Граф коммитов

33 Коммитов

Автор SHA1 Сообщение Дата
mrkn ecfc868bdb * ext/bigdecimal/bigdecimal.c (BigDecimal_save_limit):
return the result of a block.
* test/bigdecimal/test_bigdecimal.rb (test_save_limit):
  add a test for the above change.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@29363 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-09-29 08:55:11 +00:00
mrkn fed832d7f2 * ext/bigdecimal/bigdecimal.c (BigDecimal_save_rounding_mode):
return the result of a block.
* test/bigdecimal/test_bigdecimal.rb (test_save_rounding_mode):
  add a test for the above change.
* test/bigdecimal/test_bigdecimal.rb (test_save_exception_mode):
  add a test for the return value of BigDecimal.save_exception_mode.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@29361 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-09-29 04:32:20 +00:00
mrkn 971a57004e * ext/bigdecimal/bigdecimal.c (check_rounding_mode, BigDecimal_mode):
raise ArgumentError instead of TypeError passing invalid modes.
* test/bigdecimal/test_bigdecimal.rb (test_mode, test_round):
  change against the above modifications.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@29304 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-09-19 17:38:18 +00:00
mrkn fb7fe14575 * ext/bigdecimal/bigdecimal.c (check_rounding_mode): added for
converting symbol to rounding mode number.
* ext/bigdecimal/bigdecimal.c (BigDecimal_mode, BigDecimal_round):
  support to specify rounding mode by symbol.
* test/bigdecimal/test_bigdecimal.rb (test_mode, test_round):
  add tests for avobe changes.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@29295 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-09-18 21:06:08 +00:00
mrkn 3cbda570c2 * ext/bigdecimal/bigdecimal.c: fix rounding algorithms for half-down
and half-even.  This change is based on the patch created by Matthew
  Willson, the reporter of this bug.  [Bug #3803] [ruby-core:32136]
* test/bigdecimal/test_bigdecimal.rb: add tests for above changes.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@29293 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-09-18 20:19:38 +00:00
mrkn a2bf9203e8 * ext/bigdecimal/bigdecimal.c (BigDecimal_save_exception_mode,
BigDecimal_save_rounding_mode, BigDecimal_save_limit): added.
* test/bigdecimal/test_bigdecimal.rb: added tests for the above
  features.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@29127 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-08-28 08:52:14 +00:00
mrkn 3dc6edf793 * ext/bigdecimal/bigdecimal.c (Init_bigdecimal, rmpd_set_thread_local_exception_mode, VpGetException, VpSetException): thread-local exception mode.
* ext/bigdecimal/bigdecimal.c (Init_bigdecimal, rmpd_set_thread_local_precision_limit, VpGetPrecLimit, VpSetPrecLimit): thread-local precision limit.
* ext/bigdecimal/bigdecimal.c (Init_bigdecimal, rmpd_set_thread_local_rounding_mode, VpGetRoundMode, VpSetRoundMode, VpException, VpInternalRound): thread-local rounding mode.
* ext/bigdecimal/bigdecimal.c (BigDecimal_mode, BigDecimal_round, VpIsRoundMode, VpGetRoundMode, VpSetRoundMode, VpActiveRound, VpMidRound, VpLeftRound), ext/bigdecimal/bigdecimal.h: use unsigned short for rounding mode.
* test/bigdecimal/test_bigdecimal.rb (test_mode): add test for setting rounding mode.
* test/bigdecimal/test_bigdecimal.rb (test_thread_local_mode): add test for setting mode thread-locally.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@29099 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-08-26 00:14:18 +00:00
mrkn af8cae1e7f * ext/bigdecimal/bigdecimal.c (BigDecimal_to_f): to_f must underflow when the exponent is less than DBL_MIN_10_EXP - BASE_FIG.
* test/bigdecimal/test_bigdecimal.rb (test_to_f): added test for the above changes.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@29078 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-08-23 16:22:04 +00:00
mrkn d86da0fcf1 * ext/bigdecimal/extconf.rb, ext/bigdecimal/bigdecimal.h (labs, llabs): support environments missing labs and llabs.
* ext/bigdecimal/bigdecimal.h (vabs): added.
* ext/bigdecimal/extconf.rb, ext/bigdecimal/bigdecimal.h, ext/bigdecimal/bigdecimal.c, test/bigdecimal/test_bigdecimal.rb (TestBigDecimal#test_new): replace U_LONG, S_LONG, S_INT, and U_INT with appropreate standard or ruby-provided types.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@28992 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-08-14 15:22:43 +00:00
mrkn 1c3e07f0d6 * ext/bigdecimal/bigdecimal.c (Init_bigdecimal): add two new constants BigDecimal::INFINITY and BigDecimal::NAN.
* ext/bigdecimal/lib/bigdecimal/math.rb (BigMath.exp): modify the behaviors for infinity arguments as same as Math.exp.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@28533 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-07-03 07:47:00 +00:00
mrkn 74365313f7 * ext/bigdecimal/lib/bigdecimal/math.rb (atan), test/bigdecimal/test_bigmath.rb (test_atan): explicitly specify the precision for calculating a reciprocal number of an argument. [Bug #3267]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@27734 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-05-11 05:11:59 +00:00
nobu 64dee0063b * test/bigdecimal/test_bigdecimal.rb (TestBigDecimal#test_gc):
reduced heavy test.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@27708 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-05-09 16:00:37 +00:00
nobu c4546f190a * test/bigdecimal/test_bigdecimal.rb (TestBigDecimal#test_gc):
fixed test subject.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@27702 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-05-09 14:38:56 +00:00
nobu 66abf4ad68 * ext/bigdecimal/bigdecimal.c (VpAlloc): ensure buf does not get
collected.  based on a patch masaya tarui at [ruby-dev:41213].


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@27665 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-05-08 02:07:43 +00:00
naruse 54476a6024 * ext/bigdecimal/bigdecimal.c (VpCtoV): fix to check overflow.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@27647 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-05-06 19:28:12 +00:00
nobu 9803f4f55a * test/bigdecimal/testbase.rb (teardown): should reset all modes.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@27496 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-04-26 01:55:51 +00:00
naruse e41936441b Surpress warnings.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@27235 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-04-06 03:01:52 +00:00
mame 05d80e2fc6 * ext/bigdecimal/bigdecimal.c (VpCtoV): BigDecimal("0E200000000000")
was Infinity, not 0.

* test/bigdecimal/test_bigdecimal.rb: add a test for above.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@26428 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-01-26 12:52:49 +00:00
marcandre c07e7d167d * ext/bigdecimal/bigdecimal.c (BigDecimalCmp): Fix comparisons [ruby-core:26646]
* test/bigdecimal/test_bigdecimal.rb (class): Fix and improve tests.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@25765 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2009-11-14 00:17:07 +00:00
nobu 39ea1c8c7c * ext/bigdecimal/lib/bigdecimal/math.rb (atan): atan(Infinity) is
PI/2.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@25063 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2009-09-23 17:21:26 +00:00
nobu d9bfa822ea * ext/bigdecimal/bigdecimal.c (GetVpValue): support conversion from
Rational.  [ruby-core:25697]


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@25025 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2009-09-21 19:47:39 +00:00
nobu cd4f59bc5e * ext/bigdecimal/lib/bigdecimal/math.rb (sin, cos, atan, exp, log):
improved precision and performance.  based on a patch from Makoto
  Yamashita in [ruby-core:25600] and [ruby-core:25602].


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@25013 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2009-09-20 15:07:38 +00:00
kazu 3d923f01bf * test/bigdecimal/test_bigdecimal.rb (TestBigDecimal#test_hash):
shut up warning.  see [ruby-dev:37437]


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@20805 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2008-12-17 02:40:53 +00:00
matz 4d3f3af836 * ext/bigdecimal/bigdecimal.c (BigDecimal_DoDivmod): bigdecimal
division (including modulo) should raise ZeroDivisionError as
  integer division. [incompatible]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@20383 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2008-11-27 17:20:08 +00:00
matz 0434b0f098 * ext/bigdecimal/bigdecimal.c (VpException): bigdecimal zero
division should raise FloatDomainError if mode
  VP_EXCEPTION_ZERODIVIDE is set.  [ruby-dev:37204]

* ext/bigdecimal/bigdecimal.c (BigDecimal_mode): should handle
  VP_EXCEPTION_ZERODIVIDE.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@20366 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2008-11-26 14:04:58 +00:00
matz a426373520 * ext/bigdecimal/bigdecimal.c (VpAlloc): avoid ALLOCA_N() to avoid
segmentation fault caused by (insanely) long decimal values.
  [ruby-dev:37189] fix #794

* ext/bigdecimal/bigdecimal.c (BigDecimal_dump, BigDecimal_to_i,
  BigDecimal_to_f, BigDecimal_to_s, BigDecimal_split,
  BigDecimal_inspect): ditto.

* ext/bigdecimal/bigdecimal.c (VpToString): small performance
  improvement.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@20359 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2008-11-25 18:02:42 +00:00
matz 6b02097f79 * test/bigdecimal/test_bigdecimal.rb (TestBigDecimal#test_sqrt_bigdecimal):
test updated.  a patch from TAKANO Mitsuhiro <takano32 at jus.or.jp> in
  [ruby-dev:36736].

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19820 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2008-10-17 09:17:11 +00:00
matz 2de01f01a7 * test/bigdecimal/test_bigdecimal.rb (test_sqrt_bigdecimal): test
updated.  a patch from TAKANO Mitsuhiro <takano32 at jus.or.jp>
  in [ruby-dev:36669].

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19694 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2008-10-06 00:02:44 +00:00
matz 1eeb3e151e * test/bigdecimal/test_bigdecimal.rb (test_sqrt_bigdecimal): wrong
precision specification.  a patch from TAKANO Mitsuhiro
  <takano32 at jus.or.jp> in [ruby-dev:36634].

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19642 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2008-09-30 08:06:20 +00:00
nobu 61b43de755 * test/bigdecimal/test_bigdecimal.rb (test_inspect): fixed CPU bit
size dependent test.  pointed out by TAKANO Mitsuhiro <takano32 AT
  jus.or.jp>.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19586 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2008-09-26 10:01:50 +00:00
yugui 396650e0bd * prec.c: removed. Precision will be redesigned and be back again.
c.f. [ruby-dev:36352].

* common.mk (COMMON_OBJS): removed prec.o.

* inits.c (rb_call_inits): removed Init_Precision.

* numeric.c (Init_Numeric): removed inclusion of Precision.
  removed #induced_from from each class.

* rational.c: ditto.

* ext/bigdecimal/bigdecimal.c: ditto.

* lib/rdoc/knwon_classes.rb: removed the entry for Precision.

* test/ruby/test_prec.rb: removed.

* test/ruby/test_integer.rb: removed tests for Precision.

* test/ruby/test_fixnum.rb: ditto.

* test/ruby/test_float.rb: ditto.

* test/ruby/test_rational.rb: ditto.

* test/ruby/test_complex.rb: ditto.

* test/bigdecimal/test_bigdecimal.rb: ditto.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19430 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2008-09-19 13:10:54 +00:00
mame e040ff5530 * ext/bigdecimal/bigdecimal.c (VpCtoV): 1E1000...000 is interpreted as
Infinity.  [ruby-dev:36159]

* ext/bigdecimal/bigdecimal.c (VpPower): Infinity ** 1 returns
  Infinity instead of NaN.  [ruby-dev:36159]

* test/bigdecimal/test_bigdecimal.rb: add tests for above.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19419 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2008-09-18 13:48:21 +00:00
mame 2fc415a166 * test/bigdecimal/test_bigdecimal.rb: add tests to achieve over 90%
test coverage of bigdecimal.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19401 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2008-09-17 17:22:15 +00:00