From bb9231c8402e9ae99f98a868429e73c6340da304 Mon Sep 17 00:00:00 2001 From: Carmelo Amoroso Date: Thu, 10 Jul 2008 10:30:28 +0000 Subject: Added missing tests for locale with multibyte encoding --- test/locale-mbwc/dat_strcoll.c | 209 +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 209 insertions(+) create mode 100644 test/locale-mbwc/dat_strcoll.c (limited to 'test/locale-mbwc/dat_strcoll.c') diff --git a/test/locale-mbwc/dat_strcoll.c b/test/locale-mbwc/dat_strcoll.c new file mode 100644 index 000000000..e12037ca0 --- /dev/null +++ b/test/locale-mbwc/dat_strcoll.c @@ -0,0 +1,209 @@ +/* + * TEST SUITE FOR MB/WC FUNCTIONS IN C LIBRARY + * + * FILE: dat_strcoll.c + * + * STRCOLL: int strcoll (const char *s1, const char *s2); + */ + +/* + NOTE: + + If a return value is expected to be 0, set ret_flg=1 and the + expected value = 0. If a return value is expected to be a + positive/negative value, set ret_flg=0, and set the expected value + = +1/-1. + There is inconsistensy between tst_strcoll() and tst_wcscoll()(it + has cmp_flg) for input data. I'll fix it. + + Assuming en_US to be en_US.ascii. (maybe, should be iso8859-1). + + + + ASCII CODE : A,B,C, ... , a, b, c, ... B,a:-1 a,B:+1 + DICTIONARY : A,a,B,b,C,c,.... a,B:-1 B,a:+1 */ + +TST_STRCOLL tst_strcoll_loc [] = { + { + { Tstrcoll, TST_LOC_de }, + { + { /*input.*/ { "ÄBCDEFG", "ÄBCDEFG" }, /* #1 */ + /*expect*/ { 0,1,0, }, + }, + { /*input.*/ { "XX Ä XX", "XX B XX" }, /* #2 */ + /*expect*/ { 0,0,-1, }, + }, + { /*input.*/ { "XX B XX", "XX Ä XX" }, /* #3 */ + /*expect*/ { 0,0,+1, }, + }, + { /*input.*/ { "B", "a" }, /* #4 */ + /*expect*/ { 0,0,+1, }, + }, + { /*input.*/ { "a", "B" }, /* #5 */ + /*expect*/ { 0,0,-1, }, + }, + { /*input.*/ { "b", "A" }, /* #6 */ + /*expect*/ { 0,0,+1, }, + }, + { /*input.*/ { "A", "b" }, /* #7 */ + /*expect*/ { 0,0,-1, }, + }, + { /*input.*/ { "ä", "B" }, /* #8 */ + /*expect*/ { 0,0,-1, }, + }, + { /*input.*/ { "B", "ä" }, /* #9 */ + /*expect*/ { 0,0,+1, }, + }, + { .is_last = 1 } /* Last element. */ + } + }, + { + { Tstrcoll, TST_LOC_en }, + { + { /*input.*/ { "ABCDEFG", "ABCDEFG" }, /* #1 */ + /*expect*/ { 0,1,0, }, + }, + { /*input.*/ { "XX a XX", "XX B XX" }, /* #2 */ + /*expect*/ { 0,0,-1, }, + }, + { /*input.*/ { "XX B XX", "XX a XX" }, /* #3 */ + /*expect*/ { 0,0,+1, }, + }, + { + /* */ + /*input.*/ { "B", "a" }, /* #4 */ +#ifdef SHOJI_IS_RIGHT + /*expect*/ { 0,0,-1, }, +#else + /* XXX We are not testing the C locale. */ + /*expect*/ { 0,0,+1, }, +#endif + }, + { + /* */ + /*input.*/ { "a", "B" }, /* #5 */ +#ifdef SHOJI_IS_RIGHT + /*expect*/ { 0,0,+1, }, +#else + /* XXX We are not testing the C locale. */ + /*expect*/ { 0,0,-1, }, +#endif + }, + { /*input.*/ { "b", "A" }, /* #6 */ + /*expect*/ { 0,0,+1, }, + }, + { /*input.*/ { "A", "b" }, /* #7 */ + /*expect*/ { 0,0,-1, }, + }, +#ifdef NO_WAIVER + /* XXX I do not yet know whether strcoll really should reject + characters outside the multibyte character range. */ + { + /* #8 */ /* */ + /*input.*/ { "\244\242\244\244\244\246\244\250\244\252", "ABCDEFG" }, + /*expect*/ { EINVAL,0,0, }, + }, + { + /* #9 */ /* */ + /*input.*/ { "ABCZEFG", "\244\242\244\244\244\246\244\250\244\252" }, + /*expect*/ { EINVAL,0,0, }, + }, +#endif + { .is_last = 1 } /* Last element. */ + } + }, +#if 0 + { + { Tstrcoll, TST_LOC_eucJP }, + { + { /*input.*/ { "\244\242\244\244\244\246\244\250\244\252", + "\244\242\244\244\244\246\244\250\244\252" }, /* #1 */ + /*expect*/ { 0,1,0, }, + }, + { /*input.*/ { "\244\242\244\244\244\246\244\250\244\252", + "\244\242\244\244\244\363\244\250\244\252" }, /* #2 */ + /*expect*/ { 0,0,-1, }, + }, + { /*input.*/ { "\244\242\244\244\244\363\244\250\244\252", + "\244\242\244\244\244\246\244\250\244\252" }, /* #3 */ + /*expect*/ { 0,0,+1, }, + }, + { /*input.*/ { "B", "a" }, /* #4 */ + /*expect*/ { 0,0,-1, }, + }, + { /*input.*/ { "a", "B" }, /* #5 */ + /*expect*/ { 0,0,+1, }, + }, + { /*input.*/ { "b", "A" }, /* #6 */ + /*expect*/ { 0,0,+1, }, + }, + { /*input.*/ { "A", "b" }, /* #7 */ + /*expect*/ { 0,0,-1, }, + }, +#ifdef NO_WAIVER + /* XXX I do not yet know whether strcoll really should reject + characters outside the multibyte character range. */ + { + /* */ + /*input.*/ { "\200\216\217", "ABCDEFG" }, /* #8 */ + /*expect*/ { EINVAL,0,0, }, + }, + { + /* */ + /*input.*/ { "ABCZEFG", "\200\216\217" }, /* #9 */ + /*expect*/ { EINVAL,0,0, }, + }, +#endif + { .is_last = 1 } /* Last element. */ + } + }, +#else + { + { Tstrcoll, TST_LOC_ja_UTF8 }, + { + { /*input.*/ { "\343\201\202\343\201\204\343\201\206\343\201\210\343\201\212", + "\343\201\202\343\201\204\343\201\206\343\201\210\343\201\212" }, /* #1 */ + /*expect*/ { 0,1,0, }, + }, + { /*input.*/ { "\343\201\202\343\201\204\343\201\206\343\201\210\343\201\212", + "\343\201\202\343\201\204\343\202\223\343\201\210\343\201\212" }, /* #2 */ + /*expect*/ { 0,0,-1, }, + }, + { /*input.*/ { "\343\201\202\343\201\204\343\202\223\343\201\210\343\201\212", + "\343\201\202\343\201\204\343\201\206\343\201\210\343\201\212" }, /* #3 */ + /*expect*/ { 0,0,+1, }, + }, + { /*input.*/ { "B", "a" }, /* #4 */ + /*expect*/ { 0,0,-1, }, + }, + { /*input.*/ { "a", "B" }, /* #5 */ + /*expect*/ { 0,0,+1, }, + }, + { /*input.*/ { "b", "A" }, /* #6 */ + /*expect*/ { 0,0,+1, }, + }, + { /*input.*/ { "A", "b" }, /* #7 */ + /*expect*/ { 0,0,-1, }, + }, +#ifdef NO_WAIVER + /* XXX I do not yet know whether strcoll really should reject + characters outside the multibyte character range. */ + { + /* */ + /*input.*/ { "\200\216\217", "ABCDEFG" }, /* #8 */ + /*expect*/ { EINVAL,0,0, }, + }, + { + /* */ + /*input.*/ { "ABCZEFG", "\200\216\217" }, /* #9 */ + /*expect*/ { EINVAL,0,0, }, + }, +#endif + { .is_last = 1 } /* Last element. */ + } + }, +#endif + { + { Tstrcoll, TST_LOC_end } + } +}; -- cgit v1.2.3