freshlybakedca.ke
Git fork
UTF-8: introduce i18n.logoutputencoding.
It is plausible for somebody to want to view the commit log in a
different encoding from i18n.commitencoding -- the project's
policy may be UTF-8 and the user may be using a commit message
hook to run iconv to conform to that policy (and either not have
i18n.commitencoding to default to UTF-8 or have it explicitly
set to UTF-8). Even then, Latin-1 may be more convenient for
the usual pager and the terminal the user uses.
The new variable i18n.logoutputencoding is used in preference to
i18n.commitencoding to decide what encoding to recode the log
output in when git-log and friends formats the commit log message.
Signed-off-by: Junio C Hamano <junkio@cox.net>
+4
Documentation/config.txt
+4
Documentation/config.txt
···
248
248
browser (and possibly at other places in the future or in other
249
249
porcelains). See e.g. gitlink:git-mailinfo[1]. Defaults to 'utf-8'.
250
250
251
+
i18n.logOutputEncoding::
252
+
Character encoding the commit messages are converted to when
253
+
running `git-log` and friends.
254
+
251
255
log.showroot::
252
256
If true, the initial commit will be shown as a big creation event.
253
257
This is equivalent to a diff against an empty tree.
+3
-1
builtin-commit-tree.c
+3
-1
builtin-commit-tree.c
···
118
118
parents++;
119
119
}
120
120
121
-
encoding_is_utf8 = !strcmp(git_commit_encoding, "utf-8");
121
+
/* Not having i18n.commitencoding is the same as having utf-8 */
122
+
encoding_is_utf8 = (!git_commit_encoding ||
123
+
!strcmp(git_commit_encoding, "utf-8"));
122
124
123
125
init_buffer(&buffer, &size);
124
126
add_buffer(&buffer, &size, "tree %s\n", sha1_to_hex(tree_sha1));
+2
-5
builtin-log.c
+2
-5
builtin-log.c
···
33
33
const char *arg = argv[i];
34
34
if (!strncmp(arg, "--encoding=", 11)) {
35
35
arg += 11;
36
-
if (MAX_ENCODING_LENGTH <= strlen(arg))
37
-
die(" Value of output encoding '%s' too long",
38
-
arg);
39
36
if (strcmp(arg, "none"))
40
-
strcpy(git_commit_encoding, arg);
37
+
git_log_output_encoding = strdup(arg);
41
38
else
42
-
git_commit_encoding[0] = 0;
39
+
git_log_output_encoding = "";
43
40
}
44
41
else
45
42
die("unrecognized argument: %s", arg);
+2
-1
builtin-mailinfo.c
+2
-1
builtin-mailinfo.c
···
806
806
if (!strcmp(argv[1], "-k"))
807
807
keep_subject = 1;
808
808
else if (!strcmp(argv[1], "-u"))
809
-
metainfo_charset = git_commit_encoding;
809
+
metainfo_charset = (git_commit_encoding
810
+
? git_commit_encoding : "utf-8");
810
811
else if (!strncmp(argv[1], "--encoding=", 11))
811
812
metainfo_charset = argv[1] + 11;
812
813
else
+2
-2
cache.h
+2
-2
cache.h
···
416
416
extern char git_default_email[MAX_GITNAME];
417
417
extern char git_default_name[MAX_GITNAME];
418
418
419
-
#define MAX_ENCODING_LENGTH 64
420
-
extern char git_commit_encoding[MAX_ENCODING_LENGTH];
419
+
extern char *git_commit_encoding;
420
+
extern char *git_log_output_encoding;
421
421
422
422
extern int copy_fd(int ifd, int ofd);
423
423
extern void write_or_die(int fd, const void *buf, size_t count);
+15
-12
commit.c
+15
-12
commit.c
···
592
592
593
593
static char *logmsg_reencode(const struct commit *commit)
594
594
{
595
-
char *encoding = get_header(commit, "encoding");
595
+
char *encoding;
596
596
char *out;
597
+
char *output_encoding = (git_log_output_encoding
598
+
? git_log_output_encoding
599
+
: git_commit_encoding);
597
600
598
-
if (!encoding || !strcmp(encoding, git_commit_encoding))
601
+
if (!output_encoding)
599
602
return NULL;
600
-
out = reencode_string(commit->buffer, git_commit_encoding, encoding);
603
+
encoding = get_header(commit, "encoding");
604
+
if (!encoding || !strcmp(encoding, output_encoding)) {
605
+
free(encoding);
606
+
return NULL;
607
+
}
608
+
out = reencode_string(commit->buffer, output_encoding, encoding);
601
609
free(encoding);
602
610
if (!out)
603
611
return NULL;
···
618
626
int parents_shown = 0;
619
627
const char *msg = commit->buffer;
620
628
int plain_non_ascii = 0;
621
-
char *reencoded = NULL;
629
+
char *reencoded = logmsg_reencode(commit);
622
630
623
-
if (*git_commit_encoding) {
624
-
reencoded = logmsg_reencode(commit);
625
-
if (reencoded) {
626
-
msg = reencoded;
627
-
len = strlen(msg);
628
-
}
629
-
}
631
+
if (reencoded)
632
+
msg = reencoded;
630
633
631
634
if (fmt == CMIT_FMT_ONELINE || fmt == CMIT_FMT_EMAIL)
632
635
indent = 0;
···
643
646
for (in_body = i = 0; (ch = msg[i]) && i < len; i++) {
644
647
if (!in_body) {
645
648
/* author could be non 7-bit ASCII but
646
-
* the log may so; skip over the
649
+
* the log may be so; skip over the
647
650
* header part first.
648
651
*/
649
652
if (ch == '\n' &&
+7
-1
config.c
+7
-1
config.c
···
309
309
}
310
310
311
311
if (!strcmp(var, "i18n.commitencoding")) {
312
-
strlcpy(git_commit_encoding, value, sizeof(git_commit_encoding));
312
+
git_commit_encoding = strdup(value);
313
313
return 0;
314
314
}
315
+
316
+
if (!strcmp(var, "i18n.logoutputencoding")) {
317
+
git_log_output_encoding = strdup(value);
318
+
return 0;
319
+
}
320
+
315
321
316
322
if (!strcmp(var, "pager.color") || !strcmp(var, "color.pager")) {
317
323
pager_use_color = git_config_bool(var,value);
+1
contrib/completion/git-completion.bash
+1
contrib/completion/git-completion.bash
+2
-1
environment.c
+2
-1
environment.c
···
18
18
int log_all_ref_updates;
19
19
int warn_ambiguous_refs = 1;
20
20
int repository_format_version;
21
-
char git_commit_encoding[MAX_ENCODING_LENGTH] = "utf-8";
21
+
char *git_commit_encoding;
22
+
char *git_log_output_encoding;
22
23
int shared_repository = PERM_UMASK;
23
24
const char *apply_default_whitespace;
24
25
int zlib_compression_level = Z_DEFAULT_COMPRESSION;
+104
t/t3900-i18n-commit.sh
+104
t/t3900-i18n-commit.sh
···
1
+
#!/bin/sh
2
+
#
3
+
# Copyright (c) 2006 Junio C Hamano
4
+
#
5
+
6
+
test_description='commit and log output encodings'
7
+
8
+
. ./test-lib.sh
9
+
10
+
compare_with () {
11
+
git-show -s "$1" | sed -e '1,/^$/d' -e 's/^ //' -e '$d' >current &&
12
+
diff -u current "$2"
13
+
}
14
+
15
+
test_expect_success setup '
16
+
: >F &&
17
+
git-add F &&
18
+
T=$(git-write-tree) &&
19
+
C=$(git-commit-tree $T <../t3900/1-UTF-8.txt) &&
20
+
git-update-ref HEAD $C &&
21
+
git-tag C0
22
+
'
23
+
24
+
test_expect_success 'no encoding header for base case' '
25
+
E=$(git-cat-file commit C0 | sed -ne "s/^encoding //p") &&
26
+
test z = "z$E"
27
+
'
28
+
29
+
for H in ISO-8859-1 EUCJP ISO2022JP
30
+
do
31
+
test_expect_success "$H setup" '
32
+
git-repo-config i18n.commitencoding $H &&
33
+
git-checkout -b $H C0 &&
34
+
echo $H >F &&
35
+
git-commit -a -F ../t3900/$H.txt
36
+
'
37
+
done
38
+
39
+
for H in ISO-8859-1 EUCJP ISO2022JP
40
+
do
41
+
test_expect_success "check encoding header for $H" '
42
+
E=$(git-cat-file commit '$H' | sed -ne "s/^encoding //p") &&
43
+
test "z$E" = "z'$H'"
44
+
'
45
+
done
46
+
47
+
test_expect_success 'repo-config to remove customization' '
48
+
git-repo-config --unset-all i18n.commitencoding &&
49
+
if Z=$(git-repo-config --get-all i18n.commitencoding)
50
+
then
51
+
echo Oops, should have failed.
52
+
false
53
+
else
54
+
test z = "z$Z"
55
+
fi &&
56
+
git-repo-config i18n.commitencoding utf-8
57
+
'
58
+
59
+
test_expect_success 'ISO-8859-1 should be shown in UTF-8 now' '
60
+
compare_with ISO-8859-1 ../t3900/1-UTF-8.txt
61
+
'
62
+
63
+
for H in EUCJP ISO2022JP
64
+
do
65
+
test_expect_success "$H should be shown in UTF-8 now" '
66
+
compare_with '$H' ../t3900/2-UTF-8.txt
67
+
'
68
+
done
69
+
70
+
test_expect_success 'repo-config to add customization' '
71
+
git-repo-config --unset-all i18n.commitencoding &&
72
+
if Z=$(git-repo-config --get-all i18n.commitencoding)
73
+
then
74
+
echo Oops, should have failed.
75
+
false
76
+
else
77
+
test z = "z$Z"
78
+
fi
79
+
'
80
+
81
+
for H in ISO-8859-1 EUCJP ISO2022JP
82
+
do
83
+
test_expect_success "$H should be shown in itself now" '
84
+
git-repo-config i18n.commitencoding '$H' &&
85
+
compare_with '$H' ../t3900/'$H'.txt
86
+
'
87
+
done
88
+
89
+
test_expect_success 'repo-config to tweak customization' '
90
+
git-repo-config i18n.logoutputencoding utf-8
91
+
'
92
+
93
+
test_expect_success 'ISO-8859-1 should be shown in UTF-8 now' '
94
+
compare_with ISO-8859-1 ../t3900/1-UTF-8.txt
95
+
'
96
+
97
+
for H in EUCJP ISO2022JP
98
+
do
99
+
test_expect_success "$H should be shown in UTF-8 now" '
100
+
compare_with '$H' ../t3900/2-UTF-8.txt
101
+
'
102
+
done
103
+
104
+
test_done