forked from latte-int/latte
-
Notifications
You must be signed in to change notification settings - Fork 0
/
NEWS
179 lines (94 loc) · 4.76 KB
/
NEWS
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
Changes in release 1.7.6, 2021-03-17:
This is a portability release.
* Fix for new NTL versions
* Find cddlib headers in new installation location (cddlib/...)
* Remove problematic code and compiler options
Changes in release 1.7.5, 2018-09-10:
This is a portability release.
* Fix compilation errors for NTL version >= 9: don't depend on NTL_STD_CXX macro
* Fix runtime errors for NTL builds with NTL_RANGE_CHECK=on and for NTL version >= 11
Changes in release 1.7.4, 2018-09-03:
This is a bug fix and portability release.
* Fix compile errors with clang.
* Make it an error, not a warning, when cddlib is not found at configure time.
* Make test suite pass when configured without 4ti2.
* Relax test for location of gmp.h, needed for Ubuntu
Changes in release 1.7.3:
* count and integrate commands now correctly return a non-zero error code on failure
* Fixes problems when building on Macs
Changes in release 1.7.2:
* Maple versions of some LattE commands are now available; see code/maple/README.
* Fixes problems when building on Macs
Changes in release 1.7.1:
* Fixes problems when building on Macs
* Improved the top-ehrhart-knapsack command
Changes in release 1.7:
* New program for computing the highest coefficients of Ehrhart
quasipolynomials for knapsack polytopes.
Changes in release 1.6:
* New program for computing top Ehrhart coefficients (see manual).
* Fix problems with whitespace in filenames or installation directory names.
Changes since release 1.4:
* New program for computing volumes and integrating polynomials or powers of linear forms over polytopes.
* Computing volume by triangulating
integrate --triangulate --valuation=volume
* Computing volume by cone decomposition
integrate --cone-decompose --valuation=volume
* Computing integrals by triangulating
integrate --triangulate --valuation=integrate [--monomials=file | --linear-forms=file]
* Computing integrals by cone decomposition
integrate --cone-decompose --valuation=integrate [--monomials=file | --linear-forms=file]
Changes since the official release 1.2:
* Use GNU Autoconf and Automake.
* Portability fixes.
* Code clean-up.
* Some performance improvements.
* New exponential substitution
count --exp
* New primal irrational decomposition
count --irr
* New non-unimodular enumeration
count --maxdet=N
* New all-primal irrational decomposition
count --all-primal
* New method for computing Ehrhart polynomials of integral polytopes
in the primal space
count --all-primal --ehrhart-polynomial
* The homogenized method now allows V-representation input
count homog vrep
* The functionality of `ehrhart' has been merged into `count':
count --ehrhart-series FILENAME
(replaces: ehrhart FILENAME)
count --simplified-ehrhart-series FILENAME
(replaces: ehrhart simplify FILENAME)
count --ehrhart-taylor=N FILENAME
(replaces: ehrhart N FILENAME)
(The `ehrhart' program is still available, but it does not accept
the new command-line options of `count'.)
* The 4ti2 program (http://www.4ti2.de) can be used instead of cddlib and CDD+
to compute the vertex cones of polytopes and triangulations and duals of cones.
In many cases, 4ti2 is faster.
./configure --with-4ti2=PATH-TO-4ti2
count --triangulation=4ti2 --dualization=4ti2 --compute-vertex-cones=4ti2
* New command-line option --redundancy-check={none,cddlib,full-cddlib}.
- "full-cddlib" (the default) uses cddlib to compute an irredundant
system of linear equations and inequalities describing the
polyhedron. This corresponds to the traditional LattE behavior;
it can be expensive.
- "cddlib" (used to be the default in the 1.2+mk-0.9.x series) uses
cddlib to compute some implicit linearities only; it often fails
but is faster than full-cddlib.
- "none" does nothing, the input description of the polytope should
be irredundant.
* New programs, `latte2ine' and `latte2ext', are built that perform
the trivial conversion from LattE-style input files to CDD-style
input files.
latte2ine < FILENAME.latte > FILENAME.ine
latte2ext < FILENAME.vrep.latte > FILENAME.ext
* New command-line option --multivariate-generating-function.
It allows to compute the multivariate generating function of a
rational polyhedron. For unbounded polyhedra, one needs to combine
it with --compute-vertex-cones=4ti2, since the traditional method in
LattE (using CDD+) currently refuses to handle unbounded polyhedra.
count --compute-vertex-cones=4ti2 --multivariate-generating-function FILENAME
writes the multivariate generating function (in Maple notation) to "FILENAME.rat".