Author: sebor
Date: Sat May 31 14:57:21 2008
New Revision: 662096
URL: http://svn.apache.org/viewvc?rev=662096&view=rev
Log:
2008-05-31 Martin Sebor <[EMAIL PROTECTED]>
* tests/support/18.numeric.special.int.cpp (run_test): Removed
informational output (the same detail can be obtained by running
the test with the --trace option).
Replaced numeric references to clauses in the standard with their
names that don't change from one version of the spec to the next.
Documented why numeric_limits::traps is being exercised elsewhere
(see rev 661910).
(run_test): Replaced the use of "_RWSTD_LONG_LONG" with "long long"
in diagnostic messages.
Modified:
stdcxx/branches/4.2.x/tests/support/18.numeric.special.int.cpp
Modified: stdcxx/branches/4.2.x/tests/support/18.numeric.special.int.cpp
URL:
http://svn.apache.org/viewvc/stdcxx/branches/4.2.x/tests/support/18.numeric.special.int.cpp?rev=662096&r1=662095&r2=662096&view=diff
==============================================================================
--- stdcxx/branches/4.2.x/tests/support/18.numeric.special.int.cpp (original)
+++ stdcxx/branches/4.2.x/tests/support/18.numeric.special.int.cpp Sat May 31
14:57:21 2008
@@ -487,106 +487,104 @@
ASSERT (is_specialized, "%#b");
- // 18.2.1.2, p1
- rw_info (0, 0, __LINE__,
- "std::numeric_limits<%s>::min() == [EMAIL PROTECTED]",
- tname, fmt, (Traits::min ()));
-
+ // [numeric.special], p1
rw_assert ((limT::min)() == (Traits::min)(), 0, __LINE__,
"std::numeric_limits<%s>::min() == [EMAIL PROTECTED], got
[EMAIL PROTECTED]",
tname, fmt, (Traits::min)(), fmt, (limT::min)());
- // 18.2.1.2, p4
- rw_info (0, 0, __LINE__,
- "std::numeric_limits<%s>::max() == [EMAIL PROTECTED]",
- tname, fmt, (Traits::max ()));
-
+ // [numeric.special], p4
rw_assert ((limT::max)() == (Traits::max)(), 0, __LINE__,
"std::numeric_limits<%s>::max() == [EMAIL PROTECTED], got
[EMAIL PROTECTED]",
tname, fmt, (Traits::max)(), fmt, (limT::max)());
- // 18.2.1.2, p6
+ // [numeric.special], p6
ASSERT (digits, "%d");
- // 18.2.1.2, p9
+ // [numeric.special], p9
ASSERT (digits10, "%d");
rw_assert (limT::digits10 == Traits::compute_digits10 (), 0, __LINE__,
"std::numeric_limits<%s>::digits10 == %d (computed), got %d",
tname, Traits::compute_digits10 (), limT::digits10);
- // 18.2.1.2, p11
+ // [numeric.special], p11
ASSERT (is_signed, "%b");
- // 18.2.1.2, p13
+ // [numeric.special], p13
ASSERT (is_integer, "%b");
- // 18.2.1.2, p15
+ // [numeric.special], p15
ASSERT (is_exact, "%b");
- // 18.2.1.2, p17
+ // [numeric.special], p17
ASSERT (radix, "%i");
- // 18.2.1.2, p20
+ // [numeric.special], p20
ASSERT_0 (epsilon (), fmt);
- // 18.2.1.2, p22
+ // [numeric.special], p22
ASSERT_0 (round_error (), fmt);
- // 18.2.1.2, p23
+ // [numeric.special], p23
ASSERT_0 (min_exponent, "%i");
- // 18.2.1.2, p25
+ // [numeric.special], p25
ASSERT_0 (min_exponent10, "%i");
- // 18.2.1.2, p27
+ // [numeric.special], p27
ASSERT_0 (max_exponent, "%i");
- // 18.2.1.2, p29
+ // [numeric.special], p29
ASSERT_0 (max_exponent10, "%i");
- // 18.2.1.2, p31
+ // [numeric.special], p31
ASSERT_0 (has_infinity, "%b");
- // 18.2.1.2, p34
+ // [numeric.special], p34
ASSERT_0 (has_quiet_NaN, "%b");
- // 18.2.1.2, p37
+ // [numeric.special], p37
ASSERT_0 (has_signaling_NaN, "%b");
- // 18.2.1.2, p40
+ // [numeric.special], p40
ASSERT_0 (has_denorm, "%b");
- // 18.2.1.2, p42
+ // [numeric.special], p42
ASSERT_0 (has_denorm_loss, "%b");
- // 18.2.1.2, p43
+ // [numeric.special], p43
ASSERT_0 (infinity (), fmt);
- // 18.2.1.2, p45
+ // [numeric.special], p45
ASSERT_0 (quiet_NaN (), fmt);
- // 18.2.1.2, p47
+ // [numeric.special], p47
ASSERT_0 (signaling_NaN (), fmt);
- // 18.2.1.2, p49
+ // [numeric.special], p49
ASSERT_0 (denorm_min (), fmt);
- // 18.2.1.2, p52
+ // [numeric.special], p52
ASSERT_0 (is_iec559, "%b");
- // 18.2.1.2, p54
+ // [numeric.special], p54
ASSERT (is_bounded, "%b");
- // 18.2.1.2, p56
+ // [numeric.special], p56
rw_assert (limT::is_modulo == Traits::is_modulo (), 0, __LINE__,
"std::numeric_limits<%s>::is_modulo == %#b",
tname, Traits::is_modulo ());
- // 18.2.1.2, p61
+ // [numeric.special], p59
+ // numeric_limits::traps exercised separately in 18.limits.traps.cpp
+ // to prevent runtime errors (SIGFPE) from hiding potential assertion
+ // failures in the rest of the test
+ // ASSERT_0 (traps, "%b");
+
+ // [numeric.special], p61
ASSERT_0 (tinyness_before, "%b");
- // 18.2.1.2, p63
+ // [numeric.special], p63
rw_assert (limT::round_style == int (std::round_toward_zero), 0, __LINE__,
"std::numeric_limits<%s>::round_style == %d",
tname, std::round_toward_zero);
@@ -621,8 +619,8 @@
#ifdef _RWSTD_LONG_LONG
- TEST (_RWSTD_LONG_LONG, "%lli");
- TEST (unsigned _RWSTD_LONG_LONG, "%llu");
+ run_test ((_RWSTD_LONG_LONG*)0, "long long", "%lli");
+ run_test ((unsigned _RWSTD_LONG_LONG*)0, "unsigned long long", "%llu");
#endif // _RWSTD_LONG_LONG
