
-			if (!isspace(ident.name_end[0]))
-				strbuf_addch(sb, ' ');
-		}
-		strbuf_add(sb, ident.name_end, namelen - display_name_length);
-		strbuf_addch(sb, '\n');
+
+		strbuf_addf(sb, " <%s>\n", mail.buf);
 	} else {
-		strbuf_addf(sb, "%s: %.*s%.*s\n", what,
+		strbuf_addf(sb, "%s: %.*s%s <%s>\n", what,
 			      (pp->fmt == CMIT_FMT_FULLER) ? 4 : 0,
-			      "    ", namelen, line);
+			      "    ", name.buf, mail.buf);
 	}
+
+	strbuf_release(&mail);
+	strbuf_release(&name);
+
 	switch (pp->fmt) {
 	case CMIT_FMT_MEDIUM:
 		strbuf_addf(sb, "Date:   %s\n", show_date(time, tz, pp->date_mode));
-- 
1.7.9.5

[PATCH v2 6/7] log: Add --use-mailmap option

[Antoine Pelisse]

Add the --use-mailmap option to log commands. It allows
to display names from mailmap file when displaying logs,
whatever the format used.

Signed-off-by: Antoine Pelisse <apelisse@gmail.com>
---
 Documentation/git-log.txt |    5 +++++
 builtin/log.c             |    9 ++++++++-
 2 files changed, 13 insertions(+), 1 deletion(-)

diff --git a/Documentation/git-log.txt b/Documentation/git-log.txt
index 585dac4..a99be97 100644
--- a/Documentation/git-log.txt
+++ b/Documentation/git-log.txt
@@ -47,6 +47,11 @@ OPTIONS
 	Print out the ref name given on the command line by which each
 	commit was reached.
 
+--use-mailmap::
+	Use mailmap file to map author and committer names and email
+	to canonical real names and email addresses. See
+	linkgit:git-shortlog[1].
+
 --full-diff::
 	Without this flag, "git log -p <path>..." shows commits that
 	touch the specified paths, and diffs about the same specified
diff --git a/builtin/log.c b/builtin/log.c
index e7b7db1..d2bd8ce 100644
--- a/builtin/log.c
+++ b/builtin/log.c
@@ -22,6 +22,7 @@
 #include "branch.h"
 #include "streaming.h"
 #include "version.h"
+#include "mailmap.h"
 
 /* Set a default date-time format for git log ("log.date" config variable) */
 static const char *default_date_mode = NULL;
@@ -94,11 +95,12 @@ static void cmd_log_init_finish(int argc, const char **argv, const char *prefix,
 			 struct rev_info *rev, struct setup_revision_opt *opt)
 {
 	struct userformat_want w;
-	int quiet = 0, source = 0;
+	int quiet = 0, source = 0, mailmap = 0;
 
 	const struct option builtin_log_options[] = {
 		OPT_BOOLEAN(0, "quiet", &quiet, N_("suppress diff output")),
 		OPT_BOOLEAN(0, "source", &source, N_("show source")),
+		OPT_BOOLEAN(0, "use-mailmap", &mailmap, N_("Use mail map file")),
 		{ OPTION_CALLBACK, 0, "decorate", NULL, NULL, N_("decorate options"),
 		  PARSE_OPT_OPTARG, decorate_callback},
 		OPT_END()
@@ -136,6 +138,11 @@ static void cmd_log_init_finish(int argc, const char **argv, const char *prefix,
 	if (source)
 		rev->show_source = 1;
 
+	if (mailmap) {
+		rev->mailmap = xcalloc(1, sizeof(struct string_list));
+		read_mailmap(rev->mailmap, NULL);
+	}
+
 	if (rev->pretty_given && rev->commit_format == CMIT_FMT_RAW) {
 		/*
 		 * "log --pretty=raw" is special; ignore UI oriented
-- 
1.7.9.5

[PATCH v2 7/7] test: Add test for --use-mailmap option

[Antoine Pelisse]

The new option '--use-mailmap' can be used to make
sure that mailmap file is used to convert name
when running log commands.

The test is simple and checks that the Author line
is correctly replaced when running log.

Signed-off-by: Antoine Pelisse <apelisse@gmail.com>
---
 t/t4203-mailmap.sh |   14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/t/t4203-mailmap.sh b/t/t4203-mailmap.sh
index 1f182f6..db043dc 100755
--- a/t/t4203-mailmap.sh
+++ b/t/t4203-mailmap.sh
@@ -239,6 +239,20 @@ test_expect_success 'Log output (complex mapping)' '
 	test_cmp expect actual
 '
 
+cat >expect <<\EOF
+Author: CTO <cto@company.xx>
+Author: Santa Claus <santa.claus@northpole.xx>
+Author: Santa Claus <santa.claus@northpole.xx>
+Author: Other Author <other@author.xx>
+Author: Other Author <other@author.xx>
+Author: Some Dude <some@dude.xx>
+Author: A U Thor <author@example.com>
+EOF
+test_expect_success 'Log output with --use-mailmap' '
+	git log --use-mailmap | grep Author >actual &&
+	test_cmp expect actual
+'
+
 # git blame
 cat >expect <<\EOF
 ^OBJI (A U Thor     DATE 1) one
-- 
1.7.9.5

Re: [PATCH v2 0/7] Allow git log to use mailmap file

[Junio C Hamano]

Antoine Pelisse <apelisse@gmail.com> writes:

> Implement the feature suggested by Rich Mindwinter and Junio C Hamano
> (and following his advices)
>
>
>
> Allows git show/log commands to map author and committer
> names and emails using the mailmap file.
>
> Updates related to this second series:
>   - All tests are successful after each patch
>   - Use split_ident_line in shortlog.c
>   - Documentation has been added to git-log.txt
>   - Test has been added to check that we use the file
>   - Lots of improvements in the way strbufs are used
>   - New interface to map_user()
>   - Bunch of other fixes

The updated map_user() and its users look much nicer now.  Applying
your patches with "git am --whitespace=error" spots a few style
violations, though.

> git glog --committer/--author is still not looking for mailmap user
> names.

I think we should stop using the header grep mechanism for these and
instead keep two separate grep expressions in struct rev_info and
use them in revision.c::commit_match().  The unification of header
filter and message grep filter done in 2d10c55 (git log: Unify
header_filter and message_filter into one., 2006-09-20) may not have
been a good idea.

Thanks.