3 <!DOCTYPE article PUBLIC "-//Norman Walsh//DTD DocBk XML V3.1.4//EN"
4 "file:///usr/share/xml/docbook/schema/dtd/4.4/docbookx.dtd">
9 <articleinfo xmlns:xi="http://www.w3.org/2001/XInclude">
10 <title>Installation Guide</title>
11 <abstract>This article includes download, build, and installation
12 instructions.</abstract>
15 <edit by="David Hilvert" in-year="2007"/>
16 <edit by="David Hilvert" in-year="2006"/>
17 <edit by="David Hilvert" in-year="2005"/>
18 <edit by="David Hilvert" in-year="2004"/>
19 <edit by="David Hilvert" in-year="2003"/>
20 <edit by="David Hilvert" in-year="2002"/>
25 <p>Official binaries are not available for this alpha release.</p>
33 <s><t>Extraction and Use</t>
35 <p> The zip file available at the above URL is an archive containing the ALE
36 executable; a zip archival utility can be used to extract the executable from
39 <p> Once extracted, since ALE is a command-line program, it may be necessary to
40 open a command prompt window to use ALE successfully. For more information
41 about command-line usage and options, see the User Manual. </p>
47 <s><t>Compiling from source</t>
56 <s><t>Build prerequisites</t>
58 <s><t>GCC C++ compiler</t>
60 <p> GCC version 3.2.3 with C++ support is known to build ALE successfully. Older
61 versions of GCC may fail to compile some versions of ALE.</p>
65 <s><t>Prerequisites for building on Windows</t>
67 <p> Successfully running the configuration script on Windows (e.g., for native
68 builds) may require installation of Cygwin. For more information on Cygwin,
71 <ul><li><ulink url="http://www.cygwin.com/"/>
74 <p>Alternatively, this package can be cross-compiled for Windows on a non-Windows
75 architecture, provided that the appropriate cross-compilers and libraries are
84 <p>Download the source archive file from the URL indicated in the section
85 'URLs', and invoke:</p>
87 <ll>tar xzf <sourcepacktargz/>
91 <p>Invoking 'ls configure' afterward should reveal an executable script
92 'configure'. This script is used to determine the capabilities of the system
93 that will run the software.</p>
97 <s><t>Running configure</t>
99 <p>This package uses a configuration script, 'configure', generated by the GNU
100 project build tools ("autotools"). For most cases, it should be sufficient to
108 <p>In cases where the defaults are not adequate, configuration options can be
109 passed to the configuration script.</p>
112 <s><t>Configuration options</t>
114 <p>Configuration options can be passed to the 'configure' script prior to build
115 and installation, as:</p>
117 <ll>./configure --config-option1 --config-option2 ...
122 <p>Options specific to this package are described in the following subsections.
123 General configuration options are described in the file 'INSTALL' included in
124 the source package.</p>
126 <s><t>ImageMagick Support</t>
128 <p>Enabling ImageMagick support allows ALE to make use of the file-handling
129 capabilities of ImageMagick ( <ulink url="http://imagemagick.org/"/> ). This
130 includes reading and writing a variety of file formats, including, among many
131 others, PNG and JPEG (two formats commonly used in web publishing) as well as
132 common interchange formats such as PPM and TIFF. ImageMagick 6.0.6 is known to
133 work with this version of ALE; certain older versions of ImageMagick may cause
134 compile, link, or runtime errors. By default, support is automatically
137 <ll>--with-imagemagick=auto Auto-detect ImageMagick support [default]
138 --with-imagemagick=yes Enable ImageMagick support
139 --with-imagemagick=no Disable ImageMagick support
143 <s><t>Color Data Precision</t>
145 <p>ALE's internal data structures store each color channel within a pixel as
146 floating-point data; similarly, arithmetic operations on color channels are
147 performed in floating point. The available data-type configurations are
148 listed below, with typical data sizes indicated in parentheses. </p>
150 <ll>--with-colors=single Use C++ 'float' data type (32-bit) [default]
151 --with-colors=double Use C++ 'double' data type (64-bit)
156 Coordinate Data Precision
159 <p>In general, coordinate transformations used by ALE produce results outside of
160 the set of integers; hence, coordinates are treated as floating-point data.
161 The two available data-type configurations are listed below, with typical data
162 sizes indicated in parentheses.</p>
164 <ll>--with-coords=single Use C++ 'float' data type (32-bit) [default]
165 --with-coords=double Use C++ 'double' data type (64-bit)
174 Threads can improve performance on multiprocessor machines. POSIX threads are
178 <ll>--enable-threads=auto Auto-detect thread support. [default]
179 --enable-threads=yes Enable thread support.
180 --enable-threads=no Disable thread support.
189 This feature is required in order to use the --fl option; it also determines
190 the approach to calculation of discrete Fourier transforms used by the
191 Irani-Peleg renderer. For more information on FFTW, see http://fftw.org/.
194 <ll>--with-fftw3=auto Auto-detect FFTW3. [default]
195 --with-fftw3=yes Use FFTW3 discrete Fourier transform.
196 --with-fftw3=no Use built-in discrete Fourier transform.
206 This build option is required in order to use the --wmx option. If
207 --enable-wmx is specified, it is expected that fork() and execlp() calls
211 <ll>--enable-wmx=auto Automatically detect whether to enable --wmx
212 --enable-wmx=yes Enable --wmx
213 --enable-wmx=no Do not enable --wmx
220 <p>This build option is required to obtain additional output from --ui=tty. If
221 TIOCGWINSZ is not defined in <sys/ioctl.h>, however, additional output will not
222 be provided even if --enable-getsize is specified at build time.
225 <ll>--enable-getsize=auto Check for TIOCGWINSZ at build time. [default]
226 --enable-getsize=yes Configure to attempt to use TIOCGWINSZ
227 --enable-getsize=no Do not attempt to use TIOCGWINSZ
237 Finding bugs within ALE, or within custom modifications to ALE, can be
238 simplified by enabling run-time assertions. These checks enforce conditions
239 believed to be necessary or desirable for correct program operation. If a
240 condition is not satisfied, ALE will abort and display a message indicating the
244 <ll>--enable-assertions Enable assertions. [default]
245 --disable-assertions Disable assertions.
251 <s><t>Cross-compilation</t>
253 <p>For cross-compilation, it may be necessary to specify both host and build
254 machine types, using the configure --host and --build options. For more
255 information on these options, see the file 'INSTALL' in the source package.</p>
258 <s><t>Further details</t>
260 <p>General operation of the configure script is described in more detail in the
261 file 'INSTALL' in the source package.</p>
267 <s><t>Further Documentation</t>
269 <p>This alpha release only includes standard package documentation files.</p>
274 <s><t>On-line resources</t>
276 <ll>Home Page http://auricle.dyndns.org/ALE/
277 darcs repository http://auricle.dyndns.org/darcs/ale/
278 Mailing list archive http://ventricle.dyndns.org/pipermail/ale/
281 <s><t>Mailing list</t>
283 <p>Bug reports, feature requests, patch and release announcements, and related
284 discussion can be sent to ale@ventricle.dyndns.org. Archives of past postings
285 and subscription information are available at the URLs indicated below.</p>
287 <ll>Posting address ale@ventricle.dyndns.org
288 Subscription info http://ventricle.dyndns.org/cgi-bin/mailman/listinfo/ale
289 Archives http://ventricle.dyndns.org/pipermail/ale/