Mercurial > hg > xemacs-beta
comparison src/ChangeLog @ 4888:c27efc9acb5a
merge
author | Ben Wing <ben@xemacs.org> |
---|---|
date | Wed, 27 Jan 2010 00:37:59 -0600 |
parents | a47abe9c47f2 1e9078742fa7 |
children | 276e07b3cc93 |
comparison
equal
deleted
inserted
replaced
4887:a47abe9c47f2 | 4888:c27efc9acb5a |
---|---|
163 * number.c (Fnumerator, Fdenominator, Fcanonicalize_number): | 163 * number.c (Fnumerator, Fdenominator, Fcanonicalize_number): |
164 Be more careful to return integers in their canonical forms here, | 164 Be more careful to return integers in their canonical forms here, |
165 and to give the same answer all the time when treating a rational | 165 and to give the same answer all the time when treating a rational |
166 that is itself is an integer as a ratio. | 166 that is itself is an integer as a ratio. |
167 | 167 |
168 2010-01-24 Aidan Kehoe <kehoea@parhasard.net> | |
169 | |
170 Fix problems with #'eql, extended number types, and the hash table | |
171 implementation; change the Bintegerp bytecode to fixnump semantics | |
172 even on bignum builds, since #'integerp can have a fast | |
173 implementation in terms of #'fixnump for most of its extant uses, | |
174 but not vice-versa. | |
175 | |
176 * lisp.h: Always #include number.h; we want the macros provided in | |
177 it, even if the various number types are not available. | |
178 * number.h (NON_FIXNUM_NUMBER_P): New macro, giving 1 when its | |
179 argument is of non-immediate number type. Equivalent to FLOATP if | |
180 WITH_NUMBER_TYPES is not defined. | |
181 | |
182 * elhash.c (lisp_object_eql_equal, lisp_object_eql_hash): | |
183 Use NON_FIXNUM_NUMBER_P in these functions, instead of FLOATP, | |
184 giving more correct behaviour in the presence of the extended | |
185 number types. | |
186 * bytecode.c (Bfixnump, execute_optimized_program): | |
187 Rename Bintegerp to Bfixnump; change its semantics to reflect the | |
188 new name on builds with bignum support. | |
189 | |
190 * data.c (Ffixnump, Fintegerp, syms_of_data, vars_of_data): | |
191 Always make #'fixnump available, even on non-BIGNUM builds; | |
192 always implement #'integerp in this file, even on BIGNUM builds. | |
193 Move most-positive-fixnum, most-negative-fixnum here from | |
194 number.c, so they are Lisp constants even on builds without number | |
195 types, and attempts to change or bind them error. | |
196 Use the NUMBERP and INTEGERP macros even on builds without | |
197 extended number types. | |
198 * data.c (fixnum_char_or_marker_to_int): | |
199 Rename this function from integer_char_or_marker_to_int, to better | |
200 reflect the arguments it accepts. | |
201 | |
202 * number.c (Fevenp, Foddp, syms_of_number): | |
203 Never provide #'integerp in this file. Remove #'oddp, | |
204 #'evenp; their implementations are overridden by those in cl.el. | |
205 * number.c (vars_of_number): | |
206 most-positive-fixnum, most-negative-fixnum are no longer here. | |
207 | |
208 >>>>>>> other | |
168 2010-01-17 Vin Shelton <acs@xemacs.org> | 209 2010-01-17 Vin Shelton <acs@xemacs.org> |
169 | 210 |
170 * fileio.c (check_writable): Remove cast from 2010-01-14 | 211 * fileio.c (check_writable): Remove cast from 2010-01-14 |
171 | 212 |
172 * syswindows.h: Use consts in macros where appropriate in lieu of | 213 * syswindows.h: Use consts in macros where appropriate in lieu of |