| blob | 3b227772a97562fa5879f9144f7541d0351dcda5 |
1 version: 0.27
2 date: Sun May 4 23:42:49 CEST 2008
3 changes:
4 - add Parker's library for DFA based counting
5 - support nested sums based exact summation
6 - support Laurent expansion based exact summation
7 - handle equalities and factors in summation
8 - support range propagation for computing bounds
9 - support computation of exact bound through iteration
10 - fix conversion from GiNaC's numeric to Values
11 ---
12 version: 0.26
13 date: Wed Jan 16 23:45:17 CET 2008
14 changes:
15 - support computation of generating functions in barvinok_enumerate_e
16 - support polymake 2.3
17 - support GiNaC 1.3.8 and newer
18 - support summation using Bernoulli formula
19 - add and use zsolve from 4ti2 for computing Hilbert bases of cones
20 - support use of piplib for solving LPs
21 - add polyhedron_integer_hull and polytope_minimize
22 - add polytope_lattice_width for computing the lattice widths of polytopes
23 ---
24 version: 0.25
25 date: Tue Sep 11 22:02:22 CEST 2007
26 changes:
27 - add PolyLib and piplib as submodules
28 - remove compatibility with old versions of PolyLib and piplib
29 - support use of piplib for basis reduction
30 - support summation using local Euler-Maclaurin formula for 2D polytopes
31 - support TOPCOM based chamber decompositions
32 - simplify primal decomposition
33 ---
34 version: 0.24
35 date: Sat Jun 30 23:37:31 CEST 2007
36 changes:
37 - support exponential (Todd) substitution for non-parametric problems
38 - support primal and stopped decomposition on parametric problems
39 - support nested sums polynomial approximation of quasi-polynomials
40 - add barvinok_summate for summating a polynomial over a polytope
41 - speed improvements in barvinok_count and polyhedron_sample
42 ---
43 version: 0.23
44 date: Sat Apr 28 20:10:35 CEST 2007
45 changes:
46 - use arpg for argument parsing in most commands
47 - |-
48 support polynomial approximations of quasi-polynomials, including
49 volume computation, Meister's method and replacing fractional parts
50 by constants
51 - add barvinok_maximize for computing upper (and lower) bounds
52 - support non-parametric primal and stopped Barvinok decomposition
53 - periodic number representation now run-time configurable
54 - drop support for ancient PolyLib versions
55 - simplified partitioning of chamber complex
56 - add evalue_eval, an exact version of copmute_evalue
57 - improved handling of large generating functions
58 - integrate barvinok_series functionality into barvinok_enumerate
59 - support use of cdd for solving LPs in basis reduction
60 - initial support for computing bounds over quasi-polynomials
61 - recursive Bernstein expansion
62 - corrections and improvements of lexmin
63 - optionally print quasi-polynomial as list of polynomials
64 - add some documentation
65 ---
66 version: 0.22
67 date: Tue Oct 31 23:15:15 CET 2006
68 changes:
69 - include bernstein library with interface to barvinok and occ
70 - some occ clean-ups
71 - barvinok_count now only returns -1 for unbounded polyhedra with integer points
72 - preliminary implementation of projection via neighborhood complexes
73 - implement generalized basis reduction
74 - add polyhedron_sample
75 - add polytope_scan
76 - add lexmin
77 - some internal reorganization
78 - some gen_fun clean-ups and optimizations
79 - plug some memory leaks
80 - fix some corner cases
81 - shared libraries are no longer built by default
82 - introduce barvinok_options to pass to functions instead of just MaxRays
83 - use same LLL reduction parameter that LattE uses
84 - add some documentation
85 ---
86 version: 0.21
87 date: Thu Jun 22 20:49:11 CEST 2006
88 changes:
89 - add some documentation
90 - add barvinok_ehrhart
91 - add h_star_vector polymake client
92 - fix some corner cases
93 - more experimental support for enumerating unions of polytopes
94 - rename cdd2polylib.pl to latte2polylib.pl
95 ---
96 version: 0.20
97 date: Wed Feb 22 12:01:23 CET 2006
98 changes:
99 - fix PIP-based reduction by removing some code and assuming a recent PIP
100 - some memory leaks plugged
101 - optionally use Omega as a preprocessor in barvinok_enumerate_e
102 - integrate verif_ehrhart_e into barvinok_enumerate_e
103 - barvinok_series handles more problem with equalities
104 - add gen_fun::Hadamard_product
105 - experimental support for enumerating unions of polytopes
106 ---
107 version: 0.19
108 date: Fri Jan 20 22:14:44 CET 2006
109 changes:
110 - include Omega Calculator interface occ
111 - print_enode creates parameter names if none are provided
112 - only use PIP to project out some variables
113 - some header file fixes
114 - C89 fixes again
115 ---
116 version: 0.18
117 date: Thu Sep 15 11:27:19 CEST 2005
118 changes:
119 - factor polyhedra into independent parts before enumeration
120 - make "random vector" specialization more robust
121 - speed up verif_ehrhart by only checking inside the context
122 - remove {Param,}Polyhedron_Reduce
123 - rename triangularize_cone to triangulate_cone
124 - generalize computation of lexmin/max using PIP
125 - compute the number of lexsmaller elements using PolyLib's LexSmaller
126 ---
127 version: 0.17
128 date: Mon Aug 1 12:23:50 CEST 2005
129 changes:
130 - fix some configuration problems
131 - include minimal polymake client
132 ---
133 version: 0.16
134 date: Thu Jun 30 12:10:29 CEST 2005
135 changes:
136 - implement "breadth-first" incremental specialization
137 - implement non-incremental version for series computation
138 - make non-incremental specialization the default again
139 - check for bad PolyLib version at runtime
140 - compiles on Windows (mingw32)
141 - add version option
142 - allow crosscompilation
143 - compute the coefficients of an Ehrhart series
144 - fix conversion of PIP output to polyhedra
145 - allow direct verification of Ehrhart series
146 ---
147 version: 0.15
148 date: Sun Dec 12 23:00:40 CET 2004
149 changes:
150 - export Polyhedron_Project
151 - allow verification of step-functions computed from Ehrhart series
152 - optionally compute step-functions incrementally
153 - disable possibly incorrect optimization
154 - minor optimizations in Ehrhart series computations
155 ---
156 version: 0.14
157 date: Wed Nov 17 19:22:23 CET 2004
158 changes:
159 - use code for computing series to count non-parametric polytopes
160 - add conversion from series to explicit function
161 - fix removal of redundant equalities
162 ---
163 version: 0.13
164 date: Sat Oct 16 12:13:30 CEST 2004
165 changes:
166 - use slightly larger random numbers
167 - change basis to remove existential variable
168 - fix bug in reduction
169 - handle empty PIP solution
170 - add utility to remove equalities
171 ---
172 version: 0.12
173 date: Sun Oct 3 20:47:11 CEST 2004
174 changes:
175 - reduce memory requirements
176 - add cdd2polylib.pl
177 - add patch for NTL 5.3.2
178 - experimental support for Ehrhart series
179 - compile fix on systems that lack <getopt.h>
180 ---
181 version: 0.11
182 date: Sat Sep 18 14:49:40 CEST 2004
183 changes:
184 - fix minor bugs
185 - find and use PIP if available
186 - don't overallocate matrices on newer PolyLibs
187 - verification program for enumeration of integer projections
188 - summation of enumerations
189 - avoid gccisms
190 - support older versions of GMP
191 ---
192 version: 0.10
193 date: Tue Jul 6 22:09:11 CEST 2004
194 changes:
195 - install header files
196 - modulo representation changed to fractionals
197 - perform some simplifications on the modulo representation
198 - barvinok_enumerate returns a malloc'ed evalue
199 - a few bugs in barvinok_enumerate_e removed
200 - optionally print an indication of the size of an enumeration
201 - emask added
202 - bug in triangularization fixed
203 - bug in polyhedron reduction fixed
204 - verif_ehrhart.c should compile on C89 systems again
205 ---
206 version: 0.09
207 changes:
208 - bug in polyhedron reduction fixed
209 ---
210 version: 0.08
211 date: Sun Feb 22 21:15:00 CET 2004
212 changes:
213 - remove remaining part of Kristof's Ehrhart manipulation code
214 - optimize modulo expression based on validity domain
215 - support partition evalues
216 - factor out 1D polytopes
217 - validity domains can now be unions of polytopes
218 - optionally convert modulo representation to lookup-table
219 - rename indicator to relation
220 - add DomainIncludes
221 ---
222 version: 0.07
223 changes:
224 - support "indicator" evalues
225 - support equalities for modulo representation
226 ---
227 version: 0.06
228 changes:
229 - remove part of Kristof's Ehrhart manipulation code
230 - partial support for modulo representation
231 - support cvs version of PolyLib
232 ---
233 version: 0.05
234 date: Thu Nov 27 13:35:59 CET 2003
235 changes:
236 - |-
237 include code from Kristof Beyls for manipulating Ehrhart
238 polynomials
239 - |-
240 implement barvinok_enumerate for constructing Ehrhart
241 polynomials using barvinok's algorithm
242 ---
243 version: 0.04
244 date: Thu Nov 6 01:36:25 CET 2003
245 changes:
246 - |-
247 allow specification of max #constraints
248 for those without automatically growing chernikova tables
249 - fix a few small bugs
250 - some minor optimizations
251 ---
252 version: 0.03
253 date: Mon Nov 3 16:35:57 CET 2003
254 changes:
255 - fix a number of bugs
256 - add tests
257 ---
258 version: 0.02
259 date: Mon Oct 20 14:15:45 CEST 2003
260 changes:
261 - change name from count to barvinok_count
262 - return -1 if polyhedron is unbounded
263 - trivially optimize counting in presence of some independent constraints