Merge branch 'en/fast-export-encoding'

The "git fast-export/import" pair has been taught to handle commits
with log messages in encoding other than UTF-8 better.

* en/fast-export-encoding:
  fast-export: do automatic reencoding of commit messages only if requested
  fast-export: differentiate between explicitly UTF-8 and implicitly UTF-8
  fast-export: avoid stripping encoding header if we cannot reencode
  fast-import: support 'encoding' commit header
  t9350: fix encoding test to actually test reencoding

1 files changed, 49 insertions, 6 deletions

diff --git a/builtin/fast-export.c b/builtin/fast-export.c
index 9e28348..c22cef3 100644
--- a/builtin/fast-export.c
+++ b/builtin/fast-export.c
@@ -33,6 +33,7 @@ static const char *fast_export_usage[] = {
 static int progress;
 static enum { SIGNED_TAG_ABORT, VERBATIM, WARN, WARN_STRIP, STRIP } signed_tag_mode = SIGNED_TAG_ABORT;
 static enum { TAG_FILTERING_ABORT, DROP, REWRITE } tag_of_filtered_mode = TAG_FILTERING_ABORT;
+static enum { REENCODE_ABORT, REENCODE_YES, REENCODE_NO } reencode_mode = REENCODE_ABORT;
 static int fake_missing_tagger;
 static int use_done_feature;
 static int no_data;
@@ -77,6 +78,31 @@ static int parse_opt_tag_of_filtered_mode(const struct option *opt,
 	return 0;
 }
 
+static int parse_opt_reencode_mode(const struct option *opt,
+				   const char *arg, int unset)
+{
+	if (unset) {
+		reencode_mode = REENCODE_ABORT;
+		return 0;
+	}
+
+	switch (git_parse_maybe_bool(arg)) {
+	case 0:
+		reencode_mode = REENCODE_NO;
+		break;
+	case 1:
+		reencode_mode = REENCODE_YES;
+		break;
+	default:
+		if (!strcasecmp(arg, "abort"))
+			reencode_mode = REENCODE_ABORT;
+		else
+			return error("Unknown reencoding mode: %s", arg);
+	}
+
+	return 0;
+}
+
 static struct decoration idnums;
 static uint32_t last_idnum;
 
@@ -453,7 +479,7 @@ static const char *find_encoding(const char *begin, const char *end)
 	bol = memmem(begin, end ? end - begin : strlen(begin),
 		     needle, strlen(needle));
 	if (!bol)
-		return git_commit_encoding;
+		return NULL;
 	bol += strlen(needle);
 	eol = strchrnul(bol, '\n');
 	*eol = '\0';
@@ -633,18 +659,32 @@ static void handle_commit(struct commit *commit, struct rev_info *rev,
 	}
 
 	mark_next_object(&commit->object);
-	if (anonymize)
+	if (anonymize) {
 		reencoded = anonymize_commit_message(message);
-	else if (!is_encoding_utf8(encoding))
-		reencoded = reencode_string(message, "UTF-8", encoding);
+	} else if (encoding) {
+		switch(reencode_mode) {
+		case REENCODE_YES:
+			reencoded = reencode_string(message, "UTF-8", encoding);
+			break;
+		case REENCODE_NO:
+			break;
+		case REENCODE_ABORT:
+			die("Encountered commit-specific encoding %s in commit "
+			    "%s; use --reencode=[yes|no] to handle it",
+			    encoding, oid_to_hex(&commit->object.oid));
+		}
+	}
 	if (!commit->parents)
 		printf("reset %s\n", refname);
 	printf("commit %s\nmark :%"PRIu32"\n", refname, last_idnum);
 	if (show_original_ids)
 		printf("original-oid %s\n", oid_to_hex(&commit->object.oid));
-	printf("%.*s\n%.*s\ndata %u\n%s",
+	printf("%.*s\n%.*s\n",
 	       (int)(author_end - author), author,
-	       (int)(committer_end - committer), committer,
+	       (int)(committer_end - committer), committer);
+	if (!reencoded && encoding)
+		printf("encoding %s\n", encoding);
+	printf("data %u\n%s",
 	       (unsigned)(reencoded
 			  ? strlen(reencoded) : message
 			  ? strlen(message) : 0),
@@ -1088,6 +1128,9 @@ int cmd_fast_export(int argc, const char **argv, const char *prefix)
 		OPT_CALLBACK(0, "tag-of-filtered-object", &tag_of_filtered_mode, N_("mode"),
 			     N_("select handling of tags that tag filtered objects"),
 			     parse_opt_tag_of_filtered_mode),
+		OPT_CALLBACK(0, "reencode", &reencode_mode, N_("mode"),
+			     N_("select handling of commit messages in an alternate encoding"),
+			     parse_opt_reencode_mode),
 		OPT_STRING(0, "export-marks", &export_filename, N_("file"),
 			     N_("Dump marks to this file")),
 		OPT_STRING(0, "import-marks", &import_filename, N_("file"),