[v6,2/2] POSIX locale covers every byte [BZ# 29511]
Checks
| Context |
Check |
Description |
| dj/TryBot-apply_patch |
success
|
Patch applied to master at the time it was sent
|
| dj/TryBot-32bit |
success
|
Build for i686
|
Commit Message
This is a logistically trivial patch,
largely duplicating the extant ASCII code with the error path changed
There are two user-facing changes:
* nl_langinfo(CODESET) is "POSIX" instead of "ANSI_X3.4-1968"
* mbrtowc() and friends return b if b <= 0x7F else <UDF00>+b
Since Issue 7 TC 2/Issue 8, the C/POSIX locale, effectively,
(a) is 1-byte, stateless, and contains 256 characters
(b) they collate in byte order
(c) the first 128 characters are equivalent to ASCII (like previous)
cf. https://www.austingroupbugs.net/view.php?id=663 for a summary of
changes to the standard;
in short, this means that mbrtowc() must never fail and must return
b if b <= 0x7F else ab+c for all bytes b
where c is some constant >=0x80
and a is a positive integer constant
By strategically picking c=<UDF00> we land at the tail-end of the
Unicode Low Surrogate Area at DC00-DFFF, described as
> Isolated surrogate code points have no interpretation;
> consequently, no character code charts or names lists
> are provided for this range.
and match musl
Signed-off-by: Ahelenia Ziemiańska <nabijaczleweli@nabijaczleweli.xyz>
---
new in v2: nothing
new in v3: POSIX charset, __wcsmbs_gconv_fcts_c comment fixed
new in v4: nothing
new in v5: nothing
new in v6: clean rebase, rephrase message
iconv/Makefile | 2 +-
iconv/gconv_builtin.h | 8 ++
iconv/gconv_int.h | 7 ++
iconv/gconv_posix.c | 96 ++++++++++++++++++++
iconv/tst-iconv_prog.sh | 43 +++++++++
iconvdata/tst-tables.sh | 1 +
inet/tst-idna_name_classify.c | 6 +-
locale/tst-C-locale.c | 69 ++++++++++++++
localedata/charmaps/POSIX | 136 ++++++++++++++++++++++++++++
localedata/locales/POSIX | 143 +++++++++++++++++++++++++++++-
stdio-common/tst-printf-bz25691.c | 2 +
wcsmbs/wcsmbsload.c | 14 +--
12 files changed, 516 insertions(+), 11 deletions(-)
create mode 100644 iconv/gconv_posix.c
create mode 100644 localedata/charmaps/POSIX
Comments
* наб:
> This is a logistically trivial patch,
> largely duplicating the extant ASCII code with the error path changed
I wouldn't say it's trivial in the commit message. 8-)
> There are two user-facing changes:
> * nl_langinfo(CODESET) is "POSIX" instead of "ANSI_X3.4-1968"
> * mbrtowc() and friends return b if b <= 0x7F else <UDF00>+b
>
> Since Issue 7 TC 2/Issue 8, the C/POSIX locale, effectively,
> (a) is 1-byte, stateless, and contains 256 characters
> (b) they collate in byte order
> (c) the first 128 characters are equivalent to ASCII (like previous)
> cf. https://www.austingroupbugs.net/view.php?id=663 for a summary of
> changes to the standard;
> in short, this means that mbrtowc() must never fail and must return
> b if b <= 0x7F else ab+c for all bytes b
> where c is some constant >=0x80
> and a is a positive integer constant
>
> By strategically picking c=<UDF00> we land at the tail-end of the
> Unicode Low Surrogate Area at DC00-DFFF, described as
> > Isolated surrogate code points have no interpretation;
> > consequently, no character code charts or names lists
> > are provided for this range.
> and match musl
Sadly this doesn't match Python and PEP 540:
>>> b'\x80'.decode('UTF-8', errors='surrogateescape')
'\udc80'
I believe the implementation translates this to 0xDF80 instead.
Not sure what is more important here, musl compatibility or Python
compatibility. Cc:ing Victor in case he as comments. I should probably
ask on the musl list as well as how this divergence came to pass.
This change definitely needs a NEWS entry.
The mechanics of the patch look okay to me, just a few nits below.
> diff --git a/iconv/gconv_int.h b/iconv/gconv_int.h
> index 1c6745043e..45ab1edfad 100644
> --- a/iconv/gconv_int.h
> +++ b/iconv/gconv_int.h
> @@ -281,6 +281,8 @@ extern int __gconv_compare_alias (const char *name1, const char *name2)
>
> __BUILTIN_TRANSFORM (__gconv_transform_ascii_internal);
> __BUILTIN_TRANSFORM (__gconv_transform_internal_ascii);
> +__BUILTIN_TRANSFORM (__gconv_transform_posix_internal);
> +__BUILTIN_TRANSFORM (__gconv_transform_internal_posix);
> __BUILTIN_TRANSFORM (__gconv_transform_utf8_internal);
> __BUILTIN_TRANSFORM (__gconv_transform_internal_utf8);
> __BUILTIN_TRANSFORM (__gconv_transform_ucs2_internal);
> @@ -299,6 +301,11 @@ __BUILTIN_TRANSFORM (__gconv_transform_utf16_internal);
> only ASCII characters. */
> extern wint_t __gconv_btwoc_ascii (struct __gconv_step *step, unsigned char c);
>
> +/* Specialized conversion function for a single byte to INTERNAL,
> + identity-mapping bytes [0, 0x7F], and moving [0x80, 0xFF] into the end
> + of the Low Surrogate Area at [U+DF80, U+DFFF]. */
> +extern wint_t __gconv_btwoc_posix (struct __gconv_step *step, unsigned char c);
> +
> #endif
Missing attribute_hidden. Yes, it's also missing from
__gconv_btwoc_ascii. The linker probably papers over it.
>
> __END_DECLS
> diff --git a/iconv/gconv_posix.c b/iconv/gconv_posix.c
> new file mode 100644
> index 0000000000..dcb13fbb43
> --- /dev/null
> +++ b/iconv/gconv_posix.c
> @@ -0,0 +1,96 @@
> +/* Simple transformations functions.
I think this line should say something about surrogate-escape encoding
for the POSIX locale.
> +#define MIN_NEEDED_INPUT MIN_NEEDED_FROM
> +#define MIN_NEEDED_OUTPUT MIN_NEEDED_TO
> +#define LOOPFCT FROM_LOOP
> +#define BODY \
> + { \
> + uint32_t val = *((const uint32_t *) inptr); \
> + if (__glibc_unlikely ((val > 0x7f && val < 0xdf80) || val > 0xdfff)) \
> + { \
> + UNICODE_TAG_HANDLER (val, 4); \
> + STANDARD_TO_LOOP_ERR_HANDLER (4); \
> + } \
> + else \
> + { \
> + if (__glibc_unlikely (val > 0x7f)) \
> + val -= 0xdf00; \
> + *outptr++ = val; \
> + inptr += sizeof (uint32_t); \
> + } \
> + }
I suggest to drop the last __glibc_unlikely here because it's
input-dependent.
> +#define LOOP_NEED_FLAGS
> +#include <iconv/loop.c>
> +#include <iconv/skeleton.c>
> diff --git a/iconv/tst-iconv_prog.sh b/iconv/tst-iconv_prog.sh
> index b3d8bf5110..a24d8d2207 100644
> --- a/iconv/tst-iconv_prog.sh
> +++ b/iconv/tst-iconv_prog.sh
> @@ -285,3 +285,46 @@ for errorcommand in "${errorarray[@]}"; do
> execute_test
> check_errtest_result
> done
> +
> +allbytes ()
> +{
> + for (( i = 0; i <= 255; i++ )); do
> + printf '\'"$(printf "%o" "$i")"
> + done
> +}
> +
> +allucs4be ()
> +{
> + for (( i = 0; i <= 127; i++ )); do
> + printf '\0\0\0\'"$(printf "%o" "$i")"
> + done
> + for (( i = 128; i <= 255; i++ )); do
> + printf '\0\0\xdf\'"$(printf "%o" "$i")"
> + done
> +}
> +
> +check_posix_result ()
> +{
> + if [ $? -eq 0 ]; then
> + result=PASS
> + else
> + result=FAIL
> + fi
> +
> + echo "$result: from \"$1\", to: \"$2\""
> +
> + if [ "$result" != "PASS" ]; then
> + exit 1
> + fi
> +}
> +
> +check_posix_encoding ()
> +{
> + eval PROG=\"$ICONV\"
> + allbytes | $PROG -f POSIX -t UCS-4BE | cmp -s - <(allucs4be)
> + check_posix_result POSIX UCS-4BE
> + allucs4be | $PROG -f UCS-4BE -t POSIX | cmp -s - <(allbytes)
> + check_posix_result UCS-4BE POSIX
> +}
> +
> +check_posix_encoding
> diff --git a/iconvdata/tst-tables.sh b/iconvdata/tst-tables.sh
> index 4207b44175..33a02158ac 100755
> --- a/iconvdata/tst-tables.sh
> +++ b/iconvdata/tst-tables.sh
> @@ -31,6 +31,7 @@ cat <<EOF |
> # Keep this list in the same order as gconv-modules.
> #
> # charset name table name comment
> + POSIX
> ASCII ANSI_X3.4-1968
> ISO646-GB BS_4730
> ISO646-CA CSA_Z243.4-1985-1
> diff --git a/inet/tst-idna_name_classify.c b/inet/tst-idna_name_classify.c
> index bfd34eee31..b379481844 100644
> --- a/inet/tst-idna_name_classify.c
> +++ b/inet/tst-idna_name_classify.c
> @@ -37,11 +37,11 @@ do_test (void)
> puts ("info: C locale tests");
> locale_insensitive_tests ();
> TEST_COMPARE (__idna_name_classify ("abc\200def"),
> - idna_name_encoding_error);
> + idna_name_nonascii);
> TEST_COMPARE (__idna_name_classify ("abc\200\\def"),
> - idna_name_encoding_error);
> + idna_name_nonascii_backslash);
> TEST_COMPARE (__idna_name_classify ("abc\377def"),
> - idna_name_encoding_error);
> + idna_name_nonascii);
>
> puts ("info: en_US.ISO-8859-1 locale tests");
> if (setlocale (LC_CTYPE, "en_US.ISO-8859-1") == 0)
This seems to be okay, there is further test coverage for
idna_name_encoding_error.
> diff --git a/locale/tst-C-locale.c b/locale/tst-C-locale.c
> index 6bd0367069..f30396ae12 100644
> --- a/locale/tst-C-locale.c
> +++ b/locale/tst-C-locale.c
> @@ -229,6 +229,75 @@ run_test (const char *locname)
> STRTEST (YESSTR, "");
> STRTEST (NOSTR, "");
>
> +#define CONVTEST(b, v) \
> + { \
> + unsigned char bs[] = {b, 0}; \
> + mbstate_t ctx = {}; \
> + wchar_t wc = -1; \
> + size_t sz = mbrtowc(&wc, (char *) bs, 1, &ctx); \
Missing space before '(' (also in other cases below).
Not sure if the macros are needed, maybe write one loop for each
direction with a condition in it?
> diff --git a/stdio-common/tst-printf-bz25691.c b/stdio-common/tst-printf-bz25691.c
> index 44844e71c3..e66242b58f 100644
> --- a/stdio-common/tst-printf-bz25691.c
> +++ b/stdio-common/tst-printf-bz25691.c
> @@ -30,6 +30,8 @@
> static int
> do_test (void)
> {
> + setlocale(LC_CTYPE, "C.UTF-8");
> +
> mtrace ();
>
> /* For 's' conversion specifier with 'l' modifier the array must be
What's the rationale for this change? If it is really required, you
must also update stdio-common/Makefile with a new dependency on
$(gen-locales).
Thanks,
Florian
* Florian Weimer:
> * наб:
>
>> This is a logistically trivial patch,
>> largely duplicating the extant ASCII code with the error path changed
>
> I wouldn't say it's trivial in the commit message. 8-)
>
>> There are two user-facing changes:
>> * nl_langinfo(CODESET) is "POSIX" instead of "ANSI_X3.4-1968"
>> * mbrtowc() and friends return b if b <= 0x7F else <UDF00>+b
>>
>> Since Issue 7 TC 2/Issue 8, the C/POSIX locale, effectively,
>> (a) is 1-byte, stateless, and contains 256 characters
>> (b) they collate in byte order
>> (c) the first 128 characters are equivalent to ASCII (like previous)
>> cf. https://www.austingroupbugs.net/view.php?id=663 for a summary of
>> changes to the standard;
>> in short, this means that mbrtowc() must never fail and must return
>> b if b <= 0x7F else ab+c for all bytes b
>> where c is some constant >=0x80
>> and a is a positive integer constant
>>
>> By strategically picking c=<UDF00> we land at the tail-end of the
>> Unicode Low Surrogate Area at DC00-DFFF, described as
>> > Isolated surrogate code points have no interpretation;
>> > consequently, no character code charts or names lists
>> > are provided for this range.
>> and match musl
>
> Sadly this doesn't match Python and PEP 540:
>
>>>> b'\x80'.decode('UTF-8', errors='surrogateescape')
> '\udc80'
>
> I believe the implementation translates this to 0xDF80 instead.
>
> Not sure what is more important here, musl compatibility or Python
> compatibility. Cc:ing Victor in case he as comments. I should probably
> ask on the musl list as well as how this divergence came to pass.
Raised on the musl list here:
Choice of wchar_t mapping for non-ASCII bytes in the POSIX locale
<https://www.openwall.com/lists/musl/2022/11/10/1>
> This change definitely needs a NEWS entry.
(With this I meant the change overall, not the encoding.)
Thanks,
Florian
18-day bump :)
On Thu, Nov 10, 2022 at 09:10:57AM +0100, Florian Weimer wrote:
> Raised on the musl list here:
> Choice of wchar_t mapping for non-ASCII bytes in the POSIX locale
> <https://www.openwall.com/lists/musl/2022/11/10/1>
That thread seems to've been exhausted (at least I don't see anything
fresh in the archive) ‒ should I just resend with the comments for v7
applied, or do you have a mapping range you'd rather see given those
givens?
наб
* наб:
> 18-day bump :)
>
> On Thu, Nov 10, 2022 at 09:10:57AM +0100, Florian Weimer wrote:
>> Raised on the musl list here:
>> Choice of wchar_t mapping for non-ASCII bytes in the POSIX locale
>> <https://www.openwall.com/lists/musl/2022/11/10/1>
>
> That thread seems to've been exhausted (at least I don't see anything
> fresh in the archive) ‒ should I just resend with the comments for v7
> applied, or do you have a mapping range you'd rather see given those
> givens?
I still can't make up my mind. I think the options are:
* Some sort of custom encoding (like you posted).
* Latin-1
* UTF-8 with surrogate escape encoding (and encouraging POSIX to change again)
What argues in favor of the last point is that many, many people are
using C.UTF-8 nowadays. And effectively disabling wide/multibyte
conversion until you call setlocale does not seem particularly useful.
I have a feeling of déjà vu regarding this—I think I have investigated
non-setlocale defaults for wide/multibyte conversion in the past, but
can't find the previous discussion.
Thanks,
Florian
Hi!
On Fri, Dec 02, 2022 at 06:36:26PM +0100, Florian Weimer wrote:
> * наб:
> > On Thu, Nov 10, 2022 at 09:10:57AM +0100, Florian Weimer wrote:
> >> Raised on the musl list here:
> >> Choice of wchar_t mapping for non-ASCII bytes in the POSIX locale
> >> <https://www.openwall.com/lists/musl/2022/11/10/1>
> >
> > That thread seems to've been exhausted (at least I don't see anything
> > fresh in the archive) ‒ should I just resend with the comments for v7
> > applied, or do you have a mapping range you'd rather see given those
> > givens?
>
> I still can't make up my mind. I think the options are:
>
> * Some sort of custom encoding (like you posted).
For which there's Prior Art in both other libcs and implementations
of similar mechanisms in unrelated software, making it just about what
users expect, and the lowest-energy conversion to what POSIX
has mandated for close to 8 years now.
> * Latin-1
Sorry, what? The Latin-1 that's so poorly defined W3C requires,
per spec, that Latin-1 charset specs be ignored? The Latin-1 they got
wrong so badly they made two subsequent standards, only one of which
compatible? The Latin-1 that has some random subset of germanic and
maybe like french if you squint and that's apparently fine? The iron
curtain has fallen, for better or for worse, since the 80s. If that's
the "solution", then leave "C" 7-bit. I'm gonna assume that's a joke.
(Also, people would then try to "use" it, and then (a) you've lost,
but (b) the collation sequence is gonna be wrong always because
it no longer represents any given language
(though apparently it's "fine" if they collate in any random order,
so it's legal per spec to make it just spanish, I think;
this is somehow even worse, and I may be misunderstanding
POSIX 7.3.2.6 because it'd mean that other parts of the standard
that use and recommend "LC_ALL=C utility ..." to process bytes,
like for sort, are also wrong?).)
> * UTF-8 with surrogate escape encoding (and encouraging POSIX to change again)
Well it's not gonna ‒ at least I don't think it is ‒ given that I don't
think it /changed/ anything actually? Issue 7-2008 7.2 says
> The tables in Locale Definition describe the characteristics and
> behavior of the POSIX locale for data consisting entirely of
> characters from the portable character set and the control character
> set. For other characters, the behavior is unspecified.
And just TC2 specified what had been unspecified behaviour I think?
Implementations had freedom to do whatever, including UTF-8, until 2016.
Naturally, as we're seeing now, not one has exercised that freedom.
If glibc /did/ do POSIX=C=C.UTF-8 before then,
then maybe we'd see a different result, but it hadn't, so we didn't.
> What argues in favor of the last point is that many, many people are
> using C.UTF-8 nowadays.
Great! They can continue to use C.UTF-8. They have had to opt in to
their preferred encoding like everyone else, and they will continue.
0 changes observed here.
> And effectively disabling wide/multibyte
> conversion until you call setlocale does not seem particularly useful.
"Mangling input data until explicitly disabled" is worse than
"input data is data, and you can make it characters".
Don't take me for not-a-UTF-8-maximalist, but, y'know,
it will never, unfortunately, be all I see,
and being able to completely opt out of additional input processing
would be nice; we're kinda close now with the current hard-7-bit ASCII,
and making it, essentially, I Can't Believe It's Not Just Bytes!,
per pt. 1, would eliminate even more head-aches IME.
Putting pro-verbial KOI-8 or [your grandma's favourite encoding] through
the UTF-8 grinder is much worse than just degrading to strcmp(),
but at this point I think I'm rambling, and spilled enough ink;
your call to make, at the end of the day.
наб
@@ -25,7 +25,7 @@ include ../Makeconfig
headers = iconv.h gconv.h
routines = iconv_open iconv iconv_close \
gconv_open gconv gconv_close gconv_db gconv_conf \
- gconv_builtin gconv_simple gconv_trans gconv_cache
+ gconv_builtin gconv_simple gconv_posix gconv_trans gconv_cache
routines += gconv_dl gconv_charset
vpath %.c ../locale/programs ../intl
@@ -89,6 +89,14 @@ BUILTIN_TRANSFORMATION ("INTERNAL", "ANSI_X3.4-1968//", 1, "=INTERNAL->ascii",
__gconv_transform_internal_ascii, NULL, 4, 4, 1, 1)
+BUILTIN_TRANSFORMATION ("POSIX//", "INTERNAL", 1, "=posix->INTERNAL",
+ __gconv_transform_posix_internal, __gconv_btwoc_posix,
+ 1, 1, 4, 4)
+
+BUILTIN_TRANSFORMATION ("INTERNAL", "POSIX//", 1, "=INTERNAL->posix",
+ __gconv_transform_internal_posix, NULL, 4, 4, 1, 1)
+
+
#if BYTE_ORDER == BIG_ENDIAN
BUILTIN_ALIAS ("UNICODEBIG//", "ISO-10646/UCS2/")
BUILTIN_ALIAS ("UCS-2BE//", "ISO-10646/UCS2/")
@@ -281,6 +281,8 @@ extern int __gconv_compare_alias (const char *name1, const char *name2)
__BUILTIN_TRANSFORM (__gconv_transform_ascii_internal);
__BUILTIN_TRANSFORM (__gconv_transform_internal_ascii);
+__BUILTIN_TRANSFORM (__gconv_transform_posix_internal);
+__BUILTIN_TRANSFORM (__gconv_transform_internal_posix);
__BUILTIN_TRANSFORM (__gconv_transform_utf8_internal);
__BUILTIN_TRANSFORM (__gconv_transform_internal_utf8);
__BUILTIN_TRANSFORM (__gconv_transform_ucs2_internal);
@@ -299,6 +301,11 @@ __BUILTIN_TRANSFORM (__gconv_transform_utf16_internal);
only ASCII characters. */
extern wint_t __gconv_btwoc_ascii (struct __gconv_step *step, unsigned char c);
+/* Specialized conversion function for a single byte to INTERNAL,
+ identity-mapping bytes [0, 0x7F], and moving [0x80, 0xFF] into the end
+ of the Low Surrogate Area at [U+DF80, U+DFFF]. */
+extern wint_t __gconv_btwoc_posix (struct __gconv_step *step, unsigned char c);
+
#endif
__END_DECLS
new file mode 100644
@@ -0,0 +1,96 @@
+/* Simple transformations functions.
+ Copyright (C) 2022 Free Software Foundation, Inc.
+ This file is part of the GNU C Library.
+
+ The GNU C Library is free software; you can redistribute it and/or
+ modify it under the terms of the GNU Lesser General Public
+ License as published by the Free Software Foundation; either
+ version 2.1 of the License, or (at your option) any later version.
+
+ The GNU C Library is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ Lesser General Public License for more details.
+
+ You should have received a copy of the GNU Lesser General Public
+ License along with the GNU C Library; if not, see
+ <https://www.gnu.org/licenses/>. */
+
+
+#include <gconv_int.h>
+
+
+/* Specialized conversion function for a single byte to INTERNAL,
+ identity-mapping bytes [0, 0x7F], and moving [0x80, 0xFF] into the end
+ of the Low Surrogate Area at [U+DF80, U+DFFF]. */
+wint_t
+__gconv_btwoc_posix (struct __gconv_step *step, unsigned char c)
+{
+ if (c < 0x80)
+ return c;
+ else
+ return 0xdf00 + c;
+}
+
+
+/* Convert from {[0, 0x7F] => ISO 646-IRV; [0x80, 0xFF] => [U+DF80, U+DFFF]}
+ to the internal (UCS4-like) format. */
+#define DEFINE_INIT 0
+#define DEFINE_FINI 0
+#define MIN_NEEDED_FROM 1
+#define MIN_NEEDED_TO 4
+#define FROM_DIRECTION 1
+#define FROM_LOOP posix_internal_loop
+#define TO_LOOP posix_internal_loop /* This is not used. */
+#define FUNCTION_NAME __gconv_transform_posix_internal
+#define ONE_DIRECTION 1
+
+#define MIN_NEEDED_INPUT MIN_NEEDED_FROM
+#define MIN_NEEDED_OUTPUT MIN_NEEDED_TO
+#define LOOPFCT FROM_LOOP
+#define BODY \
+ { \
+ if (__glibc_unlikely (*inptr > '\x7f')) \
+ *((uint32_t *) outptr) = 0xdf00 + *inptr++; \
+ else \
+ *((uint32_t *) outptr) = *inptr++; \
+ outptr += sizeof (uint32_t); \
+ }
+#include <iconv/loop.c>
+#include <iconv/skeleton.c>
+
+
+/* Convert from the internal (UCS4-like) format to
+ {ISO 646-IRV => [0, 0x7F]; [U+DF80, U+DFFF] => [0x80, 0xFF]}. */
+#define DEFINE_INIT 0
+#define DEFINE_FINI 0
+#define MIN_NEEDED_FROM 4
+#define MIN_NEEDED_TO 1
+#define FROM_DIRECTION 1
+#define FROM_LOOP internal_posix_loop
+#define TO_LOOP internal_posix_loop /* This is not used. */
+#define FUNCTION_NAME __gconv_transform_internal_posix
+#define ONE_DIRECTION 1
+
+#define MIN_NEEDED_INPUT MIN_NEEDED_FROM
+#define MIN_NEEDED_OUTPUT MIN_NEEDED_TO
+#define LOOPFCT FROM_LOOP
+#define BODY \
+ { \
+ uint32_t val = *((const uint32_t *) inptr); \
+ if (__glibc_unlikely ((val > 0x7f && val < 0xdf80) || val > 0xdfff)) \
+ { \
+ UNICODE_TAG_HANDLER (val, 4); \
+ STANDARD_TO_LOOP_ERR_HANDLER (4); \
+ } \
+ else \
+ { \
+ if (__glibc_unlikely (val > 0x7f)) \
+ val -= 0xdf00; \
+ *outptr++ = val; \
+ inptr += sizeof (uint32_t); \
+ } \
+ }
+#define LOOP_NEED_FLAGS
+#include <iconv/loop.c>
+#include <iconv/skeleton.c>
@@ -285,3 +285,46 @@ for errorcommand in "${errorarray[@]}"; do
execute_test
check_errtest_result
done
+
+allbytes ()
+{
+ for (( i = 0; i <= 255; i++ )); do
+ printf '\'"$(printf "%o" "$i")"
+ done
+}
+
+allucs4be ()
+{
+ for (( i = 0; i <= 127; i++ )); do
+ printf '\0\0\0\'"$(printf "%o" "$i")"
+ done
+ for (( i = 128; i <= 255; i++ )); do
+ printf '\0\0\xdf\'"$(printf "%o" "$i")"
+ done
+}
+
+check_posix_result ()
+{
+ if [ $? -eq 0 ]; then
+ result=PASS
+ else
+ result=FAIL
+ fi
+
+ echo "$result: from \"$1\", to: \"$2\""
+
+ if [ "$result" != "PASS" ]; then
+ exit 1
+ fi
+}
+
+check_posix_encoding ()
+{
+ eval PROG=\"$ICONV\"
+ allbytes | $PROG -f POSIX -t UCS-4BE | cmp -s - <(allucs4be)
+ check_posix_result POSIX UCS-4BE
+ allucs4be | $PROG -f UCS-4BE -t POSIX | cmp -s - <(allbytes)
+ check_posix_result UCS-4BE POSIX
+}
+
+check_posix_encoding
@@ -31,6 +31,7 @@ cat <<EOF |
# Keep this list in the same order as gconv-modules.
#
# charset name table name comment
+ POSIX
ASCII ANSI_X3.4-1968
ISO646-GB BS_4730
ISO646-CA CSA_Z243.4-1985-1
@@ -37,11 +37,11 @@ do_test (void)
puts ("info: C locale tests");
locale_insensitive_tests ();
TEST_COMPARE (__idna_name_classify ("abc\200def"),
- idna_name_encoding_error);
+ idna_name_nonascii);
TEST_COMPARE (__idna_name_classify ("abc\200\\def"),
- idna_name_encoding_error);
+ idna_name_nonascii_backslash);
TEST_COMPARE (__idna_name_classify ("abc\377def"),
- idna_name_encoding_error);
+ idna_name_nonascii);
puts ("info: en_US.ISO-8859-1 locale tests");
if (setlocale (LC_CTYPE, "en_US.ISO-8859-1") == 0)
@@ -229,6 +229,75 @@ run_test (const char *locname)
STRTEST (YESSTR, "");
STRTEST (NOSTR, "");
+#define CONVTEST(b, v) \
+ { \
+ unsigned char bs[] = {b, 0}; \
+ mbstate_t ctx = {}; \
+ wchar_t wc = -1; \
+ size_t sz = mbrtowc(&wc, (char *) bs, 1, &ctx); \
+ if (sz != !!b) \
+ { \
+ printf ("mbrtowc(%02hhx) width in locale %s wrong " \
+ "(is %zd, should be %d)\n", *bs, locname, sz, !!b); \
+ result = 1; \
+ } \
+ if (wc != v) \
+ { \
+ printf ("mbrtowc(%02hhx) value in locale %s wrong " \
+ "(is %x, should be %x)\n", *bs, locname, wc, v); \
+ result = 1; \
+ } \
+ }
+ for(int i = 0; i <= 0x7f; ++i)
+ CONVTEST(i, i);
+ for(int i = 0x80; i <= 0xff; ++i)
+ CONVTEST(i, 0xdf00 + i);
+
+#define DECONVTEST(v, b) \
+ { \
+ unsigned char ob = -1; \
+ mbstate_t ctx = {}; \
+ size_t sz = wcrtomb((char *) &ob, v, &ctx); \
+ if (sz != 1) \
+ { \
+ printf ("wcrtomb(%x) width in locale %s wrong " \
+ "(is %zd, should be 1)\n", v, locname, sz); \
+ result = 1; \
+ } \
+ if (ob != b) \
+ { \
+ printf ("wcrtomb(%x) value in locale %s wrong " \
+ "(is %hhx, should be %hhx)\n", v, locname, ob, b); \
+ result = 1; \
+ } \
+ }
+#define DECONVERR(v) \
+ { \
+ unsigned char ob = -1; \
+ mbstate_t ctx = {}; \
+ size_t sz = wcrtomb((char *) &ob, v, &ctx); \
+ if (sz != (size_t) -1) \
+ { \
+ printf ("wcrtomb(%x) width in locale %s wrong " \
+ "(is %zd, should be (size_t )-1)\n", v, locname, sz); \
+ result = 1; \
+ } \
+ if (ob != (unsigned char) -1) \
+ { \
+ printf ("wcrtomb(%x) value in locale %s wrong " \
+ "(is %hhx, should be unchanged)\n", v, locname, ob); \
+ result = 1; \
+ } \
+ }
+ for(int i = 0; i <= 0x7f; ++i)
+ DECONVTEST(i, i);
+ for(int i = 0x80; i < 0xdf00; ++i)
+ DECONVERR(i);
+ for(int i = 0x80; i <= 0xff; ++i)
+ DECONVTEST(0xdf00 + i, i);
+ for(int i = 0xe000; i <= 0xffff; ++i)
+ DECONVERR(i);
+
/* Test the new locale mechanisms. */
loc = newlocale (LC_ALL_MASK, locname, NULL);
if (loc == NULL)
new file mode 100644
@@ -0,0 +1,136 @@
+<code_set_name> POSIX
+<comment_char> %
+<escape_char> /
+% source: cf. localedata/locales/POSIX, LC_COLLATE
+
+CHARMAP
+<U0000> /x00 NULL (NUL)
+<U0001> /x01 START OF HEADING (SOH)
+<U0002> /x02 START OF TEXT (STX)
+<U0003> /x03 END OF TEXT (ETX)
+<U0004> /x04 END OF TRANSMISSION (EOT)
+<U0005> /x05 ENQUIRY (ENQ)
+<U0006> /x06 ACKNOWLEDGE (ACK)
+<U0007> /x07 BELL (BEL)
+<U0008> /x08 BACKSPACE (BS)
+<U0009> /x09 CHARACTER TABULATION (HT)
+<U000A> /x0a LINE FEED (LF)
+<U000B> /x0b LINE TABULATION (VT)
+<U000C> /x0c FORM FEED (FF)
+<U000D> /x0d CARRIAGE RETURN (CR)
+<U000E> /x0e SHIFT OUT (SO)
+<U000F> /x0f SHIFT IN (SI)
+<U0010> /x10 DATALINK ESCAPE (DLE)
+<U0011> /x11 DEVICE CONTROL ONE (DC1)
+<U0012> /x12 DEVICE CONTROL TWO (DC2)
+<U0013> /x13 DEVICE CONTROL THREE (DC3)
+<U0014> /x14 DEVICE CONTROL FOUR (DC4)
+<U0015> /x15 NEGATIVE ACKNOWLEDGE (NAK)
+<U0016> /x16 SYNCHRONOUS IDLE (SYN)
+<U0017> /x17 END OF TRANSMISSION BLOCK (ETB)
+<U0018> /x18 CANCEL (CAN)
+<U0019> /x19 END OF MEDIUM (EM)
+<U001A> /x1a SUBSTITUTE (SUB)
+<U001B> /x1b ESCAPE (ESC)
+<U001C> /x1c FILE SEPARATOR (IS4)
+<U001D> /x1d GROUP SEPARATOR (IS3)
+<U001E> /x1e RECORD SEPARATOR (IS2)
+<U001F> /x1f UNIT SEPARATOR (IS1)
+<U0020> /x20 SPACE
+<U0021> /x21 EXCLAMATION MARK
+<U0022> /x22 QUOTATION MARK
+<U0023> /x23 NUMBER SIGN
+<U0024> /x24 DOLLAR SIGN
+<U0025> /x25 PERCENT SIGN
+<U0026> /x26 AMPERSAND
+<U0027> /x27 APOSTROPHE
+<U0028> /x28 LEFT PARENTHESIS
+<U0029> /x29 RIGHT PARENTHESIS
+<U002A> /x2a ASTERISK
+<U002B> /x2b PLUS SIGN
+<U002C> /x2c COMMA
+<U002D> /x2d HYPHEN-MINUS
+<U002E> /x2e FULL STOP
+<U002F> /x2f SOLIDUS
+<U0030> /x30 DIGIT ZERO
+<U0031> /x31 DIGIT ONE
+<U0032> /x32 DIGIT TWO
+<U0033> /x33 DIGIT THREE
+<U0034> /x34 DIGIT FOUR
+<U0035> /x35 DIGIT FIVE
+<U0036> /x36 DIGIT SIX
+<U0037> /x37 DIGIT SEVEN
+<U0038> /x38 DIGIT EIGHT
+<U0039> /x39 DIGIT NINE
+<U003A> /x3a COLON
+<U003B> /x3b SEMICOLON
+<U003C> /x3c LESS-THAN SIGN
+<U003D> /x3d EQUALS SIGN
+<U003E> /x3e GREATER-THAN SIGN
+<U003F> /x3f QUESTION MARK
+<U0040> /x40 COMMERCIAL AT
+<U0041> /x41 LATIN CAPITAL LETTER A
+<U0042> /x42 LATIN CAPITAL LETTER B
+<U0043> /x43 LATIN CAPITAL LETTER C
+<U0044> /x44 LATIN CAPITAL LETTER D
+<U0045> /x45 LATIN CAPITAL LETTER E
+<U0046> /x46 LATIN CAPITAL LETTER F
+<U0047> /x47 LATIN CAPITAL LETTER G
+<U0048> /x48 LATIN CAPITAL LETTER H
+<U0049> /x49 LATIN CAPITAL LETTER I
+<U004A> /x4a LATIN CAPITAL LETTER J
+<U004B> /x4b LATIN CAPITAL LETTER K
+<U004C> /x4c LATIN CAPITAL LETTER L
+<U004D> /x4d LATIN CAPITAL LETTER M
+<U004E> /x4e LATIN CAPITAL LETTER N
+<U004F> /x4f LATIN CAPITAL LETTER O
+<U0050> /x50 LATIN CAPITAL LETTER P
+<U0051> /x51 LATIN CAPITAL LETTER Q
+<U0052> /x52 LATIN CAPITAL LETTER R
+<U0053> /x53 LATIN CAPITAL LETTER S
+<U0054> /x54 LATIN CAPITAL LETTER T
+<U0055> /x55 LATIN CAPITAL LETTER U
+<U0056> /x56 LATIN CAPITAL LETTER V
+<U0057> /x57 LATIN CAPITAL LETTER W
+<U0058> /x58 LATIN CAPITAL LETTER X
+<U0059> /x59 LATIN CAPITAL LETTER Y
+<U005A> /x5a LATIN CAPITAL LETTER Z
+<U005B> /x5b LEFT SQUARE BRACKET
+<U005C> /x5c REVERSE SOLIDUS
+<U005D> /x5d RIGHT SQUARE BRACKET
+<U005E> /x5e CIRCUMFLEX ACCENT
+<U005F> /x5f LOW LINE
+<U0060> /x60 GRAVE ACCENT
+<U0061> /x61 LATIN SMALL LETTER A
+<U0062> /x62 LATIN SMALL LETTER B
+<U0063> /x63 LATIN SMALL LETTER C
+<U0064> /x64 LATIN SMALL LETTER D
+<U0065> /x65 LATIN SMALL LETTER E
+<U0066> /x66 LATIN SMALL LETTER F
+<U0067> /x67 LATIN SMALL LETTER G
+<U0068> /x68 LATIN SMALL LETTER H
+<U0069> /x69 LATIN SMALL LETTER I
+<U006A> /x6a LATIN SMALL LETTER J
+<U006B> /x6b LATIN SMALL LETTER K
+<U006C> /x6c LATIN SMALL LETTER L
+<U006D> /x6d LATIN SMALL LETTER M
+<U006E> /x6e LATIN SMALL LETTER N
+<U006F> /x6f LATIN SMALL LETTER O
+<U0070> /x70 LATIN SMALL LETTER P
+<U0071> /x71 LATIN SMALL LETTER Q
+<U0072> /x72 LATIN SMALL LETTER R
+<U0073> /x73 LATIN SMALL LETTER S
+<U0074> /x74 LATIN SMALL LETTER T
+<U0075> /x75 LATIN SMALL LETTER U
+<U0076> /x76 LATIN SMALL LETTER V
+<U0077> /x77 LATIN SMALL LETTER W
+<U0078> /x78 LATIN SMALL LETTER X
+<U0079> /x79 LATIN SMALL LETTER Y
+<U007A> /x7a LATIN SMALL LETTER Z
+<U007B> /x7b LEFT CURLY BRACKET
+<U007C> /x7c VERTICAL LINE
+<U007D> /x7d RIGHT CURLY BRACKET
+<U007E> /x7e TILDE
+<U007F> /x7f DELETE (DEL)
+<UDF80>..<UDFFF> /x80
+END CHARMAP
@@ -97,6 +97,20 @@ END LC_CTYPE
LC_COLLATE
% This is the POSIX Locale definition for the LC_COLLATE category.
% The order is the same as in the ASCII code set.
+% Values above <DEL> (<U007F>) inserted in order, per Issue 7 TC2,
+% XBD, 7.3.2, LC_COLLATE Category in the POSIX Locale:
+% > All characters not explicitly listed here shall be inserted
+% > in the character collation order after the listed characters
+% > and shall be assigned unique primary weights. If the listed
+% > characters have ASCII encoding, the other characters shall
+% > be in ascending order according to their coded character set values
+% Since Issue 7 TC2 (XBD, 6.2 Character Encoding):
+% > The POSIX locale shall contain 256 single-byte characters [...]
+% (cf. bug 663, 674).
+% this is in contrast to previous issues, which limited the POSIX
+% locale to the Portable Character Set (7-bit ASCII).
+% We use the end of the Low Surrogate Area to contain these,
+% yielding [<UDF80>, <UDFFF>]
order_start forward
<U0000>
<U0001>
@@ -226,7 +240,134 @@ order_start forward
<U007D>
<U007E>
<U007F>
-UNDEFINED
+<UDF80>
+<UDF81>
+<UDF82>
+<UDF83>
+<UDF84>
+<UDF85>
+<UDF86>
+<UDF87>
+<UDF88>
+<UDF89>
+<UDF8A>
+<UDF8B>
+<UDF8C>
+<UDF8D>
+<UDF8E>
+<UDF8F>
+<UDF90>
+<UDF91>
+<UDF92>
+<UDF93>
+<UDF94>
+<UDF95>
+<UDF96>
+<UDF97>
+<UDF98>
+<UDF99>
+<UDF9A>
+<UDF9B>
+<UDF9C>
+<UDF9D>
+<UDF9E>
+<UDF9F>
+<UDFA0>
+<UDFA1>
+<UDFA2>
+<UDFA3>
+<UDFA4>
+<UDFA5>
+<UDFA6>
+<UDFA7>
+<UDFA8>
+<UDFA9>
+<UDFAA>
+<UDFAB>
+<UDFAC>
+<UDFAD>
+<UDFAE>
+<UDFAF>
+<UDFB0>
+<UDFB1>
+<UDFB2>
+<UDFB3>
+<UDFB4>
+<UDFB5>
+<UDFB6>
+<UDFB7>
+<UDFB8>
+<UDFB9>
+<UDFBA>
+<UDFBB>
+<UDFBC>
+<UDFBD>
+<UDFBE>
+<UDFBF>
+<UDFC0>
+<UDFC1>
+<UDFC2>
+<UDFC3>
+<UDFC4>
+<UDFC5>
+<UDFC6>
+<UDFC7>
+<UDFC8>
+<UDFC9>
+<UDFCA>
+<UDFCB>
+<UDFCC>
+<UDFCD>
+<UDFCE>
+<UDFCF>
+<UDFD0>
+<UDFD1>
+<UDFD2>
+<UDFD3>
+<UDFD4>
+<UDFD5>
+<UDFD6>
+<UDFD7>
+<UDFD8>
+<UDFD9>
+<UDFDA>
+<UDFDB>
+<UDFDC>
+<UDFDD>
+<UDFDE>
+<UDFDF>
+<UDFE0>
+<UDFE1>
+<UDFE2>
+<UDFE3>
+<UDFE4>
+<UDFE5>
+<UDFE6>
+<UDFE7>
+<UDFE8>
+<UDFE9>
+<UDFEA>
+<UDFEB>
+<UDFEC>
+<UDFED>
+<UDFEE>
+<UDFEF>
+<UDFF0>
+<UDFF1>
+<UDFF2>
+<UDFF3>
+<UDFF4>
+<UDFF5>
+<UDFF6>
+<UDFF7>
+<UDFF8>
+<UDFF9>
+<UDFFA>
+<UDFFB>
+<UDFFC>
+<UDFFD>
+<UDFFE>
+<UDFFF>
order_end
%
END LC_COLLATE
@@ -30,6 +30,8 @@
static int
do_test (void)
{
+ setlocale(LC_CTYPE, "C.UTF-8");
+
mtrace ();
/* For 's' conversion specifier with 'l' modifier the array must be
@@ -33,10 +33,10 @@ static const struct __gconv_step to_wc =
.__shlib_handle = NULL,
.__modname = NULL,
.__counter = INT_MAX,
- .__from_name = (char *) "ANSI_X3.4-1968//TRANSLIT",
+ .__from_name = (char *) "POSIX",
.__to_name = (char *) "INTERNAL",
- .__fct = __gconv_transform_ascii_internal,
- .__btowc_fct = __gconv_btwoc_ascii,
+ .__fct = __gconv_transform_posix_internal,
+ .__btowc_fct = __gconv_btwoc_posix,
.__init_fct = NULL,
.__end_fct = NULL,
.__min_needed_from = 1,
@@ -53,8 +53,8 @@ static const struct __gconv_step to_mb =
.__modname = NULL,
.__counter = INT_MAX,
.__from_name = (char *) "INTERNAL",
- .__to_name = (char *) "ANSI_X3.4-1968//TRANSLIT",
- .__fct = __gconv_transform_internal_ascii,
+ .__to_name = (char *) "POSIX",
+ .__fct = __gconv_transform_internal_posix,
.__btowc_fct = NULL,
.__init_fct = NULL,
.__end_fct = NULL,
@@ -67,7 +67,9 @@ static const struct __gconv_step to_mb =
};
-/* For the default locale we only have to handle ANSI_X3.4-1968. */
+/* The default/"POSIX"/"C" locale is an 8-bit-clean mapping
+ with ANSI_X3.4-1968 in the first 128 characters;
+ we lift the remaining bytes by <UDF00>. */
const struct gconv_fcts __wcsmbs_gconv_fcts_c =
{
.towc = (struct __gconv_step *) &to_wc,