Sophie

Sophie

distrib > Mandriva > 9.1 > i586 > by-pkgid > 2f55e9965d93904f4bc397ff79857f26 > files > 76

htdig-3.2.0-0.6mdk.i586.rpm

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">
<html>
  <head>
	<title>
	  ht://Dig: Installation
	</title>
  </head>
  <body bgcolor="#eef7ff">
	<h1>
	  Installation
	</h1>
	<p>
	  ht://Dig Copyright &copy; 1995-2001 <a href="THANKS.html">The ht://Dig Group</a><br>
	  Please see the file <a href="COPYING">COPYING</a> for
	  license information.
	</p>
	<hr size="4" noshade>
	<p>
	  This document will attempt to show the steps needed to build
	  and install the ht://Dig system.<br>
	  The main sections are:
	</p>
	<ul>
	  <li>
		<a href="where.html">Getting the software</a>
	  </li>
	  <li>
		<a href="#extract">Extracting the software</a>
	  </li>
	  <li>
		<a href="#configure">Configure</a>
	  </li>
	  <li>
		<a href="#compile">Compile</a>
	  </li>
	  <li>
		<a href="#test">Test</a>
	  </li>
	  <li>
		<a href="#install">Install</a>
	  </li>
	  <li>
		<a href="#sharelib">Shared Libraries</a>
	  </li>
	</ul>
	<hr noshade>
	<h2>
	  <a name="extract">Extracting the software</a>
	</h2>
	<p>
	  The distribution of ht://Dig is in the form of a gzipped tar
	  file. The name of the file will be something like <code>
	  htdig-3.2.0.tar.gz</code>. To extract, you can use either the GNU
	  tar command as follows:
	</p>
	<blockquote>
	  % <strong>tar xzf <em>tarfile.tar.gz</em></strong>
	</blockquote>
	<p>
	  If you do not have GNU tar, you can do the following:
	</p>
	<blockquote>
	  % <strong>gunzip -c <em>tarfile.tar.gz</em> | tar xf -</strong>
	</blockquote>
	<p>
	  In either case, a new directory will be created under which the
	  distribution will be extracted. The directory it creates will
	  be <code>htdig-</code> followed by the version number.
	</p>
	<hr noshade>
	<h2>
	  <a name="configure">Configure</a>
	</h2>
	<p>
	  Once the distribution has been extracted, change to the newly
	  created directory.<br>
	  In this directory you will need execute the <code>configure</code> program:
	<blockquote>
	  % <strong>./configure</strong>
	</blockquote>
	<p>
	  This program will attempt to determine what your particular
	  system can and can't do.
	</p>
<!--
	<p>
	  The <strong>configure</strong> program may produce some warnings about
	  libguile and guile. Ignore those warnings since they do not
	  affect the ht://Dig program at all.
	</p>
-->
	<p>
	  Before running configure you can optionally add dependency information
	  to all the Makefiles with the command
	</p>
	<blockquote>
	  % <strong>automake</strong>
	</blockquote>
	<p>
	  This only needs to be done if you are going to change any of
	  the sources to ht://Dig and requires <strong>GCC</strong>.
	</p>
	<p>
	  The <strong>configure</strong> program has some ht://Dig specific options
	  in addition to the standard ones that you get when running
	  <pre>
	  ./configure --help
	  </pre>
	</p>
	<dl>
	  <dt>
		<code>--prefix=DIR</code>
	  </dt>
	  <dd>
		This is where all of the ht://Dig parts will be installed.
		Various other variables will use this value as their base.
		[default=/opt/www]
	  </dd>
	  <dt>
		<code>--bindir=DIR</code>
	  </dt>
	  <dd>
		All the ht://Dig executable programs will go here
		[default=/opt/www/bin].
	  </dd>
	  <dt>
		<code>--with-config-dir=DIR</code>
	  </dt>
	  <dd>
		where your config directory is [default=/opt/www/conf]
	  </dd>
	  <dt>
		<code>--with-default-config-file=FILE</code>
	  </dt>
	  <dd>
		Point this to where all the tools will look for the
		configuration file [default=/opt/www/conf/htdig.conf].
	  </dd>
	  <dt>
		<code>--with-common-dir=DIR</code>
	  </dt>
	  <dd>
		This directory is for files which can be shared between
		different search databases [default=/opt/www/share/htdig].
	  </dd>
	  <dt>
		<code>--with-database-dir=DIR</code>
	  </dt>
	  <dd>
		Set this to the directory where the search databases are
		going to be. (Make sure there is plenty of space on the
		partition you put this on!) [default=/opt/www/var/htdig].
	  </dd>
	  <dt>
		<code>--with-cgi-bin-dir=DIR</code>
	  </dt>
	  <dd>
		The directory where your HTTP server looks for CGI
		programs. This is where htsearch will get installed
		[default=/opt/www/cgi-bin].
	  </dd>
	  <dt>
		<code>--with-image-dir=DIR</code>
	  </dt>
	  <dd>
		Define this to be a place that can be accessed by your web
		server. This is where a couple of images will be installed
		[default=/opt/www/htdocs/htdig].
	  </dd>
	  <dt>
		<code>--with-image-url-prefix=LOCATION</code>
	  </dt>
	  <dd>
		This is the URL which points to the directory specified by
		the <code>--with-image-dir=DIR</code> option above
		[default=/htdig].
	  </dd>
	  <dt>
		<code>--with-search-dir=FILE</code>
	  </dt>
	  <dd>
		where the sample search form should be installed
		[default=/opt/www/htdocs/htdig].
	  </dd>
	  <dt>
		<code>--with-search-form=FILE</code>
	  </dt>
	  <dd>
		The name of the file in which the sample search form will
		be installed, relative to the directory specified with
		<code>--with-search-dir=DIR</code> [default=search.html]. 
	  </dd>
	</dl>
	<hr noshade>
	<h2>
	  <a name="compile">Compile</a>
	</h2>
	<p>
	  The configure program will have created Makefiles in all the
	  important directories. 
	<p>
	  If <strong>make</strong> is not able to deal with the generated Makefiles,
	  you should probably obtain 
	  <a href="ftp://ftp.gnu.org/pub/gnu/">GNU make</a>.
	</p>
	<p>
	  Now build the complete system with
	</p>
	<blockquote>
	  % <strong>make</strong>
	</blockquote>
	<p>
	  (Sit back and relax for a while...)
	</p>
	<p>
	  If the compilation failed with the error that it cannot find
	  libht.a, the most likely problem is that your system does not
	  have libstdc++ installed. Please check the
	  <a href="require.html">system requirements</a> for details on this.
	</p>
	<hr noshade>
	<h2>
	  <a name="test">Test</a>
	</h2>
	<p>
	  Once built, you can run tests if you provided the
	  <code>--enable-tests</code> at configure time. These tests
	  require that you have a working
	  <a href=http://www.apache.org/>Apache</a> daemon available with
	  version >= 1.3.1. To run the tests execute:
	</p>
	<p>
	  If your <a href=http://www.apache.org/>Apache</a> daemon is 
	  installed in a non standard place, specify it with the 
	  <code>--with-apache=PATH</code> configure option.
	</p>
	<blockquote>
	  % <strong>make check</strong>
	</blockquote>
	<hr noshade>
	<h2>
	  <a name="install">Install</a>
	</h2>
	<p>
	  Everything should have built at this point. To install the
	  software, you need to execute
	</p>
	<blockquote>
	  % <strong>make install</strong>
	</blockquote>
	<p>
	  This will perform several tasks. It will first attempt to
	  create the directories that you specified to <strong>./configure</strong>.
	  It will then copy the following programs to the <code>
	  --bindir=DIR</code> directory:
	</p>
	<ul>
	  <li>
		htdig
	  </li>
	  <li>
		htmerge
	  </li>
	  <li>
		htfuzzy
	  </li>
	  <li>
		htnotify
	  </li>
	  <li>
		htdump
	  </li>
	  <li>
		htstat
	  </li>
	  <li>
		htload
	  </li>
	</ul>
	<p>
	  It will also copy the htsearch program to your 
	  <code>--with-cgi-bin-dir=DIR</code> directory.
	</p>
	<p>
	  After this, several files will be customized and installed.
	  Here is a list of the files that get installed:
	</p>
	  <blockquote>
		<dl>
		  <dt>
			<em><img src="bdot.gif" width=9 height=9 alt="*"><code>--with-config-dir=DIR</code></em>/htdig.conf
		  </dt>
		  <dd>
			A minimal config file which can be used to create a
			search database for http://www.htdig.org/
		  </dd>
		  <dt>
			<em><img src="bdot.gif" width=9 height=9 alt="*"><code>--with-search-form=FILE</code></em>
		  </dt>
		  <dd>
			A sample HTML document that contains a search form.
		  </dd>
		  <dt>
			<em><img src="bdot.gif" width=9 height=9 alt="*"> <code>--with-common-dir=DIR</code></em>/footer.html
		  </dt>
		  <dd>
			A sample HTML document that can be used as the search
			results footer.
		  </dd>
		  <dt>
			<em><img src="bdot.gif" width=9 height=9 alt="*"> <code>--with-common-dir=DIR</code></em>/header.html
		  </dt>
		  <dd>
			A sample HTML document that can be used as the search
			results header.
		  </dd>
		  <dt>
			<em><img src="bdot.gif" width=9 height=9 alt="*"> <code>--with-common-dir=DIR</code></em>/nomatch.html
		  </dt>
		  <dd>
			A sample HTML document that can be used if nothing was
			found.
		  </dd>
		  <dt>
			<em><img src="bdot.gif" width=9 height=9 alt="*"> <code>--with-common-dir=DIR</code></em>/syntax.html
		  </dt>
		  <dd>
			A sample HTML document that will be displayed if the
			user entered an illegal boolean expression. found.
		  </dd>
		  <dt>
			<em><img src="bdot.gif" width=9 height=9 alt="*"> <code>--with-common-dir=DIR</code></em>/english.0
		  </dt>
		  <dd>
			Default list of words with affixes that is used by
			htfuzzy.
		  </dd>
		  <dt>
			<em><img src="bdot.gif" width=9 height=9 alt="*"> <code>--with-common-dir=DIR</code></em>/english.aff
		  </dt>
		  <dd>
			Default affix rule database that is used by htfuzzy.
		  </dd>
		  <dt>
			<em><img src="bdot.gif" width=9 height=9 alt="*"> <code>--with-image-dir=DIR</code></em>/star.gif
		  </dt>
		  <dd>
			The default star image that is used to rank matches.
		  </dd>
		  <dt>
			<em><img src="bdot.gif" width=9 height=9 alt="*"> <code>--with-image-dir=DIR</code></em>/star_blank.gif
		  </dt>
		  <dd>
			The default placeholder image that is the same size as
			the star, but is blank. This is used to align the
			results in the short listing.
		  </dd>
		  <dt>
			<em><img src="bdot.gif" width=9 height=9 alt="*"> <code>--with-image-dir=DIR</code></em>/htdig.gif
		  </dt>
		  <dd>
			The nifty ht://Dig logo.
		  </dd>
		  <dt>
			<em><img src="bdot.gif" width=9 height=9 alt="*"> <code>--with-image-dir=DIR</code></em>/button*.gif
		  </dt>
		  <dd>
			Sample images used to show the search result pages.
		  </dd>
		  <dt>
			<em><img src="bdot.gif" width=9 height=9 alt="*"> <code>--bindir=DIR</code></em>/rundig
		  </dt>
		  <dd>
			A sample shell script which will create a database.
		  </dd>
		  <dt>
			<em><img src="bdot.gif" width=9 height=9 alt="*"> <code>--prefix=DIR</code>/lib/htdig</em>/*.{a,so}
		  </dt>
		  <dd>
			The shared and static libraries.
		  </dd>
		  <dt>
			<em><img src="bdot.gif" width=9 height=9 alt="*"> <code>--prefix=DIR</code>/include/htdig</em>/*.h
		  </dt>
		  <dd>
			The header files that allow anyone to develop a program
			based on the internals of htdig.
		  </dd>
		</dl>
	  </blockquote>
	<p>
	  Note that these files will <em>never</em> replace any existing
	  files that may already be installed.
	</p>
	<p>
	  It is also important to note that these files are mostly
	  examples. As they stand, they will work, but you probably
	  want to modify them to reflect your needs.
	</p>
	<p>
	  After the installation, you will be ready to test out
	  everything. You can use the <code>rundig</code> script to make a
	  test database of the online documentation at
	  http://www.htdig.org/
	</p>
	<p>
	  The only thing left to do is to modify the <code>
	  htdig.conf</code> config file which was placed in <em>
	  <code>--with-config-dir=DIR</code></em>/htdig.conf. The <a href="attrs.html">
	  Configuration</a> manual has the details on how what
	  attributes are needed.
	</p>
	<hr noshade>
	<h2>
	  <a name="sharelib">Shared Libraries</a>
	</h2>
	<p>
	  By default ht://Dig is compiled with shared libraries. If
	  running on a platform other than Linux or FreeBSD, this may
	  be a problem. We recommend that you compile with the 
	  <code>--disable-shared</code> configure option.
	</p>
	<p>
	  If you installed with shared libraries, you must be sure the
	  system will find them. Usually it's done by adding the 
	  prefix/lib/htdig directory to the LD_LIBRARY_PATH.
	  This is really system dependent and you must check your 
	  documentation.  
	</p>
	<hr size="4" noshade>
	  <a href="author.html">Andrew Scherpbier &lt;andrew@contigo.com&gt; &amp; the ht://Dig Group</a>

	Last modified: $Date: 2001/02/15 17:05:33 $

  </body>
</html>