fast-import: support 'encoding' commit header

Since git supports commit messages with an encoding other than UTF-8,
allow fast-import to import such commits.  This may be useful for folks
who do not want to reencode commit messages from an external system, and
may also be useful to achieve reversible history rewrites (e.g. sha1sum
<-> sha256sum transitions or subtree work) with git repositories that
have used specialized encodings in their commit history.

Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>

Documentation/git-fast-import.txt

@@ -388,6 +388,7 @@ change to the project.
 	original-oid?
 	('author' (SP <name>)? SP LT <email> GT SP <when> LF)?
 	'committer' (SP <name>)? SP LT <email> GT SP <when> LF
+	('encoding' SP <encoding>)?
 	data
 	('from' SP <commit-ish> LF)?
 	('merge' SP <commit-ish> LF)?
@@ -455,6 +456,12 @@ that was selected by the --date-format=<fmt> command-line option.
 See ``Date Formats'' above for the set of supported formats, and
 their syntax.
 
+`encoding`
+^^^^^^^^^^
+The optional `encoding` command indicates the encoding of the commit
+message.  Most commits are UTF-8 and the encoding is omitted, but this
+allows importing commit messages into git without first reencoding them.
+
 `from`
 ^^^^^^
 The `from` command is used to specify the commit to initialize

fast-import.c

@@ -2585,6 +2585,7 @@ static void parse_new_commit(const char *arg)
 	struct branch *b;
 	char *author = NULL;
 	char *committer = NULL;
+	const char *encoding = NULL;
 	struct hash_list *merge_list = NULL;
 	unsigned int merge_count;
 	unsigned char prev_fanout, new_fanout;
@@ -2607,6 +2608,8 @@ static void parse_new_commit(const char *arg)
 	}
 	if (!committer)
 		die("Expected committer but didn't get one");
+	if (skip_prefix(command_buf.buf, "encoding ", &encoding))
+		read_next_command();
 	parse_data(&msg, 0, NULL);
 	read_next_command();
 	parse_from(b);
@@ -2670,9 +2673,13 @@ static void parse_new_commit(const char *arg)
 	}
 	strbuf_addf(&new_data,
 		"author %s\n"
-		"committer %s\n"
-		"\n",
+		"committer %s\n",
 		author ? author : committer, committer);
+	if (encoding)
+		strbuf_addf(&new_data,
+			"encoding %s\n",
+			encoding);
+	strbuf_addch(&new_data, '\n');
 	strbuf_addbuf(&new_data, &msg);
 	free(author);
 	free(committer);

t/t9300-fast-import.sh

@@ -3299,4 +3299,24 @@ test_expect_success !MINGW 'W: get-mark & empty orphan commit with erroneous thi
 	sed -e s/LFs/LLL/ W-input | tr L "\n" | test_must_fail git fast-import
 '
 
+###
+### series X (other new features)
+###
+
+test_expect_success 'X: handling encoding' '
+	test_tick &&
+	cat >input <<-INPUT_END &&
+	commit refs/heads/encoding
+	committer $GIT_COMMITTER_NAME <$GIT_COMMITTER_EMAIL> $GIT_COMMITTER_DATE
+	encoding iso-8859-7
+	data <<COMMIT
+	INPUT_END
+
+	printf "Pi: \360\nCOMMIT\n" >>input &&
+
+	git fast-import <input &&
+	git cat-file -p encoding | grep $(printf "\360") &&
+	git log -1 --format=%B encoding | grep $(printf "\317\200")
+'
+
 test_done