Граф коммитов

353 Коммитов

Автор SHA1 Сообщение Дата
marcandre 3dc28710c9 * numeric.c (flo_round): Fix criteria for 32 bits platform
part 2 of [bug #5276]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@33198 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-09-05 21:44:42 +00:00
marcandre 75c42535c9 * numeric.c (flo_round): Make Float#round round big values [bug #5272]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@33186 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-09-04 20:14:29 +00:00
marcandre 269de36cd0 * numeric.c: Extract integer rounding into int_round_0
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@33185 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-09-04 20:14:00 +00:00
marcandre 9d2f5dff62 * numeric.c (int_round): Integer#round always returns an Integer [Bug #5271]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@33183 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-09-04 19:28:54 +00:00
nobu c4eb2983ba * numeric.c (flo_round): substitute machine dependent magic number.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@33158 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-09-01 16:07:16 +00:00
marcandre 64e0cae274 * numeric.c (flo_round): Avoid overflow by optimizing for trivial cases
[Bug #5227]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@33140 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-08-31 04:13:00 +00:00
mrkn 640420f704 * numeric.c (bit_coerce): A Fixnum and a Bignum are only permitted for
bitwise arithmetic with a Fixnum.  #1792
* test/ruby/test_fixnum.rb: add tests for the above change.
* bignum.c (bit_coerce): A Fixnum and a Bignum are only permitted for
 bitwise arithmetic with a Bignum.  #1792
* test/ruby/test_bignum.rb: add tests for the above change.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@33108 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-08-29 01:07:15 +00:00
marcandre 85e24491a6 * numeric.c (int_round): Fix Integer#round [ruby-core:39096]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@33061 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-08-24 22:57:55 +00:00
marcandre ee0d942ed9 * numeric.c: Rdoc fix
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@33060 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-08-24 22:57:42 +00:00
ngoto 9fb5dca9c9 * numeric.c (rb_infinity, rb_nan): use WORDS_BIGENDIAN to get endian.
fix [Bug #5160] [ruby-dev:44356]


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32863 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-08-05 13:28:50 +00:00
usa 12adf25a80 * numeric.c (rb_num2ull): use FIX2LONG instead of FIX2ULONG. see
rb_num2ulong().


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32512 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-07-11 06:03:41 +00:00
nobu 03c618daf0 * numeric.c (ULLONG_MAX): fallback definition.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32471 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-07-09 07:54:38 +00:00
naruse 136c117fc1 * numeric.c (rb_num2ull): use own switch sentense.
Current implementation can't convert 18446744073709551615.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32433 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-07-07 06:23:40 +00:00
akr e7996eb3cc * internal.h: declare internal functions here.
* node.h: declare NODE dependent internal functions here.

* iseq.h: declare rb_iseq_t dependent internal functions here.

* vm_core.h: declare rb_thread_t dependent internal functions here.

* bignum.c, class.c, compile.c, complex.c, cont.c, dir.c, encoding.c,
  enumerator.c, error.c, eval.c, file.c, gc.c, hash.c, inits.c, io.c,
  iseq.c, load.c, marshal.c, math.c, numeric.c, object.c, parse.y,
  proc.c, process.c, range.c, rational.c, re.c, ruby.c, string.c,
  thread.c, time.c, transcode.c, variable.c, vm.c,
  tool/compile_prelude.rb: don't declare internal functions declared
  in above headers.  include above headers if required.

  Note that rb_thread_mark() was declared as
  void rb_thread_mark(rb_thread_t *th) in cont.c but defined as
  void rb_thread_mark(void *ptr) in vm.c.  Now it is declared as
  the later in internal.h.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32156 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-06-17 22:43:38 +00:00
nobu 6f58118e4c * numeric.c (rb_enc_uint_chr): fix message format. Bug#4869
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@31999 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-06-11 07:52:34 +00:00
usa 799fafc03b * include/ruby/missing.h, numeric.c (round): moved prototype of round()
from numeric.c to missing.h.  (note: round() is C99 feature, so ruby
  provides it if not exists in C runtime.)


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@31911 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-06-03 11:58:11 +00:00
naruse 1d5431374d * insns.def (opt_mult): as r31805, volatile it.
Without this, clang -O fails calculation.

* numeric.c (fix_mul): ditto.

* rational.c (f_imul): ditto.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@31822 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-05-30 06:53:21 +00:00
naruse 4e4649e13c * numeric.c (int_pow): make sure to assign the result of x * z.
If xz is optimized out, the value won't overflow.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@31805 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-05-30 01:28:37 +00:00
nobu 3b58e17d7b revert r31783.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@31788 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-05-29 22:06:54 +00:00
nobu 7c99a7df85 * numeric.c (flo_round): use absolute value as divisor.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@31783 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-05-29 14:22:49 +00:00
nagachika 4173fb0cb5 * numeric.c: add #include "interna.h" for rb_big_uminus() prototype.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@31779 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-05-29 08:55:54 +00:00
nobu a3ffe9698b * numeric.c (flo_round): fix for negative value.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@31778 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-05-29 06:09:08 +00:00
nobu 3a47cf3395 * remove trailing spaces.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@31573 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-05-15 11:55:52 +00:00
nagachika a55d803ed3 * numeric.c (flo_coerce): fix a typo in documentation.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@31547 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-05-13 12:38:49 +00:00
shyouhei eb5de217f1 Add #flo_coerce documentation.
Signed-off-by: URABE, Shyouhei <shyouhei@ruby-lang.org>



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@31545 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-05-13 10:27:10 +00:00
naruse 074818f3b0 * numeric.c (int_ord): remove K&R style.
patched by Daehyub Kim. https://github.com/ruby/ruby/pull/17

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@31490 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-05-09 02:53:55 +00:00
usa 9d5f651a98 * numeric.c (ruby_float_step): wrong loop condition.
fixes [ruby-core:35753], reported by Joey Zhou.

* test/ruby/test_range.rb (TestRange#test_step_ruby_core_35753):
  test above change.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@31286 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-04-14 14:50:46 +00:00
nobu 9b66922d7f * numeric.c (flo_round): fix inaccurate results.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@31158 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-03-22 23:07:36 +00:00
naruse 2c1936cf87 * numeric.c (flo_round): use pow instead of while-loop. fixes #4510
patched by Alex Young [ruby-core:35526]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@31146 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2011-03-22 01:08:49 +00:00
usa 91d2ee737d * numeric.c (Init_Numeric): fixed a potential bug when using bccwin32
ruby with Microsoft's dll, though we already gave up of supporting
  bccwin32. [ruby-core:33503]


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@29998 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-12-01 13:05:17 +00:00
naruse ded94d1c8c * numeric.c (ruby_float_step): fix Numeric#step with infinity unit
doesn't works well. [ruby-core:32779]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@29490 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-10-13 19:17:49 +00:00
nobu 9d0c732f0d * numeric.c (check_uint): get rid of overflow on LLP64 platforms.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@29486 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-10-13 13:53:22 +00:00
naruse 74dcda0d41 * numeric.c (int_chr): raise error when the value is negative.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@29483 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-10-13 12:13:53 +00:00
naruse e79fbf3dfe * numeric.c (rb_num_to_uint): fix 32bit logic.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@29481 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-10-13 05:58:35 +00:00
naruse a01e628318 * numeric.c (rb_num_to_uint): added to check the range of arguments.
Mainly for negative value with NUM2UINT on 32bit environment.

* string.c (rb_str_concat): use rb_num_to_uint.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@29480 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-10-13 03:56:31 +00:00
nobu 2f0b8e2793 * numeric.c (rb_num2fix): result of rb_num2long is SIGNED_VALUE.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@29465 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-10-12 14:54:06 +00:00
naruse 2d8612e966 * numeric (check_uint): the mask must refer to VALUE.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@29452 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-10-12 08:56:00 +00:00
naruse 232257238a * numeric (check_uint): set MSB for negative value.
* numeric (rb_num2uint): return value's type of rb_num2ulong
  is VALUE.

* numeric (int_chr): variable i can't be negative.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@29451 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-10-12 08:50:30 +00:00
naruse e960fffda1 * numeric.c (rb_enc_uint_chr): split from int_chr.
* numeric.c (int_chr): use rb_enc_uint_chr.

* include/ruby/encoding.h (rb_enc_uint_chr): added.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@29446 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-10-12 06:18:08 +00:00
naruse 6913b2734a * numeric.c (int_chr): a codepoint of Ruby M17N must be 32bit
unsigned int; GB18030 uses MSB. Also note that OnigCodePoint
  is defined as uisigned int.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@29445 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-10-12 05:14:23 +00:00
naruse 000f1c5b75 Add links about floating point. [ruby-core:31849]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@29131 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-08-29 00:42:05 +00:00
usa 592d885caa * numeric.c (flo_cmp): typo.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@28763 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-07-27 07:27:48 +00:00
nobu 7bfa1146e2 * numeric.c (flo_cmp): honor the result of infinite? method of the
other.  [ruby-core:31470]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@28751 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-07-24 20:37:31 +00:00
mame b79e96ae62 * numeric.c (fix_divide): must not use rb_rational_new1 for coercion
because it returns an argument itself when canonical mode is set.
  [ruby-core:31279]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@28652 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-07-15 12:55:23 +00:00
shyouhei ce9c952937 Merge branch 'mybranch' of git://github.com/orangea/ruby into trunk
Conflicts:
	ChangeLog

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@28569 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-07-07 10:59:44 +00:00
akr 26202fba93 * numeric.c (rb_num2long): accept LONG_MAX < x < LONG_MAX+1 and
LONG_MIN-1 < x < LONG_MIN as well because they are converted
  into the valid range of long by truncation.
  (rb_num2ulong): accept ULONG_MAX < x < ULONG_MAX+1 and
  LONG_MIN-1 < x < LONG_MIN as well.
  (rb_num2ll): accept LLONG_MAX < x < LLONG_MAX+1 and
  LLONG_MIN-1 < x < LLONG_MIN.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@28518 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-07-01 20:39:32 +00:00
akr 8bab1aff2c * numeric.c (rb_num2ulong): fix the lower limit for float.
[ruby-dev:41361]


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@28516 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-07-01 13:34:13 +00:00
drbrain 79aac05474 Clarification of what '*' matches. Patch by John Wells <john.wells at greatworx.com>
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@28158 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-06-04 21:22:47 +00:00
nobu eb5f96d70c * numeric.c (rb_num2ulong): explicit cast to suppress a warning.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@27953 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-05-22 08:14:02 +00:00
nobu e207a3912c * marshal.c (w_float): use dtoa directly instead of stripping
needless trailing .0.

* numeric.c (flo_to_s): reverted.  [ruby-dev:41341]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@27937 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2010-05-21 03:26:00 +00:00