vecmath-cpp/README at main · kenjihiranabe/vecmath-cpp · GitHub

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
THIS IS OBSOLETE. READ README.md/README-ja.md
I KEEP THIS FOR THE HISTORY REASON.

-----------------------------------------------------
 C++ port of unofficial javax.vecmath package
         by Kenji Hiranabe

                                3/5 ,1999 version 1.1-1.0
                               5/25 ,1999 version 1.1-1.1
                               6/12 ,1999 version 1.1-1.2
                               10/5 ,1999 version 1.2-1.3
                              10/22 ,1999 version 1.2-1.4
-----------------------------------------------------

* INTRODUCTION

This is a C++ port of unofficial free implementation
(source code release) of Java(TM) 3D API 1.2
vecmath package.

This package supports vector/point transformation math
often utilized in computer graphics.

The included classes are the followings.

    * Tuples

      Tuple2<T>      Tuple3<T>     Tuple4<T>
        |             |             |
        +Vector2<T>   +Vector3<T>   +Vector4<T>
        +Point2<T>    +Point3<T>    +Point4<T>
        +TexCoord2<T> +Color3<T>    +Quat4<T>
                                    +Color4<T>
      AxisAngle4<T>

    * Matrices

      Matrix3<T>     Matrix4<T>

These classes are (optionally) in the namespace of 'kh_vecmath'.

You can download the zipped or tar+gzipped source code from;

       http://www.esm.co.jp/divisions/open-sys/java/vecmath/

* FEATURES

 ** Highest priority to SPEED. No 'new' operators, no arrays, no virual functions.
 ** All files are *.h template header files. No need for linking libraries.
 ** Has a test suite.
 ** Aliasing-safe. i.e. v.cross(v,v1) or m.mul(m1,m) works.
 ** optional IO stream and  string conversion support.
 ** Java3D 1.2 specification is implemented, but no GVector and GMatrix yet.
 ** Works with standard C++ library (STL).

* INSTALL

1. unzip and untar the downloaded file, like;
      % gunzip vecmath-c++-1.1-1.1.tar.gz
      % tar xvf vecmath-c++-1.1-1.1.tar
2. place all *.h files into your header directory.
      % cp vecmath-c++/*.h <yourdir>
3. write your C++ code, hello-vecmath.cpp

    #include <iostream>
    #include "vecmath.h"
    using namespace kh_vecmath;   // depends on vm_config.h
    int main(int, char**) {
       Matrix3d m(1,2,3,4,5,6,7,8,9);
       Vector3d v(2,3,4);
       cout << m*v;
       return 0;
    }

4. compile, and run. don't forget to set include path to <yourdir>
   when compilation.

on unix,
    % g++ -I<yourdir> hello-vecmath.cpp -o hello-vecmath
    % hello-vecmath

5. (20,47,74) is printed, voila !
6. if needed, configure "vm_conf.h".
7. if needed, compile test programs(test-*.cpp) and run.
on unix,
    % cp Makefile.unix Makefile
    % make test

on Windows with Visual Studio, use 'vecmath.dsw'for workspace.

* TESTED PLATFORMS AND COMPILERS

(1) Sun Solaris2.6 on SPARCStation with gcc2.7.2(+ libstdc++-2.8.1.1),
   gcc2.8.1 and egcs1.1.2

for gcc2.7.2, gcc2.8.1 namespace is not supported. vm_conf.h will automaticlly
turns off VM_INCLUDE_NAMESPACE and eliminates namespace kh_vecmath.

(2) Visual C++ 6.0 sp2
  (Thanks > "Lucas Ammon" <lammon@iam.unibe.ch>)

(3) Windows 95 + Cygwin b20 egcs-2.91.66 19990314 (egcs-1.1.2 release)
  (Thanks > Takayuki MATSUOKA <matsuoka@osaka.whoopeecamp.co.jp>)

(4) IRIX6.4 + MIPSPro7.2
  (Thanks > Katsuaki KAWACHI <kawachi@cim.pe.u-tokyo.ac.jp>)

(5) Visual C++ 5.0
  (Thanks > Takayuki MATSUOKA <matsuoka@osaka.whoopeecamp.co.jp>)
  Add "/D TESTALL" and "/I .\\" to preprocessor macro when building
  test project.

..... please report on your platforms ! .....


* TIPS

** If you don't want namespace, #undef VM_INCLUDE_NAMESPAE in vm_conf.h.

** If you don't want toString() method, #undef VM_INCLUDE_TOSTRING
 in vm_conf.h. This may solve compilation errors in other compilers.

** If you don't want cout << operation, #undef VM_INCLUDE_IO
 in vm_conf.h. This may solve compilation errors in other compilers.

** All the math functions such as sin,cos,atan2 are encapsulated
in VmUtil.h. If you have your own faster version, adjust the file.

* FEEDBACK

Any bug reports or comments are welcome.
If you think this package is useful, PLEASE give me feedback.
I would like to hear what kind of applications you are working on !

---
  Eiwa System Management, Inc.  http://www.esm.co.jp/
  Kenji Hiranabe                mailto:hiranabe@esm.co.jp