From 8906a5101bf62b719ddafa376fd4e0805e35617b Mon Sep 17 00:00:00 2001 From: Bruno Haible Date: Thu, 30 Mar 2023 23:15:52 +0200 Subject: [PATCH] mbsnrtowcs: Fix behaviour in the C locale. * m4/mbsnrtowcs.m4 (gl_FUNC_MBSNRTOWCS): Invoke gl_MBRTOWC_C_LOCALE. If mbrtowc is buggy in the C locale, override also mbsnrtowcs. * modules/mbsnrtowcs (Files): Add m4/mbrtowc.m4. * tests/test-mbsnrtowcs.c (main): Add a test of the C locale, based on tests/test-mbsrtowcs.c. * tests/test-mbsnrtowcs5.sh: New file, based on tests/test-mbrtowc5.sh. * modules/mbsnrtowcs-tests (Files): Add it. (Makefile.am): Test it. * doc/posix-functions/mbsnrtowcs.texi: Mention the C locale behaviour bug. --- ChangeLog | 14 ++++++ doc/posix-functions/mbsnrtowcs.texi | 4 ++ m4/mbsnrtowcs.m4 | 9 +++- modules/mbsnrtowcs | 1 + modules/mbsnrtowcs-tests | 5 ++- tests/test-mbsnrtowcs.c | 66 +++++++++++++++++++++++++++++ tests/test-mbsnrtowcs5.sh | 9 ++++ 7 files changed, 106 insertions(+), 2 deletions(-) create mode 100755 tests/test-mbsnrtowcs5.sh diff --git a/ChangeLog b/ChangeLog index 2fe1679061..4afba58a3c 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,17 @@ +2023-03-30 Bruno Haible + + mbsnrtowcs: Fix behaviour in the C locale. + * m4/mbsnrtowcs.m4 (gl_FUNC_MBSNRTOWCS): Invoke gl_MBRTOWC_C_LOCALE. If + mbrtowc is buggy in the C locale, override also mbsnrtowcs. + * modules/mbsnrtowcs (Files): Add m4/mbrtowc.m4. + * tests/test-mbsnrtowcs.c (main): Add a test of the C locale, based on + tests/test-mbsrtowcs.c. + * tests/test-mbsnrtowcs5.sh: New file, based on tests/test-mbrtowc5.sh. + * modules/mbsnrtowcs-tests (Files): Add it. + (Makefile.am): Test it. + * doc/posix-functions/mbsnrtowcs.texi: Mention the C locale behaviour + bug. + 2023-03-30 Bruno Haible mbsrtowcs: Fix behaviour in the C locale. diff --git a/doc/posix-functions/mbsnrtowcs.texi b/doc/posix-functions/mbsnrtowcs.texi index c6defd2c20..3a420c673a 100644 --- a/doc/posix-functions/mbsnrtowcs.texi +++ b/doc/posix-functions/mbsnrtowcs.texi @@ -14,6 +14,10 @@ FreeBSD 5.2.1, NetBSD 5.0, OpenBSD 3.8, Minix 3.1.8, HP-UX 11, IRIX 6.5, Solaris @item This function produces invalid wide characters on some platforms: Solaris 11.4. +@item +In the C or POSIX locales, this function can return @code{(size_t) -1} +and set @code{errno} to @code{EILSEQ}: +glibc 2.35. @end itemize Portability problems not fixed by Gnulib: diff --git a/m4/mbsnrtowcs.m4 b/m4/mbsnrtowcs.m4 index 34dcf30e63..7cab5f79ed 100644 --- a/m4/mbsnrtowcs.m4 +++ b/m4/mbsnrtowcs.m4 @@ -1,4 +1,4 @@ -# mbsnrtowcs.m4 serial 8 +# mbsnrtowcs.m4 serial 9 dnl Copyright (C) 2008, 2010-2023 Free Software Foundation, Inc. dnl This file is free software; the Free Software Foundation dnl gives unlimited permission to copy and/or distribute it, @@ -29,6 +29,13 @@ AC_DEFUN([gl_FUNC_MBSNRTOWCS], *yes) ;; *) REPLACE_MBSNRTOWCS=1 ;; esac + if test $REPLACE_MBSNRTOWCS = 0; then + gl_MBRTOWC_C_LOCALE + case "$gl_cv_func_mbrtowc_C_locale_sans_EILSEQ" in + *yes) ;; + *) REPLACE_MBSNRTOWCS=1 ;; + esac + fi fi fi ]) diff --git a/modules/mbsnrtowcs b/modules/mbsnrtowcs index 5b66cc2b8f..10260246ea 100644 --- a/modules/mbsnrtowcs +++ b/modules/mbsnrtowcs @@ -7,6 +7,7 @@ lib/mbsnrtowcs-impl.h lib/mbsrtowcs-state.c m4/mbsnrtowcs.m4 m4/mbstate_t.m4 +m4/mbrtowc.m4 Depends-on: wchar diff --git a/modules/mbsnrtowcs-tests b/modules/mbsnrtowcs-tests index c7c528448f..7c5de260ba 100644 --- a/modules/mbsnrtowcs-tests +++ b/modules/mbsnrtowcs-tests @@ -3,6 +3,7 @@ tests/test-mbsnrtowcs1.sh tests/test-mbsnrtowcs2.sh tests/test-mbsnrtowcs3.sh tests/test-mbsnrtowcs4.sh +tests/test-mbsnrtowcs5.sh tests/test-mbsnrtowcs.c tests/signature.h tests/macros.h @@ -24,7 +25,9 @@ gt_LOCALE_JA gt_LOCALE_ZH_CN Makefile.am: -TESTS += test-mbsnrtowcs1.sh test-mbsnrtowcs2.sh test-mbsnrtowcs3.sh test-mbsnrtowcs4.sh +TESTS += \ + test-mbsnrtowcs1.sh test-mbsnrtowcs2.sh test-mbsnrtowcs3.sh \ + test-mbsnrtowcs4.sh test-mbsnrtowcs5.sh TESTS_ENVIRONMENT += \ LOCALE_FR='@LOCALE_FR@' \ LOCALE_FR_UTF8='@LOCALE_FR_UTF8@' \ diff --git a/tests/test-mbsnrtowcs.c b/tests/test-mbsnrtowcs.c index e3456b4e2b..2d0e6a7521 100644 --- a/tests/test-mbsnrtowcs.c +++ b/tests/test-mbsnrtowcs.c @@ -281,6 +281,72 @@ main (int argc, char *argv[]) } break; + case '5': + /* C or POSIX locale. */ + { + char input[] = "n/a"; + memset (&state, '\0', sizeof (mbstate_t)); + + src = input; + temp_state = state; + ret = mbsnrtowcs (NULL, &src, 4, unlimited ? BUFSIZE : 1, &temp_state); + ASSERT (ret == 3); + ASSERT (src == input); + ASSERT (mbsinit (&state)); + + src = input; + ret = mbsnrtowcs (buf, &src, 4, unlimited ? BUFSIZE : 1, &state); + ASSERT (ret == (unlimited ? 3 : 1)); + ASSERT (src == (unlimited ? NULL : input + 1)); + ASSERT (buf[0] == 'n'); + if (unlimited) + { + ASSERT (buf[1] == '/'); + ASSERT (buf[2] == 'a'); + ASSERT (buf[3] == 0); + ASSERT (buf[4] == (wchar_t) 0xBADFACE); + } + else + ASSERT (buf[1] == (wchar_t) 0xBADFACE); + ASSERT (mbsinit (&state)); + } + { + int c; + char input[2]; + + memset (&state, '\0', sizeof (mbstate_t)); + for (c = 0; c < 0x100; c++) + if (c != 0) + { + /* We are testing all nonnull bytes. */ + input[0] = c; + input[1] = '\0'; + + src = input; + ret = mbsnrtowcs (NULL, &src, 2, unlimited ? BUFSIZE : 1, &state); + ASSERT (ret == 1); + ASSERT (src == input); + ASSERT (mbsinit (&state)); + + buf[0] = buf[1] = (wchar_t) 0xBADFACE; + src = input; + ret = mbsnrtowcs (buf, &src, 2, unlimited ? BUFSIZE : 1, &state); + /* POSIX:2018 says: "In the POSIX locale an [EILSEQ] error + cannot occur since all byte values are valid characters." */ + ASSERT (ret == 1); + ASSERT (src == (unlimited ? NULL : input + 1)); + if (c < 0x80) + /* c is an ASCII character. */ + ASSERT (buf[0] == c); + else + /* On most platforms, the bytes 0x80..0xFF map to U+0080..U+00FF. + But on musl libc, the bytes 0x80..0xFF map to U+DF80..U+DFFF. */ + ASSERT (buf[0] == (btowc (c) == 0xDF00 + c ? btowc (c) : c)); + ASSERT (mbsinit (&state)); + } + } + break; + default: return 1; } diff --git a/tests/test-mbsnrtowcs5.sh b/tests/test-mbsnrtowcs5.sh new file mode 100755 index 0000000000..b121fb1a69 --- /dev/null +++ b/tests/test-mbsnrtowcs5.sh @@ -0,0 +1,9 @@ +#!/bin/sh + +# Test whether the POSIX locale has encoding errors. +LC_ALL=C \ +${CHECKER} ./test-mbsnrtowcs${EXEEXT} 5 || exit 1 +LC_ALL=POSIX \ +${CHECKER} ./test-mbsnrtowcs${EXEEXT} 5 || exit 1 + +exit 0 -- 2.39.5