1049 commits

Author	SHA1	Message	Date
jjgarcia	612ef5b5f5	WinSock.h is needed for struct timeval (Claus Brod)	2008-09-28 22:41:44 +00:00
jjgarcia	23a3c78611	Now frame stack overflows can also be recovered. Functions for enlarging stack sizes.	2008-09-28 22:15:24 +00:00
jjgarcia	1a8633a0fe	-0.0 and 0.0 do no longer compare EQL	2008-09-28 20:14:18 +00:00
jjgarcia	140f0611a7	The reader and the printer understand negative zeros.	2008-09-28 20:14:00 +00:00
jjgarcia	22538df177	FLOAT-SIGN returns the right value on negative zeros.	2008-09-28 20:11:58 +00:00
jjgarcia	8b1c69589a	ECL now signals a STACK-OVERFLOW condition when the binding stack grows too much	2008-09-28 20:11:38 +00:00
jjgarcia	9a2b12e34c	struct timeval is now defined in windows as well.	2008-09-19 17:23:30 +00:00
jjgarcia	7839fb9987	Windows.h is needed for shellapi.h	2008-09-19 17:23:16 +00:00
jjgarcia	44f137a3a4	New configuration flag, --without-fpe, to disable floating point exception code in platforms that have buggy implementations of feenableexcept().	2008-09-09 20:11:04 +00:00
jjgarcia	f40e78e063	Add patches that were missing	2008-09-04 20:50:17 +00:00
jjgarcia	d94977d507	Missing header for mingw.	2008-09-01 19:13:14 +00:00
jjgarcia	881af8a9eb	Let sharp-equal (#n=) reader macros work inside a sharp-dot.	2008-09-01 19:07:50 +00:00
jjgarcia	fff7731bec	By using WIN32_LEAN_AND_MEAN we removed some critical headers	2008-08-28 19:10:13 +00:00
jjgarcia	d6e1da448a	Fixed typo	2008-08-25 14:03:38 +00:00
jjgarcia	2b08e138cf	In addition to HOME, ECL now recognizes the environment variables HOMEDRIVE and HOMEPATH	2008-08-25 14:00:36 +00:00
jjgarcia	2c4e4afaf4	In reading functions, the value of :RECURSIVE-P should not interfere with that of :EOF-ERROR-P	2008-08-24 13:35:01 +00:00
jjgarcia	1f20a4a62f	ECL now is capable of detecting whether two calls to dlopen/LoadLibrary/etc returned the same handle.	2008-08-24 10:59:16 +00:00
jjgarcia	7fa40e8693	More robust mechanism for detecting whether a foreign library has been already loaded	2008-08-24 10:38:17 +00:00
jjgarcia	8e508390bf	Bytecode objects created inside an eval-when had an invalid name.	2008-08-24 10:23:29 +00:00
jjgarcia	2b2ae457ee	Fixed typo	2008-08-24 09:01:05 +00:00
jjgarcia	7db6f0810f	SI:LOAD-FOREIGN-MODE does not apply TRUENAME on its argument	2008-08-24 00:04:37 +00:00
jjgarcia	4c95540c38	The null stream should be writable, because it is used by the pretty printer	2008-08-19 12:19:19 +00:00
jjgarcia	ba7f0da4a7	Null stream should be closed initially	2008-08-19 12:09:56 +00:00
jjgarcia	eb548bfd4a	Fix two problems with RUN-PROGRAM: Missing :WAIT keyword and cl_core.null_stream not being input/output	2008-08-19 12:06:35 +00:00
jjgarcia	4adaa528de	Missing keyword :WAIT, required by RUN-PROGRAM	2008-08-19 11:51:18 +00:00
jjgarcia	fdfe31a38c	Sockets must be closed using closesocket in Windows	2008-08-19 11:44:23 +00:00
jjgarcia	bc6ae2146f	Hash reader macros #n# and #n=# break with specialized arrays	2008-08-19 10:27:06 +00:00
jjgarcia	28f0b90e25	Improved the code for using a smaller stack frame.	2008-08-05 21:22:18 +00:00
jjgarcia	808927c054	Do not pass more than 63 arguments in the C stack.	2008-08-05 14:08:15 +00:00
jjgarcia	2ba2041d88	Avoid the hash clearing loop when it has no elements	2008-08-04 17:23:55 +00:00
jjgarcia	270f13d2f3	Int may not be as large as unsigned long.	2008-08-03 15:38:22 +00:00
jjgarcia	767365d964	Rounding of numbers is now performed towards closest even integer, not odd as before.	2008-08-03 15:25:59 +00:00
jjgarcia	2eae8071fa	Final fix for a problem with buggy tanf().	2008-08-03 14:54:56 +00:00
jjgarcia	1743a8a68a	Second iteration of a solution for inaccurate float/fixnum comparison	2008-08-03 14:11:21 +00:00
jjgarcia	7c9e3cfbd9	Fxed typo.	2008-08-03 13:51:50 +00:00
jjgarcia	cef04b005c	When double has as many bits as a fixnum, comparisons between both types give wrong results because of rounding errors.	2008-08-03 10:32:19 +00:00
jjgarcia	a072c1dc88	Hack to avoid the problem of tanf(PI/4) overflowing in GLIBC.	2008-08-03 08:36:51 +00:00
jjgarcia	f83ba9a73d	Increased precision and fixed problems with long-float/long double and special functions	2008-08-02 18:31:10 +00:00
jjgarcia	860d38f2bb	Upgrading of imaginary part in a complex number did not work if the real part had type long-float	2008-08-02 18:19:39 +00:00
jjgarcia	1b9b2d0720	CONJUGATE did not support long double types	2008-08-02 17:57:59 +00:00
jjgarcia	ebefa6ed13	Increased precision of PI	2008-08-02 16:24:34 +00:00
jjgarcia	37923e4ac5	Fixed ATAN for builds without long floats. PI has to be of type LONG-FLOAT	2008-08-02 15:25:58 +00:00
jjgarcia	29907ce57c	Increased accuracy of SIN, SINH and ATAN when working with long floats	2008-08-02 15:23:11 +00:00
jjgarcia	a23460d397	SIN and SINH, when applied on long doubles, used only float accuracy.	2008-08-02 14:33:54 +00:00
jjgarcia	543b954b84	ERROR is a macro in MSVC	2008-08-02 10:00:22 +00:00
jjgarcia	e0f91caf6e	Fixed typo	2008-08-02 09:36:00 +00:00
jjgarcia	ae4766c079	When using long double, use no specialized arrays for LONG-FLOAT	2008-07-25 15:40:28 +00:00
jjgarcia	0c86faece3	Use isfinite/isnan as macros, like C99 prescribes. Needed for long double support in OS X	2008-07-25 14:59:31 +00:00
jjgarcia	924825d457	Missing break statement, caused IMAGPART to return NUL on long doubles	2008-07-25 14:59:09 +00:00
jjgarcia	5c60196300	Support for large files	2008-07-25 09:54:41 +00:00