mailinfo: ignore in-body header that we do not care aboutjc/mailinfo

"git mailinfo" (hence "git am") understands some well-known headers,
like "Subject: ", "Date: " and "From: ", placed at the beginning of
the message body (and the "--scissors" can discard the part of the
body before a scissors-mark).  However, some people throw other
kinds of header-looking things there, expecting them to be
discarded.

Finding and discarding anything that looks like RFC2822 header is
not a right solution.  The body of the message may start with a line
that begins with a word followed by a colon that is a legitimate
part of the message that should not be discarded.

Instead, keep reading non-blank lines once we see an in-body header
at the beginning and discard them.  Nobody will be insane enough to
reorder the headers to read like this:

    Garbage-non-in-body-header: here
    Subject: in-body subject

    Here is the body of the commit log.

but it is common for lazy or misguided people to leave non-header
materials in-body like this:

    From: Junio C Hamano <gitster@pobox.com>
    Date: Mon, 28 Sep 2015 19:19:27 -0700
    Subject: [PATCH] Git 2.6.1
    MIME-Version: 1.0

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

6 files changed, 90 insertions, 32 deletions

diff --git a/mailinfo.c b/mailinfo.c
index e157ca6eb5..74d19aec0a 100644
--- a/mailinfo.c
+++ b/mailinfo.c
@@ -628,17 +628,71 @@ static int is_scissors_line(const struct strbuf *line)
 		gap * 2 < perforation);
 }
 
+static int is_rfc2822_header(const struct strbuf *line)
+{
+	/*
+	 * The section that defines the loosest possible
+	 * field name is "3.6.8 Optional fields".
+	 *
+	 * optional-field = field-name ":" unstructured CRLF
+	 * field-name = 1*ftext
+	 * ftext = %d33-57 / %59-126
+	 */
+	int ch;
+	char *cp = line->buf;
+
+	/* Count mbox From headers as headers */
+	if (starts_with(cp, "From ") || starts_with(cp, ">From "))
+		return 1;
+
+	while ((ch = *cp++)) {
+		if (ch == ':')
+			return 1;
+		if ((33 <= ch && ch <= 57) ||
+		    (59 <= ch && ch <= 126))
+			continue;
+		break;
+	}
+	return 0;
+}
+
 static int handle_commit_msg(struct mailinfo *mi, struct strbuf *line)
 {
+	int is_empty_line;
+
 	assert(!mi->filter_stage);
 
-	if (mi->header_stage) {
-		if (!line->len || (line->len == 1 && line->buf[0] == '\n'))
+	is_empty_line = (!line->len || (line->len == 1 && line->buf[0] == '\n'));
+	if (mi->header_stage == 1) {
+		/*
+		 * Haven't seen a known in-body header; discard an empty line.
+		 */
+		if (is_empty_line)
 			return 0;
 	}
 
 	if (mi->use_inbody_headers && mi->header_stage) {
-		mi->header_stage = check_header(mi, line, mi->s_hdr_data, 0);
+		int is_known_header = check_header(mi, line, mi->s_hdr_data, 0);
+
+		if (mi->header_stage == 2) {
+			/*
+			 * an empty line after the in-body header block,
+			 * or a line obviously not an attempt to invent
+			 * an unsupported in-body header.
+			 */
+			if (is_empty_line || !is_rfc2822_header(line))
+				mi->header_stage = 0;
+			if (is_empty_line)
+				return 0;
+			/* otherwise do not discard the line, but keep going */
+		} else if (is_known_header) {
+			/* We know we are in the in-body header block now. */
+			mi->header_stage = 2;
+		} else if (mi->header_stage != 2) {
+			/* garbage and we are not in the in-body header block */
+			mi->header_stage = 0;
+		}
+
 		if (mi->header_stage)
 			return 0;
 	} else
@@ -699,34 +753,6 @@ static void handle_filter(struct mailinfo *mi, struct strbuf *line)
 	}
 }
 
-static int is_rfc2822_header(const struct strbuf *line)
-{
-	/*
-	 * The section that defines the loosest possible
-	 * field name is "3.6.8 Optional fields".
-	 *
-	 * optional-field = field-name ":" unstructured CRLF
-	 * field-name = 1*ftext
-	 * ftext = %d33-57 / %59-126
-	 */
-	int ch;
-	char *cp = line->buf;
-
-	/* Count mbox From headers as headers */
-	if (starts_with(cp, "From ") || starts_with(cp, ">From "))
-		return 1;
-
-	while ((ch = *cp++)) {
-		if (ch == ':')
-			return 1;
-		if ((33 <= ch && ch <= 57) ||
-		    (59 <= ch && ch <= 126))
-			continue;
-		break;
-	}
-	return 0;
-}
-
 static int read_one_header_line(struct strbuf *line, FILE *in)
 {
 	struct strbuf continuation = STRBUF_INIT;
diff --git a/t/t5100-mailinfo.sh b/t/t5100-mailinfo.sh
index e97cfb2ab8..3ce041bf65 100755
--- a/t/t5100-mailinfo.sh
+++ b/t/t5100-mailinfo.sh
@@ -11,7 +11,8 @@ test_expect_success 'split sample box' \
 	'git mailsplit -o. "$TEST_DIRECTORY"/t5100/sample.mbox >last &&
 	last=`cat last` &&
 	echo total is $last &&
-	test `cat last` = 17'
+	test `cat last` = 18
+'
 
 check_mailinfo () {
 	mail=$1 opt=$2
diff --git a/t/t5100/info0018 b/t/t5100/info0018
new file mode 100644
index 0000000000..ec671fc2ad
--- /dev/null
+++ b/t/t5100/info0018
@@ -0,0 +1,5 @@
+Author: A U Thor
+Email: a.u.thor@example.com
+Subject: A E I O U
+Date: Mon, 17 Sep 2012 14:23:49 -0700
+
diff --git a/t/t5100/msg0018 b/t/t5100/msg0018
new file mode 100644
index 0000000000..2ee0900850
--- /dev/null
+++ b/t/t5100/msg0018
@@ -0,0 +1,2 @@
+New content here
+
diff --git a/t/t5100/patch0018 b/t/t5100/patch0018
new file mode 100644
index 0000000000..35cf84c9a1
--- /dev/null
+++ b/t/t5100/patch0018
@@ -0,0 +1,6 @@
+diff --git a/foo b/foo
+index e69de29..d95f3ad 100644
+--- a/foo
++++ b/foo
+@@ -0,0 +1 @@
++New content
diff --git a/t/t5100/sample.mbox b/t/t5100/sample.mbox
index 8b2ae064c3..d7c58783d2 100644
--- a/t/t5100/sample.mbox
+++ b/t/t5100/sample.mbox
@@ -406,6 +406,7 @@ Subject: re: [PATCH] another patch
 
 From: A U Thor <a.u.thor@example.com>
 Subject: [PATCH] another patch
+
 >Here is an empty patch from A U Thor.
 
 Hey you forgot the patch!
@@ -699,3 +700,20 @@ index e69de29..d95f3ad 100644
 +++ b/foo
 @@ -0,0 +1 @@
 +New content
+From nobody Mon Sep 17 00:00:00 2001
+From: A U Thor <a.u.thor@example.com>
+Subject: Re: some discussion title
+Date: Mon, 17 Sep 2012 14:23:49 -0700
+
+Subject: A E I O U
+MIME-VERSION: 1.0
+Garbage: Not a valid in-body header
+
+New content here
+
+diff --git a/foo b/foo
+index e69de29..d95f3ad 100644
+--- a/foo
++++ b/foo
+@@ -0,0 +1 @@
++New content