Merge branch 'jc/utf8'

* jc/utf8:
  t3900: test conversion to non UTF-8 as well
  Rename t3900 test vector file
  UTF-8: introduce i18n.logoutputencoding.
  Teach log family --encoding
  i18n.logToUTF8: convert commit log message to UTF-8
  Move encoding conversion routine out of mailinfo to utf8.c

Conflicts:

	commit.c

Documentation/config.txt

@@ -267,6 +267,10 @@ i18n.commitEncoding::
 	browser (and possibly at other places in the future or in other
 	porcelains). See e.g. gitlink:git-mailinfo[1]. Defaults to 'utf-8'.
 
+i18n.logOutputEncoding::
+	Character encoding the commit messages are converted to when
+	running `git-log` and friends.
+
 log.showroot::
 	If true, the initial commit will be shown as a big creation event.
 	This is equivalent to a diff against an empty tree.

builtin-commit-tree.c

@@ -92,6 +92,7 @@ int cmd_commit_tree(int argc, const char **argv, const char *prefix)
 	char comment[1000];
 	char *buffer;
 	unsigned int size;
+	int encoding_is_utf8;
 
 	setup_ident();
 	git_config(git_default_config);
@@ -117,6 +118,10 @@ int cmd_commit_tree(int argc, const char **argv, const char *prefix)
 			parents++;
 	}
 
+	/* Not having i18n.commitencoding is the same as having utf-8 */
+	encoding_is_utf8 = (!git_commit_encoding ||
+			    !strcmp(git_commit_encoding, "utf-8"));
+
 	init_buffer(&buffer, &size);
 	add_buffer(&buffer, &size, "tree %s\n", sha1_to_hex(tree_sha1));
 
@@ -130,7 +135,11 @@ int cmd_commit_tree(int argc, const char **argv, const char *prefix)
 
 	/* Person/date information */
 	add_buffer(&buffer, &size, "author %s\n", git_author_info(1));
-	add_buffer(&buffer, &size, "committer %s\n\n", git_committer_info(1));
+	add_buffer(&buffer, &size, "committer %s\n", git_committer_info(1));
+	if (!encoding_is_utf8)
+		add_buffer(&buffer, &size,
+				"encoding %s\n", git_commit_encoding);
+	add_buffer(&buffer, &size, "\n");
 
 	/* And add the comment */
 	while (fgets(comment, sizeof(comment), stdin) != NULL)
@@ -138,7 +147,7 @@ int cmd_commit_tree(int argc, const char **argv, const char *prefix)
 
 	/* And check the encoding */
 	buffer[size] = '\0';
-	if (!strcmp(git_commit_encoding, "utf-8") && !is_utf8(buffer))
+	if (encoding_is_utf8 && !is_utf8(buffer))
 		fprintf(stderr, commit_utf8_warn);
 
 	if (!write_sha1_file(buffer, size, commit_type, commit_sha1)) {

builtin-log.c

@@ -20,6 +20,8 @@ void add_head(struct rev_info *revs);
 static void cmd_log_init(int argc, const char **argv, const char *prefix,
 		      struct rev_info *rev)
 {
+	int i;
+
 	rev->abbrev = DEFAULT_ABBREV;
 	rev->commit_format = CMIT_FMT_DEFAULT;
 	rev->verbose_header = 1;
@@ -27,8 +29,18 @@ static void cmd_log_init(int argc, const char **argv, const char *prefix,
 	argc = setup_revisions(argc, argv, rev, "HEAD");
 	if (rev->diffopt.pickaxe || rev->diffopt.filter)
 		rev->always_show_header = 0;
-	if (argc > 1)
-		die("unrecognized argument: %s", argv[1]);
+	for (i = 1; i < argc; i++) {
+		const char *arg = argv[i];
+		if (!strncmp(arg, "--encoding=", 11)) {
+			arg += 11;
+			if (strcmp(arg, "none"))
+				git_log_output_encoding = strdup(arg);
+			else
+				git_log_output_encoding = "";
+		}
+		else
+			die("unrecognized argument: %s", arg);
+	}
 }
 
 static int cmd_log_walk(struct rev_info *rev)

builtin-mailinfo.c

@@ -4,6 +4,7 @@
  */
 #include "cache.h"
 #include "builtin.h"
+#include "utf8.h"
 
 static FILE *cmitmsg, *patchfile, *fin, *fout;
 
@@ -510,40 +511,18 @@ static int decode_b_segment(char *in, char *ot, char *ep)
 
 static void convert_to_utf8(char *line, char *charset)
 {
-#ifndef NO_ICONV
-	char *in, *out;
-	size_t insize, outsize, nrc;
-	char outbuf[4096]; /* cheat */
 	static char latin_one[] = "latin1";
 	char *input_charset = *charset ? charset : latin_one;
-	iconv_t conv = iconv_open(metainfo_charset, input_charset);
+	char *out = reencode_string(line, metainfo_charset, input_charset);
 
-	if (conv == (iconv_t) -1) {
-		static int warned_latin1_once = 0;
-		if (input_charset != latin_one) {
-			fprintf(stderr, "cannot convert from %s to %s\n",
-				input_charset, metainfo_charset);
-			*charset = 0;
-		}
-		else if (!warned_latin1_once) {
-			warned_latin1_once = 1;
-			fprintf(stderr, "tried to convert from %s to %s, "
-				"but your iconv does not work with it.\n",
-				input_charset, metainfo_charset);
-		}
+	if (!out) {
+		fprintf(stderr, "cannot convert from %s to %s\n",
+			input_charset, metainfo_charset);
+		*charset = 0;
 		return;
 	}
-	in = line;
-	insize = strlen(in);
-	out = outbuf;
-	outsize = sizeof(outbuf);
-	nrc = iconv(conv, &in, &insize, &out, &outsize);
-	iconv_close(conv);
-	if (nrc == (size_t) -1)
-		return;
-	*out = 0;
-	strcpy(line, outbuf);
-#endif
+	strcpy(line, out);
+	free(out);
 }
 
 static int decode_header_bq(char *it)
@@ -827,7 +806,8 @@ int cmd_mailinfo(int argc, const char **argv, const char *prefix)
 		if (!strcmp(argv[1], "-k"))
 			keep_subject = 1;
 		else if (!strcmp(argv[1], "-u"))
-			metainfo_charset = git_commit_encoding;
+			metainfo_charset = (git_commit_encoding
+					    ? git_commit_encoding : "utf-8");
 		else if (!strncmp(argv[1], "--encoding=", 11))
 			metainfo_charset = argv[1] + 11;
 		else

cache.h

@@ -416,8 +416,8 @@ extern int check_repository_format_version(const char *var, const char *value);
 extern char git_default_email[MAX_GITNAME];
 extern char git_default_name[MAX_GITNAME];
 
-#define MAX_ENCODING_LENGTH 64
-extern char git_commit_encoding[MAX_ENCODING_LENGTH];
+extern char *git_commit_encoding;
+extern char *git_log_output_encoding;
 
 extern int copy_fd(int ifd, int ofd);
 extern void write_or_die(int fd, const void *buf, size_t count);

commit.c

@@ -2,6 +2,7 @@
 #include "tag.h"
 #include "commit.h"
 #include "pkt-line.h"
+#include "utf8.h"
 
 int save_commit_buffer = 1;
 
@@ -597,10 +598,61 @@ static int add_merge_info(enum cmit_fmt fmt, char *buf, const struct commit *com
 	return offset;
 }
 
-unsigned long pretty_print_commit(enum cmit_fmt fmt, const struct commit *commit,
-				  unsigned long len, char *buf, unsigned long space,
+static char *get_header(const struct commit *commit, const char *key)
+{
+	int key_len = strlen(key);
+	const char *line = commit->buffer;
+
+	for (;;) {
+		const char *eol = strchr(line, '\n'), *next;
+
+		if (line == eol)
+			return NULL;
+		if (!eol) {
+			eol = line + strlen(line);
+			next = NULL;
+		} else
+			next = eol + 1;
+		if (!strncmp(line, key, key_len) && line[key_len] == ' ') {
+			int len = eol - line - key_len;
+			char *ret = xmalloc(len);
+			memcpy(ret, line + key_len + 1, len - 1);
+			ret[len - 1] = '\0';
+			return ret;
+		}
+		line = next;
+	}
+}
+
+static char *logmsg_reencode(const struct commit *commit)
+{
+	char *encoding;
+	char *out;
+	char *output_encoding = (git_log_output_encoding
+				 ? git_log_output_encoding
+				 : git_commit_encoding);
+
+	if (!output_encoding)
+		return NULL;
+	encoding = get_header(commit, "encoding");
+	if (!encoding || !strcmp(encoding, output_encoding)) {
+		free(encoding);
+		return NULL;
+	}
+	out = reencode_string(commit->buffer, output_encoding, encoding);
+	free(encoding);
+	if (!out)
+		return NULL;
+	return out;
+}
+
+unsigned long pretty_print_commit(enum cmit_fmt fmt,
+				  const struct commit *commit,
+				  unsigned long len,
+				  char *buf, unsigned long space,
 				  int abbrev, const char *subject,
-				  const char *after_subject, int relative_date)
+				  const char *after_subject,
+				  int relative_date)
 {
 	int hdr = 1, body = 0;
 	unsigned long offset = 0;
@@ -608,6 +660,10 @@ unsigned long pretty_print_commit(enum cmit_fmt fmt, const struct commit *commit
 	int parents_shown = 0;
 	const char *msg = commit->buffer;
 	int plain_non_ascii = 0;
+	char *reencoded = logmsg_reencode(commit);
+
+	if (reencoded)
+		msg = reencoded;
 
 	if (fmt == CMIT_FMT_ONELINE || fmt == CMIT_FMT_EMAIL)
 		indent = 0;
@@ -624,7 +680,7 @@ unsigned long pretty_print_commit(enum cmit_fmt fmt, const struct commit *commit
 		for (in_body = i = 0; (ch = msg[i]) && i < len; i++) {
 			if (!in_body) {
 				/* author could be non 7-bit ASCII but
-				 * the log may so; skip over the
+				 * the log may be so; skip over the
 				 * header part first.
 				 */
 				if (ch == '\n' &&
@@ -755,6 +811,8 @@ unsigned long pretty_print_commit(enum cmit_fmt fmt, const struct commit *commit
 	if (fmt == CMIT_FMT_EMAIL && !body)
 		buf[offset++] = '\n';
 	buf[offset] = '\0';
+
+	free(reencoded);
 	return offset;
 }
 

config.c

@@ -309,10 +309,16 @@ int git_default_config(const char *var, const char *value)
 	}
 
 	if (!strcmp(var, "i18n.commitencoding")) {
-		strlcpy(git_commit_encoding, value, sizeof(git_commit_encoding));
+		git_commit_encoding = strdup(value);
 		return 0;
 	}
 
+	if (!strcmp(var, "i18n.logoutputencoding")) {
+		git_log_output_encoding = strdup(value);
+		return 0;
+	}
+
+
 	if (!strcmp(var, "pager.color") || !strcmp(var, "color.pager")) {
 		pager_use_color = git_config_bool(var,value);
 		return 0;

contrib/completion/git-completion.bash

@@ -711,6 +711,7 @@ _git_repo_config ()
 		core.compression
 		core.legacyHeaders
 		i18n.commitEncoding
+		i18n.logOutputEncoding
 		diff.color
 		color.diff
 		diff.renameLimit

environment.c

@@ -18,7 +18,8 @@ int prefer_symlink_refs;
 int log_all_ref_updates;
 int warn_ambiguous_refs = 1;
 int repository_format_version;
-char git_commit_encoding[MAX_ENCODING_LENGTH] = "utf-8";
+char *git_commit_encoding;
+char *git_log_output_encoding;
 int shared_repository = PERM_UMASK;
 const char *apply_default_whitespace;
 int zlib_compression_level = Z_DEFAULT_COMPRESSION;

revision.h

@@ -72,6 +72,7 @@ struct rev_info {
 	const char	*ref_message_id;
 	const char	*add_signoff;
 	const char	*extra_headers;
+	const char	*log_reencode;
 
 	/* Filter by commit log message */
 	struct grep_opt	*grep_filter;

t/t3900-i18n-commit.sh

@@ -0,0 +1,115 @@
+#!/bin/sh
+#
+# Copyright (c) 2006 Junio C Hamano
+#
+
+test_description='commit and log output encodings'
+
+. ./test-lib.sh
+
+compare_with () {
+	git-show -s "$1" | sed -e '1,/^$/d' -e 's/^    //' -e '$d' >current &&
+	diff -u current "$2"
+}
+
+test_expect_success setup '
+	: >F &&
+	git-add F &&
+	T=$(git-write-tree) &&
+	C=$(git-commit-tree $T <../t3900/1-UTF-8.txt) &&
+	git-update-ref HEAD $C &&
+	git-tag C0
+'
+
+test_expect_success 'no encoding header for base case' '
+	E=$(git-cat-file commit C0 | sed -ne "s/^encoding //p") &&
+	test z = "z$E"
+'
+
+for H in ISO-8859-1 EUCJP ISO-2022-JP
+do
+	test_expect_success "$H setup" '
+		git-repo-config i18n.commitencoding $H &&
+		git-checkout -b $H C0 &&
+		echo $H >F &&
+		git-commit -a -F ../t3900/$H.txt
+	'
+done
+
+for H in ISO-8859-1 EUCJP ISO-2022-JP
+do
+	test_expect_success "check encoding header for $H" '
+		E=$(git-cat-file commit '$H' | sed -ne "s/^encoding //p") &&
+		test "z$E" = "z'$H'"
+	'
+done
+
+test_expect_success 'repo-config to remove customization' '
+	git-repo-config --unset-all i18n.commitencoding &&
+	if Z=$(git-repo-config --get-all i18n.commitencoding)
+	then
+		echo Oops, should have failed.
+		false
+	else
+		test z = "z$Z"
+	fi &&
+	git-repo-config i18n.commitencoding utf-8
+'
+
+test_expect_success 'ISO-8859-1 should be shown in UTF-8 now' '
+	compare_with ISO-8859-1 ../t3900/1-UTF-8.txt
+'
+
+for H in EUCJP ISO-2022-JP
+do
+	test_expect_success "$H should be shown in UTF-8 now" '
+		compare_with '$H' ../t3900/2-UTF-8.txt
+	'
+done
+
+test_expect_success 'repo-config to add customization' '
+	git-repo-config --unset-all i18n.commitencoding &&
+	if Z=$(git-repo-config --get-all i18n.commitencoding)
+	then
+		echo Oops, should have failed.
+		false
+	else
+		test z = "z$Z"
+	fi
+'
+
+for H in ISO-8859-1 EUCJP ISO-2022-JP
+do
+	test_expect_success "$H should be shown in itself now" '
+		git-repo-config i18n.commitencoding '$H' &&
+		compare_with '$H' ../t3900/'$H'.txt
+	'
+done
+
+test_expect_success 'repo-config to tweak customization' '
+	git-repo-config i18n.logoutputencoding utf-8
+'
+
+test_expect_success 'ISO-8859-1 should be shown in UTF-8 now' '
+	compare_with ISO-8859-1 ../t3900/1-UTF-8.txt
+'
+
+for H in EUCJP ISO-2022-JP
+do
+	test_expect_success "$H should be shown in UTF-8 now" '
+		compare_with '$H' ../t3900/2-UTF-8.txt
+	'
+done
+
+for J in EUCJP ISO-2022-JP
+do
+	git-repo-config i18n.logoutputencoding $J
+	for H in EUCJP ISO-2022-JP
+	do
+		test_expect_success "$H should be shown in $J now" '
+			compare_with '$H' ../t3900/'$J'.txt
+		'
+	done
+done
+
+test_done

t/t3900/1-UTF-8.txt

@@ -0,0 +1,3 @@
+ÄËÑÏÖ
+
+Ábçdèfg

t/t3900/2-UTF-8.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/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/ISO-2022-JP.txt

@@ -0,0 +1,4 @@
+$B$O$l$R$[$U(B
+
+$B$7$F$$$k$N$,!"$$$k$N$G!#(B
+$B_@IM$[$l$W$j$]$l$^$S$0$j$m$X!#(B

t/t3900/ISO-8859-1.txt

@@ -0,0 +1,3 @@
+ﾄﾋﾑﾏﾖ
+
+ﾁb軼鑁g

utf8.c

@@ -276,3 +276,57 @@ void print_wrapped_text(const char *text, int indent, int indent2, int width)
 		}
 	}
 }
+
+/*
+ * Given a buffer and its encoding, return it re-encoded
+ * with iconv.  If the conversion fails, returns NULL.
+ */
+#ifndef NO_ICONV
+char *reencode_string(const char *in, const char *out_encoding, const char *in_encoding)
+{
+	iconv_t conv;
+	size_t insz, outsz, outalloc;
+	char *out, *outpos, *cp;
+
+	if (!in_encoding)
+		return NULL;
+	conv = iconv_open(out_encoding, in_encoding);
+	if (conv == (iconv_t) -1)
+		return NULL;
+	insz = strlen(in);
+	outsz = insz;
+	outalloc = outsz + 1; /* for terminating NUL */
+	out = xmalloc(outalloc);
+	outpos = out;
+	cp = (char *)in;
+
+	while (1) {
+		size_t cnt = iconv(conv, &cp, &insz, &outpos, &outsz);
+
+		if (cnt == -1) {
+			size_t sofar;
+			if (errno != E2BIG) {
+				free(out);
+				iconv_close(conv);
+				return NULL;
+			}
+			/* insz has remaining number of bytes.
+			 * since we started outsz the same as insz,
+			 * it is likely that insz is not enough for
+			 * converting the rest.
+			 */
+			sofar = outpos - out;
+			outalloc = sofar + insz * 2 + 32;
+			out = xrealloc(out, outalloc);
+			outpos = out + sofar;
+			outsz = outalloc - sofar - 1;
+		}
+		else {
+			*outpos = '\0';
+			break;
+		}
+	}
+	iconv_close(conv);
+	return out;
+}
+#endif

utf8.h

@@ -5,4 +5,10 @@ int utf8_width(const char **start);
 int is_utf8(const char *text);
 void print_wrapped_text(const char *text, int indent, int indent2, int len);
 
+#ifndef NO_ICONV
+char *reencode_string(const char *in, const char *out_encoding, const char *in_encoding);
+#else
+#define reencode_string(a,b,c) NULL
+#endif
+
 #endif