Fix wchar tests for zero length conversions.
Same as the previous change for the uchar tests: the tests are wrong to match bionic's wrong implementation. Fix the test to encode the bug for now while I get the tests into good shape, then I'll be back to fix the bugs and remove the test differences. Bug: None Test: this is a test Change-Id: I1123660994f755f8bac1f2656f6890d5a43310b3
This commit is contained in:
Dan Albert
parent
686e67d077
commit
512469a858
1 changed files with 94 additions and 34 deletions
|
|
@ -38,6 +38,31 @@ static locale_t SAFE_LC_GLOBAL_LOCALE = duplocale(LC_GLOBAL_LOCALE);
|
|||
static locale_t SAFE_LC_GLOBAL_LOCALE = LC_GLOBAL_LOCALE;
|
||||
#endif
|
||||
|
||||
// C23 7.31.6.3.2 (mbrtowc) says:
|
||||
//
|
||||
// Returns:
|
||||
//
|
||||
// 0 if the next n or fewer bytes complete the multibyte character that
|
||||
// corresponds to the null wide character (which is the value stored).
|
||||
//
|
||||
// (size_t)(-2) if the next n bytes contribute to an incomplete (but
|
||||
// potentially valid) multibyte character, and all n bytes have been
|
||||
// processed (no value is stored).
|
||||
//
|
||||
// Bionic historically interpreted the behavior when n is 0 to be the next 0
|
||||
// bytes decoding to the null. That's a pretty bad interpretation, and both
|
||||
// glibc and musl return -2 for that case.
|
||||
//
|
||||
// The tests currently checks the incorrect behavior for bionic because gtest
|
||||
// doesn't support explicit xfail annotations. The behavior difference here
|
||||
// should be fixed, but danalbert wants to add more tests before tackling the
|
||||
// bugs.
|
||||
#ifdef __ANDROID__
|
||||
constexpr size_t kExpectedResultForZeroLength = 0U;
|
||||
#else
|
||||
constexpr size_t kExpectedResultForZeroLength = static_cast<size_t>(-2);
|
||||
#endif
|
||||
|
||||
TEST(wchar, sizeof_wchar_t) {
|
||||
EXPECT_EQ(4U, sizeof(wchar_t));
|
||||
EXPECT_EQ(4U, sizeof(wint_t));
|
||||
|
|
@ -45,7 +70,7 @@ TEST(wchar, sizeof_wchar_t) {
|
|||
|
||||
TEST(wchar, mbrlen) {
|
||||
char bytes[] = { 'h', 'e', 'l', 'l', 'o', '\0' };
|
||||
EXPECT_EQ(0U, mbrlen(&bytes[0], 0, nullptr));
|
||||
EXPECT_EQ(kExpectedResultForZeroLength, mbrlen(&bytes[0], 0, nullptr));
|
||||
EXPECT_EQ(1U, mbrlen(&bytes[0], 1, nullptr));
|
||||
|
||||
EXPECT_EQ(1U, mbrlen(&bytes[4], 1, nullptr));
|
||||
|
|
@ -264,63 +289,98 @@ TEST(wchar, wcsstr_80199) {
|
|||
TEST(wchar, mbtowc) {
|
||||
wchar_t out[8];
|
||||
|
||||
// bionic has the same misunderstanding of the result for a zero-length
|
||||
// conversion for mbtowc as it does for all the other multibyte conversion
|
||||
// functions but mbtowc returns different values than all the others:
|
||||
//
|
||||
// C23 7.24.7.2.4:
|
||||
//
|
||||
// If s is a null pointer, the mbtowc function returns a nonzero or zero
|
||||
// value, if multibyte character encodings, respectively, do or do not have
|
||||
// state-dependent encodings. If s is not a null pointer, the mbtowc function
|
||||
// either returns 0 (if s points to the null character), or returns the number
|
||||
// of bytes that are contained in the converted multibyte character (if the
|
||||
// next n or fewer bytes form a valid multibyte character), or returns -1 (if
|
||||
// they do not form a valid multibyte character).
|
||||
|
||||
#ifdef __BIONIC__
|
||||
int expected_result_for_zero_length = 0;
|
||||
#else
|
||||
int expected_result_for_zero_length = -1;
|
||||
#endif
|
||||
|
||||
out[0] = 'x';
|
||||
ASSERT_EQ(0, mbtowc(out, "hello", 0));
|
||||
ASSERT_EQ('x', out[0]);
|
||||
EXPECT_EQ(expected_result_for_zero_length, mbtowc(out, "hello", 0));
|
||||
EXPECT_EQ('x', out[0]);
|
||||
|
||||
ASSERT_EQ(0, mbtowc(out, "hello", 0));
|
||||
ASSERT_EQ(0, mbtowc(out, "", 0));
|
||||
ASSERT_EQ(1, mbtowc(out, "hello", 1));
|
||||
ASSERT_EQ(L'h', out[0]);
|
||||
EXPECT_EQ(expected_result_for_zero_length, mbtowc(out, "hello", 0));
|
||||
EXPECT_EQ(0, mbtowc(out, "", 0));
|
||||
EXPECT_EQ(1, mbtowc(out, "hello", 1));
|
||||
EXPECT_EQ(L'h', out[0]);
|
||||
|
||||
ASSERT_EQ(0, mbtowc(nullptr, "hello", 0));
|
||||
ASSERT_EQ(0, mbtowc(nullptr, "", 0));
|
||||
ASSERT_EQ(1, mbtowc(nullptr, "hello", 1));
|
||||
EXPECT_EQ(expected_result_for_zero_length, mbtowc(nullptr, "hello", 0));
|
||||
EXPECT_EQ(0, mbtowc(nullptr, "", 0));
|
||||
EXPECT_EQ(1, mbtowc(nullptr, "hello", 1));
|
||||
|
||||
ASSERT_EQ(0, mbtowc(nullptr, nullptr, 0));
|
||||
EXPECT_EQ(0, mbtowc(nullptr, nullptr, 0));
|
||||
}
|
||||
|
||||
TEST(wchar, mbrtowc) {
|
||||
wchar_t out[8];
|
||||
|
||||
out[0] = 'x';
|
||||
ASSERT_EQ(0U, mbrtowc(out, "hello", 0, nullptr));
|
||||
ASSERT_EQ('x', out[0]);
|
||||
EXPECT_EQ(kExpectedResultForZeroLength, mbrtowc(out, "hello", 0, nullptr));
|
||||
EXPECT_EQ('x', out[0]);
|
||||
|
||||
ASSERT_EQ(0U, mbrtowc(out, "hello", 0, nullptr));
|
||||
ASSERT_EQ(0U, mbrtowc(out, "", 0, nullptr));
|
||||
ASSERT_EQ(1U, mbrtowc(out, "hello", 1, nullptr));
|
||||
ASSERT_EQ(L'h', out[0]);
|
||||
EXPECT_EQ(kExpectedResultForZeroLength, mbrtowc(out, "hello", 0, nullptr));
|
||||
EXPECT_EQ(kExpectedResultForZeroLength, mbrtowc(out, "", 0, nullptr));
|
||||
EXPECT_EQ(1U, mbrtowc(out, "hello", 1, nullptr));
|
||||
EXPECT_EQ(L'h', out[0]);
|
||||
|
||||
ASSERT_EQ(0U, mbrtowc(nullptr, "hello", 0, nullptr));
|
||||
ASSERT_EQ(0U, mbrtowc(nullptr, "", 0, nullptr));
|
||||
ASSERT_EQ(1U, mbrtowc(nullptr, "hello", 1, nullptr));
|
||||
EXPECT_EQ(kExpectedResultForZeroLength, mbrtowc(nullptr, "hello", 0, nullptr));
|
||||
EXPECT_EQ(kExpectedResultForZeroLength, mbrtowc(nullptr, "", 0, nullptr));
|
||||
EXPECT_EQ(1U, mbrtowc(nullptr, "hello", 1, nullptr));
|
||||
|
||||
ASSERT_EQ(0U, mbrtowc(nullptr, nullptr, 0, nullptr));
|
||||
EXPECT_EQ(0U, mbrtowc(nullptr, nullptr, 0, nullptr));
|
||||
|
||||
ASSERT_STREQ("C.UTF-8", setlocale(LC_CTYPE, "C.UTF-8"));
|
||||
EXPECT_STREQ("C.UTF-8", setlocale(LC_CTYPE, "C.UTF-8"));
|
||||
uselocale(LC_GLOBAL_LOCALE);
|
||||
|
||||
// 1-byte UTF-8.
|
||||
ASSERT_EQ(1U, mbrtowc(out, "abcdef", 6, nullptr));
|
||||
ASSERT_EQ(L'a', out[0]);
|
||||
EXPECT_EQ(1U, mbrtowc(out, "abcdef", 6, nullptr));
|
||||
EXPECT_EQ(L'a', out[0]);
|
||||
// 2-byte UTF-8.
|
||||
ASSERT_EQ(2U, mbrtowc(out, "\xc2\xa2" "cdef", 6, nullptr));
|
||||
ASSERT_EQ(static_cast<wchar_t>(0x00a2), out[0]);
|
||||
EXPECT_EQ(2U, mbrtowc(out,
|
||||
"\xc2\xa2"
|
||||
"cdef",
|
||||
6, nullptr));
|
||||
EXPECT_EQ(static_cast<wchar_t>(0x00a2), out[0]);
|
||||
// 3-byte UTF-8.
|
||||
ASSERT_EQ(3U, mbrtowc(out, "\xe2\x82\xac" "def", 6, nullptr));
|
||||
ASSERT_EQ(static_cast<wchar_t>(0x20ac), out[0]);
|
||||
EXPECT_EQ(3U, mbrtowc(out,
|
||||
"\xe2\x82\xac"
|
||||
"def",
|
||||
6, nullptr));
|
||||
EXPECT_EQ(static_cast<wchar_t>(0x20ac), out[0]);
|
||||
// 4-byte UTF-8.
|
||||
ASSERT_EQ(4U, mbrtowc(out, "\xf0\xa4\xad\xa2" "ef", 6, nullptr));
|
||||
ASSERT_EQ(static_cast<wchar_t>(0x24b62), out[0]);
|
||||
EXPECT_EQ(4U, mbrtowc(out,
|
||||
"\xf0\xa4\xad\xa2"
|
||||
"ef",
|
||||
6, nullptr));
|
||||
EXPECT_EQ(static_cast<wchar_t>(0x24b62), out[0]);
|
||||
#if defined(__BIONIC__) // glibc allows this.
|
||||
// Illegal 5-byte UTF-8.
|
||||
ASSERT_EQ(static_cast<size_t>(-1), mbrtowc(out, "\xf8\xa1\xa2\xa3\xa4" "f", 6, nullptr));
|
||||
ASSERT_EQ(EILSEQ, errno);
|
||||
EXPECT_EQ(static_cast<size_t>(-1), mbrtowc(out,
|
||||
"\xf8\xa1\xa2\xa3\xa4"
|
||||
"f",
|
||||
6, nullptr));
|
||||
EXPECT_EQ(EILSEQ, errno);
|
||||
#endif
|
||||
// Illegal over-long sequence.
|
||||
ASSERT_EQ(static_cast<size_t>(-1), mbrtowc(out, "\xf0\x82\x82\xac" "ef", 6, nullptr));
|
||||
ASSERT_EQ(EILSEQ, errno);
|
||||
EXPECT_EQ(static_cast<size_t>(-1), mbrtowc(out,
|
||||
"\xf0\x82\x82\xac"
|
||||
"ef",
|
||||
6, nullptr));
|
||||
EXPECT_EQ(EILSEQ, errno);
|
||||
}
|
||||
|
||||
TEST(wchar, mbrtowc_valid_non_characters) {
|
||||
|
|
|
|||
Loading…
Reference in a new issue