| blob | d39ac03e2055afdb857ef5b6d6cc61d041da9a49 |
1 version: 0.30
2 date: Wed Mar 10 18:10:33 CET 2010
3 changes:
4 - add isl interface and new isl based calculator
5 - barvinok_enumerate now returns NaN on infinite problems
6 - minor compilation and bug fixes
7 - support newer versions of TOPCOM
8 ---
9 version: 0.29
10 date: Wed Jul 22 15:14:01 CEST 2009
11 changes:
12 - new version of Laurent expansion based exact summation
13 - minor compilation and bug fixes
14 - add semigroup_holes application
15 - add vector_partition_chambers application
16 ---
17 version: 0.28
18 date: Thu Oct 16 21:56:01 CEST 2008
19 changes:
20 - add omega as submodule
21 - minor compilation and bug fixes
22 ---
23 version: 0.27
24 date: Sun May 4 23:42:49 CEST 2008
25 changes:
26 - add Parker's library for DFA based counting
27 - support nested sums based exact summation
28 - support Laurent expansion based exact summation
29 - handle equalities and factors in summation
30 - support range propagation for computing bounds
31 - support computation of exact bound through iteration
32 - fix conversion from GiNaC's numeric to Values
33 ---
34 version: 0.26
35 date: Wed Jan 16 23:45:17 CET 2008
36 changes:
37 - support computation of generating functions in barvinok_enumerate_e
38 - support polymake 2.3
39 - support GiNaC 1.3.8 and newer
40 - support summation using Bernoulli formula
41 - add and use zsolve from 4ti2 for computing Hilbert bases of cones
42 - support use of piplib for solving LPs
43 - add polyhedron_integer_hull and polytope_minimize
44 - add polytope_lattice_width for computing the lattice widths of polytopes
45 ---
46 version: 0.25
47 date: Tue Sep 11 22:02:22 CEST 2007
48 changes:
49 - add PolyLib and piplib as submodules
50 - remove compatibility with old versions of PolyLib and piplib
51 - support use of piplib for basis reduction
52 - support summation using local Euler-Maclaurin formula for 2D polytopes
53 - support TOPCOM based chamber decompositions
54 - simplify primal decomposition
55 ---
56 version: 0.24
57 date: Sat Jun 30 23:37:31 CEST 2007
58 changes:
59 - support exponential (Todd) substitution for non-parametric problems
60 - support primal and stopped decomposition on parametric problems
61 - support nested sums polynomial approximation of quasi-polynomials
62 - add barvinok_summate for summating a polynomial over a polytope
63 - speed improvements in barvinok_count and polyhedron_sample
64 ---
65 version: 0.23
66 date: Sat Apr 28 20:10:35 CEST 2007
67 changes:
68 - use arpg for argument parsing in most commands
69 - |-
70 support polynomial approximations of quasi-polynomials, including
71 volume computation, Meister's method and replacing fractional parts
72 by constants
73 - add barvinok_maximize for computing upper (and lower) bounds
74 - support non-parametric primal and stopped Barvinok decomposition
75 - periodic number representation now run-time configurable
76 - drop support for ancient PolyLib versions
77 - simplified partitioning of chamber complex
78 - add evalue_eval, an exact version of copmute_evalue
79 - improved handling of large generating functions
80 - integrate barvinok_series functionality into barvinok_enumerate
81 - support use of cdd for solving LPs in basis reduction
82 - initial support for computing bounds over quasi-polynomials
83 - recursive Bernstein expansion
84 - corrections and improvements of lexmin
85 - optionally print quasi-polynomial as list of polynomials
86 - add some documentation
87 ---
88 version: 0.22
89 date: Tue Oct 31 23:15:15 CET 2006
90 changes:
91 - include bernstein library with interface to barvinok and occ
92 - some occ clean-ups
93 - barvinok_count now only returns -1 for unbounded polyhedra with integer points
94 - preliminary implementation of projection via neighborhood complexes
95 - implement generalized basis reduction
96 - add polyhedron_sample
97 - add polytope_scan
98 - add lexmin
99 - some internal reorganization
100 - some gen_fun clean-ups and optimizations
101 - plug some memory leaks
102 - fix some corner cases
103 - shared libraries are no longer built by default
104 - introduce barvinok_options to pass to functions instead of just MaxRays
105 - use same LLL reduction parameter that LattE uses
106 - add some documentation
107 ---
108 version: 0.21
109 date: Thu Jun 22 20:49:11 CEST 2006
110 changes:
111 - add some documentation
112 - add barvinok_ehrhart
113 - add h_star_vector polymake client
114 - fix some corner cases
115 - more experimental support for enumerating unions of polytopes
116 - rename cdd2polylib.pl to latte2polylib.pl
117 ---
118 version: 0.20
119 date: Wed Feb 22 12:01:23 CET 2006
120 changes:
121 - fix PIP-based reduction by removing some code and assuming a recent PIP
122 - some memory leaks plugged
123 - optionally use Omega as a preprocessor in barvinok_enumerate_e
124 - integrate verif_ehrhart_e into barvinok_enumerate_e
125 - barvinok_series handles more problem with equalities
126 - add gen_fun::Hadamard_product
127 - experimental support for enumerating unions of polytopes
128 ---
129 version: 0.19
130 date: Fri Jan 20 22:14:44 CET 2006
131 changes:
132 - include Omega Calculator interface occ
133 - print_enode creates parameter names if none are provided
134 - only use PIP to project out some variables
135 - some header file fixes
136 - C89 fixes again
137 ---
138 version: 0.18
139 date: Thu Sep 15 11:27:19 CEST 2005
140 changes:
141 - factor polyhedra into independent parts before enumeration
142 - make "random vector" specialization more robust
143 - speed up verif_ehrhart by only checking inside the context
144 - remove {Param,}Polyhedron_Reduce
145 - rename triangularize_cone to triangulate_cone
146 - generalize computation of lexmin/max using PIP
147 - compute the number of lexsmaller elements using PolyLib's LexSmaller
148 ---
149 version: 0.17
150 date: Mon Aug 1 12:23:50 CEST 2005
151 changes:
152 - fix some configuration problems
153 - include minimal polymake client
154 ---
155 version: 0.16
156 date: Thu Jun 30 12:10:29 CEST 2005
157 changes:
158 - implement "breadth-first" incremental specialization
159 - implement non-incremental version for series computation
160 - make non-incremental specialization the default again
161 - check for bad PolyLib version at runtime
162 - compiles on Windows (mingw32)
163 - add version option
164 - allow crosscompilation
165 - compute the coefficients of an Ehrhart series
166 - fix conversion of PIP output to polyhedra
167 - allow direct verification of Ehrhart series
168 ---
169 version: 0.15
170 date: Sun Dec 12 23:00:40 CET 2004
171 changes:
172 - export Polyhedron_Project
173 - allow verification of step-functions computed from Ehrhart series
174 - optionally compute step-functions incrementally
175 - disable possibly incorrect optimization
176 - minor optimizations in Ehrhart series computations
177 ---
178 version: 0.14
179 date: Wed Nov 17 19:22:23 CET 2004
180 changes:
181 - use code for computing series to count non-parametric polytopes
182 - add conversion from series to explicit function
183 - fix removal of redundant equalities
184 ---
185 version: 0.13
186 date: Sat Oct 16 12:13:30 CEST 2004
187 changes:
188 - use slightly larger random numbers
189 - change basis to remove existential variable
190 - fix bug in reduction
191 - handle empty PIP solution
192 - add utility to remove equalities
193 ---
194 version: 0.12
195 date: Sun Oct 3 20:47:11 CEST 2004
196 changes:
197 - reduce memory requirements
198 - add cdd2polylib.pl
199 - add patch for NTL 5.3.2
200 - experimental support for Ehrhart series
201 - compile fix on systems that lack <getopt.h>
202 ---
203 version: 0.11
204 date: Sat Sep 18 14:49:40 CEST 2004
205 changes:
206 - fix minor bugs
207 - find and use PIP if available
208 - don't overallocate matrices on newer PolyLibs
209 - verification program for enumeration of integer projections
210 - summation of enumerations
211 - avoid gccisms
212 - support older versions of GMP
213 ---
214 version: 0.10
215 date: Tue Jul 6 22:09:11 CEST 2004
216 changes:
217 - install header files
218 - modulo representation changed to fractionals
219 - perform some simplifications on the modulo representation
220 - barvinok_enumerate returns a malloc'ed evalue
221 - a few bugs in barvinok_enumerate_e removed
222 - optionally print an indication of the size of an enumeration
223 - emask added
224 - bug in triangularization fixed
225 - bug in polyhedron reduction fixed
226 - verif_ehrhart.c should compile on C89 systems again
227 ---
228 version: 0.09
229 changes:
230 - bug in polyhedron reduction fixed
231 ---
232 version: 0.08
233 date: Sun Feb 22 21:15:00 CET 2004
234 changes:
235 - remove remaining part of Kristof's Ehrhart manipulation code
236 - optimize modulo expression based on validity domain
237 - support partition evalues
238 - factor out 1D polytopes
239 - validity domains can now be unions of polytopes
240 - optionally convert modulo representation to lookup-table
241 - rename indicator to relation
242 - add DomainIncludes
243 ---
244 version: 0.07
245 changes:
246 - support "indicator" evalues
247 - support equalities for modulo representation
248 ---
249 version: 0.06
250 changes:
251 - remove part of Kristof's Ehrhart manipulation code
252 - partial support for modulo representation
253 - support cvs version of PolyLib
254 ---
255 version: 0.05
256 date: Thu Nov 27 13:35:59 CET 2003
257 changes:
258 - |-
259 include code from Kristof Beyls for manipulating Ehrhart
260 polynomials
261 - |-
262 implement barvinok_enumerate for constructing Ehrhart
263 polynomials using barvinok's algorithm
264 ---
265 version: 0.04
266 date: Thu Nov 6 01:36:25 CET 2003
267 changes:
268 - |-
269 allow specification of max #constraints
270 for those without automatically growing chernikova tables
271 - fix a few small bugs
272 - some minor optimizations
273 ---
274 version: 0.03
275 date: Mon Nov 3 16:35:57 CET 2003
276 changes:
277 - fix a number of bugs
278 - add tests
279 ---
280 version: 0.02
281 date: Mon Oct 20 14:15:45 CEST 2003
282 changes:
283 - change name from count to barvinok_count
284 - return -1 if polyhedron is unbounded
285 - trivially optimize counting in presence of some independent constraints