| blob | b2c8fc0ad32e1da847a18895237f7e2040f440b3 |
1 version: 0.21
2 date: Thu Jun 22 20:49:11 CEST 2006
3 changes:
4 - add some documentation
5 - add barvinok_ehrhart
6 - add h_star_vector polymake client
7 - fix some corner cases
8 - more experimental support for enumerating unions of polytopes
9 - rename cdd2polylib.pl to latte2polylib.pl
10 ---
11 version: 0.20
12 date: Wed Feb 22 12:01:23 CET 2006
13 changes:
14 - fix PIP-based reduction by removing some code and assuming a recent PIP
15 - some memory leaks plugged
16 - optionally use Omega as a preprocessor in barvinok_enumerate_e
17 - integrate verif_ehrhart_e into barvinok_enumerate_e
18 - barvinok_series handles more problem with equalities
19 - add gen_fun::Hadamard_product
20 - experimental support for enumerating unions of polytopes
21 ---
22 version: 0.19
23 date: Fri Jan 20 22:14:44 CET 2006
24 changes:
25 - include Omega Calculator interface occ
26 - print_enode creates parameter names if none are provided
27 - only use PIP to project out some variables
28 - some header file fixes
29 - C89 fixes again
30 ---
31 version: 0.18
32 date: Thu Sep 15 11:27:19 CEST 2005
33 changes:
34 - factor polyhedra into independent parts before enumeration
35 - make "random vector" specialization more robust
36 - speed up verif_ehrhart by only checking inside the context
37 - remove {Param,}Polyhedron_Reduce
38 - rename triangularize_cone to triangulate_cone
39 - generalize computation of lexmin/max using PIP
40 - compute the number of lexsmaller elements using PolyLib's LexSmaller
41 ---
42 version: 0.17
43 date: Mon Aug 1 12:23:50 CEST 2005
44 changes:
45 - fix some configuration problems
46 - include minimal polymake client
47 ---
48 version: 0.16
49 date: Thu Jun 30 12:10:29 CEST 2005
50 changes:
51 - implement "breadth-first" incremental specialization
52 - implement non-incremental version for series computation
53 - make non-incremental specialization the default again
54 - check for bad PolyLib version at runtime
55 - compiles on Windows (mingw32)
56 - add version option
57 - allow crosscompilation
58 - compute the coefficients of an Ehrhart series
59 - fix conversion of PIP output to polyhedra
60 - allow direct verification of Ehrhart series
61 ---
62 version: 0.15
63 date: Sun Dec 12 23:00:40 CET 2004
64 changes:
65 - export Polyhedron_Project
66 - allow verification of step-functions computed from Ehrhart series
67 - optionally compute step-functions incrementally
68 - disable possibly incorrect optimization
69 - minor optimizations in Ehrhart series computations
70 ---
71 version: 0.14
72 date: Wed Nov 17 19:22:23 CET 2004
73 changes:
74 - use code for computing series to count non-parametric polytopes
75 - add conversion from series to explicit function
76 - fix removal of redundant equalities
77 ---
78 version: 0.13
79 date: Sat Oct 16 12:13:30 CEST 2004
80 changes:
81 - use slightly larger random numbers
82 - change basis to remove existential variable
83 - fix bug in reduction
84 - handle empty PIP solution
85 - add utility to remove equalities
86 ---
87 version: 0.12
88 date: Sun Oct 3 20:47:11 CEST 2004
89 changes:
90 - reduce memory requirements
91 - add cdd2polylib.pl
92 - add patch for NTL 5.3.2
93 - experimental support for Ehrhart series
94 - compile fix on systems that lack <getopt.h>
95 ---
96 version: 0.11
97 date: Sat Sep 18 14:49:40 CEST 2004
98 changes:
99 - fix minor bugs
100 - find and use PIP if available
101 - don't overallocate matrices on newer PolyLibs
102 - verification program for enumeration of integer projections
103 - summation of enumerations
104 - avoid gccisms
105 - support older versions of GMP
106 ---
107 version: 0.10
108 date: Tue Jul 6 22:09:11 CEST 2004
109 changes:
110 - install header files
111 - modulo representation changed to fractionals
112 - perform some simplifications on the modulo representation
113 - barvinok_enumerate returns a malloc'ed evalue
114 - a few bugs in barvinok_enumerate_e removed
115 - optionally print an indication of the size of an enumeration
116 - emask added
117 - bug in triangularization fixed
118 - bug in polyhedron reduction fixed
119 - verif_ehrhart.c should compile on C89 systems again
120 ---
121 version: 0.09
122 changes:
123 - bug in polyhedron reduction fixed
124 ---
125 version: 0.08
126 date: Sun Feb 22 21:15:00 CET 2004
127 changes:
128 - remove remaining part of Kristof's Ehrhart manipulation code
129 - optimize modulo expression based on validity domain
130 - support partition evalues
131 - factor out 1D polytopes
132 - validity domains can now be unions of polytopes
133 - optionally convert modulo representation to lookup-table
134 - rename indicator to relation
135 - add DomainIncludes
136 ---
137 version: 0.07
138 changes:
139 - support "indicator" evalues
140 - support equalities for modulo representation
141 ---
142 version: 0.06
143 changes:
144 - remove part of Kristof's Ehrhart manipulation code
145 - partial support for modulo respresentation
146 - support cvs version of PolyLib
147 ---
148 version: 0.05
149 date: Thu Nov 27 13:35:59 CET 2003
150 changes:
151 - |-
152 include code from Kristof Beyls for manipulating Ehrhart
153 polynomials
154 - |-
155 implement barvinok_enumerate for constructing Ehrhart
156 polynomials using barvinok's algorithm
157 ---
158 version: 0.04
159 date: Thu Nov 6 01:36:25 CET 2003
160 changes:
161 - |-
162 allow specification of max #constraints
163 for those without automatically growing chernikova tables
164 - fix a few small bugs
165 - some minor optimizations
166 ---
167 version: 0.03
168 date: Mon Nov 3 16:35:57 CET 2003
169 changes:
170 - fix a number of bugs
171 - add tests
172 ---
173 version: 0.02
174 date: Mon Oct 20 14:15:45 CEST 2003
175 changes:
176 - change name from count to barvinok_count
177 - return -1 if polyhedron is unbounded
178 - trivially optimize counting in presence of some independent constraints