I wrote:
> In Gnulib, a workaround for this issue was added to the 'mbrtowc' module
> in 2016. But this is not the only function affected by this issue.
That's not all. *wprintf also invokes mbrtowc, for the %c directive and
the %s directive. It is affected by the glibc bug:
==================== foo.c ===================
#include <stdio.h>
#include <wchar.h>
int main ()
{
wchar_t buf[10];
int ret;
ret = swprintf (buf, 10, L"%c", 0x80);
printf ("ret = %d\n", ret);
ret = swprintf (buf, 10, L"%s", "\200");
printf ("ret = %d\n", ret);
}
==============================================
$ gcc -Wall foo.c && ./a.out
ret = -1
ret = -1
After adding corresponding unit tests to
test-vasnwprintf-posix.c, the %c tests indeed fail.
For the %s directive, vasnprintf.c already contains its own code,
that invokes the overridden mbrlen and mbrtowc functions, therefore
the %s tests already pass.
2023-04-01 Bruno Haible <br...@clisp.org>
vasnwprintf-posix: Fix behaviour in the C locale.
* lib/vasnprintf.c (VASNPRINTF): If NEED_WPRINTF_DIRECTIVE_C is set,
implement the 'c' directive here.
* m4/vasnprintf.m4 (gl_PREREQ_VASNWPRINTF): Invoke gl_MBRTOWC_C_LOCALE.
If mbrtowc is buggy in the C locale, define NEED_WPRINTF_DIRECTIVE_C.
* modules/vasnwprintf (Files): Add m4/mbrtowc.m4.
* tests/test-vasnwprintf-posix.c (test_function): Add tests of %s and %c
in the C locale.
* doc/posix-functions/fwprintf.texi: Mention the C locale behaviour bug.
* doc/posix-functions/swprintf.texi: Likewise.
* doc/posix-functions/vfwprintf.texi: Likewise.
* doc/posix-functions/vswprintf.texi: Likewise.
* doc/posix-functions/vwprintf.texi: Likewise.
* doc/posix-functions/wprintf.texi: Likewise.
diff --git a/doc/posix-functions/fwprintf.texi
b/doc/posix-functions/fwprintf.texi
index 2b4600de11..db42363cfa 100644
--- a/doc/posix-functions/fwprintf.texi
+++ b/doc/posix-functions/fwprintf.texi
@@ -34,6 +34,10 @@
argument of @code{strerror(errno)} (or a version of @code{strerror_r})
instead.
@item
+In the C or POSIX locales, the @code{%c} and @code{%s} conversions may fail
+on some platforms:
+glibc 2.35.
+@item
When formatting an integer with grouping flag, this function inserts thousands
separators even in the "C" locale on some platforms:
NetBSD 5.1.
diff --git a/doc/posix-functions/swprintf.texi
b/doc/posix-functions/swprintf.texi
index b53c96c6d7..cd20d0e702 100644
--- a/doc/posix-functions/swprintf.texi
+++ b/doc/posix-functions/swprintf.texi
@@ -63,6 +63,10 @@
argument of @code{strerror(errno)} (or a version of @code{strerror_r})
instead.
@item
+In the C or POSIX locales, the @code{%c} and @code{%s} conversions may fail
+on some platforms:
+glibc 2.35.
+@item
When formatting an integer with grouping flag, this function inserts thousands
separators even in the "C" locale on some platforms:
NetBSD 5.1.
diff --git a/doc/posix-functions/vfwprintf.texi
b/doc/posix-functions/vfwprintf.texi
index 22e4a59212..0712269aa4 100644
--- a/doc/posix-functions/vfwprintf.texi
+++ b/doc/posix-functions/vfwprintf.texi
@@ -34,6 +34,10 @@
argument of @code{strerror(errno)} (or a version of @code{strerror_r})
instead.
@item
+In the C or POSIX locales, the @code{%c} and @code{%s} conversions may fail
+on some platforms:
+glibc 2.35.
+@item
When formatting an integer with grouping flag, this function inserts thousands
separators even in the "C" locale on some platforms:
NetBSD 5.1.
diff --git a/doc/posix-functions/vswprintf.texi
b/doc/posix-functions/vswprintf.texi
index 4083cfd2a1..7512a2d023 100644
--- a/doc/posix-functions/vswprintf.texi
+++ b/doc/posix-functions/vswprintf.texi
@@ -36,6 +36,10 @@
argument of @code{strerror(errno)} (or a version of @code{strerror_r})
instead.
@item
+In the C or POSIX locales, the @code{%c} and @code{%s} conversions may fail
+on some platforms:
+glibc 2.35.
+@item
When formatting an integer with grouping flag, this function inserts thousands
separators even in the "C" locale on some platforms:
NetBSD 5.1.
diff --git a/doc/posix-functions/vwprintf.texi
b/doc/posix-functions/vwprintf.texi
index 7b029c9ceb..1e364288f6 100644
--- a/doc/posix-functions/vwprintf.texi
+++ b/doc/posix-functions/vwprintf.texi
@@ -37,6 +37,10 @@
argument of @code{strerror(errno)} (or a version of @code{strerror_r})
instead.
@item
+In the C or POSIX locales, the @code{%c} and @code{%s} conversions may fail
+on some platforms:
+glibc 2.35.
+@item
When formatting an integer with grouping flag, this function inserts thousands
separators even in the "C" locale on some platforms:
NetBSD 5.1.
diff --git a/doc/posix-functions/wprintf.texi b/doc/posix-functions/wprintf.texi
index 56106b309e..c4d58fab2e 100644
--- a/doc/posix-functions/wprintf.texi
+++ b/doc/posix-functions/wprintf.texi
@@ -37,6 +37,10 @@
argument of @code{strerror(errno)} (or a version of @code{strerror_r})
instead.
@item
+In the C or POSIX locales, the @code{%c} and @code{%s} conversions may fail
+on some platforms:
+glibc 2.35.
+@item
When formatting an integer with grouping flag, this function inserts thousands
separators even in the "C" locale on some platforms:
NetBSD 5.1.
diff --git a/lib/vasnprintf.c b/lib/vasnprintf.c
index 90e52d8a96..111d898a94 100644
--- a/lib/vasnprintf.c
+++ b/lib/vasnprintf.c
@@ -3543,6 +3543,77 @@ VASNPRINTF (DCHAR_T *resultbuf, size_t *lengthp,
}
}
#endif
+#if NEED_WPRINTF_DIRECTIVE_C && WIDE_CHAR_VERSION
+ else if (dp->conversion == 'c'
+ && a.arg[dp->arg_index].type != TYPE_WIDE_CHAR)
+ {
+ /* Implement the 'c' directive ourselves, in order to avoid
+ EILSEQ in the "C" locale. */
+ int flags = dp->flags;
+ size_t width;
+
+ width = 0;
+ if (dp->width_start != dp->width_end)
+ {
+ if (dp->width_arg_index != ARG_NONE)
+ {
+ int arg;
+
+ if (!(a.arg[dp->width_arg_index].type == TYPE_INT))
+ abort ();
+ arg = a.arg[dp->width_arg_index].a.a_int;
+ width = arg;
+ if (arg < 0)
+ {
+ /* "A negative field width is taken as a '-' flag
+ followed by a positive field width." */
+ flags |= FLAG_LEFT;
+ width = -width;
+ }
+ }
+ else
+ {
+ const FCHAR_T *digitp = dp->width_start;
+
+ do
+ width = xsum (xtimes (width, 10), *digitp++ - '0');
+ while (digitp != dp->width_end);
+ }
+ }
+
+ /* %c in vasnwprintf. See the specification of fwprintf. */
+ {
+ char arg = (char) a.arg[dp->arg_index].a.a_char;
+ mbstate_t state;
+ wchar_t wc;
+
+ memset (&state, '\0', sizeof (mbstate_t));
+ int count = mbrtowc (&wc, &arg, 1, &state);
+ if (count < 0)
+ /* Invalid or incomplete multibyte character. */
+ goto fail_with_EILSEQ;
+
+ if (1 < width && !(flags & FLAG_LEFT))
+ {
+ size_t n = width - 1;
+ ENSURE_ALLOCATION (xsum (length, n));
+ DCHAR_SET (result + length, ' ', n);
+ length += n;
+ }
+
+ ENSURE_ALLOCATION (xsum (length, 1));
+ result[length++] = wc;
+
+ if (1 < width && (flags & FLAG_LEFT))
+ {
+ size_t n = width - 1;
+ ENSURE_ALLOCATION (xsum (length, n));
+ DCHAR_SET (result + length, ' ', n);
+ length += n;
+ }
+ }
+ }
+#endif
#if NEED_PRINTF_DIRECTIVE_B || NEED_PRINTF_DIRECTIVE_UPPERCASE_B
else if (0
# if NEED_PRINTF_DIRECTIVE_B
@@ -6816,7 +6887,7 @@ VASNPRINTF (DCHAR_T *resultbuf, size_t *lengthp,
errno = ENOMEM;
goto fail_with_errno;
-#if ENABLE_UNISTDIO || ((!USE_SNPRINTF || WIDE_CHAR_VERSION ||
!HAVE_SNPRINTF_RETVAL_C99 || USE_MSVC__SNPRINTF || NEED_PRINTF_DIRECTIVE_LS ||
ENABLE_WCHAR_FALLBACK) && HAVE_WCHAR_T) || ((NEED_PRINTF_DIRECTIVE_LC ||
ENABLE_WCHAR_FALLBACK) && HAVE_WINT_T && !WIDE_CHAR_VERSION)
+#if ENABLE_UNISTDIO || ((!USE_SNPRINTF || WIDE_CHAR_VERSION ||
!HAVE_SNPRINTF_RETVAL_C99 || USE_MSVC__SNPRINTF || NEED_PRINTF_DIRECTIVE_LS ||
ENABLE_WCHAR_FALLBACK) && HAVE_WCHAR_T) || ((NEED_PRINTF_DIRECTIVE_LC ||
ENABLE_WCHAR_FALLBACK) && HAVE_WINT_T && !WIDE_CHAR_VERSION) ||
(NEED_WPRINTF_DIRECTIVE_C && WIDE_CHAR_VERSION)
fail_with_EILSEQ:
errno = EILSEQ;
goto fail_with_errno;
diff --git a/m4/vasnprintf.m4 b/m4/vasnprintf.m4
index 929f3d80af..86133b30b1 100644
--- a/m4/vasnprintf.m4
+++ b/m4/vasnprintf.m4
@@ -1,4 +1,4 @@
-# vasnprintf.m4 serial 47
+# vasnprintf.m4 serial 48
dnl Copyright (C) 2002-2004, 2006-2023 Free Software Foundation, Inc.
dnl This file is free software; the Free Software Foundation
dnl gives unlimited permission to copy and/or distribute it,
@@ -105,6 +105,15 @@ AC_DEFUN_ONCE([gl_PREREQ_VASNWPRINTF]
that contains null wide characters.])
;;
esac
+ gl_MBRTOWC_C_LOCALE
+ case "$gl_cv_func_mbrtowc_C_locale_sans_EILSEQ" in
+ *yes) ;;
+ *)
+ AC_DEFINE([NEED_WPRINTF_DIRECTIVE_C], [1],
+ [Define if the vasnwprintf implementation needs special code for
+ the 'c' directive.])
+ ;;
+ esac
gl_MUSL_LIBC
gl_PREREQ_VASNXPRINTF
])
diff --git a/modules/vasnwprintf b/modules/vasnwprintf
index 82d4aa4668..70f3fb2c95 100644
--- a/modules/vasnwprintf
+++ b/modules/vasnwprintf
@@ -19,6 +19,7 @@ m4/stdint_h.m4
m4/inttypes_h.m4
m4/vasnprintf.m4
m4/printf.m4
+m4/mbrtowc.m4
m4/math_h.m4
m4/exponentd.m4
m4/musl.m4
diff --git a/tests/test-vasnwprintf-posix.c b/tests/test-vasnwprintf-posix.c
index 74a26079d3..125b9386a4 100644
--- a/tests/test-vasnwprintf-posix.c
+++ b/tests/test-vasnwprintf-posix.c
@@ -3757,6 +3757,28 @@ test_function (wchar_t * (*my_asnwprintf) (wchar_t *,
size_t *, const wchar_t *,
free (result);
}
+ /* On Android ≥ 5.0, the default locale is the "C.UTF-8" locale, not the
+ "C" locale. Furthermore, when you attempt to set the "C" or "POSIX"
+ locale via setlocale(), what you get is a "C" locale with UTF-8 encoding,
+ that is, effectively the "C.UTF-8" locale. */
+#ifndef __ANDROID__
+ { /* The conversion of %s to wide characters is done as if through repeated
+ invocations of mbrtowc(), and in the "C" and "POSIX" locales, "an
+ [EILSEQ] error cannot occur since all byte values are valid characters",
+ says POSIX:2018. */
+ int c;
+
+ for (c = 0; c < 0x100; c++)
+ {
+ char s[2] = { c, '\0' };
+ size_t length;
+ wchar_t *result = my_asnwprintf (NULL, &length, L"%s", s);
+ ASSERT (result != NULL);
+ free (result);
+ }
+ }
+#endif
+
#if HAVE_WCHAR_T
static wchar_t L_xyz[4] = { 'x', 'y', 'z', 0 };
@@ -3960,6 +3982,26 @@ test_function (wchar_t * (*my_asnwprintf) (wchar_t *,
size_t *, const wchar_t *,
free (result);
}
+ /* On Android ≥ 5.0, the default locale is the "C.UTF-8" locale, not the
+ "C" locale. Furthermore, when you attempt to set the "C" or "POSIX"
+ locale via setlocale(), what you get is a "C" locale with UTF-8 encoding,
+ that is, effectively the "C.UTF-8" locale. */
+#ifndef __ANDROID__
+ { /* The conversion of %c to wide character is done as if through btowc(),
+ and in the "C" and "POSIX" locales, "btowc() shall not return WEOF if
+ c has a value in the range 0 to 255 inclusive", says POSIX:2018. */
+ int c;
+
+ for (c = 0; c < 0x100; c++)
+ {
+ size_t length;
+ wchar_t *result = my_asnwprintf (NULL, &length, L"%c", c);
+ ASSERT (result != NULL);
+ free (result);
+ }
+ }
+#endif
+
#if HAVE_WCHAR_T
static wint_t L_x = (wchar_t) 'x';
