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
|
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"
"http://www.w3.org/TR/REC-html40/loose.dtd">
<!-- $Id: start.html 111 2000-07-31 21:40:02Z garland $ -->
<html>
<head>
<title>libgfx: Getting Started</title>
<link rel=stylesheet href="cdoc.css" type="text/css">
<meta name="Author" content="Michael Garland">
</head>
<body>
<h2>Getting Started </h2>
<p>The <tt>libgfx</tt> library can be used for writing both
console-oriented and GUI programs. All programs, whether console or
GUI, should begin by including the standard header:
<pre>
#include <gfx/gfx.h>
</pre>
<h3>The Standard Header: gfx.h</h3>
<p>The first task of the standard header file is to provide a consist
code environment. It begins by including a set of standard C++
headers <tt><cstdlib></tt>, <tt><cmath></tt>,
<tt><climits></tt>, and <tt><iostream></tt>.
It then makes sure that various common symbols, such as <tt>bool</tt>,
<tt>M_PI</tt>, and <tt>HUGE</tt> are defined.
<h4>Numerical Procedures</h4>
<p>Graphics programs typically involve a substantial amount of
mathematical calculation. Indeed, much the the <tt>libgfx</tt>
library is devoted to supporting things such as matrix/vector
computations. Certain mathematical procedures are common enough, and
simple enough, that they are included in the standard header.
<p>First, there are the procedures for generating random numbers:
<pre>
inline double random1(); <i>// Random number between 0 and 1</i>
inline char random_byte(); <i>// Random byte between 0 and 255</i>
</pre>
These functions use the internal <tt>random()</tt> procedure if it's
available, or <tt>rand()</tt> if not.
<p>Next, are procedures for comparing floating point numbers:
<pre>
const double FEQ_EPS = 1e-6;
const double FEQ_EPS2 = 1e-12;
inline bool FEQ(double a, double b, double eps=FEQ_EPS);
inline bool FEQ2(double a, double b, double eps=FEQ_EPS2);
</pre>
The <tt>FEQ()</tt> procedures return <tt>true</tt> if <tt>a</tt> and
<tt>b</tt> are within <tt>eps</tt> of each other.
<h4>Timing Procedures</h4>
<p>To characterize the performance of a program, it is often useful to
measure it's running time. For this reason, <tt>libgfx</tt> provides
some basic facilities for measuring time.
<pre>
extern double get_cpu_time();
</pre>
The function returns the number of seconds on the CPU clock. Based on
the platform you're using, this function may use various system
services to compute this clock. Therefore, it does not use a
consistent measure across all systems; however, the returned quantity
will usually be the number of seconds since the system was turned on.
<p>A more convenient way to measure running time is to use the
following macro:
<pre>
#define TIMING(t, cmd) { t=get_cpu_time(); cmd; t=get_cpu_time() - t; }
</pre>
Given a procedure <tt>test_proc()</tt> whose performance we want to
measure, we can use the following code:
<pre>
double running_time;
TIME(running_time, test_proc());
cout << "The running time was: " << running_time << " seconds." << endl;
</pre>
This will print the running time of <tt>test_proc()</tt> on the console.
<h4>Configuration Definitions</h4>
<p>The final task of the standard header is to include the
<tt>libgfx</tt> configuration header. This header is generated when
the library is compiled. On Unix-like platforms, the header is named
<tt><gfx/config.h></tt> and is generated automatically by the
configuration script. For Microsoft Visual Studio platforms, a
hand-coded header such as <tt><gfx/config-vc5.h></tt> is used
instead. The symbols defined in this header can be used to detect the
presence of various language features and external libraries.
The symbols of interest to external programs are summarized below:
<br><br>
<table align=center width=85% border=0 cellpadding=5>
<tr valign=top><th align=left>Symbol</th><th align=left>Defined when ...</th></tr>
<tr valign=top><td><tt>HAVE_LIBTIFF</tt></td>
<td>library supporting TIFF image I/O is available.</td></tr>
<tr valign=top><td><tt>HAVE_LIBTIFF_LZW</tt></td>
<td><tt>libtiff</tt> supports patented LZW compression.</td></tr>
<tr valign=top><td><tt>HAVE_LIBPNG</tt></td>
<td>library supporting PNG image I/O is available.</td></tr>
<tr valign=top><td><tt>HAVE_LIBJPEG</tt></td>
<td>library supporting JPEG image I/O is available.</td></tr>
<tr valign=top><td><tt>HAVE_OPENGL</tt></td>
<td>OpenGL is available. Value is name of implementation
(e.g., "OpenGL" or "Mesa").</td></tr>
</table>
<p>These configuration symbols should be tested with <tt>#ifdef</tt>
directives, as in the following example:
<pre>
#ifdef HAVE_LIBTIFF
generate_tiff_output();
#endif
</pre>
</body>
</html>
|
