Merge branch 'dd/sequencer-utf8'

Handling of commit objects that use non UTF-8 encoding during
"rebase -i" has been improved.

* dd/sequencer-utf8:
  sequencer: reencode commit message for am/rebase --show-current-patch
  sequencer: reencode old merge-commit message
  sequencer: reencode squashing commit's message
  sequencer: reencode revert/cherry-pick's todo list
  sequencer: reencode to utf-8 before arrange rebase's todo list
  t3900: demonstrate git-rebase problem with multi encoding
  configure.ac: define ICONV_OMITS_BOM if necessary
  t0028: eliminate non-standard usage of printf

configure.ac

@@ -844,12 +844,61 @@ AC_MSG_CHECKING([for old iconv()])
 AC_COMPILE_IFELSE([OLDICONVTEST_SRC],
 	[AC_MSG_RESULT([no])],
 	[AC_MSG_RESULT([yes])
+	AC_DEFINE(HAVE_OLD_ICONV, 1)
 	OLD_ICONV=UnfortunatelyYes])
 
 GIT_UNSTASH_FLAGS($ICONVDIR)
 
 GIT_CONF_SUBST([OLD_ICONV])
 
+#
+# Define ICONV_OMITS_BOM if you are on a system which
+# iconv omits bom for utf-{16,32}
+if test -z "$NO_ICONV"; then
+AC_CACHE_CHECK([whether iconv omits bom for utf-16 and utf-32],
+ [ac_cv_iconv_omits_bom],
+[
+old_LIBS="$LIBS"
+if test -n "$NEEDS_LIBICONV"; then
+	LIBS="$LIBS -liconv"
+fi
+
+AC_RUN_IFELSE(
+	[AC_LANG_PROGRAM([AC_INCLUDES_DEFAULT
+	#include <iconv.h>
+	#ifdef HAVE_OLD_ICONV
+	typedef const char *iconv_ibp;
+	#else
+	typedef char *iconv_ibp;
+	#endif
+	],
+	[[
+	int v;
+	iconv_t conv;
+	char in[] = "a"; iconv_ibp pin = in;
+	char out[20] = ""; char *pout = out;
+	size_t isz = sizeof in;
+	size_t osz = sizeof out;
+
+	conv = iconv_open("UTF-16", "UTF-8");
+	iconv(conv, &pin, &isz, &pout, &osz);
+	iconv_close(conv);
+	v = (unsigned char)(out[0]) + (unsigned char)(out[1]);
+	return v != 0xfe + 0xff;
+	]])],
+	[ac_cv_iconv_omits_bom=no],
+	[ac_cv_iconv_omits_bom=yes])
+
+LIBS="$old_LIBS"
+])
+if test "x$ac_cv_iconv_omits_bom" = xyes; then
+	ICONV_OMITS_BOM=Yes
+else
+	ICONV_OMITS_BOM=
+fi
+GIT_CONF_SUBST([ICONV_OMITS_BOM])
+fi
+
 ## Checks for typedefs, structures, and compiler characteristics.
 AC_MSG_NOTICE([CHECKS for typedefs, structures, and compiler characteristics])
 #

sequencer.c

@@ -1574,6 +1574,7 @@ static int update_squash_messages(struct repository *r,
 	struct strbuf buf = STRBUF_INIT;
 	int res;
 	const char *message, *body;
+	const char *encoding = get_commit_output_encoding();
 
 	if (opts->current_fixup_count > 0) {
 		struct strbuf header = STRBUF_INIT;
@@ -1600,7 +1601,7 @@ static int update_squash_messages(struct repository *r,
 			return error(_("need a HEAD to fixup"));
 		if (!(head_commit = lookup_commit_reference(r, &head)))
 			return error(_("could not read HEAD"));
-		if (!(head_message = get_commit_buffer(head_commit, NULL)))
+		if (!(head_message = logmsg_reencode(head_commit, NULL, encoding)))
 			return error(_("could not read HEAD's commit message"));
 
 		find_commit_subject(head_message, &body);
@@ -1621,7 +1622,7 @@ static int update_squash_messages(struct repository *r,
 		unuse_commit_buffer(head_commit, head_message);
 	}
 
-	if (!(message = get_commit_buffer(commit, NULL)))
+	if (!(message = logmsg_reencode(commit, NULL, encoding)))
 		return error(_("could not read commit message of %s"),
 			     oid_to_hex(&commit->object.oid));
 	find_commit_subject(message, &body);
@@ -2562,14 +2563,17 @@ static int walk_revs_populate_todo(struct todo_list *todo_list,
 	enum todo_command command = opts->action == REPLAY_PICK ?
 		TODO_PICK : TODO_REVERT;
 	const char *command_string = todo_command_info[command].str;
+	const char *encoding;
 	struct commit *commit;
 
 	if (prepare_revs(opts))
 		return -1;
 
+	encoding = get_log_output_encoding();
+
 	while ((commit = get_revision(opts->revs))) {
 		struct todo_item *item = append_new_todo(todo_list);
-		const char *commit_buffer = get_commit_buffer(commit, NULL);
+		const char *commit_buffer = logmsg_reencode(commit, NULL, encoding);
 		const char *subject;
 		int subject_len;
 
@@ -2966,7 +2970,8 @@ static int make_patch(struct repository *r,
 
 	strbuf_addf(&buf, "%s/message", get_dir(opts));
 	if (!file_exists(buf.buf)) {
-		const char *commit_buffer = get_commit_buffer(commit, NULL);
+		const char *encoding = get_commit_output_encoding();
+		const char *commit_buffer = logmsg_reencode(commit, NULL, encoding);
 		find_commit_subject(commit_buffer, &subject);
 		res |= write_message(subject, strlen(subject), buf.buf, 1);
 		unuse_commit_buffer(commit, commit_buffer);
@@ -3368,7 +3373,8 @@ static int do_merge(struct repository *r,
 	}
 
 	if (commit) {
-		const char *message = get_commit_buffer(commit, NULL);
+		const char *encoding = get_commit_output_encoding();
+		const char *message = logmsg_reencode(commit, NULL, encoding);
 		const char *body;
 		int len;
 
@@ -4149,9 +4155,10 @@ static int commit_staged_changes(struct repository *r,
 				 */
 				struct commit *commit;
 				const char *path = rebase_path_squash_msg();
+				const char *encoding = get_commit_output_encoding();
 
 				if (parse_head(r, &commit) ||
-				    !(p = get_commit_buffer(commit, NULL)) ||
+				    !(p = logmsg_reencode(commit, NULL, encoding)) ||
 				    write_message(p, strlen(p), path, 0)) {
 					unuse_commit_buffer(commit, p);
 					return error(_("could not write file: "
@@ -5167,7 +5174,7 @@ int todo_list_rearrange_squash(struct todo_list *todo_list)
 		*commit_todo_item_at(&commit_todo, item->commit) = item;
 
 		parse_commit(item->commit);
-		commit_buffer = get_commit_buffer(item->commit, NULL);
+		commit_buffer = logmsg_reencode(item->commit, NULL, "UTF-8");
 		find_commit_subject(commit_buffer, &subject);
 		format_subject(&buf, subject, " ");
 		subject = subjects[i] = strbuf_detach(&buf, &subject_len);

t/t0028-working-tree-encoding.sh

@@ -17,7 +17,7 @@ test_lazy_prereq NO_UTF32_BOM '
 write_utf16 () {
 	if test_have_prereq NO_UTF16_BOM
 	then
-		printf '\xfe\xff'
+		printf '\376\377'
 	fi &&
 	iconv -f UTF-8 -t UTF-16
 }
@@ -25,7 +25,7 @@ write_utf16 () {
 write_utf32 () {
 	if test_have_prereq NO_UTF32_BOM
 	then
-		printf '\x00\x00\xfe\xff'
+		printf '\0\0\376\377'
 	fi &&
 	iconv -f UTF-8 -t UTF-32
 }

t/t3434-rebase-i18n.sh

@@ -0,0 +1,84 @@
+#!/bin/sh
+#
+# Copyright (c) 2019 Doan Tran Cong Danh
+#
+
+test_description='rebase with changing encoding
+
+Initial setup:
+
+1 - 2              master
+ \
+  3 - 4            first
+   \
+    5 - 6          second
+'
+
+. ./test-lib.sh
+
+compare_msg () {
+	iconv -f "$2" -t "$3" "$TEST_DIRECTORY/t3434/$1" >expect &&
+	git cat-file commit HEAD >raw &&
+	sed "1,/^$/d" raw >actual &&
+	test_cmp expect actual
+}
+
+test_expect_success setup '
+	test_commit one &&
+	git branch first &&
+	test_commit two &&
+	git switch first &&
+	test_commit three &&
+	git branch second &&
+	test_commit four &&
+	git switch second &&
+	test_commit five &&
+	test_commit six
+'
+
+test_expect_success 'rebase --rebase-merges update encoding eucJP to UTF-8' '
+	git switch -c merge-eucJP-UTF-8 first &&
+	git config i18n.commitencoding eucJP &&
+	git merge -F "$TEST_DIRECTORY/t3434/eucJP.txt" second &&
+	git config i18n.commitencoding UTF-8 &&
+	git rebase --rebase-merges master &&
+	compare_msg eucJP.txt eucJP UTF-8
+'
+
+test_expect_failure 'rebase --rebase-merges update encoding eucJP to ISO-2022-JP' '
+	git switch -c merge-eucJP-ISO-2022-JP first &&
+	git config i18n.commitencoding eucJP &&
+	git merge -F "$TEST_DIRECTORY/t3434/eucJP.txt" second &&
+	git config i18n.commitencoding ISO-2022-JP &&
+	git rebase --rebase-merges master &&
+	compare_msg eucJP.txt eucJP ISO-2022-JP
+'
+
+test_rebase_continue_update_encode () {
+	old=$1
+	new=$2
+	msgfile=$3
+	test_expect_success "rebase --continue update from $old to $new" '
+		(git rebase --abort || : abort current git-rebase failure) &&
+		git switch -c conflict-$old-$new one &&
+		echo for-conflict >two.t &&
+		git add two.t &&
+		git config i18n.commitencoding $old &&
+		git commit -F "$TEST_DIRECTORY/t3434/$msgfile" &&
+		git config i18n.commitencoding $new &&
+		test_must_fail git rebase -m master &&
+		test -f .git/rebase-merge/message &&
+		git stripspace <.git/rebase-merge/message >two.t &&
+		git add two.t &&
+		git rebase --continue &&
+		compare_msg $msgfile $old $new &&
+		: git-commit assume invalid utf-8 is latin1 &&
+		test_cmp expect two.t
+	'
+}
+
+test_rebase_continue_update_encode ISO-8859-1 UTF-8 ISO8859-1.txt
+test_rebase_continue_update_encode eucJP UTF-8 eucJP.txt
+test_rebase_continue_update_encode eucJP ISO-2022-JP eucJP.txt
+
+test_done

t/t3434/ISO8859-1.txt

@@ -0,0 +1,3 @@
+ﾄﾋﾑﾏﾖ
+
+ﾁb軼鑁g

t/t3434/eucJP.txt

@@ -0,0 +1,4 @@
+はれひほふ
+
+しているのが、いるので。
+濱浜ほれぷりぽれまびぐりろへ。

t/t3900-i18n-commit.sh

@@ -204,4 +204,41 @@ test_commit_autosquash_flags eucJP fixup
 
 test_commit_autosquash_flags ISO-2022-JP squash
 
+test_commit_autosquash_multi_encoding () {
+	flag=$1
+	old=$2
+	new=$3
+	msg=$4
+	test_expect_success "commit --$flag into $old from $new" '
+		git checkout -b $flag-$old-$new C0 &&
+		git config i18n.commitencoding $old &&
+		echo $old >>F &&
+		git commit -a -F "$TEST_DIRECTORY"/t3900/$msg &&
+		test_tick &&
+		echo intermediate stuff >>G &&
+		git add G &&
+		git commit -a -m "intermediate commit" &&
+		test_tick &&
+		git config i18n.commitencoding $new &&
+		echo $new-$flag >>F &&
+		git commit -a --$flag HEAD^ &&
+		git rebase --autosquash -i HEAD^^^ &&
+		git rev-list HEAD >actual &&
+		test_line_count = 3 actual &&
+		iconv -f $old -t UTF-8 "$TEST_DIRECTORY"/t3900/$msg >expect &&
+		if test $flag = squash; then
+			subject="$(head -1 expect)" &&
+			printf "\nsquash! %s\n" "$subject" >>expect
+		fi &&
+		git cat-file commit HEAD^ >raw &&
+		(sed "1,/^$/d" raw | iconv -f $new -t utf-8) >actual &&
+		test_cmp expect actual
+	'
+}
+
+test_commit_autosquash_multi_encoding fixup UTF-8 ISO-8859-1 1-UTF-8.txt
+test_commit_autosquash_multi_encoding squash ISO-8859-1 UTF-8 ISO8859-1.txt
+test_commit_autosquash_multi_encoding squash eucJP ISO-2022-JP eucJP.txt
+test_commit_autosquash_multi_encoding fixup ISO-2022-JP UTF-8 ISO-2022-JP.txt
+
 test_done