Andrei Alexandrescu
36220eb549
eliminated cyclic dependency
2009-04-06 05:44:43 +00:00
Andrei Alexandrescu
a394eddf1d
minor
2009-04-06 05:44:13 +00:00
Andrei Alexandrescu
072f7cdb4f
added Unqual template
2009-04-06 05:43:22 +00:00
Andrei Alexandrescu
4f3d7f3d88
minor
2009-04-06 05:06:27 +00:00
Andrei Alexandrescu
cc1440f20d
strip, stripl, stripr, startsWith, endsWith now work with any string type
2009-04-06 05:04:47 +00:00
Andrei Alexandrescu
c5cdcfc3ec
minor
2009-04-06 04:58:45 +00:00
Andrei Alexandrescu
3a1f910906
minor
2009-04-06 04:48:50 +00:00
Andrei Alexandrescu
34879c7c75
minor
2009-04-06 04:41:49 +00:00
Andrei Alexandrescu
7325018c47
Added support for parameterless delegates
2009-04-06 04:40:21 +00:00
Andrei Alexandrescu
b2f892a285
Added support for parameterless delegates
2009-04-06 04:35:25 +00:00
Andrei Alexandrescu
d30427c814
cosmetic
2009-04-06 04:30:13 +00:00
Sean Kelly
777cdd3726
std.stdio will declare fopen64 for linux if it isn't declared by an import module.
2009-03-31 01:10:49 +00:00
Walter Bright
eb37fa62a6
fix phobos build breaks from new druntime
2009-03-30 21:01:32 +00:00
Sean Kelly
6f7fc647bd
Added sys.posix.sys.types to the list of public imports for ssize_t.
2009-03-30 20:04:19 +00:00
Don Clugston
935dbdf282
double/float overloads in std.math are now pure nothrow.
...
(required fix for bug2756).
2009-03-30 08:15:02 +00:00
Don Clugston
4937d6c6f1
Removed unnecessary assert(0) to allow compilation with -w.
2009-03-30 08:12:36 +00:00
Walter Bright
ae02a262d8
bugzilla 2752
2009-03-30 06:58:10 +00:00
Walter Bright
a0ce8d1f66
fix isIntegral, isFloatingPoint, add unit tests
2009-03-30 03:04:18 +00:00
Walter Bright
245844d89a
druntime.lib moved
2009-03-29 23:43:50 +00:00
Walter Bright
a97e9c4fb8
fix unittest
2009-03-29 18:01:11 +00:00
Brad Roberts
22c437c8b3
Apply attachment from bug 2570 from Aziz to cleanup some bad html in ddoc comments
2009-03-29 06:32:11 +00:00
Walter Bright
afc7d37ef0
bugzilla 2763
2009-03-27 08:24:39 +00:00
Sean Kelly
035f8f39c1
Resolved differences between Posix and Windows declarations for BSD socket routines by standardizing on the Posix declaration (which is likely the correct one). This required reverting some changes to std.socket and changing the type of SOCKET from uint to int, which shoudl be fine since they're both 4 bytes anyway.
2009-03-26 18:04:44 +00:00
Sean Kelly
0b1bf3b1d5
2009-03-26 15:13:53 +00:00
Sean Kelly
e108928d21
* Modified all std modules to use core.sys.posix in place of std.c.linux.
...
* Transformed std.c.linux.* into wrappers around core.sys.posix modules where appropriate.
* Added std.c.osx as the OSX equivalent of std.c.linux.
* Added std.c.osx.socket for OSX-specific socket declarations not covered by the Posix spec.
* Altered all makefiles to zip, etc, the new std.c.osx package.
2009-03-26 03:02:32 +00:00
Sean Kelly
7de36d8138
Posix targets should now build correctly against the core.stdc chages.
2009-03-25 14:53:41 +00:00
Don Clugston
5ae67f23ac
* Disabled pure nothrow for double & float functions, because of bug 2756
...
* More extensive unittests for nextUp.
Trivial tidying up:
* Changed all references to std.math.isXXX to the new names instead of the aliased old names.
* Moved license to end of std.math.
2009-03-24 08:49:38 +00:00
Sean Kelly
c4a3aab657
The std.c modules now simply wrap the corresponding core.stdc modules to expose standard C declarations. std.c.os remains unchanged thus far, but will likely be altered as well. This unfortunately eliminates the auto doc generation for the std.c modules, since the modules are now largely empty. If the user wants to know what they contain, how the functions work, etc, I suggest for now simply referring to a C library spec--the core.stdc modules conform almost exactly to the C99 library definition.
2009-03-20 18:52:45 +00:00
Don Clugston
47ddb78118
* All math functions are now pure nothrow, except those which call the C library. (This assumes that the 'pure float' problem will be solved, perhaps by one of the two methods I've proposed).
...
* New implementation of std.math.hypot() which is faster, and now works correctly for subnormal numbers.
2009-03-18 09:19:02 +00:00
Don Clugston
b94f6e1c20
static const -> enum, to allow these templates to be used in pure functions.
2009-03-18 09:12:43 +00:00
Walter Bright
f14ca1c010
fix bugzilla 2727
2009-03-10 18:13:37 +00:00
Don Clugston
eed37a809b
Correct a bug in frexp() for the GDC doubledouble case. (Copied from Tango).
2009-03-09 15:35:52 +00:00
Walter Bright
d5a4d17c40
commented out failing unit test
2009-03-05 09:17:17 +00:00
Don Clugston
5fc7dab498
Fixes the failing unit tests.
2009-03-05 08:04:35 +00:00
Walter Bright
095813650f
comment out unittests
2009-03-05 06:45:47 +00:00
Walter Bright
ce0dc8aa42
fix unittest compile fail
2009-03-04 19:19:59 +00:00
Walter Bright
63a892d41e
added overload for enums
2009-03-01 21:10:13 +00:00
Walter Bright
9e96cfb642
add std.c.math to phobos
2009-02-27 22:01:00 +00:00
Don Clugston
d3b2166a1b
Allow phobos to compile with -w.
2009-02-27 15:31:48 +00:00
Don Clugston
a9ee82e9fc
Added comments, fixed a broken unit test, simplify frexp().
2009-02-27 08:05:13 +00:00
Don Clugston
eb61deb82e
Tango compatibility.
...
* IEEE754 names for isNaN, isFinite, isNormal, isSubnormal, isInfinity. Aliases for the old names have been retained.
* The non-functional nan(char[]) is replaced with NaN, getNaNpayload.
* Restored float, double versions of the trig & exp functions
* finally got rid of the private mfeq() function
2009-02-26 19:44:09 +00:00
Don Clugston
cde0563c68
Added NaN payload functions (from Tango).
...
Removed std.math.nan, which was a no-op.
2009-02-26 10:02:43 +00:00
Don Clugston
99e533ab68
Faster complex trig functions. Marked math as a System module.
2009-02-26 08:13:21 +00:00
Walter Bright
18404fdeea
add purity and nothrow
2009-02-24 11:02:48 +00:00
Don Clugston
2d098cbc5e
Removed the last bits of HTML from the ddoc comments.
2009-02-24 08:13:34 +00:00
Don Clugston
02b2dc3fe5
Replaced html with ddoc tags
2009-02-24 07:55:57 +00:00
Don Clugston
f76e9c5ff1
* exp2() and expm1() also 3 times faster.
2009-02-23 20:05:01 +00:00
Don Clugston
cabeb2fac6
* exp() is now 3 times faster.
...
* sinh, cosh, tanh now use D implementations rather than the C standard library.
2009-02-20 14:17:21 +00:00
Don Clugston
b85951e49a
This dead module should have been removed long ago
2009-02-20 13:27:15 +00:00
Walter Bright
209a765a40
2.025
2009-02-14 20:18:44 +00:00
