Parent: [e41216] (diff)

Child: [c03f46] (diff)

Download this file

usermanual.sgml    3969 lines (3336 with data), 179.1 kB

<!DOCTYPE BOOK PUBLIC "-//FreeBSD//DTD DocBook V4.1-Based Extension//EN" [

<!ENTITY RCL "<application>Recoll</application>">
<!ENTITY RCLAPPS "<ulink url='http://www.recoll.org/features.html'>Recoll helper applications page</ulink>">
<!ENTITY RCLVERSION "1.15">
<!ENTITY XAP "<application>Xapian</application>">
]>
 
<book lang="en">
  
  <bookinfo>
    <title>Recoll user manual</title>

    <author>
      <firstname>Jean-Francois</firstname>
      <surname>Dockes</surname>
      <affiliation>
        <address><email>jfd@recoll.org</email></address>
      </affiliation>
    </author>

    <copyright>
      <year>2005-2011</year>
      <holder role="mailto:jfd@recoll.org">Jean-Francois
      Dockes</holder>
    </copyright>

    <releaseinfo>$Id: usermanual.sgml,v 1.71 2008-12-15 09:33:49 dockes Exp $</releaseinfo>

    <abstract>
      <para>This document introduces full text search notions
      and describes the installation and use of the &RCL;
      application. It currently describes &RCL; &RCLVERSION;.</para>
    </abstract>


  </bookinfo>
  
  <chapter id="rcl.introduction">
    <title>Introduction</title>

    <sect1 id="rcl.introduction.tryit">
      <title>Giving it a try</title>
      
      <para>If you do not like reading manuals (who does?) and would
      like to give &RCL; a try, just perform <link
      linkend="rcl.install.binary">installation</link> and start the
      <command>recoll</command> user interface, which will index your
      home directory by default, allowing you to search immediately after
      indexing completes.</para>

      <para>Do not do this if your home directory contains a huge
        number of documents and you do not want to wait or are very
        short on disk space. In this case, you may first want to customize
        the <link linkend="rcl.indexing.config">configuration</link>
        to restrict the indexed area.</para> 

      <para>Also be aware that you may need to install the
        appropriate <link linkend="rcl.install.external"> supporting
        applications</link> for document types that need them (for
        example <application>antiword</application> for ms-word
        files).</para>
      
    <sect1 id="rcl.introduction.search"> 
      <title>Full text search</title>

      <para>&RCL; is a full text search application. Full text search
        applications let you find your data by content rather
        than by external attributes (like a file name). More
        specifically, they will let you specify words (terms) that
        should or should not appear in the text you are looking for,
        and return a list of matching documents, ordered so that the
        most <emphasis>relevant</emphasis> documents will appear
        first.</para>

      <para>You do not need to remember in what file or email message you
        stored a given piece of information. You just ask for related
        terms, and the tool will return a list of documents where
        those terms are prominent, in a similar way to Internet search
        engines.</para>

      <para>A search application tries to determine which documents are
        most relevant to the search terms you provide. Computer algorithms
        for determining relevance can be very complex, and in general are
        inferior to the power of the human mind to rapidly determine
        relevance. The quality of relevance guessing is probably the most
        important aspect when evaluating a search application.</para>

      <para>In many cases, you are looking for all the forms of a
        word, not for a specific form or spelling. These different forms
        may include plurals, different tenses for a verb, or terms derived
        from the same root or <emphasis>stem</emphasis> (example: floor,
        floors, floored, flooring...). Search applications usually expand
        queries to all such related terms (words that reduce to the same
        stem) and also provide a way to disable this expansion if you are
        actually searching for a specific form.</para>

      <para>Stemming, by itself, does not accommodate for misspellings or
        phonetic searches. &RCL; supports these features through a specific
        tool (the <literal>term explorer</literal>) which will let you
        explore the set of index terms along different modes.</para>


    </sect1>

      <sect1 id="rcl.introduction.recoll">
      <title>Recoll overview</title>

      <para>&RCL; uses the 
      <ulink url="http://www.xapian.org">&XAP;</ulink> information retrieval
      library as its storage and retrieval engine. &XAP; is a very
      mature package using <ulink
      url="http://www.xapian.org/docs/intro_ir.html">a sophisticated
      probabilistic ranking model</ulink>. &RCL; provides the mechanisms
      and interface to get data into and out of the system.</para>

      <para>In practice, &XAP; works by remembering where terms appear
      in your document files. The acquisition process is called
      indexing. </para> 

      <para>The resulting index can be big (roughly the size of the
        original document set), but it is not a document
        archive. &RCL; can only display documents that still exist at
        the place from which they were indexed. (Actually, there is a
        way to reconstruct a document from the information in the
        index, but the result is not nice, as all formatting,
        punctuation and capitalization are lost).</para>

      <para>&RCL; stores all internal data in <application>Unicode
      UTF-8</application> format, and it can index files with
      different character sets, encodings, and languages into the same
      index. It has input filters for many document types.</para>
      
      <para>Stemming depends on the document language. &RCL; stores
      the unstemmed versions of terms and uses auxiliary databases for
      term expansion. It can switch stemming languages, or add a
      language, without re-indexing.  Storing documents in different
      languages in the same index is possible, and useful in
      practice, but does introduce possibilities of confusion. &RCL;
      currently makes no attempt at automatic language recognition.</para>

      <para>&RCL; has many parameters which define exactly what to
        index, and how to classify and decode the source
        documents. These are kept in <link
        linkend="rcl.indexing.config">configuration files</link>. A
        default configuration is copied into a standard location
        (usually something like
        <filename>/usr/[local/]share/recoll/examples</filename>)
        during installation. The default parameters from this file may
        be overridden by values that you set inside your personal
        configuration, found by default in the
        <filename>.recoll</filename> sub-directory of your home
        directory. The default configuration will index your home
        directory with default parameters and should be sufficient for
        giving &RCL; a try, but you may want to adjust it
        later.</para>

      <para><link linkend="rcl.indexing.periodic.exec">Indexing</link>
      is started automatically the first time you execute the
      <command>recoll</command> search graphical user interface, or by
      executing the <command>recollindex</command> command.</para>

      <para><link linkend="rcl.search">Searches</link> are usually
        performed inside the <command>recoll</command> graphical user
        interface (GUI) program, which has many options to help you find
        what you are looking for. However, there are other ways to perform
        &RCL; searches: mostly a <link linkend="rcl.search.commandline">
          command line tool</link>, a 
        <link linkend="rcl.program.api.python">
          <application>Python</application>
          programming interface</link>, and a <link linkend="rcl.searchkio">
          <application>KDE</application> KIO slave module</link>.</para>

    </sect1>
  </chapter>


  <chapter id="rcl.indexing">
    <title>Indexing</title>

    <sect1 id="rcl.indexing.introduction">
      <title>Introduction</title>

      <para>Indexing is the process by which the set of documents is
      analyzed and the data entered into the database. &RCL; indexing
      is normally incremental: documents will only be processed if
      they have been modified. On the first execution, of course, all
      documents will need processing. A full index build can be forced
      later by specifying an option to the indexing command
      (<command>recollindex -z</command>).</para> 

      <para>&RCL; indexing can be performed with two different
       methods:</para>

      <itemizedlist>

        <listitem>
          <formalpara><title>Periodic indexing:</title>
            <para>indexing takes place at discrete
              times, by executing the <command>recollindex</command>
              command. The typical usage is to have a nightly indexing run 
              <link linkend="rcl.indexing.periodic.automat">programmed</link>
              into your <command>cron</command> file.</para>
          </formalpara>
        </listitem>

        <listitem>
          <formalpara><title>Real time indexing:</title>
            <para>indexing takes place as soon as a file is created or
              changed. <command>recollindex</command> runs as a daemon
              and uses a file system alteration monitor such as 
              <application>inotify</application>, 
            <application>Fam</application> or
            <application>Gamin</application>
            to detect file changes.</para>
          </formalpara>
        </listitem>
      </itemizedlist>

      <para>The choice between the two methods is mostly a matter of
        preference, and they can be combined by setting up multiple
        indexes (ie: use periodic indexing on a big documentation
        directory, and real time indexing on a small home
        directory). Monitoring a big file system tree can consume
        significant system resources.<para>

      <para>&RCL; knows about quite a few different document
        types. The parameters for document types recognition and
        processing are set in 
        <link linkend="rcl.indexing.config">configuration files</link>.</para>

      <para>Most file types, like HTML or word processing files, only hold
        one document. Some file types, like mail folder files or zip
        archives, can hold many individually indexed documents, which may
        in turn be themselves compound ones. Such hierarchies can go quite
        deep, and &RCL; has no problem processing, for example, an ms-word
        document which would be an attachment to an email message part of
        a folder file archived inside a zip file...</para>

      <para>&RCL; indexing processes plain text, HTML, openoffice
        and e-mail files internally (a few more actually).</para>

      <para>Other file types (ie: postscript, pdf, ms-word, rtf ...) 
        need external applications for preprocessing. The list is in the
        <link linkend="rcl.install.external"> installation</link>
        section. After every indexing operation, &RCL; updates a list of
        commands that would be needed for indexing existing files
        types. This list can be displayed from the
        <command>recoll</command> <guilabel>File</guilabel> menu. It is
        stored in the <filename>missing</filename> text file
        inside the configuration directory.</para>

      <para>Without further configuration, &RCL; will index all
        appropriate files from your home directory, with a reasonable
        set of defaults.</para>

      <para>In some cases, it may be interesting to index different
	areas of the file system to separate databases. You can do this
	by using multiple configuration directories, each indexing a
	file system area to a specific database. See the 
	<link linkend="rcl.search.multidb">section about using multiple
	  databases</link> for more information on multiple configurations
	and indexes. </para>

      <para>In the rare case where the index becomes corrupted (which can
	signal itself by weird search results or crashes), the index files
	need to be erased before restarting a clean indexing pass. Just delete
	the <filename>xapiandb</filename> directory (see 
	<link linkend="rcl.indexing.storage">next section</link>), or, 
	alternatively, start the next <command>recollindex</command> with the 
	<literal>-z</literal> option, which will reset the database before
	indexing.</para>


    </sect1>

    <sect1 id="rcl.indexing.storage">
      <title>Index storage</title>

      <para>The default location for the index data is the
      <filename>xapiandb</filename> subdirectory of the &RCL;
      configuration directory, typically
      <filename>$HOME/.recoll/xapiandb/</filename>. This can be
      changed via two different methods (with different purposes):</para>

      <itemizedlist>

          <listitem><para>You can specify a different configuration
          directory by setting the <literal>RECOLL_CONFDIR</literal>
          environment variable, or using the <literal>-c</literal>
          option to the &RCL; commands. This method would typically be
          used to index different areas of the file system to
          different indexes. For example, if you were to issue the
          following commands: 
          <programlisting>
export RECOLL_CONFDIR=~/.indexes-email
recoll
          </programlisting> Then &RCL; would use configuration files
          stored in <filename>~/.indexes-email/</filename> and,
          (unless specified otherwise in
          <filename>recoll.conf</filename>) would look for 
          the index in <filename>~/.indexes-email/xapiandb/</filename>.

          <para>Using multiple configuration directories and 
          <link linkend="rcl.install.config.recollconf">configuration
          options</link> allows you to tailor multiple configurations
          and indexes to handle whatever subset of the available data
          that you wish to make searchable.</para>

          </listitem>
          <listitem><para>You can also specify a different storage
          location for the index by setting the <literal>dbdir</literal>
          parameter in the configuration file 
          (see the <link linkend="rcl.install.config.recollconf">configuration
          section</link>). This method would mainly be of use if you
          wanted to keep the configuration directory in its default location,
          but desired another location for the index, typically out of
          disk occupation concerns.</para>
          </listitem>

        </itemizedlist>

      <para>The size of the index is determined by the document set size,
        but the ratio can vary a lot. For a typical mixed
        set of documents, the index size will often be close to
        the data set size. In specific cases (a set of compressed
        mbox files for example), the index can become much bigger than
        the documents. It may also be much smaller if the documents
        contain a lot of images or other non-indexed data (an extreme
        example being a set of mp3 files where only the tags would be
        indexed).</para>

      <para>Of course, images, sound and video do not increase the
        index size, which means that it will be quite typical nowadays
        (2006), that even a big index will be negligible against the
        total amount of data on the computer.</para>
      
      <para>The index data directory (<filename>xapiandb</filename>)
	only contains data that can be completely rebuilt by an index
	run, and it can always be destroyed safely.</para>

      <sect2 id="rcl.indexing.storage.format">
        <title>Xapian index formats</title>

        <para>If your first installation of &RCL; was 1.9.0 or more
          recent, you can skip this section.</para>

        <para>&XAP; has had two possible index formats for quite some
          time. The "old" one named <literal>Quartz</literal>, and the
          new one named <literal>Flint</literal>. &XAP; 0.9 used
          <literal>Quartz</literal> by default, but could use
          <literal>Flint</literal> if a specific environment variable
          (<literal>XAPIAN_PREFER_FLINT</literal>) was set. &XAP; 1.0
          still supports <literal>Quartz</literal> but will use
          <literal>Flint</literal> by default for new index
          creations.</para>

        <para>The number of disk accesses performed during indexing
          has been much optimized in the new <literal>Flint</literal>
          engine and you may see indexing times improved by 50% in some
          cases (compared to <literal>Quartz</literal>), typically for
          big indexes where disk accesses dominate the indexing
          time. There is also a more modest improvement of index
          size.</para>

        <para>&XAP; will not convert automatically an existing index
          from the <literal>Quartz</literal> to the
          <literal>Flint</literal> format. If you have an older index
          and want to take advantage of the new format (which can be
          done without setting the environment variable as of &RCL;
          1.8.2 and &XAP; 1.0.0), you will have to explicitly delete
          the old index, then run a normal indexing process.</para>

        <para>Unfortunately, using the <literal>-z</literal> option to
          <command>recollindex</command> is not sufficient to change the
          format, you have to delete all files inside the index
          directory (typically <filename>~/.recoll/xapiandb</filename>)
          before starting indexing.</para>

      </sect2>

      <sect2 id="rcl.indexing.storage.security">
        <title>Security aspects</title>

        <para>The &RCL; index does not hold copies of the indexed
          documents. But it does hold enough data to allow for an almost
          complete reconstruction. If confidential data is indexed,
          access to the database directory should be restricted. </para>

        <para>As of version 1.4, &RCL; will create the configuration
          directory with a mode of 0700 (access by owner only). As the
          index data directory is by default a sub-directory of the
          configuration directory, this should result in appropriate
          protection.</para> 

        <para>If you use another setup, you should think of the kind
          of protection you need for your index, set the directory
          and files access modes appropriately, and also maybe adjust
          the <literal>umask</literal> used during index updates.</para>
        

      </sect2>

    </sect1>

    <sect1 id="rcl.indexing.config">
      <title>Indexing configuration</title>

      <para>Variables set inside the 
        <link linkend="rcl.install.config">&RCL; configuration files</link>
        control which areas of the file system are indexed, and how
        files are processed. These variables can be set either by
        editing the text files or using the dialogs in the
        <command>recoll</command> GUI.</para>

      <para>You can also use <link linkend="rcl.search.multidb">multiple 
          indexes</link> defined by separate configurations, typically to 
        separate personal and shared indexes, or to take advantage of
        the organization of your data to improve search precision.</para> 

      <para>The first time you start <command>recoll</command>, you
        will be asked whether or not you would like it to build the
        index. If you want to adjust the configuration before indexing,
        just click <guilabel>Cancel</guilabel> at this point, which will get
        you into the configuration interface. If you exit, 
        <filename>recoll</filename> will have created a ~/.recoll directory
        containing empty configuration files, which you can edit by hand.</para>

      <para>The configuration is documented inside the 
        <link linkend="rcl.install.config">installation chapter</link> 
        of this document, or in the recoll.conf(5) man page, but the most
        current information will most likely be the comments inside the
        sample file. The most immediately useful variable you may
        interested in is probably 
        <link linkend="rcl.install.config.recollconf.topdirs">topdirs</link>,
        which determines what subtrees get indexed.</para>

      <para>The applications needed to index file types other than
        text, HTML or email (ie: pdf, postscript, ms-word...) are
        described in the <link linkend="rcl.install.external">external
          packages section</link></para>

      <sect2 id="rcl.indexing.config.gui">
        <title>The indexing configuration GUI</title>

        <para>Most parameters for a given indexing configuration can
        be set from a <command>recoll</command> GUI running on this
        configuration (either as default, or by setting
        <literal>RECOLL_CONFDIR</literal> or the <literal>-c</literal>
        option.)</para> 

        <para>The interface is started from the
        <guilabel>Preferences</guilabel> menu. It has two main
        panels. The first panel allows setting global variables, like
        the list of top directories or the list of skipped paths. The
        second panel allows setting variables that can be redefined
        for subdirectories. This second panel has an initially empty list of
        customisation directories, to which you can add. The variables
        are then set for the currently selected directory (or at the top
        level if the empty line is selected).</para>

        <para>The meaning for most entries in the interface is 
        self-evident and documented by a <literal>ToolTip</literal>
        popup on the text label. For more detail, you will need to
        refer to the <link linkend="rcl.install.config">configuration
        section</link> of this guide.</para>

        <para>The configuration tool normally respects the comments
        and most of the formatting inside the configuration file, so
        that it is quite possible to use it on hand-edited files,
        which you might nevertheless want to backup first...</para>

      </sect2>

    </sect1>

    <sect1 id="rcl.indexing.beaglequeue">
      <title>Using Beagle WEB browser plugins</title>

      <para><application>Beagle</application> is (was?) a concurrent desktop
	indexer, built on Lucene and the Mono project (C#), for which a
	number of add-on browser plugins were written. These work by
	copying visited web pages to an indexing queue directory, which the
	indexer then processes. Especially, there is a
	<application>Firefox</application> extension.</para>

      <para>If, for any reason, you so happen to prefer &RCL; to
	<application>Beagle</application>, you can still use the
	<application>Firefox</application> plugin, which is written in
	Javascript and completely independant of C#, Beagle, Lucene..., and
	set &RCL; to process the <application>Beagle</application> queue
	directory. This supposes that <application>Beagle</application> is
	not running, else both programs will fight for the same
	files.</para>

      <para>This feature can be enabled in the GUI indexing configuration
	panel, or by editing the configuration file (set
	<literal>processbeaglequeue</literal> to 1).</para>

      <para>There are more recent instructions about how to find and
      install the <application>Firefox</application> extension on the 
        <ulink url="https://bitbucket.org/medoc/recoll/wiki/IndexBeagleWeb">
         Recoll wiki</ulink>.</para>

    </sect1>

    <sect1 id="rcl.indexing.periodic">
      <title>Periodic indexing</title>

      <sect2 id="rcl.indexing.periodic.exec">
        <title>Running indexing</title>

        <para>Indexing is performed either by the
          <command>recollindex</command> program, or by the
          indexing thread inside the <command>recoll</command>
          program (use the <guimenu>File</guimenu> menu). Both programs
          will use the <literal>RECOLL_CONFDIR</literal>
          variable or accept a <literal>-c</literal>
          <replaceable>confdir</replaceable> option to specify a non-default
          configuration directory.</para>

        <para>Reasons to use either the indexing thread or the
        <command>recollindex</command> command:
          <itemizedlist>
            <listitem><para>Starting the indexing thread is more convenient,
                being just one click away.</para>
            </listitem>
            <listitem><para>The <command>recollindex</command> command has
                more options, especially the one to reset the index
                (<literal>-z</literal>).</para>
            </listitem>
            <listitem><para>The <command>recollindex</command> command will
                not take down your GUI if it crashes (a rare occurrence,
                but who knows...)</para>
            </listitem>
            <listitem><para>The <command>recollindex</command> command uses
                <command>setpriority/nice</command> to lower its priority while
                indexing 
                (it will also use <command>ionice</command> when this becomes
                more widely available), the thread can't do it, else it would
                also slow down the user/search interface.</para>
            </listitem>
          </itemizedlist>
          I'll let the reader decide where my heart belongs...</para>

        <para>If the <command>recoll</command> program finds no index
          when it starts, it will automatically start indexing (except
          if canceled).</para>

        <para>The <command>recollindex</command> indexing process can be
          interrupted by sending an 
          interrupt (^C, SIGINT) or terminate (SIGTERM) signal. Some time may
          elapse before the process exits, because it needs to properly flush
          and close the index. The indexing thread can be equivalently
          stopped from the menu.</para>

	<para>After such an interruption, the index will be somewhat
	  inconsistent because some operations which are normally performed
	  at the end of the indexing pass will have been skipped (for
	  exemple, the stemming and spelling databases will be inexistant
	  or out of date). You just need to restart indexing at a later
	  time to restore consistency. The indexing will restart at the
          interruption point (the full file tree will be traversed,
          but files that were indexed up to the interruption and are still
          up to date will not need to be reindexed).</para>

        <para><command>recollindex</command> has a number of other options
          which are described in its man page.</para>

        <para>Of special interest maybe are the <literal>-i</literal> and
          <literal>-f</literal> options. <literal>-i</literal> allows
          indexing an explicit list of files (given as command line
          parameters or read on stdin). <literal>-f</literal> tells
          <command>recollindex</command> to ignore file selection
          parameters from the configuration. Together, these options allow
          building a custom file selection process for some area of the
          file system, by adding the top directory to the
          <literal>skippedPaths</literal> list and using an appropriate
          file selection method to build the file list to be fed to
          <literal>recollindex&nbsp;-if</literal> .</para>

        <para><literal>recollindex&nbsp;-i</literal> will not descend into
          directory parameters, but just add them as index entries. It is
          up to the external file selection method to build the complete
          file list.</para>
      </sect2>

      <sect2 id="rcl.indexing.periodic.automat">
        <title>Using <command>cron</command> to automate
          indexing</title>

        <para>The most common way to set up indexing is to have a cron
          task execute it every night. For example the following
          <filename>crontab</filename> entry would do it every day at
          3:30AM (supposing <command>recollindex</command> is in your
          PATH):

        <programlisting>30 3 * * * recollindex > /some/tmp/dir/recolltrace 2>&1</programlisting>

	  Or, using <command>anacron</command>:
<programlisting>1  15  su mylogin -c "recollindex recollindex > /tmp/rcltraceme 2>&1"</programlisting>
        </para>

        <para>The usual command to edit your
          <filename>crontab</filename> is 
          <userinput>crontab -e</userinput> (which will usually start
          the <command>vi</command> editor to edit the file). You may
          have more sophisticated tools available on your
          system.</para>

      </sect2>
    </sect1>

    <sect1 id="rcl.indexing.monitor">
      <title>Real time indexing</title>

      <para>Real time monitoring/indexing is performed by starting the
      <command>recollindex -m</command> command. With this option,
      <command>recollindex</command> will detach from the terminal and
      become a daemon, permanently monitoring file changes and updating
      the index.</para>

      <para>The real time indexing support can be customised during package 
       <link linkend="rcl.install.building.build">configuration</link>
      with the <literal>--with[out]-fam</literal> or
      <literal>--with[out]-inotify</literal> options.  The default is
      currently to include inotify monitoring on systems that support
      it.</para>

      <para>The <filename>rclmon.sh</filename> script can be used to
      easily start and stop the daemon. It can be found in the
      <filename>examples</filename> directory (typically
      <filename>/usr/local/[share/]recoll/examples</filename>).</para>

      <para>Starting the daemon is normally performed as part
      of the user session script. For example, my out of fashion
      xdm-based session has a <filename>.xsession</filename> script
      with the following lines at the end:</para>

      <programlisting>recollconf=$HOME/.recoll-home
recolldata=/usr/local/share/recoll
RECOLL_CONFDIR=$recollconf $recolldata/examples/rclmon.sh start

fvwm 

</programlisting>

      <para>The indexing daemon gets started, then the window manager,
      for which the session waits.</para> <para>By default the
      indexing daemon will monitor the state of the X11 session, and
      exit when it finishes, it is not necessary to kill it
      explicitly. (The X11 server monitoring can be disabled with option
      <literal>-x</literal> to <command>recollindex</command>).
      </para>

      <para>Under KDE, you can place a small script to start
      <command>recollindex -m</command> under
      <filename>$HOME/.kde/Autostart</filename>. This will be executed
      when the session begins.</para>
      
      <para>There is a similar mechanism under Gnome (find the session
      control tool in the menus and use the "Startup programs" tab).</para>

      <para>By default, the messages from the indexing daemon will be
        discarded. You may want to change this by setting the
        <literal>daemlogfilename</literal> and
        <literal>daemloglevel</literal> configuration parameters. Also the
        log file will only be truncated when the daemon starts. If the
        daemon runs permanently, the log file may grow quite big, depending
        on the log level.</para>

      <para>While it is convenient that data is indexed in real time,
        repeated indexing can generate a significant load on the
        system when files such as email folders change. Also,
        monitoring large file trees by itself significantly taxes
        system resources. You probably do not want to enable it if
        your system is short on resources. Periodic indexing is
        adequate in most cases.</para>

    </sect1>

  </chapter>

  <chapter id="rcl.search">
    <title>Searching</title>

  <sect1 id="rcl.search.gui">
    <title>Searching with the Qt graphical user interface</title>

    <para>The <command>recoll</command> program provides the main user
      interface for searching. It is based on the
      <application>Qt</application> library.</para>

    <para><command>recoll</command> has two search modes:</para>
    <itemizedlist>
      <listitem><para>Simple search (the default, on the main screen) has
        a single entry field where you can enter multiple words.</para>
      </listitem>
      <listitem><para>Advanced search (a panel accessed through the
        <guilabel>Tools</guilabel> menu or the toolbox bar icon) has
        multiple entry fields, which you may use to build a logical
        condition, with additional filtering on file type and location
        in the file system.</para>
      </listitem>
    </itemizedlist>

    <para>In most cases, you can enter the terms as you
        think them, even if they contain embedded punctuation or other
        non-textual characters. For
        exemple, &RCL; can handle things like e-mail addresses, or
        arbitrary cut and paste from another text window, punctation
        and all.</para>

    <para>The main case where you should enter text differently from
      how it is printed is for east-asian languages (Chinese,
      Japanese, Korean). Words composed of single or multiple
      characters should be entered separated by white space in this
      case (they would typically be printed without white
      space).</para>

    <sect2 id="rcl.search.simple">
      <title>Simple search</title>

      <procedure>
        <step><para>Start the <command>recoll</command> program.</para>
        </step>
        <step><para>Possibly choose a search mode: <guilabel>Any
        term</guilabel>, <guilabel>All terms</guilabel>, 
        <guilabel>File name</guilabel> or
        <guilabel>Query language</guilabel>.</para>
        </step>
        <step><para>Enter search term(s) in the text field at the top of the
        window.</para>
        </step>
        <step><para>Click the <guilabel>Search</guilabel> button or
        hit the <keycap>Enter</keycap> key to start the search.</para>
        </step>
      </procedure>

      <para>The initial default search mode is <guilabel>Query
        language</guilabel>. Without special directives, this will look for
        documents containing all of the search terms (the ones with more
        terms will get better scores), just like the <guilabel>All
        terms</guilabel> mode which will ignore such
        directives. <guilabel>Any term</guilabel> will search for documents
        where at least one of the terms appear. </para>

      <para>The <guilabel>Query Language</guilabel> features are
      described in <link linkend="rcl.search.lang">a separate
      section</link>.</para>  

      <para><guilabel>File name</guilabel> will specifically look for file
        names. The entry will be split at white space characters,
        and each fragment will be separately expanded, then the search will
        be for file names matching all fragments (this is new in 1.15,
        older releases did an OR of the whole thing which did not make
        sense). Things to know:
        <itemizedlist>
            <listitem><para>The search is case- and accent-insensitive.</para>
            </listitem>
            <listitem><para>Fragments without any wild card
            character and not capitalized will be prepended and appended
            with '*' (ie: <replaceable>etc</replaceable> ->
            <replaceable>*etc*</replaceable>, but
            <replaceable>Etc</replaceable> ->
            <replaceable>etc</replaceable>). Of course it does not make
            sense to have multiple fragments if one of them is capitalized
            (as this one will require an exact match).</para> 
            </listitem>
            <listitem><para>If you want to search for a pattern including
            white space, use double quotes (ie: <replaceable>"admin
            note*"</replaceable>).</para> 
            </listitem>
            <listitem><para>If you have a big index (many files),
            excessively generic fragments may result in inefficient
            searches.</para>
            </listitem>
            <listitem><para>As an example, <replaceable>inst
            recoll</replaceable> would match
            <replaceable>recollinstall.in</replaceable> (and quite a few
            others...).</para> 
            </listitem>
          </itemizedlist>
        The point of having a separate file name
        search is that wild card expansion can be performed more
        efficiently on a relatively small subset of the index (allowing
        wild cards on the left of terms without excessive penality).</para>

      <para>All search modes allow wildcards inside terms
        (<literal>*</literal>, <literal>?</literal>,
        <literal>[]</literal>). You may want to have a look at the
        <link linkend="rcl.search.wildcards">section about wildcards</link>
        for more information about this.</para>

      <para>You can search for exact phrases (adjacent words in a
      given order) by enclosing the input inside double quotes. Ex:
     <literal>"virtual reality"</literal>.</para>

      <para>Character case has no influence on search, except that you
      can disable stem expansion for any term by capitalizing it. Ie:
      a search for <literal>floor</literal> will also normally look for 
      <literal>flooring</literal>, <literal>floored</literal>, etc., but
      a search for <literal>Floor</literal> will only look for
      <literal>floor</literal>, in any character case. Stemming can
      also be disabled globally in the preferences. </para>

      <para>&RCL; remembers the last few searches that you
        performed. You can use the simple search text entry widget (a
        combobox) to recall them (click on the thing at the right of the
        text field). Please note, however, that only the search texts
        are remembered, not the mode (all/any/file name).</para>

      <para>Typing <keycap>Esc</keycap> <keycap>Space</keycap> while
        entering a word in the simple search entry will open a window
        with possible completions for the word. The completions are
        extracted from the database.</para>

      <para>Double-clicking on a word in the result list or a preview
      window will insert it into the simple search entry field.</para>

      <para>You can cut and paste any text into an <guilabel>All
      terms</guilabel> or <guilabel>Any term</guilabel> search field,
      punctuation, newlines and all - except for wildcard characters
      (single <literal>?</literal> characters are ok). &RCL; will process
      it and produce a meaningful search. This is what most differentiates
      this mode from the <guilabel>Query Language</guilabel> mode, where
      you have to care about the syntax.</para>

      <para>You can use the <link linkend="rcl.search.complex">
       <guilabel>Tools</guilabel> / <guilabel>Advanced search</guilabel>
        </link> dialog for more complex searches.</para>
    </sect2>

    <sect2 id="rcl.search.reslist">
      <title>The default result list</title>

      <para>After starting a search, a list of results will instantly
       be displayed in the main list window.</para> 

      <para>By default, the document list is presented in order of
       relevance (how well the system estimates that the document
       matches the query). You can sort the result by ascending or
       descending date by using the vertical arrows in the toolbar (the old
       sort tool is gone after release 1.15, because the new <link
       linkend="rcl.search.restable">result table</link> has much better
       capability).</para> 

      <para>Clicking on the
       <literal>Preview</literal> link for an entry will open an
       internal preview window for the document. Further
       <literal>Preview</literal> clicks for the same search will open
       tabs in the existing preview window. You can use
       <keycap>Shift</keycap>+Click to force the creation of another
       preview window, which may be useful to view the documents side
       by side. (You can also browse successive results in a single
       preview window by typing
       <keycap>Shift</keycap>+<keycap>ArrowUp/Down</keycap> in the
       window).</para> 

      <para>Clicking the <literal>Open</literal> link will attempt to 
       start an external viewer. The viewer for each document type can be
       configured through the user preferences dialog, or by editing the
       <filename>mimeview</filename> configuration file. You can also check
       the <guilabel>Use desktop preferences</guilabel> option in the user
       preferences dialog to use the desktop defaults for all
       documents. This is probably the best option if you are using a well
       configured Gnome or KDE desktop.</para>

      <para>The <literal>Preview</literal> and <literal>Open</literal>
       edit links may not be present for all entries, meaning that
       &RCL; has no configured way to preview a given file type (which
       was indexed by name only), or no configured external editor for
       the file type. This can sometimes be adjusted simply by tweaking
       the <link linkend="rcl.install.config.mimemap">
             <filename>mimemap</filename></link> and  
       <link linkend="rcl.install.config.mimeview">
       <filename>mimeview</filename></link> configuration files (the latter
       can be modified with the user preferences dialog).</para> 

      <para>The format of the result list entries is entirely
      configurable by using the preference dialog to 
      <link linkend="rcl.search.custom.reslistpara">edit an HTML
      fragment</link>.  

      <para>You can click on the <literal>Query details</literal> link
        at the top of the results page to see the query actually 
        performed, after stem expansion and other processing.</para>

      <para>Double-clicking on any word inside the result list or a
      preview window will insert it into the simple search text.</para>

      <para>The result list is divided into pages (the size of which
       you can change in the preferences). Use the arrow buttons in the
       toolbar or the links at the bottom of the page to browse the
       results.</para>


      <sect3 id="rcl.search.resultlist.menu">
        <title>The result list right-click menu</title>

        <para>Apart from the preview and edit links, you can display a
          pop-up menu by right-clicking over a paragraph in the result
         list. This menu has the following entries:</para>

        <itemizedlist>
          <listitem><para><guilabel>Preview</guilabel></para></listitem>
          <listitem><para><guilabel>Open</guilabel></para></listitem>
          <listitem><para><guilabel>Copy File Name</guilabel></para></listitem>
          <listitem><para><guilabel>Copy Url</guilabel></para></listitem>
          <listitem><para><guilabel>Save to File</guilabel></para></listitem>
          <listitem><para><guilabel>Find similar</guilabel></para></listitem>
          <listitem><para><guilabel>Preview Parent
          document</guilabel></para></listitem> 
          <listitem><para><guilabel>Open Parent
          document</guilabel></para></listitem> 
        </itemizedlist>

        <para>The <guilabel>Preview</guilabel> and
          <guilabel>Open</guilabel> entries do the same thing as the 
          corresponding links.</para>

        <para>The <guilabel>Copy File Name</guilabel> and
        <guilabel>Copy Url</guilabel> copy the relevant data to the
        clipboard, for later pasting.</para> 

        <para><guilabel>Save to File</guilabel> allows saving the
        contents of a result document to a chosen file. This entry
        will only appear if the document does not correspond to an
        existing file, but is a subdocument inside such a file (ie: an
        email attachment). It is especially useful to extract attachments
        with no associated editor.</para> 

        <para>The <guilabel>Find similar</guilabel> entry will select
          a number of relevant term from the current document and enter
          them into the simple search field. You can then start a simple
          search, with a good chance of finding documents related to the
          current result.</para>

        <para>The <guilabel>Parent document</guilabel> entries will
          appear for documents which are not actually files but are part
          of, or attached to, a higher level document. This entry is mainly
          useful for email attachments and permits viewing the message to
          which the document is attached. Note that the entry will also
          appear for an email which is part of an mbox folder file, but
          that you can't actually visualize the folder (there will be an
          error dialog if you try). &RCL; is unfortunately not yet smart
          enough to disable the entry in this case. In other cases, the
          <guilabel>Open</guilabel> option makes sense, for exemple to
          start a <application>chm</application> viewer on the parent
          document for a help page.</para>

      </sect3>
    </sect2>

    <sect2 id="rcl.search.restable">
      <title>The result table</title>

        <para>In &RCL; 1.15 and newer, the results can be displayed in 
        spreadsheet-like fashion. You can switch to this presentation by
        clicking the table-like icon in the toolbar (this is a toggle,
        click again to restore the list).</para>

        <para>Clicking on the column headers will allow sorting by the
        values in the column. You can click again to invert the order, and
        use the header right-click menu to reset sorting to the default
        relevance order (you can also use the sort-by-date arrows to do
        this).</para>

        <para>Both the list and the table display the same underlying
          results. The sort order set from the table is still active if you
          switch back to the list mode. You can click twice on a date sort
          arrow to reset it from there.</para>

        <para>The header right-click menu allows adding or deleting
          columns. The columns can be resized, and their order can be changed
          (by dragging). All the changes are recorded when you quit
          <command>recoll</command></para> 
        
        <para>Hovering over a table row will update the detail area at the
        bottom of the window with the corresponding values. You can click
        the row to freeze the display. The bottom area is equivalent to a
        classical result list paragraph, with links for
        starting a preview or a native application, and an equivalent
        right-click menu. Typing <keycap>Esc</keycap> (the Escape key) will
        unfreeze the display.</para> 

      </sect2>

    <sect2 id="rcl.search.preview">
      <title>The preview window</title>

      <para>The preview window opens when you first click a
      <literal>Preview</literal> link inside the result list.</para>

      <para>Subsequent preview requests for a given search open new
      tabs in the existing window (except if you hold the
      <keycap>Shift</keycap> key while clicking which will open a new
      window for side by side viewing).</para>
      
      <para>Starting another search and requesting a preview will
      create a new preview window. The old one stays open until you
      close it.</para>

      <para>You can close a preview tab by typing <keycap>^W</keycap> 
      (<keycap>Ctrl</keycap> + <keycap>W</keycap>) in the
      window. Closing the last tab for a window will also close the
      window.</para> 

      <para>Of course you can also close a preview window by using the
      window manager button in the top of the frame.</para>

      <para>You can display successive or previous documents from the
      result list inside a preview tab by typing
      <keycap>Shift</keycap>+<keycap>Down</keycap> or
      <keycap>Shift</keycap>+<keycap>Up</keycap> (<keycap>Down</keycap>
      and <keycap>Up</keycap> are the arrow keys).</para> 

      <para>The preview tabs have an internal incremental search
      function. You initiate the search either by typing a
      <keycap>/</keycap> (slash) or <keycap>CTL-F</keycap> inside the text
      area or by clicking into the <guilabel>Search for:</guilabel> text
      field and entering the search string. You can then use the
      <guilabel>Next</guilabel> and <guilabel>Previous</guilabel> buttons
      to find the next/previous occurrence. You can also type
      <keycap>F3</keycap> inside the text area to get to the next
      occurrence.</para>

      <para>If you have a search string entered and you use ^Up/^Down
      to browse the results, the search is initiated for each successive
      document. If the string is found, the cursor will be positioned
      at the first occurrence of the search string.</para>

      <para>A right-click menu in the text area allows switching
      between displaying the main text or the contents of fields
      associated to the document (ie: author, abtract, etc.). This is
      especially useful in cases where the term match did not occur in
      the main text but in one of the fields.</para>
        
      <para>You can print the current preview window contents by typing
         <keycap>^P</keycap> (<keycap>Ctrl</keycap> + <keycap>P</keycap>) in 
        the window text.</para> 
        
    </sect2>

    <sect2 id="rcl.search.complex">
      <title>Complex/advanced search</title>

      <para>The advanced search dialog helps you build more complex queries
      without memorizing the search language constructs. It can be opened
      through the <guilabel>Tools</guilabel> menu or through the main
      toolbar.</para>

      <para>The dialog has three parts:</para>

      <itemizedlist>
        <listitem><para>The top part allows constructing a query by
          combining multiple clauses of different types.
          Each entry field is configurable for the following modes:</para>

        <itemizedlist>
          <listitem><para>All terms.</para>
          </listitem>
          <listitem><para>Any term.</para>
          </listitem>
          <listitem><para>None of the terms.</para>
          </listitem>
          <listitem><para>Phrase (exact terms in order within an
          adjustable window).</para>
          </listitem>
          <listitem><para>Proximity (terms in any order within an
          adjustable window).</para>
          </listitem>
          <listitem><para>Filename search.</para>
          </listitem>
        </itemizedlist>

        <para>Additional entry fields can be created by clicking the
          <guilabel>Add clause</guilabel> button.</para>

        <para>When searching, the non-empty clauses will be
          combined either with an AND or an OR conjunction, depending on
          the choice made on the left (<guilabel>All clauses</guilabel> or
          <guilabel>Any clause</guilabel>).</para>

        <para>Entries of all types except "Phrase" and "Near" accept
          a mix of single words and phrases enclosed in double quotes. 
          Stemming and wildcard expansion will be performed as for simple
          search. </para>
        </listitem>

        <listitem><para>The next part allows filtering the
          results by their mime types.</para> 
          <para>The state of the file type selection can be saved as
            the default (the file type filter will not be activated at
            program start-up, but the lists will be in the restored
            state).</para> 
        </listitem>

        <listitem>
          <para>The bottom part allows restricting the search results to a
            sub-tree of the indexed area. You can use the
            <guilabel>Invert</guilabel> checkbox to search for files not in
            the sub-tree instead. If you use directory filtering often and on
            big subsets of the file system, you may think of setting up
            multiple indexes instead, as the performance may be
            better. </para>
        </listitem>

      </itemizedlist>


      <formalpara><title>Phrases and Proximity searches</title>
      <para>These two clauses work in similar ways, with the
      difference that proximity searches do not impose an order on the
      words. In both cases, an adjustable number (slack) of non-matched words
      may be accepted between the searched ones (use the counter on
      the left to adjust this count). For phrases, the default count
      is zero (exact match). For proximity it is ten (meaning that two search
      terms, would be matched if found within a window of twelve
      words). Examples: a phrase search for <literal>quick
      fox</literal> with a slack of 0 will match <literal>quick
      fox</literal> but not <literal>quick brown fox</literal>. With
      a slack of 1 it will match the latter, but not <literal>fox
      quick</literal>. A proximity search for <literal>quick
      fox</literal> with the default slack will match the
      latter, and also <literal>a fox is a cunning and quick animal</literal>.
      </formalpara>

      <para>Click on the <guilabel>Start Search</guilabel> button in
        the advanced search dialog, or type <keycap>Enter</keycap> in
        any text field to start the search. The button in
        the main window always performs a simple search.</para>
      <para>Click on the <literal>Show query details</literal> link at
        the top of the result page to see the query expansion.</para>

    </sect2>

    <sect2 id="rcl.search.termexplorer">
      <title>The term explorer tool</title>

      <para>&RCL; automatically manages the expansion of search terms
      to their derivatives (ie: plural/singular, verb
      inflections). But there are other cases where the exact search
      term is not known. For example, you may not remember the exact
      spelling, or only know the beginning of the name.</para>

      <para>The term explorer tool (started from the toolbar icon or
      from the <guilabel>Term explorer</guilabel> entry of the
      <guilabel>Tools</guilabel> menu) can be used to search the full index
      terms list. It has three modes of operations:</para>
        <variablelist>

          <varlistentry>
            <term>Wildcard</term>
            <listitem><para>In this mode of operation, you can enter a
            search string with shell-like wildcards (*, ?, []). ie:
            <replaceable>xapi*</replaceable> would display all index terms
            beginning with <replaceable>xapi</replaceable>. (More
            about wildcards <link
            linkend="rcl.search.wildcards">here</link>).</para></listitem> 
          </varlistentry>

          <varlistentry>
          <term>Regular expression</term>
          <listitem><para>This mode will accept a regular expression
            as input. Example:
            <replaceable>word[0-9]+</replaceable>. The expression is
            implicitely anchored at the beginning. Ie:
            <replaceable>press</replaceable> will match
            <replaceable>pression</replaceable> but not
            <replaceable>expression</replaceable>. You can use
            <replaceable>.*press</replaceable> to match the latter,
            but be aware that this will cause a full index term list
            scan, which can be quite long.</para>
          </listitem>
          </varlistentry>
          <varlistentry>

          <term>Stem expansion</term>
          <listitem><para>This mode will perform the usual stem expansion
          normally done as part user input processing. As such it is
          probably mostly useful to demonstrate the process.
          </para></listitem>
          </varlistentry>

          <varlistentry>
            <term>Spelling/Phonetic</term> <listitem><para>In this
            mode, you enter the term as you think it is spelled, and
            &RCL; will do its best to find index terms that sound like
            your entry. This mode uses the
            <application>Aspell</application> spelling application,
            which must be installed on your system for things to work
            (if your documents contain non-ascii characters, &RCL;
            needs an aspell version newer than 0.60 for UTF-8
            support). The language which is used to build the
            dictionary out of the index terms (which is done at the
            end of an indexing pass) is the one defined by your NLS
            environment. Weird things will probably happen if
            languages are mixed up.</para></listitem>
          </varlistentry>
      </variablelist>

      <para>Note that in cases where &RCL; does not know the beginning
      of the string to search for (ie a wildcard expression like
      <replaceable>*coll</replaceable>), the expansion can take quite
      a long time because the full index term list will have to be
      processed. The expansion is currently limited at 200 results for
      wildcards and regular expressions.</para>
      
      <para>Double-clicking on a term in the result list will insert
      it into the simple search entry field. You can also cut/paste
      between the result list and any entry field (the end of lines
      will be taken care of).</para>

    </sect2>

    <sect2 id="rcl.search.multidb">
      <title>Multiple databases</title>

      <para>Multiple &RCL; databases or indexes can be created by
      using several configuration directories which are usually set to
      index different areas of the file system. A specific index can
      be selected for updating or searching, using the
      <literal>RECOLL_CONFDIR</literal> environment variable or the
      <literal>-c</literal> option to <command>recoll</command> and
      <command>recollindex</command>.</para>

      <para>A <command>recollindex</command> program instance can only
      update one specific index.</para>

      <para>A <command>recoll</command> program instance is also
      associated with a specific index, which is the one to be
      updated by its indexing thread, but it can use any
      number of &RCL; indexes for searching. The external indexes
      can be selected through the <guilabel>external
      indexes</guilabel> tab in the preferences dialog.</para>

      <para>Index selection is performed in two phases. A set of all
      usable indexes must first be defined, and then the subset of
      indexes to be used for searching. Of course, these parameters
      are retained across program executions (there are kept
      separately for each &RCL; configuration). The set of all indexes
      is usually quite stable, while the active ones might typically
      be adjusted quite frequently.</para>

      <para>The main index (defined by
      <literal>RECOLL_CONFDIR</literal>) is always active. If this is
      undesirable, you can set up your base configuration to index
      an empty directory.</para>

      <para>As building the set of all indexes can be a little tedious
      when done through the user interface, you can use the
      <literal>RECOLL_EXTRA_DBS</literal> environment
      variable to provide an initial set. This might typically be
      set up by a system administrator so that every user does not
      have to do it. The variable should define a colon-separated list
      of index  directories, ie: 
     </para>
       <screen>export RECOLL_EXTRA_DBS=/some/place/xapiandb:/some/other/db</screen> 

      <para>A typical usage scenario for the multiple index feature
      would be for a system administrator to set up a central index
      for shared data, that you choose to search or not in addition to
      your personal data. Of course, there are other
      possibilities. There are many cases where you know the subset of
      files that should be searched, and where narrowing the search
      can improve the results. You can achieve approximately the same
      effect with the directory filter in advanced search, but
      multiple indexes will have much better performance and may be
      worth the trouble.</para>

    </sect2>

    <sect2 id="rcl.search.history">
      <title>Document history</title>

      <para>Documents that you actually view (with the internal preview
        or an external tool) are entered into the document history,
        which is remembered.</para> 
      <para>You can display the history list by using
        the <guilabel>Tools/</guilabel><guilabel>Doc History</guilabel> menu
        entry.</para> 
      <para>You can erase the document history by using the
      <guilabel>Erase document history</guilabel> entry in the
      <guimenu>File</guimenu> menu.

    </sect2>

    <sect2 id="rcl.search.sort">
      <title>Sorting search results and collapsing duplicates</title>

      <para>The documents in a result list are normally sorted in
        order of relevance. It is possible to specify different sort
        parameters by using the <guimenu>Sort parameters</guimenu>
        dialog (located in the <guimenu>Tools</guimenu> menu).</para>

      <para>The tool sorts a specified number of the most
        relevant documents in the result list, according to specified
        criteria. The currently available criteria are
        <emphasis>date</emphasis> and <emphasis>mime
        type</emphasis>.</para>

      <para>The sort parameters stay in effect until they are
        explicitly reset, or the program exits. An activated sort is
        indicated in the result list header.</para>

      <para>Sort parameters are remembered between program
        invocations, but result sorting is normally always inactive
        when the program starts. It is possible to keep the sorting
        activation state between program invocations by checking the
        <guilabel>Remember sort activation state</guilabel> option in
        the preferences.</para>

      <para>It is also possible to hide duplicate entries inside
        the result list (documents with the exact same contents as the
        displayed one). The test of identity is based on an MD5 hash
        of the document container, not only of the text contents (so
        that ie, a text document with an image added will not be a
        duplicate of the text only). Duplicates hiding is controlled
        by an entry in the <guilabel>Query configuration</guilabel>
        dialog, and is off by default.</para>

    </sect2>

    <sect2 id="rcl.search.tips">
      <title>Search tips, shortcuts</title>

    <sect3 id="rcl.search.tips.terms">
      <title>Terms and search expansion</title>

      <formalpara><title>Term completion</title>
        <para>Typing <keycap>Esc</keycap> <keycap>Space</keycap> in
        the simple search entry field while entering a word will
        either complete the current word if its beginning matches a
        unique term in the index, or open a window to propose a list
        of completions.</para>
      </formalpara>

      <formalpara><title>Picking up new terms from result or preview 
                   text</title>
        <para>Double-clicking on a word in the result list or in a
        preview window will copy it to the simple search entry field.</para>
      </formalpara>

      <formalpara><title>Wildcards</title>
          <para>Wildcards can be used inside search terms in all forms
            of searches. <link linkend="rcl.search.wildcards">
            More about wildcards</link>.
          </para>
      </formalpara>

      <formalpara><title>Automatic suffixes</title>
          <para>Words like <literal>odt</literal> or <literal>ods</literal>
            can be automatically turned into query language
            <literal>ext:xxx</literal> clauses. This can be enabled in the
            <guilabel>Search preferences</guilabel> panel in the GUI.
          </para>
      </formalpara>

      <formalpara><title>Disabling stem expansion</title>
      <para>Entering a capitalized word in any search field will prevent
        stem expansion (no search for
        <literal>gardening</literal> if you enter
        <literal>Garden</literal> instead of
        <literal>garden</literal>). This is the only case where
        character case should make a difference for a &RCL;
        search. You can also disable stem expansion or change the
        stemming language in the preferences.</para>
      </formalpara>

      <formalpara><title>Finding related documents</title>
        <para>Selecting the <guilabel>Find similar documents</guilabel> entry
        in the result list paragraph right-click menu will select a
        set of "interesting" terms from the current result, and insert
        them into the simple search entry field. You can then possibly
        edit the list and start a search to find documents which may
        be apparented to the current result.</para>
      </formalpara>

      <formalpara><title>File names</title>
        <para>File names are added as terms during indexing, and you can
        specify them as ordinary terms in normal search fields (&RCL; used
        to index all directories in the file path as terms. This has been
        abandoned as it did not seem really useful). Alternatively, you
        can use the specific file name search which will
        <emphasis>only</emphasis> look for file names, and may be
        faster than the generic search especially when using wildcards.</para>
      </formalpara>

      </sect3>


    <sect3 id="rcl.search.tips.phrases">
      <title>Working with phrases and proximity</title>

      <formalpara><title>Phrases and Proximity searches</title>
      <para>A phrase can be looked for by enclosing it in double
        quotes. Example: <literal>"user manual"</literal> will look
        only for occurrences of <literal>user</literal> immediately
        followed by <literal>manual</literal>. You can use the
        <guilabel>This phrase</guilabel> field of the advanced
        search dialog to the same effect. Phrases can be entered along
        simple terms in all simple or advanced search entry fields
        (except <guilabel>This exact phrase</guilabel>).</para>
      </formalpara>

      <formalpara><title>AutoPhrases</title>
      <para>This option can be set in the preferences dialog. If it is
      set, a phrase will be automatically built and added to simple
      searches when looking for <literal>Any terms</literal>. This
      will not change radically the results, but will give a relevance
      boost to the results where the search terms appear as a
      phrase. Ie: searching for <literal>virtual reality</literal>
      will still find all documents where either
      <literal>virtual</literal> or <literal>reality</literal> or 
      both appear, but those which contain <literal>virtual
      reality</literal> should appear sooner in the list.</para>

      </sect3>

    <sect3 id="rcl.search.tips.misc">
      <title>Others</title>

      <formalpara><title>Using fields</title>
        <para>You can use the <link linkend="rcl.search.lang">query
        language </link> and field specifications
        to only search certain parts of documents. This can be
        especially helpful with email, for example only searching
        emails from a specific originator:
        <literal>search tips from:helpfulgui</literal>
        </para>
      </formalpara>

      <formalpara><title>Query explanation</title>
        <para>You can get an exact description of what the query
        looked for, including stem expansion, and Boolean operators
        used, by clicking on the result list header.</para>
      </formalpara>

      <formalpara><title>Browsing the result list inside a preview 
                   window</title>
       <para>Entering <keycap>Shift-Down</keycap> or <keycap>Shift-Up</keycap>
       (<keycap>Shift</keycap> + an arrow key) in a preview window will
       display the next or the previous document from the result
       list. Any secondary search currently active will be executed on
       the new document.</para>
      </formalpara>

      <formalpara><title>Scrolling the result list from the keyboard</title>
       <para>You can use <keycap>PageUp</keycap> and <keycap>PageDown</keycap>
	 to scroll the result list, <keycap>Shift+Home</keycap> to go back
	 to the first page. These work even while the focus is in the
	 search entry.</para>
      </formalpara>
      
      <formalpara><title>Forced opening of a preview window</title>
       <para>You can use <keycap>Shift</keycap>+Click on a result list
       <literal>Preview</literal> link to force the creation of a
       preview window instead of a new tab in the existing one.</para>
      </formalpara>

      <formalpara><title>Closing previews</title> 
       <para>Entering <keycap>^W</keycap> in a tab will
        close it (and, for the last tab, close the preview
        window). Entering <keycap>Esc</keycap> will close the preview
        window and all its tabs.</para>
      </formalpara>

      <formalpara><title>Printing previews</title> 
       <para>Entering <keycap>^P</keycap> in a preview window will print 
        the currently displayed text.</para>
      </formalpara>

      <formalpara><title>Quitting</title>
      <para>Entering <keycap>^Q</keycap> almost anywhere will
        close the application.</para>
      </formalpara>
      </sect3>
    </sect2>

    <sect2 id="rcl.search.custom">
      <title>Customizing the search interface</title>

      <para>You can customize some aspects of the search interface by using
      the <guimenu>Query configuration</guimenu> entry in the
      <guimenu>Preferences</guimenu> menu.</para>

      <para>There are several tabs in the dialog, dealing with the
      interface itself, the parameters used for searching and
      returning results, and what indexes are searched.</para> 

      <formalpara id="rcl.search.custom.ui">
       <title>User interface parameters:</title>
        <para>
      <itemizedlist>

            <listitem><para><guilabel>Number of results in a result
              page</guilabel>: </para> 
            </listitem>

            <listitem><para><guilabel>Hide duplicate results</guilabel>:
            decides if result list entries are shown for identical
            documents found in different places.</para>
            </listitem>

            <listitem><para><guilabel>Highlight color for query
            terms</guilabel>: Terms from the user query are highlighted in
            the result list samples and the preview window. The color can
            be chosen here. Any Qt color string should work (ie
            <literal>red</literal>, <literal>#ff0000</literal>). The
            default is <literal>blue</literal>.</para>
            </listitem>

            <listitem><para><guilabel>Result list font</guilabel>: There is
            quite a lot of information shown in the result list, and you
            may want to customize the font and/or font size. The rest of
            the fonts used by &RCL; are determined by your generic Qt
            config (try the <command>qtconfig</command> command).</para>
            </listitem>

            <listitem><anchor id="rcl.search.custom.resultpara">
            <para><guilabel>Result paragraph format string</guilabel>:
            allows you to change the presentation of each result list
            entry. This is <link linkend="rcl.search.custom.reslistpara">
            described in its own section.</link></para>
            </listitem>

            <listitem><anchor id="rcl.search.custom.abssep">
            <para><guilabel>Abstract snippet separator</guilabel>:
            for synthetic abstracts built from index data, which are
            usually made of several snippets from different parts of the
            document, this defines the snippet separator, an ellipsis by
            default. </para>
            </listitem>

	    <listitem><para><guilabel>Maximum text size highlighted for
            preview</guilabel> Inserting highlights on search term inside
            the text before inserting it in the preview window involves
            quite a lot of processing, and can be disabled over the given
            text size to speed up loading.</para>
            </listitem>

            <listitem><para><guilabel>Use desktop preferences to choose
            document editor</guilabel>: if this is checked, the
            <command>xdg-open</command> utility will be used to open files
            when you click the <guilabel>Open</guilabel> link in the result
            list, instead of the application defined in
            <filename>mimeview</filename>. <command>xdg-open</command> will
            in term use your desktop preferences to choose an appropriate
            application.</para>
           </listitem>

            <listitem><para><guilabel>Choose editor applications</guilabel>
            this will let you choose the command started by the
            <guilabel>Open</guilabel> links inside the result list, for
            specific document types.</para>
            </listitem>
	    
	    <listitem><para><guilabel>Display category filter as
	    toolbar...</guilabel> this will let you choose if the document
	    categories are displayed as a list or a set of buttons.</para>
	    </listitem>

            <listitem><para><guilabel>Auto-start simple search on white
            space entry</guilabel>: if this is checked, a search will be
            executed each time you enter a space in the simple search input
            field. This lets you look at the result list as you enter new
            terms. This is off by default, you may like it or not...</para>
            </listitem>

	    <listitem><para><guilabel>Start with advanced search dialog open
            </guilabel> and <guilabel>Start with sort dialog
            open</guilabel>: If you use these dialogs all the time, checking
            these entries will get them to open when recoll starts.</para>
	    </listitem>
	    
	    <listitem><para><guilabel>Remember sort activation
	    state</guilabel> if set, Recoll will remember the sort tool
	    stat between invocations. It normally starts with sorting
	    disabled.</para>
	    </listitem>
	    <listitem><para><guilabel>Prefer HTML to plain text for preview

	    </guilabel> if set, Recoll will display HTML as such inside the
	    preview window. If this causes problems with the Qt HTML
	    display, you can uncheck it to display the plain text version
	    instead. </para>
	    </listitem>

	  </itemizedlist>
      </para>
      </formalpara>


      <formalpara id="rcl.search.custom.search">
	<title>Search parameters:</title>
        <para>
      <itemizedlist>

	    <listitem><para><guilabel>Stemming language</guilabel>:
            stemming obviously depends on the document's language. This
            listbox will let you chose among the stemming databases which
            were built during indexing (this is set in the <link
            linkend="rcl.install.config.recollconf">main configuration
            file</link>), or later added with <command>recollindex
            -s</command> (See the recollindex manual). Stemming languages
            which are dynamically added will be deleted at the next
            indexing pass unless they are also added in the configuration
            file.</para>
            </listitem>

            <listitem><para><guilabel>Dynamically add phrase to simple
            searches</guilabel>: a phrase will be automatically built and
            added to simple searches when looking for <literal>Any
            terms</literal>. This will give a relevance boost to the
            results where the search terms appear as a phrase (consecutive
            and in order).</para>
            </listitem>

            <listitem><para><guilabel>Replace abstracts from
            documents</guilabel>: this decides if we should synthesize and
            display an abstract in place of an explicit abstract found
            within the document itself.</para>
            </listitem>

            <listitem><para><guilabel>Dynamically build
            abstracts</guilabel>: this decides if &RCL; tries to build
            document abstracts when displaying the result list. Abstracts
            are constructed by taking context from the document
            information, around the search terms. This can slow down
            result list display significantly for big documents, and you
            may want to turn it off.</para>
            </listitem>

            <listitem><para><guilabel>Synthetic abstract size</guilabel>:
            adjust to taste...</para>
            </listitem>

            <listitem><para><guilabel>Synthetic abstract context
            words</guilabel>: how many words should be displayed around
            each term occurrence.</para>
            </listitem>

            <listitem><para><guilabel>Query language magic file name 
              suffixes</guilabel>: a list of words which automatically get
              turned into <literal>ext:xxx</literal> file name suffix clauses
              when starting a query language query (ie: <literal>doc xls
              xlsx...</literal>). This will save some typing for people who
              use file types a lot when querying.</para>
            </listitem>
      </itemizedlist>
       </para>
      </formalpara>

      <formalpara id="rcl.search.custom.extradb">
        <title>External indexes:</title> 
      <para>This panel will let you browse for additional indexes
      that you may want to search. External indexes are designated by
      their database directory (ie:
      <filename>/home/someothergui/.recoll/xapiandb</filename>,
      <filename>/usr/local/recollglobal/xapiandb</filename>).</para>

      <para>Once entered, the indexes will appear in the
        <guilabel>External indexes</guilabel> list, and you can
        chose which ones you want to use at any moment by checking or
        unchecking their entries.</para> 

      <para>Your main database (the one the current configuration
      indexes to), is always implicitly active. If this is not
      desirable, you can set up your configuration so that it indexes,
      for example, an empty directory. An alternative indexer may also
      need to implement a way of purging the index from stale data,
      </para>

    <sect3 id="rcl.search.custom.reslistpara">
      <title>The result list paragraph format</title>

        <para>The presentation of each result inside the result list can be
        customized by setting the result list paragraph format inside the
        <guilabel>User Interface</guilabel> tab of the <guilabel>Query
        configuration</guilabel>.</para> 

        <para>This is a Qt HTML string where the following printf-like
        <literal>%</literal> substitutions will be performed:

        <itemizedlist>
          <listitem>
            <formalpara><title>%A</title><para>Abstract</para></formalpara>
          </listitem>
          <listitem><formalpara><title>%D</title><para>Date</para></formalpara>
          </listitem>
          <listitem><formalpara><title>%I</title><para>Icon image name
          </para></formalpara> 
          </listitem>
          <listitem><formalpara><title>%K</title><para>Keywords (if
          any)</para></formalpara> 
          </listitem>
          <listitem><formalpara><title>%L</title><para>Preview and
          Edit links</para></formalpara> 
          </listitem>
          <listitem><formalpara><title>%M</title><para>Mime
                  type</para></formalpara> 
          </listitem>
          <listitem><formalpara><title>%N</title><para>result Number
                  </para></formalpara> 
          </listitem>
          <listitem><formalpara><title>%R</title><para>Relevance
          percentage</para></formalpara> 
          </listitem>
          <listitem><formalpara><title>%S</title><para>Size
          information</para></formalpara> 
          </listitem>
          <listitem><formalpara><title>%T</title><para>Title</para>
                    </formalpara> 
          </listitem>
          <listitem><formalpara><title>%U</title><para>Url</para></formalpara>
          </listitem>
        </itemizedlist>

        The format of the Preview and Edit links is 
        <literal>&lt;a href="P%N"&gt;</literal>
        and 
        <literal>&lt;a href="E%N"&gt;</literal>
        where <replaceable>docnum</replaceable> (%N expands to the document
        number inside the result list).</para>

        <para>In addition to the predefined values above, all strings like
        <literal>%(fieldname)</literal> will be replaced by the value of
        the field named <literal>fieldname</literal> for this
        document. Only stored fields can be accessed in this way, the value
        of indexed but not stored fields is not known at this point in the
        search process (see <link linkend="rcl.program.fields">field
        configuration</link>). There are currently very few fields stored
        by default, apart from the values above (only
        <literal>author</literal>), so this feature will need some custom
        local configuration to be useful. For example, you could look at
        the fields for the document types of interest (use the right-click
        menu inside the preview window), and add what you want to the list
        of stored fields. A candidate example would be the
        <literal>recipient</literal> field which is generated by the
        message filters.</para>

        <para>The default value for the paragraph format string is:
        <programlisting>&lt;img src="%I" align="left">%R %S %L &amp;nbsp;&amp;nbsp;&lt;b>%T&lt;/b>&lt;br>
%M&amp;nbsp;%D&amp;nbsp;&amp;nbsp;&amp;nbsp;&lt;i>%U&lt;/i>&amp;nbsp;%i&lt;br>
%A %K
        </programlisting>
        You may, for example, try the following for a more web-like
        experience:
        <programlisting>&lt;u>&lt;b>&lt;a href="P%N"&gt;%T&lt;/a&gt;&lt;/b>&lt;/u>&lt;br>
%A&lt;font color=#008000>%U - %S&lt;/font> - %L
        </programlisting>
        Or the clean looking:
        <programlisting>&lt;img src="%I" align="left">%L &lt;font color="#900000">%R&lt;/font>
&nbsp;&nbsp;&lt;b>%T&lt;/b>&lt;br>%S&nbsp;
&lt;font color="#808080">&lt;i>%U&lt;/i>&lt;/font>
&lt;table bgcolor="#e0e0e0">
&lt;tr>&lt;td>&lt;div>%A&lt;/div>&lt;/td>&lt;/tr>
&lt;/table>%K
        </programlisting>
           Note that the P%N link in the above paragraph makes the title a
           preview link.
        </para>

          <para>It is also possible to 
          <link linkend="rcl.search.custom.abssep">
             define the value of the snippet separator inside the abstract
             section</link>.</para>
      </sect3>
    </sect2>

  </sect1> <!-- search GUI -->

  <sect1 id="rcl.searchkio">
    <title>Searching with the KDE KIO slave</title>

    <sect2 id="rcl.searchkio.intro">
      <title>What's this</title>

      <para>The &RCL; KIO slave allows performing a &RCL; search
      by entering an appropriate URL in a KDE open dialog, or with an
      HTML-based interface displayed in
      <command>Konqueror</command>.</para>  

      <para>The HTML-based interface is similar to the Qt-based
      interface, but slightly less powerful for now. Its advantage is
      that you can perform your search while staying fully within the
      KDE framework: drag and drop from the result list works normally
      and you have your normal choice of applications for opening
      files.</para>

      <para>The alternative interface uses a directory view of search
      results. Due to limitations in the current KIO slave interface,
      it is currently not obviously useful (to me).</para>

      <para>The interface is described in more detail inside a help
      file which you can access by entering
      <filename>recoll:/</filename> inside the
      <command>konqueror</command> URL line (this works only if the
      recoll KIO slave has been previously installed).</para>


      <para>The instructions for building this module are located in the
      source tree. See:
      <filename>kde/kio/recoll/00README.txt</filename>. Some Linux
      distributions do package the kio-recoll module, so check before
      diving into the build process, maybe it's already out there ready for
      one-click installation.</para>
    </sect2>


    <sect2 id="rcl.searchkio.searchabledocs">
      <title>Searchable documents</title>

      <para>As a sample application, the &RCL; KIO slave could allow
      preparing a set of HTML documents (for example a manual) so that
      they become their own search interface inside
      <command>konqueror</command>.</para>

      <para>This can be done by either explicitly inserting
      <literal>&lt;a&nbsp;href="recoll:/..."&gt;</literal> links 
      around some document areas, or automatically by adding a
      very small <application>javascript</application> program to the
      documents, like the following example, which would initiate a search by
      double-clicking any term:</para>
 
     <programlisting>&lt;script language="JavaScript">
    function recollsearch() {
        var t = document.getSelection();
        window.location.href = 'recoll://search/query?qtp=a&amp;p=0&amp;q=' +
            encodeURIComponent(t);
    }
&lt;/script>
 ....
&lt;body ondblclick="recollsearch()">

</programlisting>
    </sect2>
    </sect1>


  <sect1 id="rcl.search.commandline">
    <title>Searching on the command line</title>

    <para>There are several ways to obtain search results as a text
    stream, without a graphical interface:</para>
    <itemizedlist>
      <listitem><para>By passing option <literal>-t</literal> to the
      <command>recoll</command> program.</para>
      </listitem>
      <listitem><para>By using the <command>recollq</command> program.</para>
      </listitem>
      <listitem><para>By writing a custom
      <application>Python</application> program, using the 
      <link linkend="rcl.program.api.python">Recoll Python API</link>.</para>
      </listitem>
    </itemizedlist>

    <para>The first two methods work in the same way and accept/need the same
    arguments (except for the additional <literal>-t</literal> to
    <command>recoll</command>). The query to be executed is specified
    as command line arguments.</para> 

    <para><command>recollq</command> is not built by default. You can
    use the <filename>Makefile</filename> in the
    <filename>query</filename> directory to build it. This is a very
    simple program, and if you can program a little c++, you may find it
    useful to taylor its output format to your needs.</para>

    <para><command>recollq</command> has a man page (not installed by
    default, look in the <filename>doc/man</filename> directory). The
    Usage string is as follows:</para>
<programlisting>recollq [-o|-a|-f] &lt;query string>
 Runs a recoll query and displays result lines. 
  Default: will interpret the argument(s) as a query language string
  -o Emulate the gui simple search in ANY TERM mode
  -a Emulate the gui simple search in ALL TERMS mode
  -f Emulate the gui simple search in filename mode
Common options:
    -c &lt;configdir> : specify config directory, overriding $RECOLL_CONFDIR
    -d also dump file contents
    -n &lt;cnt> limit the maximum number of results (0->no limit, default 2000)
    -b : basic. Just output urls, no mime types or titles
    -m : dump the whole document meta[] array
    -S fld : sort by field name
    -D : sort descending
</programlisting>

    <para>Sample execution:</para>
<programlisting>recollq 'ilur -nautique mime:text/html'
Recoll query: ((((ilur:(wqf=11) OR ilurs) AND_NOT (nautique:(wqf=11)
  OR nautiques OR nautiqu OR nautiquement)) FILTER Ttext/html))
4 results
text/html       [file:///Users/uncrypted-dockes/projets/bateaux/ilur/comptes.html]      [comptes.html]  18593   bytes   
text/html       [file:///Users/uncrypted-dockes/projets/nautique/webnautique/articles/ilur1/index.html] [Constructio...
text/html       [file:///Users/uncrypted-dockes/projets/pagepers/index.html]    [psxtcl/writemime/recoll]...
text/html       [file:///Users/uncrypted-dockes/projets/bateaux/ilur/factEtCie/recu-chasse-maree....
</programlisting>
      </sect1>

    <sect1 id="rcl.search.lang">
      <title>The query language</title>

      <para>The query language processor is activated in the GUI
      simple search entry when the search mode selector is set to
      <guilabel>Query Language</guilabel>. It can also be used with the KIO
      slave or the command line search. It broadly has the same
      capabilities as the complex search interface in the
      GUI. Additionally, the query language is for now the only way to
      access the important &RCL; field search capabilities.</para>

      <para>The language is roughly based on the <ulink
            url="http://www.xesam.org/main/XesamUserSearchLanguage95">
      Xesam</ulink> user search language specification.</para>

      <para>If the results of a query language search puzzle you and you
      doubt what has been actually searched for, you can use the GUI
      <literal>show query</literal> link at the top of the result list to
      check the exact query which was finally executed by Xapian.</para>

      <para>Here follows a sample request that we are going to
      explain:</para>

      <programlisting>
          author:"john doe" Beatles OR Lennon Live OR Unplugged -potatoes
      </programlisting>

      <para>This would search for all documents with 
      <replaceable>John Doe</replaceable>
      appearing as a phrase in the author field (exactly what this is
      would depend on the document type, ie: the
      <literal>From:</literal> header, for an email message),
      and containing either <replaceable>beatles</replaceable> or
      <replaceable>lennon</replaceable> and either
      <replaceable>live</replaceable> or
      <replaceable>unplugged</replaceable> but not
      <replaceable>potatoes</replaceable> (in any part of the document).</para>

      <para>An element is composed of an optional field specification,
      and a value, separated by a colon. Exemple:
      <replaceable>Beatles</replaceable>,
      <replaceable>author:balzac</replaceable>,
      <replaceable>dc:title:grandet</replaceable> </para>

      <para>The colon, if present, means "contains". Xesam defines other
      relations, which are not supported for now.</para>

      <para>All elements in the search entry are normally combined
      with an implicit AND. It is possible to specify that elements be
      OR'ed instead, as in <replaceable>Beatles</replaceable>
      <literal>OR</literal> <replaceable>Lennon</replaceable>. The
      <literal>OR</literal> must be entered literally (capitals), and
      it has priority over the AND associations:
      <replaceable>word1</replaceable>
      <replaceable>word2</replaceable> <literal>OR</literal>
      <replaceable>word3</replaceable> 
      means 
      <replaceable>word1</replaceable> AND 
      (<replaceable>word2</replaceable> <literal>OR</literal>
      <replaceable>word3</replaceable>)
      not 
      (<replaceable>word1</replaceable> AND
      <replaceable>word2</replaceable>) <literal>OR</literal>
      <replaceable>word3</replaceable>. Do not enter explicit
      parenthesis, they are not supported for now.</para>

      <para>An element preceded by a <literal>-</literal> specifies a
      term that should <emphasis>not</emphasis> appear. Pure negative
      queries are forbidden.</para>

      <para>As usual, words inside quotes define a phrase
      (the order of words is significant), so that
      <replaceable>title:"prejudice pride"</replaceable> is not the same as
      <replaceable>title:prejudice title:pride</replaceable>, and is
      unlikely to find a result.</para> 
      <para>Most Xesam phrase modifiers are unsupported, except for
      <literal>l</literal> (small ell) to disable stemming, and
      <literal>p</literal> to turn a phrase into a NEAR (unordered proximity)
      search. Exemple: <replaceable>"prejudice pride"p</replaceable></para>

      <para>&RCL; currently manages the following default fields:</para>
      <itemizedlist>
        <listitem><para><literal>title</literal>,
        <literal>subject</literal> or <literal>caption</literal> are
        synonyms which specify data to be searched for in the
        document title or subject.</para>
        </listitem>
        <listitem><para><literal>author</literal> or
        <literal>from</literal> for searching the documents originators.</para>
        </listitem>
        <listitem><para><literal>recipient</literal> or
        <literal>to</literal> for searching the documents recipients.</para>
        </listitem>
        <listitem><para><literal>keyword</literal> for searching the
        document-specified keywords (few documents actually have any).</para>
        </listitem>
        <listitem><para><literal>filename</literal> for the document's
        file name.</listitem>
        <listitem><para><literal>ext</literal> specifies the file
        name extension (Ex: <literal>ext:html</literal>)</para>
        </listitem>
      </itemizedlist>

      <para>The field syntax also supports a few field-like, but
      special, criteria:</para>
      <itemizedlist>
        <listitem><para><literal>dir</literal> for filtering the
        results on file location (Ex:
        <literal>dir:/home/me/somedir</literal>). <literal>-dir</literal>
        also works to find results out of the specified directory, only
        after release 1.15.8.</para>
        </listitem>

        <listitem><para><literal>date</literal> for searching or filtering
        on dates. The syntax for the argument is based on the ISO8601
        standard for dates and time intervals. Only dates are supported, no
        times. The general syntax is 2 elements separated by a
        <literal>/</literal> character. Each element can be a date or a
        period of time. Periods are specified as 
<literal>P</literal><replaceable>n</replaceable><literal>Y</literal><replaceable>n</replaceable><literal>M</literal><replaceable>n</replaceable><literal>D</literal>. 
       The <replaceable>n</replaceable> numbers are the respective numbers
       of years, months or days, any of which may be missing. Dates are
       specified as  
<replaceable>YYYY</replaceable>-<replaceable>MM</replaceable>-<replaceable>DD</replaceable>. 
       The days and months parts may be missing. If the
       <literal>/</literal> is present but an element is missing, the
       missing element is interpreted as the lowest or highest date in the
       index. Exemples:</para>
	  <itemizedlist>
	    <listitem><para><literal>2001-03-01/2002-05-01</literal> the
	    basic syntax for an interval of dates.</para>
	    </listitem>
	    <listitem><para><literal>2001-03-01/P1Y2M</literal> the
	    same specified with a period.</para>
	    </listitem>
	    <listitem><para><literal>2001/</literal> from the beginning of
	    2001 to the latest date in the index.</para>
	    </listitem>
	    <listitem><para><literal>2001</literal> the whole year of
	      2001</para></listitem>
	    <listitem><para><literal>P2D/</literal> means 2 days ago up to
	      now if there are no documents with dates in the future.</para>
	    </listitem>
	    <listitem><para><literal>/2003</literal> all documents from
	    2003 or older.</para>
	    </listitem>
	  </itemizedlist>
	  <para>Periods can also be specified with small letters (ie:
	  p2y).</para> 
        </listitem>

        <listitem><para><literal>mime</literal> or
        <literal>format</literal> for specifying the
        mime type. This one is quite special because you can specify
        several values which will be OR'ed (the normal default for the
        language is AND). Ex: <literal>mime:text/plain
        mime:text/html</literal>. Specifying an explicit boolean
        operator before a
        <literal>mime</literal> specification is not supported and
        will produce strange results. You can filter out certain types
        by using negation (<literal>-mime:some/type</literal>), and you can
        use wildcards in the value (<literal>mime:text/*</literal>).
        Note that <literal>mime</literal> is
        the ONLY field with an OR default. You do need to use
        <literal>OR</literal> with <literal>ext</literal> terms for
        example.</para> 
        </listitem>

        <listitem><para><literal>type</literal> or
        <literal>rclcat</literal> for specifying the category (as in
        text/media/presentation/etc.). The classification of mime
        types in categories is defined in the &RCL; configuration
        (<filename>mimeconf</filename>), and can be modified or
        extended. The default category names are those which permit
        filtering results in the main GUI screen. Categories are OR'ed
        like mime types above. This can't be negated with
        <literal>-</literal> either.</para>
        </listitem>

      </itemizedlist>

      <para>Words inside phrases and capitalized words are not
      stem-expanded. Wildcards may be used anywhere inside a term.
      Specifying a wild-card on the left of a term can produce a very
      slow search (or even an incorrect one if the expansion is
      truncated because of excessive size). Also see <link
      linkend="rcl.search.wildcards">More about wildcards</link>.</para>

      <para>The document filters used while indexing have the
      possibility to create other fields with arbitrary names, and
      aliases may be defined in the configuration, so that the exact
      field search possibilities may be different for you if someone
      took care of the customisation.</para>

    <sect2 id="rcl.search.wildcards">
      <title>More about wildcards</title>

      <para>All words entered in &RCL; search fields will be processed
      for wildcard expansion before the request is finally
      executed.</para>

      <para>The wildcard characters are:</para>

      <itemizedlist>
       <listitem><para><literal>*</literal> which matches 0 or more 
        characters.</para>
        </listitem>
        <listitem><para><literal>?</literal> which matches
           a single character.</para>
        </listitem>
        <listitem><para><literal>[]</literal> which allow
         defining sets of characters to be matched (ex:
         <literal>[</literal><userinput>abc</userinput><literal>]</literal> 
          matches a single character which may be 'a' or 'b' or 'c',
         <literal>[</literal><userinput>0-9</userinput><literal>]</literal>
         matches any number.</para>
        </listitem>
      </itemizedlist>

      <para>You should be aware of a few things before using
        wildcards.</para>

      <itemizedlist>
        <listitem><para>Using a wildcard character at the beginning of
        a word can make for a slow search because &RCL; will have to
        scan the whole index term list to find the matches.</para>
        </listitem>
        <listitem><para>Using a <literal>*</literal> at the end of a
        word can produce more matches than you would think, and
        strange search results. You can use the <link
        linkend="rcl.search.termexplorer">term explorer</link> tool to
        check what completions exist for a given term. You can also
        see exactly what search was performed by clicking on the link
        at the top of the result list. In general, for natural
        language terms, stem expansion will produce better results
        than an ending <literal>*</literal> (stem expansion is turned
        off when any wildcard character appears in the term).</para>
        </listitem>
      </itemizedlist>

    </sect2>

    </sect1> <!-- rcl.search.lang -->

    <sect1 id="rcl.search.desktop">
      <title>Desktop integration</title>

      <para>Being independant of the desktop type has its drawbacks: &RCL;
      desktop integration is minimal. Here follow a few things that may
      help.</para> 

      <sect2 id="rcl.search.shortcut">
        <title>Hotkeying recoll</title>

        <para>It is surprisingly convenient to be able to show or hide the
          &RCL; GUI with a single keystroke. Recoll comes with a small
          Python script, based on the <literal>libwnck</literal> window
          manager interface library, which will allow you to do just
          this. The detailed instructions are on
          <ulink url="http://bitbucket.org/medoc/recoll/wiki/HotRecoll">
            this wiki page</ulink>.</para>

      </sect2>

      <sect2 id="rcl.kicker-applet">
        <title>The KDE Kicker Recoll applet</title>

        <para>The &RCL; source tree contains the source code to the
        <literal>recoll_applet</literal>, a small application derived
        from the <literal>find_applet</literal>. This can be used to
        add a small &RCL; launcher to the KDE panel.</para>

        <para>The applet is not automatically built with the main &RCL;
         programs, nor is it included with the main source distribution
        (because the KDE build boilerplate makes it relatively big). You can
        download its source from the recoll.org download page. Use the
        omnipotent <userinput>configure;make;make install</userinput>
        incantation to build and install.</para>
      
        <para>You can then add the applet to the panel by right-clicking the
        panel and choosing the <guilabel>Add applet</guilabel> entry.</para>

        <para>The <literal>recoll_applet</literal> has a small text window
         where you can type a &RCL; query (in query language form), and an
         icon which can be used to restrict the search to certain types of
         files. It is quite primitive, and launches a new recoll GUI instance
         every time (even if it is already running). You may find it useful
         anyway.</para>

      </sect2>

    </sect1> <!-- rcl.search.desktop -->

  </chapter> <!-- Search -->

  <chapter id="rcl.program">
    <title>Programming interface</title>

    <para>&RCL; has an Application programming Interface, usable both
    for indexing and searching, currently accessible from the
    <application>Python</application> language.</para>

    <para>Another less radical way to extend the application is to
    write filters for new types of documents.</para>

    <para>The processing of metadata attributes for documents
    (<literal>fields</literal>) is highly configurable.</para>

    <sect1 id="rcl.program.filters">
        <title>Writing a document filter</title>

      <para>&RCL; filters are executable programs which 
        translate from a specific format (ie:
        <application>openoffice</application>,
        <application>acrobat</application>, etc.) to the &RCL;
        indexing input format, which may be
        <literal>text/plain</literal> or
        <literal>text/html</literal>.</para> 

      <para>As of &RCL; 1.13, there are two kinds of filters:
        <itemizedlist>
	  <listitem><para>Simple filters (the old ones) run once and
	  exit. They can be bare programs like
	  <application>antiword</application>, or shell-scripts using other
	  programs. They are very simple to write, just having to write the
	  text to the standard output.</para>
	  </listitem>
	  <listitem><para>Multiple filters, new in 1.13, run as long as
	  their master process (ie: recollindex) is active. They can
	  process multiple files (sparing the process startup time which
	  can be very significant), or multiple documents per file (ie: for
	  zip or chm files). They communicate with the indexer through a
	  simple protocol, but are nevertheless a bit more complicated than
	  the older kind. Most of these new filters are written in
	  <application>Python</application>, using a common module to
	  handle the protocol.</para>
	  </listitem>
	</itemizedlist>
      The following will just describe the simple filters, if you are
      programmer enough to write one of the other kind, it shouldn't be too
      difficult to make sense of one of the existing modules (ie:
      rclzip).</para> 

      <para>&RCL; simple filters are usually shell-scripts, but this is in
        no way necessary. These programs are extremely simple and most
        of the difficulty lies in extracting the text from the native
        format, not outputting what is expected by &RCL;. Happily
        enough, most document formats already have translators or text
        extractors which handle the difficult part and can be called
        from the filter. In some case the output of the translating
        program is appropriate, and no intermediate shell-script is
        needed.</para> 

        <para>Filters are called with a single argument which is the
        source file name. They should output the result to stdout.</para>

        <para>The <literal>RECOLL_FILTER_FORPREVIEW</literal>
        environment variable (values <literal>yes</literal>,
        <literal>no</literal>) tells the filter if the operation is
        for indexing or previewing. Some filters use this to output a
        slightly different format. This is not essential.</para>

      <para>The association of file types to filters is performed in
      the <filename>mimeconf</filename> file. A sample:</para>
<programlisting>

[index]
application/msword = exec antiword -t -i 1 -m UTF-8;\
     mimetype = text/plain ; charset=utf-8

application/ogg = exec rclogg

text/rtf = exec unrtf --nopict --html; charset=iso-8859-1; mimetype=text/html

application/x-chm = execm rclchm
</programlisting>

      <para>The fragment specifies that:

      <itemizedlist>
	  <listitem><para><literal>application/msword</literal> files
            are processed by executing the <command>antiword</command>
            program, which outputs
            <literal>text/plain</literal> encoded in
            <literal>utf-8</literal>.</para> 
	  </listitem>
	  
	  <listitem><para><literal>application/ogg</literal> files are
            processed by the <command>rclogg</command> script, with
            default output type (<literal>text/html</literal>, with
            encoding specified in the header, or <literal>utf-8</literal>
            by default).</para>
	  </listitem>
	  
	  <listitem><para><literal>text/rtf</literal> is processed by
            <command>unrtf</command>, which outputs
            <literal>text/html</literal>. The 
            <literal>iso-8859-1</literal> encoding is specified because it
            is not the <literal>utf-8</literal> default, and not output by
            <command>unrtf</command> in the HTML header section.</para>
	  </listitem>
	  <listitem><para><literal>application/x-chm</literal> is processed
	      by a persistant filter. This is determined by the
	      <literal>execm</literal> keyword.</para>
	  </listitem>
	</itemizedlist>
      The easiest way to write a new filter is probably to start from an
      existing one.</para> 

      <para>Filters which output <literal>text/plain</literal> text
      are generally simpler, but they cannot specify the character set
      and other metadata, so they are limited to cases where these
      elements are not needed.</para>


    <sect2 id="rcl.program.filters.html">
        <title>Filter HTML output</title>

        <para>The output HTML could be very minimal like the following
        example:</para>

        <programlisting>&lt;html>&lt;head>
&lt;meta http-equiv="Content-Type" content="text/html;charset=UTF-8">
&lt/head>
&lt;body>some text content&lt;/body>&lt;/html>
          </programlisting>

        <para>You should take care to escape some
        characters inside
          the text by transforming them into appropriate
          entities. "<literal>&amp;</literal>" should be transformed into
          "<literal>&amp;amp;</literal>", "<literal>&lt;</literal>"
          should be transformed into
          "<literal>&amp;lt;</literal>". This is not always properly
          done by translating programs which output HTML, and of
          course nerver by those which output plain text.</para>

        <para>The character set needs to be specified in the
          header. It does not need to be UTF-8 (&RCL; will take care
          of translating it), but it must be accurate for good
          results.</para>

        <para>&RCL; will also make use of other header fields if
          they are present: <literal>title</literal>,
          <literal>description</literal>,
          <literal>keywords</literal>.</para>

        <para>Filters also have the possibility to "invent" field
        names. This should be output as meta tags:</para>

        <programlisting>
&lt;meta name="somefield" content="Some textual data" /&gt;
</programlisting>

      <para> See the following section for details about configuring
      how field data is processed by the indexer.</para>

      </sect2>

    </sect1>

    <sect1 id="rcl.program.fields">
        <title>Field data processing</title>

      <para><literal>Fields</literal> are named pieces of information
      in or about documents, like <literal>title</literal>,
      <literal>author</literal>, <literal>abstract</literal>.</para> 

      <para>The field values for documents can appear in several ways
      during indexing: either output by filters as
      <literal>meta</literal> fields in the HTML header section, or
      added as attributes of the <literal>Doc</literal> object when
      using the API, or again synthetized internally by &RCL;.</para>

      <para>The &RCL; query language allows searching for text in a
      specific field.</para>

      <para>&RCL; defines a number of default fields. Additional
      ones can be output by filters, and described in the
      <filename>fields</filename> configuration file.</para>

      <para>Fields can be:</para>
      <itemizedlist>

        <listitem><para><literal>indexed</literal>, meaning that their
        terms are separately stored in inverted lists (with a specific
        prefix), and that a field-specific search is possible.</para>
        </listitem>

        <listitem><para><literal>stored</literal>, meaning that their
        value is recorded in the index data record for the document,
        and can be returned and displayed with search results.</para>
        </listitem>

      </itemizedlist>

      <para>A field can be either or both indexed and stored. This and 
      other aspects of fields handling is defined inside the
      <filename>fields</filename> configuration file.</para>

      <para>You can find more information in the 
      <link linkend="rcl.install.config.fields">section about the
      <filename>fields</filename> file</link>, or in comments inside the
      file.</para> 


    </sect1>


    <sect1 id="rcl.program.api">
      <title>API</title>

    <sect2 id="rcl.program.api.elements">
      <title>Interface elements</title>

      <para>A few elements in the interface are specific and and need
      an explanation.</para>

      <variablelist>

        <varlistentry>
          <term>udi</term> <listitem><para>An udi (unique document
            identifier) identifies a document. Because of limitations
            inside the index engine, it is restricted in length (to
            200 bytes), which is why a regular URI cannot be used. The
            structure and contents of the udi is defined by the
            application and opaque to the index engine. For example,
            the internal file system indexer uses the complete
            document path (file path + internal path), truncated to
            length, the suppressed part being replaced by a hash
            value.</para> </listitem>
        </varlistentry>

        <varlistentry> 
          <term>ipath</term> 
          
          <listitem><para>This data value (set as a field in the Doc
          object) is stored, along with the URL, but not indexed by
          &RCL;. Its contents are not interpreted, and its use is up
          to the application. For example, the &RCL; internal file
          system indexer stores the part of the document access path
          internal to the container file (<literal>ipath</literal> in
          this case is a list of subdocument sequential numbers). url
          and ipath are returned in every search result and permit
          access to the original document.</para>
          </listitem>
        </varlistentry>

        <varlistentry> 
          <term>Stored and indexed fields</term> 
          
          <listitem><para>The <filename>fields</filename> file inside
          the &RCL; configuration defines which document fields are
          either "indexed" (searchable), "stored" (retrievable with
          search results), or both.</para>
          </listitem>
        </varlistentry>

        </variablelist>

      <para>Data for an external indexer, should be stored in a
      separate index, not the one for the &RCL; internal file system
      indexer, except if the latter is not used at all). The reason
      is that the main document indexer purge pass would remove all
      the other indexer's documents, as they were not seen during
      indexing. The main indexer documents would also probably be a
      problem for the external indexer purge operation.</para>

    </sect2>

    <sect2 id="rcl.program.api.python">
      <title>Python interface</title>

      <sect3 id="rcl.program.python.intro">
        <title>Introduction</title>

        <para>&RCL; versions after 1.11 define a Python programming
          interface, both for searching and indexing.</para> 

        <para>The Python interface is not built by default and can be
          found in the source package,
          under <filename>python/recoll</filename>.</para>
	<para>In order to build the module, you should first build
	  or re-build the Recoll library using position-independant
	  objects:
<screen>
  <userinput>cd recoll-xxx/</userinput>
  <userinput>configure --enable-pic</userinput>
  <userinput>make</userinput>
</screen>
	  There is no significant disadvantage in using PIC objects
	  for the main Recoll executables, so you can use the
	  <literal>--enable-pic</literal> option for the main build
	  too.</para> 

	<para>The <filename>python/recoll/</filename> directory
	  contains the usual <filename>setup.py</filename> 
          script which you can then use to build and install the
          module:
<screen>
  <userinput>cd recoll-xxx/python/recoll</userinput>
  <userinput>python setup.py build</userinput>
  <userinput>python setup.py install</userinput>
</screen>
        </para> 

      </sect3>


      <sect3 id="rcl.program.python.manual">
        <title>Interface manual</title>

      <literalLayout>
NAME
    recoll - This is an interface to the Recoll full text indexer.

FILE
    /usr/local/lib/python2.5/site-packages/recoll.so

CLASSES
        Db
        Doc
        Query
        SearchData
    
    class Db(__builtin__.object)
     |  Db([confdir=None], [extra_dbs=None], [writable = False])
     |  
     |  A Db object holds a connection to a Recoll index. Use the connect()
     |  function to create one.
     |  confdir specifies a Recoll configuration directory (default: 
     |   $RECOLL_CONFDIR or ~/.recoll).
     |  extra_dbs is a list of external databases (xapian directories)
     |  writable decides if we can index new data through this connection
     |  
     |  Methods defined here:
     |  
     |  
     |  addOrUpdate(...)
     |      addOrUpdate(udi, doc, parent_udi=None) -> None
     |      Add or update index data for a given document
     |      The udi string must define a unique id for the document. It is not
     |      interpreted inside Recoll
     |      doc is a Doc object
     |      if parent_udi is set, this is a unique identifier for the
     |      top-level container (ie mbox file)
     |  
     |  delete(...)
     |      delete(udi) -> Bool.
     |      Purge index from all data for udi. If udi matches a container
     |      document, purge all subdocs (docs with a parent_udi matching udi).
     |  
     |  makeDocAbstract(...)
     |      makeDocAbstract(Doc, Query) -> string
     |      Build and return 'keyword-in-context' abstract for document
     |      and query.
     |  
     |  needUpdate(...)
     |      needUpdate(udi, sig) -> Bool.
     |      Check if the index is up to date for the document defined by udi,
     |      having the current signature sig.
     |  
     |  purge(...)
     |      purge() -> Bool.
     |      Delete all documents that were not touched during the just finished
     |      indexing pass (since open-for-write). These are the documents for
     |      the needUpdate() call was not performed, indicating that they no
     |      longer exist in the primary storage system.
     |  
     |  query(...)
     |      query() -> Query. Return a new, blank query object for this index.
     |  
     |  setAbstractParams(...)
     |      setAbstractParams(maxchars, contextwords).
     |      Set the parameters used to build 'keyword-in-context' abstracts
     |  
     |  ----------------------------------------------------------------------
     |  Data and other attributes defined here:
     |  
    
    class Doc(__builtin__.object)
     |  Doc()
     |  
     |  A Doc object contains index data for a given document.
     |  The data is extracted from the index when searching, or set by the
     |  indexer program when updating. The Doc object has no useful methods but
     |  many attributes to be read or set by its user. It matches exactly the
     |  Rcl::Doc c++ object. Some of the attributes are predefined, but, 
     |  especially when indexing, others can be set, the name of which will be
     |  processed as field names by the indexing configuration.
     |  Inputs can be specified as unicode or strings.
     |  Outputs are unicode objects.
     |  All dates are specified as unix timestamps, printed as strings
     |  Predefined attributes (index/query/both):
     |   text (index): document plain text
     |   url (both)
     |   fbytes (both) optional) file size in bytes
     |   filename (both)
     |   fmtime (both) optional file modification date. Unix time printed 
     |      as string
     |   dbytes (both) document text bytes
     |   dmtime (both) document creation/modification date
     |   ipath (both) value private to the app.: internal access path
     |      inside file
     |   mtype (both) mime type for original document
     |   mtime (query) dmtime if set else fmtime
     |   origcharset (both) charset the text was converted from
     |   size (query) dbytes if set, else fbytes
     |   sig (both) app-defined file modification signature. 
     |      For up to date checks
     |   relevancyrating (query)
     |   abstract (both)
     |   author (both)
     |   title (both)
     |   keywords (both)
     |  
     |  Methods defined here:
     |  
     |  
     |  ----------------------------------------------------------------------
     |  Data and other attributes defined here:
     |  
    
    class Query(__builtin__.object)
     |  Recoll Query objects are used to execute index searches. 
     |  They must be created by the Db.query() method.
     |  
     |  Methods defined here:
     |  
     |  
     |  execute(...)
     |      execute(query_string, stemming=1|0)
     |      
     |      Starts a search for query_string, a Recoll search language string
     |      (mostly Xesam-compatible).
     |      The query can be a simple list of terms (and'ed by default), or more
     |      complicated with field specs etc. See the Recoll manual.
     |  
     |  executesd(...)
     |      executesd(SearchData)
     |      
     |      Starts a search for the query defined by the SearchData object.
     |  
     |  fetchone(...)
     |      fetchone(None) -> Doc
     |      
     |      Fetches the next Doc object in the current search results.
     |  
     |  sortby(...)
     |      sortby(field=fieldname, ascending=true)
     |      Sort results by 'fieldname', in ascending or descending order.
     |      Only one field can be used, no subsorts for now.
     |      Must be called before executing the search
     |  
     |  ----------------------------------------------------------------------
     |  Data descriptors defined here:
     |  
     |  next
     |      Next index to be fetched from results. Normally increments after
     |      each fetchone() call, but can be set/reset before the call effect
     |      seeking. Starts at 0
     |  
     |  ----------------------------------------------------------------------
     |  Data and other attributes defined here:
     |  
    
    class SearchData(__builtin__.object)
     |  SearchData()
     |  
     |  A SearchData object describes a query. It has a number of global
     |  parameters and a chain of search clauses.
     |  
     |  Methods defined here:
     |  
     |  
     |  addclause(...)
     |      addclause(type='and'|'or'|'excl'|'phrase'|'near'|'sub',
     |                qstring=string, slack=int, field=string, stemming=1|0,
     |                subSearch=SearchData)
     |      Adds a simple clause to the SearchData And/Or chain, or a subquery
     |      defined by another SearchData object
     |  
     |  ----------------------------------------------------------------------
     |  Data and other attributes defined here:
     |  

FUNCTIONS
    connect(...)
        connect([confdir=None], [extra_dbs=None], [writable = False])
                 -> Db.
        
        Connects to a Recoll database and returns a Db object.
        confdir specifies a Recoll configuration directory
        (the default is built like for any Recoll program).
        extra_dbs is a list of external databases (xapian directories)
        writable decides if we can index new data through this connection


</literalLayout>
        </sect3>

      <sect3 id="rcl.program.python.examples">
        <title>Example code</title>

        <para>The following sample would query the index with a user
        language string. See the <filename>python/samples</filename>
        directory inside the &RCL; source for other examples.</para>

        <programlisting>
#!/usr/bin/env python

import recoll

db = recoll.connect()
db.setAbstractParams(maxchars=80, contextwords=2)

query = db.query()
nres = query.execute("some user question")
print "Result count: ", nres
if nres > 5:
    nres = 5
while query.next >= 0 and query.next < nres: 
    doc = query.fetchone()
    print query.next
    for k in ("title", "size"):
        print k, ":", getattr(doc, k).encode('utf-8')
    abs = db.makeDocAbstract(doc, query).encode('utf-8')
    print abs
    print



</programlisting>

      </sect3>

    </sect2>

  </chapter>


  <chapter id="rcl.install">
    <title>Installation and configuration</title>

    <sect1 id="rcl.install.binary">
      <title>Installing a binary copy</title>

      <para>There are three types of binary &RCL; installations:
	<itemizedlist>
	  <listitem><para>Through your system normal software distribution
	      framework (ie, <application>Debian/Ubuntu apt</application>,
	      <application>FreeBSD</application> ports, etc.).</para>
	  </listitem> 

	  <listitem><para>From a package downloaded from the
	      &RCL; web site.</para> 
	  </listitem> 

	  <listitem><para>From a prebuilt tree downloaded from the &RCL;
	  web site.</para> 
	  </listitem>
	</itemizedlist>

      In all cases, the strict software dependancies (ie on &XAP; or
      <application>iconv</application>) will be automatically satisfied,
      you should not have to worry about them.</para>

      <para>You will only have to check or install <link
      linkend="rcl.install.external">supporting applications</link>
      for the file types that you want to index beyond those that are
      natively processed by &RCL; (text, HTML, mail files, and a few
      others).</para>

      <para>You should also maybe have a look at the 
      <link linkend="rcl.install.config">configuration section</link>
      (but this may not be necessary for a quick test with default
      parameters). Most parameters can be more conveniently set from the
      GUI interface.</para>

      <sect2 id="rcl.install.binary.package">
        <title>Installing through a package system</title>

        <para>If you use a BSD-type port system or a prebuilt package (DEB,
        RPM, manually or through the system software configuration
        utility), just follow the usual procedure for your system.</para>

      </sect2>

      <sect2 id="rcl.install.binary.rcl">
        <title>Installing a prebuilt &RCL;</title>

        <para>The unpackaged binary versions on the &RCL; web site are
        just compressed tar files of a build tree, where only the
        useful parts were kept (executables and sample
        configuration).</para>

        <para>The executable binary files are built with a static link to
        libxapian and libiconv, to make installation easier (no
        dependencies).</para> 

        <para>After extracting the tar file, you can proceed with
        <link linkend="rcl.install.building.install">installation</link> as
        if you had built the package from source (that is, just type
        <literal>make install</literal>). The binary trees are built for
        installation to <filename>/usr/local</filename>.</para>

      </sect2>
    </sect1>

    <sect1 id="rcl.install.external">
      <title>Supporting packages</title>

      <para>&RCL; uses external applications to index some file
        types. You need to install them for the file types that you wish to
        have indexed (these are run-time optional dependencies. None is
        needed for building or running &RCL; except for indexing their
        specific file type).</para>

      <para>After an indexing pass, the commands that were found
	missing can be displayed from the <command>recoll</command>
	<guilabel>File</guilabel> menu. The list is stored in the
	<filename>missing</filename> text file inside the configuration
	directory.</para>

      <para>A list of common file types which need external
        commands follows. Many of the filters need the
        <command>iconv</command> command, which is not always listed as a
        dependancy.</para> 

      <para>Please note that, due to the relatively dynamic nature of this
	information, the most up to date version is now kept on the &RCLAPPS;
	along with links to the home pages or best source/patches pages,
	and misc tips. The list below is not updated often and may be quite
	stale.</para>

      <para>For many Linux distributions, most of the commands listed can
        be installed from the package repositories. However, the packages
        are sometimes outdated, or not the best version for &RCL;, so you
        should take a look at the &RCLAPPS; if a file
        type is important to you.</para>

      <para>As of &RCL; release 1.14, a number of XML-based formats that
        were handled by ad hoc filter code now use the
        <command>xsltproc</command> command, which usually comes with  
	  <application>libxslt</application>. These are: abiword, fb2
	  (ebooks), kword, openoffice, svg.</para> 

      <para>Now for the list:</para>
      <itemizedlist>

        <listitem><para>Openoffice files need <command>unzip</command> and
        <command>xsltproc</command>.</para></listitem>

        <listitem><para>PDF files need <command>pdftotext</command> which
        is part of the <application>Xpdf</application> or
        <application>Poppler</application> packages.</para></listitem>

        <listitem><para>Postscript files need <command>pstotext</command>. 
            The original version has an issue with shell
            character in file names, which is corrected in recent
            packages. See the the &RCLAPPS; for more detail.
          </listitem>

        <listitem><para>MS Word needs
        <command>antiword</command>. It is also useful to have
        <command>wvWare</command> installed as it may be 
        be used as a fallback for some files which
        <command>antiword</command> does not handle.</para></listitem>

        <listitem><para>MS Excel and PowerPoint need <command>
            catdoc</command>.</para></listitem>

        <listitem><para>MS Open XML (docx) needs <command>
         xsltproc</command>.</para></listitem>

        <listitem><para>Wordperfect files need <command>wpd2html</command>
        from the <application>libwpd</application> (or
        <application>libwpd-tools</application> on Ubuntu)
        package.</para></listitem>

        <listitem><para>RTF files need <command>unrtf</command>, which, in
        its standard version, has much trouble with non-western character
        sets. Check the &RCLAPPS;.</para></listitem>

        <listitem><para>TeX files need <command>untex</command> or
        <command>detex</command>. Check the &RCLAPPS; for sources if it's not
        packaged for your distribution.</para></listitem>

        <listitem><para>dvi files need <command>dvips</command>.</para>
        </listitem>

        <listitem><para>djvu files need <command>djvutxt</command> and
        <command>djvused</command> from the
        <application>DjVuLibre</application> package.</para></listitem>
          
        <listitem><para>Audio files: &RCL; releases before 1.13
          used the <command>id3info</command> command from the <application>
          id3lib</application> package to extract mp3 tag information,
          <command>metaflac</command> (standard flac tools) for flac files,
          and <command>ogginfo</command> (vorbis tools) for ogg
          files. Releases 1.14 and later use a single
          <application>Python</application> filter based 
          on <application>mutagen</application> for all audio file
          types.</para>
	</listitem>

        <listitem><para>Pictures: &RCL; uses the 
         <application>Exiftool</application>
         <application>Perl</application> package to extract tag
         information. Most image file formats are supported. Note that
         there may not be much interest in indexing the technical tags
         (image size, aperture, etc.). This is only of interest if you
         store personal tags or textual descriptions inside the image
         files.</para></listitem>

	<listitem><para>chm: files in microsoft help format need Python and
          the <application>pychm</application> module (which needs 
          <application>chmlib</application>).</para></listitem>

	<listitem><para>ICS: up to &RCL; 1.13, iCalendar files need 
        <application>Python</application>
	and the <application>icalendar</application>
	module. <application>icalendar</application> is not needed for newer
	versions,  which use internal code.</para></listitem> 

	<listitem><para>Zip archives need <application>Python</application>
	(and the standard zipfile module).</para></listitem>

	<listitem><para>Midi karaoke files need
	<application>Python</application> and the 
        <ulink url="http://pypi.python.org/pypi/midi/0.2.1">
         <application>Midi module</application></ulink></para>
        </listitem>

        <listitem><para>Konqueror webarchive format with Python (uses the
        Tarfile module).</para></listitem>

        <listitem><para>mimehtml web archive format (support based on the mail
          filter, which introduces some mild weirdness, but still
          usable).</para></listitem>

        </itemizedlist>

        <para>Text, HTML, mail folders, and Scribus files are
        processed internally. <application>Lyx</application> is used to
        index Lyx files. Many filters need <command>iconv</command> and the
        standard <command>sed</command> and <command>awk</command>.
        </para>

    </sect1>


      <sect1 id="rcl.install.building">
      <title>Building from source</title>

      <sect2 id="rcl.install.building.prereqs">
        <title>Prerequisites</title>

	<para>C++ compiler. Up to &RCL; version 1.13.04, its absence can
	manifest itself by strange messages about a missing
	iconv_open.</para>

	<para>Development files for 
        <ulink url="http://www.xapian.org">
         <application>Xapian core</application></ulink>.</para> <important><para>If you
         are building Xapian for an older CPU (before Pentium 4 or Athlon
         64), you need to add the --disable-sse flag to the configure
         command. Else all Xapian application will crash with an
         <literal>illegal instruction</literal> error.</para>
	</important>

	<para>Development files for 
         <ulink url="http://www.trolltech.com/products/qt/index.html">
         <application>Qt</application> </ulink>.</para>

	<para>Development files for <application>X11</application> and
	<application>zlib</application>.</para>

        <para>Check the <ulink url="http://www.recoll.org/download.html">
         &RCL; download page</ulink> for up to date version
         information.</para>

      <para>You will most probably be able to find a binary package for
        <application>Qt</application> for your system. You may have to
        compile &XAP; but this is not difficult (if you are using
        <application>FreeBSD</application>, there is a port).</para>

      <para>You may also need 
        <ulink
        url="http://www.gnu.org/software/libiconv/">libiconv</ulink>. &RCL;
        currently uses version 1.9 (this should not be critical). On
        <application>Linux</application> systems, the iconv interface
        is part of libc and you should not need to do anything
        special.</para>
      
      <sect2 id="rcl.install.building.build">
        <title>Building</title>

      <para>&RCL; has been built on Linux, FreeBSD, Mac OS X, and Solaris,
      most versions after 2005 should be ok, maybe some older ones too
      (Solaris 8 is ok). If you build on another system, and
        need to modify things,
        <ulink url="mailto:jfd@recoll.org">I would
        very much welcome patches</ulink>.</para>

      <para>Depending on the <application>Qt&nbsp;3</application>
      configuration on your system, you may have to set the
      <literal>QTDIR</literal> and <literal>QMAKESPECS</literal>
      variables in your environment:</para>
        <itemizedlist>
          <listitem><para><literal>QTDIR</literal> should point to the
          directory above the one that holds the qt include files (ie:
          if <filename>qt.h</filename> is
          <filename>/usr/local/qt/include/qt.h</filename>, QTDIR
          should be <filename>/usr/local/qt</filename>).</para>
          </listitem>
          <listitem><para><literal>QMAKESPECS</literal> should
          be set to the name of one of the
          <application>qt</application> mkspecs sub-directories (ie:
          linux-g++).</para> 
          </listitem>
        </itemizedlist>

        <para>On many Linux systems, <literal>QTDIR</literal> is set
        by the login scripts, and <literal>QMAKESPECS</literal> is not
        needed because there is a <filename>default</filename> link in
        <filename>mkspecs/</filename>. 

	<para>Neither <literal>QTDIR</literal> nor 
	<literal>QMAKESPECS</literal> should be needed with 
        Qt&nbsp;4, configuration details are entirely determined by 
	<command>qmake</command> (which is quite often installed as 
	<command>qmake-qt4</command>).</para> 

        <formalpara><title>Configure options:</title>
	  <para>
        <itemizedlist>
	  <listitem><para><literal>--without-aspell</literal> 
            will disable the code for phonetic matching of search
            terms. </para>
          </listitem>
	  <listitem><para><literal>--with-fam</literal> or
            <literal>--with-inotify</literal> will enable the code for
            real time indexing. Inotify support is enabled by default on
            recent Linux systems.</para>
	  </listitem>
	  <listitem><para><literal>--enable-xattr</literal> will enable
 	    code to fetch data from file extended attributes. This is only
	    useful is some application stores data in there, and also needs
	    some simple configuration (see comments in the
	    <filename>fields</filename> configuration file).</para>
	  </listitem>
	  <listitem><para><literal>--enable-camelcase</literal> will enable
 	    splitting <replaceable>camelCase</replaceable> words. This
	    is not enabled by default as it has the unfortunate
	    side-effect of making some phrase searches quite
	    confusing: ie, <literal>"MySQL manual"</literal> would be
	    matched by <literal>"MySQL manual"</literal> and
	    <literal>"my sql manual"</literal> but not <literal>"mysql
	    manual"</literal> (only inside phrase searches).</para>
	  </listitem>
	  <listitem><para><literal>--with-file-command</literal> Specify
	    the version of the 'file' command to use (ie:
            --with-file-command=/usr/local/bin/file). Can be useful to
            enable the gnu version on systems where the native one is
            bad.</para> 
	  </listitem>
	  <listitem><para><literal>--without-gui</literal> Disable the Qt
	    interface, and auxiliary uses of X11, and compile the command
	    line version.</para> 
	  </listitem>
	  <listitem><para>Of course the usual 
	  <application>autoconf</application> <command>configure</command>
	  options, like <literal>--prefix</literal> apply.</para> 
	  </listitem>
         </itemizedlist>
         </para>
	</formalpara>

      <para>Normal procedure:</para>
      <screen>
        <userinput>cd recoll-xxx</userinput>
        <userinput>configure</userinput>
        <userinput>make</userinput>
        <userinput>(practices usual hardship-repelling invocations)</userinput>
      </screen>


      <para>There is little auto-configuration. The
        <command>configure</command> script will mainly link one of
        the system-specific files in the <filename>mk</filename>
        directory to <filename>mk/sysconf</filename>. If your system
        is not known yet, it will tell you as much, and you may want
        to manually copy and modify one of the existing files (the new
        file name should be the output of <command>uname -s</command>).</para>
      </sect2>

      <sect2 id="rcl.install.building.install">
        <title>Installation</title>
      
      <para>Either type <userinput>make install</userinput> or execute
      <userinput>recollinstall
      <replaceable>prefix</replaceable></userinput>, in the root 
        of the source tree. This will copy the commands to
        <filename><replaceable>prefix</replaceable>/bin</filename>
        and the sample configuration files, scripts and other shared
        data to
        <filename><replaceable>prefix</replaceable>/share/recoll</filename>.</para>
        <para>If the installation prefix given to
        <command>recollinstall</command> is different from either the 
	system default or the value which was
        specified when executing <command>configure</command> (as in 
	<userinput>configure --prefix /some/path</userinput>), you
        will have to set the <literal>RECOLL_DATADIR</literal>
        environment variable to indicate where the shared data is to
        be found (ie for (ba)sh: 
	<userinput>export RECOLL_DATADIR=/some/path/share/recoll</userinput>).
	</para> 

        <para>You can then proceed to <link
       linkend="rcl.install.config">configuration</link>. </para>

      </sect2>
    </sect1>

    <sect1 id="rcl.install.config">
      <title>Configuration overview</title>

      <para>Most of the parameters specific to the
        <command>recoll</command> GUI are set through the
        <guilabel>Preferences</guilabel> menu and stored in the
        standard Qt place (<filename>$HOME/.qt/recollrc</filename>). 
        You probably do not want to edit this by hand.</para>

      <para>&RCL; indexing options are set inside text configuration
        files located in a configuration directory. There can be
        several such directories, each of which define the parameters
        for one index.</para>

      <para>The configuration files can be edited by hand or through
        the <guilabel>Indexing configuration</guilabel> dialog 
        (<guilabel>Preferences</guilabel> menu). The GUI tool will try
        to respect your formatting and comments as much as possible,
        so it is quite possible to use both ways.</para>

      <para>The most accurate documentation for the
        configuration parameters is given by comments inside the default
        files, and we will just give a general overview here.</para>

      <para>For each index, there are two sets of configuration
        files. System-wide configuration files are kept in a directory named
        like <filename>/usr/[local/]share/recoll/examples</filename>,
        and define default values, shared by all indexes. For each
        index, a parallel set of files defines the customized
        parameters.</para>

      <para>The default location of the configuration is the
        <filename>.recoll</filename>
        directory in your home. Most people will only use this
        directory.</para> 

      <para>This location can be changed, or others can be added with the
        <literal>RECOLL_CONFDIR</literal> environment variable or the
        -c option parameter to <command>recoll</command> and
        <command>recollindex</command>.</para>

      <para>If the <filename>.recoll</filename> directory does not
        exist when <command>recoll</command> or
        <command>recollindex</command> are started, it will be created
        with a set of empty configuration files.
        <command>recoll</command> will give you a chance to edit the
        configuration file before starting
        indexing. <command>recollindex</command> will proceed
        immediately. To avoid mistakes, the automatic directory
        creation will only occur for the
       default location, not if <literal>-c</literal> or
       <literal>RECOLL_CONFDIR</literal> were used (in the latter
       cases, you will have to create the directory).</para>
      

        <para>All configuration files share the same format. For
        example, a short extract of the main configuration file might
        look as follows:</para> 
        <programlisting>
        # Space-separated list of directories to index.
        topdirs =  ~/docs /usr/share/doc

        [~/somedirectory-with-utf8-txt-files]
        defaultcharset = utf-8
        </programlisting>

        <para>There are three kinds of lines: </para>
        <itemizedlist>
          <listitem><para>Comment (starts with
          <emphasis>#</emphasis>) or empty.</para> 
          </listitem>
          <listitem><para>Parameter affectation (<emphasis>name =
          value</emphasis>).</para> 
          </listitem>
          <listitem><para>Section definition
          ([<emphasis>somedirname</emphasis>]).</para> 
          </listitem>
        </itemizedlist>

        <para>Depending on the type of configuration file, section
        definitions either separate groups of parameters or allow
        redefining some parameters for a directory sub-tree. They stay
        in effect until another section definition, or the end of
        file, is encountered. Some of the parameters used for indexing
        are looked up hierarchically from the current directory
        location upwards. Not all parameters can be meaningfully
        redefined, this is specified for each in the next
        section. </para>

        <para>When found at the beginning of a file path, the tilde
        character (~) is expanded to the name of the user's home
        directory, as a shell would do.</para> 
        
        <para>White space is used for separation inside  lists.
        List elements with embedded spaces can be quoted using
        double-quotes.</para>

      <sect2 id="rcl.install.config.recollconf">
        <title>Main configuration file</title>

        <para><filename>recoll.conf</filename> is the main
         configuration file. It defines things like
         what to index (top directories and things to ignore), and the
         default character set to use for document types which do not
         specify it internally.</para>

        <para>The default configuration will index your home
         directory. If this is not appropriate, start
         <command>recoll</command> to create a blank 
         configuration, click <guimenu>Cancel</guimenu>, and edit
         the configuration file before restarting the command. This
         will start the initial indexing, which may take some time.</para>

        <para>Most of the following parameters can be changed from the
        <guilabel>Index Configuration</guilabel> menu in the
        <command>recoll</command> interface. Some can only be set by
        editing the configuration file.</para>

        <sect3 id="rcl.install.config.recollconf.files">
          <title>Parameters affecting what documents we index:</title>

        <variablelist>

          <varlistentry id="rcl.install.config.recollconf.topdirs">
            <term><literal>topdirs</literal></term>
            <listitem><para>Specifies the list of directories or files to
            index (recursively for directories). You can use symbolic links
            as elements of this list. See the
            <literal>followLinks</literal> option about following symbolic links
            found under the top elements (not followed by default).</para>
            </listitem>
          </varlistentry>

          <varlistentry><term><literal>skippedNames</literal></term>
            <listitem>
              <para>A space-separated list of patterns for
               names of files or directories that should be completely
               ignored. The list defined in the default file is: </para>
<programlisting>
skippedNames = #* bin CVS  Cache cache* caughtspam  tmp .thumbnails .svn \
 	       *~ .beagle .git .hg .bzr loop.ps .xsession-errors \
	       .recoll* xapiandb recollrc recoll.conf 
</programlisting>
              <para>The list can be redefined at any sub-directory in the
		indexed area.</para>
              <para>The top-level directories are not affected by this
                list (that is, a directory in <literal>topdirs</literal>
                might match and would still be indexed).</para>
                <para>The list in the default configuration does not
                exclude hidden directories (names beginning with a
                dot), which means that it may index quite a few things
                that you do not want. On the other hand, mail user
                agents like <application>thunderbird</application>
                usually store messages in hidden directories, and you
                probably want this indexed. One possible solution is to
                have <filename>.*</filename> in
                <literal>skippedNames</literal>, and add things like
                <filename>~/.thunderbird</filename> or
                <filename>~/.evolution</filename> in
                <literal>topdirs</literal>.</para> 

                <para>Not even the file names are indexed for patterns
                in this list. See the
                <literal>recoll_noindex</literal> variable in
                <filename>mimemap</filename> for an alternative
                approach which indexes the file names.</para>
            </listitem>
          </varlistentry>

          <varlistentry><term><literal>skippedPaths</literal> and
             <literal>daemSkippedPaths</literal> </term>
            <listitem>
              <para>A space-separated list of patterns for
               <emphasis>paths</emphasis> of files or directories that should be skipped.
               There is no default in the sample configuration file,
               but the code always adds the configuration and database
               directories in there.</para>
              <para><literal>skippedPaths</literal> is used both by
              batch and real time
              indexing. <literal>daemSkippedPaths</literal> can be
              used to specify things that should be indexed at
              startup, but not monitored.</para>
              <para>Example of use for skipping text files only in a
              specific directory:</para>
              <programlisting>
skippedPaths = ~/somedir/&lowast;.txt
              </programlisting>
            </listitem>
          </varlistentry>

          <varlistentry id="rcl.install.config.recollconf.followlinks">
            <term><literal>followLinks</literal></term>
            <listitem><para>Specifies if the indexer should follow
            symbolic links while walking the file tree. The default is
            to ignore symbolic links to avoid multiple indexing of
            linked files. No effort is made to avoid duplication when
            this option is set to true. This option can be set
            individually for each of the <literal>topdirs</literal>
            members by using sections. It can not be changed below the
            <literal>topdirs</literal> level.</para>
            </listitem> 
          </varlistentry>

          <varlistentry><term><literal>indexedmimetypes</literal></term>
            <listitem><para>&RCL; normally indexes any file which it
            knows how to read. This list lets you restrict the indexed
            mime types to what you specify. If the variable is
            unspecified or the list empty (the default), all supported
            types are processed.</para>
            </listitem>
          </varlistentry>

          <varlistentry><term><literal>compressedfilemaxkbs</literal></term>
            <listitem><para>Size limit for compressed (.gz or .bz2)
            files. These need to be decompressed in a temporary
            directory for identification, which can be very wasteful
            if 'uninteresting' big compressed files are present.
            Negative means no limit, 0 means no processing of any
            compressed file. Defaults to -1.</para>
            </listitem>
          </varlistentry>

          <varlistentry><term><literal>textfilemaxmbs</literal></term>
            <listitem><para>Maximum size for text files. Very big text
            files are often uninteresting logs. Set to -1 to disable
            (default 20MB).</para>  
            </listitem>
           </varlistentry>

          <varlistentry><term><literal>textfilepagekbs</literal></term>
            <listitem><para>If set to other than -1, text files will be
            indexed as multiple documents of the given page size. This may
            be useful if you do want to index very big text files as it
            will both reduce memory usage at index time and help with
            loading data to the preview window. A size of a few megabytes
            would seem reasonable (default: 1MB).</para>
            </listitem>
           </varlistentry>

          <varlistentry><term><literal>indexallfilenames</literal></term>
            <listitem><para>&RCL; indexes file names in a special
            section of the database to allow specific file names
            searches using wild cards. This parameter decides if 
            file name indexing is performed only for files with mime
            types that would qualify them for full text indexing, or
            for all files inside the selected subtrees, independently of
            mime type.</para>
            </listitem>
          </varlistentry>

          <varlistentry><term><literal>usesystemfilecommand</literal></term>
            <listitem><para>Decide if we use the <command>file -i</command>
            system command as a final step for determining the mime
            type for a file (the main procedure uses suffix
            associations as defined in the  <filename>mimemap</filename>
            file). This can be useful for files with suffix-less names,
            but it will also cause the indexing of many bogus "text"
            files.</para> 
            </listitem>
          </varlistentry>

          <varlistentry><term><literal>processbeaglequeue</literal></term>
            <listitem><para>If this is set, process the directory where
            Beagle Web browser plugins copy visited pages for indexing. Of
            course, Beagle MUST NOT be running, else things will behave
            strangely.</para>
            </listitem>
           </varlistentry>

          <varlistentry><term><literal>beaglequeuedir</literal></term>
            <listitem><para>The path to the Beagle indexing queue. This is
            hard-coded in the Beagle plugin as
            <filename>~/.beagle/ToIndex</filename> so there should be no
            need to change it.</para> 
            </listitem>
           </varlistentry>

        </variablelist>
       </sect3>

       <sect3 id="rcl.install.config.recollconf.terms">
	<title>Parameters affecting how we generate terms:</title>

        <para>Changing some of these parameters will imply a full
        reindex. Also, when using multiple indexes, it may not make sense
        to search indexes that don't share the values for these parameters,
        because they usually affect both search and index operations.</para>

        <variablelist>

          <varlistentry><term><literal>nonumbers</literal></term>
            <listitem><para>If this set to true, no terms will be generated
            for numbers. For example "123", "1.5e6", 192.168.1.4, would not
            be indexed ("value123" would still be). Numbers are often quite
            interesting to search for, and this should probably not be set
            except for special situations, ie, scientific documents with huge
            amounts of numbers in them. This can only be set for a whole
            index, not for a subtree.</para>
            </listitem>
          </varlistentry>

          <varlistentry><term><literal>nocjk</literal></term>
            <listitem><para>If this set to true, specific east asian
            (Chinese Korean Japanese) characters/word splitting is
            turned off. This will save a small amount of cpu if you
            have no CJK documents. If your document base does include
            such text but you are not interested in searching it,
            setting <literal>nocjk</literal> may be a significant time
            and space saver.</para>
            </listitem>
          </varlistentry>

          <varlistentry><term><literal>cjkngramlen</literal></term>
            <listitem><para>This lets you adjust the size of n-grams
            used for indexing CJK text. The default value of 2 is
            probably appropriate in most cases. A value of 3 would
            allow more precision and efficiency on longer words, but
            the index will be approximately twice as large.</para>
            </listitem>
          </varlistentry>
          <varlistentry><term><literal>indexstemminglanguages</literal></term>
            <listitem><para>A list of languages for which the stem
            expansion databases will be built. See recollindex(1) or
            use the <literal>recollindex -l</literal> command for
            possible values. You can add a stem expansion database for
            a different language by using <command>recollindex
            -s</command>, but it will be deleted during the next
            indexing. Only languages listed in the configuration
            file are permanent.</para>
            </listitem>
          </varlistentry>
         
          <varlistentry><term><literal>defaultcharset</literal></term>
            <listitem><para>The name of the character set used for
            files that do not contain a character set definition (ie:
            plain text files). This can be redefined for any
            sub-directory. If it is not set at all, the character set
            used is the one defined by the nls environment (LC_ALL,
            LC_CTYPE, LANG), or iso8859-1 if nothing is set.</para> 
	   </listitem>
         </varlistentry>

          <varlistentry><term><literal>maildefcharset</literal></term>
            <listitem><para>This can be used to define the default
		character set specifically for mail messages which don't
		specify it. This is mainly useful for readpst (libpst) dumps,
		which are utf-8 but do not say so.</para>
            </listitem>
          </varlistentry>

          <varlistentry><term><literal>localfields</literal></term>
            <listitem><para>This allows setting fields for all documents
            under a given directory. Typical usage would be to set an
            "rclaptg" field, to be used in <filename>mimeview</filename> to
            select a specific viewer. If several fields are to be set, they
            should be separated with a colon (':') character (which there
            is currently no way to escape). Ie:
		<literal>localfields= rclaptg=gnus:other = val</literal>, then
		select specifier viewer with
		<literal>mimetype|tag=...</literal> in
		<filename>mimeview</filename>.</para>  
            </listitem>
           </varlistentry>
        </variablelist>
       </sect3>

       <sect3 id="rcl.install.config.recollconf.storage">
	<title>Parameters affecting where and how we store things:</title>

	<variablelist>
          <varlistentry><term><literal>dbdir</literal></term>
            <listitem><para>The name of the Xapian data directory. It
            will be created if needed when the index is
            initialized. If this is not an absolute path, it will be
            interpreted relative to the configuration directory. The
            value can have embedded spaces but starting or trailing
            spaces will be trimmed. You cannot use quotes here.</para>
            </listitem>
          </varlistentry>

          <varlistentry><term><literal>maxfsoccuppc</literal></term>
            <listitem><para>Maximum file system occupation before we
            stop indexing. The value is a percentage, corresponding to
            what the "Capacity" df output column shows.  The default
            value is 0, meaning no checking. </para>
            </listitem>
          </varlistentry>

	  <varlistentry><term><literal>mboxcachedir</literal></term>
	    <listitem><para>The directory where mbox message offsets cache
	    files are held. This is normally $RECOLL_CONFDIR/mboxcache, but
	    it may be useful to share a directory between different
	    configurations.</para>
	    </listitem>
	  </varlistentry>

	  <varlistentry><term><literal>mboxcacheminmbs</literal></term>
	    <listitem><para>The minimum mbox file size over which we
		cache the offsets. There is really no sense in caching
		offsets for small files. The default is 5 MB.</para>
	    </listitem>
	   </varlistentry>

          <varlistentry><term><literal>webcachedir</literal></term>
            <listitem><para>This is only used by the Beagle web browser
            plugin indexing code, and defines where the cache for visited
            pages will live. Default:
            <filename>$RECOLL_CONFDIR/webcache</filename></para> 
            </listitem>

           </varlistentry>
          <varlistentry><term><literal>webcachemaxmbs</literal></term>
            <listitem><para>This is only used by the Beagle web browser
            plugin indexing code, and defines the maximum size for the web
            page cache. Default: 40 MB.</para> 
            </listitem>
           </varlistentry>


          <varlistentry><term><literal>idxflushmb</literal></term>
            <listitem><para>Threshold (megabytes of new text data)
            where we flush from memory to disk index. Setting this can
            help control memory usage. A value of 0 means no explicit
            flushing, letting Xapian use its own default, which is
            flushing every 10000 documents (memory usage depends on
            average document size). The default value is 10.</para>
            </listitem>
          </varlistentry>

        </variablelist>
       </sect3>


       <sect3 id="rcl.install.config.recollconf.misc">
	<title>Miscellaneous parameters:</title>

	 <variablelist>

          <varlistentry><term><literal>loglevel,daemloglevel</literal></term>
            <listitem><para>Verbosity level for recoll and
            recollindex. A value of 4 lists quite a lot of
            debug/information messages. 2 only lists errors. The
            <literal>daem</literal>version is specific to the indexing monitor
            daemon.</para>
            </listitem>
          </varlistentry>

          <varlistentry><term><literal>logfilename,
		daemlogfilename</literal></term> 
            <listitem><para>Where the messages should go. 'stderr' can
            be used as a special value, and is the default. The
            <literal>daem</literal>version is specific to the indexing monitor
            daemon.</para>
            </listitem>
          </varlistentry>

          <varlistentry><term><literal>filtermaxseconds</literal></term>
            <listitem><para>Maximum filter execution time, after which it
            is aborted. Some postscript programs just loop...</para> 
            </listitem>
           </varlistentry>

          <varlistentry><term><literal>filtersdir</literal></term>
            <listitem><para>A directory to search for the external
            filter scripts used to index some types of files. The
            value should not be changed, except if you want to modify
            one of the default scripts. The value can be redefined for
            any sub-directory. </para>
            </listitem>
          </varlistentry>

          <varlistentry><term><literal>iconsdir</literal></term>
            <listitem><para>The name of the directory where
            <command>recoll</command> result list icons are
            stored. You can change this if you want different
            images.</para>
            </listitem>
          </varlistentry>

          <varlistentry><term><literal>idxabsmlen</literal></term>
            <listitem><para>&RCL; stores an abstract for each indexed
            file inside the database. The text can come from an actual
            'abstract' section in the document or will just be the
            beginning of the document. It is stored in the index so
            that it can be displayed inside the result lists without
            decoding the original
            file. The <literal>idxabsmlen</literal> parameter defines
            the size of the stored abstract. The default value is 250 bytes.
            The search interface gives you the choice to display this
            stored text or a synthetic abstract built by extracting
            text around the search terms. If you always
            prefer the synthetic abstract, you can reduce this value
            and save a little space.
            </para>
            </listitem>
          </varlistentry>

          <varlistentry><term><literal>aspellLanguage</literal></term>
            <listitem><para>Language definitions to use when creating
            the aspell dictionary.  The value must match a set of
            aspell language definition files. You can type "aspell
            config" to see where these are installed (look for
            data-dir). The default if the variable is not set is to
            use your desktop national language environment to guess
            the value.</para>
            </listitem>
          </varlistentry>

          <varlistentry><term><literal>noaspell</literal></term>
            <listitem><para>If this is set, the aspell dictionary
            generation is turned off. Useful for cases where you don't
            need the functionality or when it is unusable because
            aspell crashes during dictionary generation.</para>
            </listitem>
          </varlistentry>

          <varlistentry><term><literal>guesscharset</literal></term>
            <listitem><para>Decide if we try to guess the character
            set of files if no internal value is available (ie: for
            plain text files). This does not work well in general, and
            should probably not be used. </para>
            </listitem>
          </varlistentry>

        </variablelist>

      </sect2>

      <sect2 id="rcl.install.config.fields">
	<title>The fields file</title>

	<para>This file contains information about dynamic fields handling
	in &RCL;. Some very basic fields have hard-wired behaviour,
	and, mostly, you should not change the original data inside the
	<filename>fields</filename> file. But you can create custom fields
	fitting your data and handle them just like they were native
	ones.</para>

	<para>The <filename>fields</filename> file has several sections,
	which each define an aspect of fields processing. Quite often,
	you'll have to modify several sections to obtain the desired
	behaviour.</para>

	<para>We will only give a short description here, you should refer
	to the comments inside the file for more detailed information.</para>

	<para>Field names should be lowercase alphabetic ASCII.</para>

        <variablelist>

          <varlistentry>
            <term>[prefixes]</term>
            <listitem><para>A field becomes indexed (searchable) by having
            a prefix defined in this section.
            </listitem>
          </varlistentry>
          <varlistentry>
            <term>[stored]</term>
            <listitem><para>A field becomes stored (displayable inside
            results) by having its name listed in this section (typically
            with an empty value).
            </listitem>
          </varlistentry>
          <varlistentry>
            <term>[aliases]</term>
            <listitem><para>This section defines lists of synonyms for the
            canonical names used inside the <literal>[prefixes]</literal>
            and <literal>[stored]</literal> sections</para>
            </listitem>
          </varlistentry>
          <varlistentry>
            <term>filter-specific sections</term>
            <listitem><para>Some filters may need specific
            configuration for handling fields. Only the mail message filter
            currently has such a section (named
            <literal>[mail]</literal>). It allows indexing arbitrary mail
            headers in addition to the ones indexed by default. Other such
            sections may appear in the future.</para>
            </listitem>
          </varlistentry>

	</variablelist>

      <para>Here follows a small example of a personal
      <filename>fields</filename> 
      file. This would extract a specific mail header and
      use it as a searchable field, with data displayable inside result
      lists. (Side note: as the mail filter does no decoding on the values,
      only plain ascii headers can be indexed, and only the
      first occurrence will be used for headers that occur several times).

<programlisting>[prefixes]
# Index mailmytag contents (with the given prefix)
mailmytag = XMTAG

[stored]
# Store mailmytag inside the document data record (so that it can be
# displayed - as %(mailmytag) - in result lists).
mailmytag = 

[mail]
# Extract the X-My-Tag mail header, and use it internally with the
# mailmytag field name
x-my-tag = mailmytag
</programlisting>
</para>


      </sect2>

      <sect2 id="rcl.install.config.mimemap">
        <title>The mimemap file</title>

        <para><filename>mimemap</filename> specifies the
        file name extension to mime type mappings.</para> 

        <para>For file names without an extension, or with an unknown
        one, the system's <command>file -i</command> command will be
        executed to determine the mime type (this can be switched off
        inside the main configuration file).</para>

        <para>The mappings can be specified on a per-subtree basis,
        which may be useful in some cases. Example:
        <application>gaim</application> logs have a
        <filename>.txt</filename> extension but 
        should be handled specially, which is possible because they
        are usually all located in one place.</para>

        <para><filename>mimemap</filename> also has a
        <literal>recoll_noindex</literal> variable which is a list of
        suffixes. Matching files will be skipped (which avoids
        unnecessary decompressions or <command>file</command>
        executions). This is partially redundant with
        <literal>skippedNames</literal> in the main configuration
        file, with a few differences: it will not affect directories,
        it cannot be made dependant on the file-system location (it is
        a configuration-wide parameter), and the file names will still
        be indexed (not even the file names are indexed for patterns
        in <literal>skippedNames</literal>.
        <literal>recoll_noindex</literal> is used mostly for things
        known to be unindexable by a given &RCL; version. Having it
        there avoids cluttering the more user-oriented and locally
        customized <literal>skippedNames</literal>.</para>

      </sect2>

      <sect2 id="rcl.install.config.mimeconf">
        <title>The mimeconf file</title>

        <para><filename>mimeconf</filename> specifies how the
         different mime types are handled for indexing, and which icons
         are displayed in the <command>recoll</command> result lists.</para>

        <para>Changing the parameters in the [index] section is
         probably not a good idea except if you are a &RCL;
         developer.</para> 

        <para>The [icons] section allows you to change the icons which
         are displayed by <command>recoll</command> in the result
         lists (the values are the basenames of the png images inside
         the <filename>iconsdir</filename> directory (specified in
         <filename>recoll.conf</filename>).</para>

      </sect2>
      <sect2 id="rcl.install.config.mimeview">
        <title>The mimeview file</title>

        <para><filename>mimeview</filename> specifies which programs
        are started when you click on an <guilabel>Open</guilabel>
        link in a result list. Ie: HTML is normally displayed using
         <application>firefox</application>, but you may prefer
         <application>Konqueror</application>, your
         <application>openoffice.org</application> 
         program might be named <command>oofice</command> instead of
         <command>openoffice</command> etc. 
         </para>

        <para>Changes to this file can be done by direct editing, or
        through the <command>recoll</command> user preferences dialog.</para>

        <para>If <guilabel>Use desktop preferences to choose document
        editor</guilabel> is checked in the &RCL; GUI user preferences, all
        <filename>mimeview</filename> entries will be ignored except the
        one labelled <literal>application/x-all</literal> (which is set to
        use <command>xdg-open</command> by default).</para>

        <para>As for the other configuration files, the normal usage
        is to have a <filename>mimeview</filename> inside your own
        configuration directory, with just the non-default entries,
        which will override those from the central configuration
        file.</para>
        <para>Please note that these entries must be placed under a
          <literal>[view]</literal> section.</para>

	<para>The keys in the file are normally mime types. You can add an
	  application tag to specialize the choice for an area of the
	  filesystem (using a <literal>localfields</literal> specification
	  in <filename>mimeconf</filename>). The syntax for the key is 
<replaceable>mimetype</replaceable><literal>|</literal><replaceable>tag</replaceable></para>

        <para>The <literal>nouncompforviewmts</literal> entry, (placed at
        the top level, outside of the <literal>[view]</literal> section),
        holds a list of mime types that should not be uncompressed before
        starting the viewer (if they are found compressed, ie:
        <replaceable>mydoc.doc.gz</replaceable>).</para>

        <para>The right side of each assignment holds a command to be
        executed for opening the file. The following substitutions are
        performed:</para> 

        <itemizedlist>
          <listitem>
            <formalpara><title>%D</title><para>Document date</para></formalpara>
          </listitem> <listitem><formalpara><title>%f</title><para>File
          name. This may be the name of a temporary file if it was
          necessary to create one (ie: to extract a subdocument from a
          container).</para></formalpara>
          </listitem>
          <listitem><formalpara><title>%F</title><para>Original file name. 
          Same as %f except if a temporary file is used.</para></formalpara>
          </listitem>
          <listitem><formalpara><title>%i</title><para>Internal path, for
          subdocuments of containers. The format depends on the container
          type. If this appears in the command line, &RCL; will not create
          a temporary file to extract the subdocument, expecting the called
          application (possibly a script) to be able to handle
          it.</para></formalpara>
          </listitem>
          <listitem><formalpara><title>%M</title><para>Mime
                  type</para></formalpara> 
          </listitem>
          <listitem><formalpara><title>%U, %u</title><para>Url.
           </para></formalpara> 
          </listitem>
        </itemizedlist>

        <para>In addition to the predefined values above, all strings like
        <literal>%(fieldname)</literal> will be replaced by the value of
        the field named <literal>fieldname</literal> for the
        document. This could be used in combination with field
        customisation to help with opening the document.</para> 

      </sect2>

      <sect2 id="rcl.install.config.examples">
        <title>Examples of configuration adjustments</title>

        <sect3 id="rcl.install.config.examples.addview">
          <title>Adding an external viewer for an non-indexed type</title>

          <para>Imagine that you have some kind of file which does not
            have indexable content, but for which you would like to have a
            functional <guilabel>Open</guilabel> link in the result list
            (when found by file name). The file names end in
            <replaceable>.blob</replaceable> and can be displayed by
            application <replaceable>blobviewer</replaceable>.</para>

          <para>You need two entries in the configuration files for this
	    to work:</para>

          <itemizedlist>
            <listitem><para>In <filename>$RECOLL_CONFDIR/mimemap</filename>
		(typically <filename>~/.recoll/mimemap</filename>), add the
		following line:<programlisting>
.blob = application/x-blobapp
</programlisting>
		Note that the mime type is made up here, and you could
		call it <replaceable>diesel/oil</replaceable> just the
		same.</para>
            </listitem>
            <listitem><para>In <filename>$RECOLL_CONFDIR/mimeview</filename>
		under the <literal>[view]</literal> section, add:</para>
              <programlisting>
application/x-blobapp = blobviewer %f
</programlisting>
              <para>We are supposing
              that <replaceable>blobviewer</replaceable> wants a file
              name parameter here, you would use <literal>%u</literal> if
              it liked URLs better.</para>
            </listitem>
          </itemizedlist>

          <para>If you just wanted to change the application used by
            &RCL; to display a mime type which it already knows, you
            would just need to edit <filename>mimeview</filename>. The
            entries you add in your personal file override those in the
            central configuration, which you do not need to
            alter. <filename>mimeview</filename> can also be modified
            from the Gui.</para>

        </sect3>

        <sect3 id="rcl.install.config.examples.addindex">
          <title>Adding indexing support for a new file type</title>

          <para>Let us now imagine that the above
            <replaceable>.blob</replaceable> files actually contain
            indexable text and that you know how to extract it with a
            command line program. Getting &RCL; to index the files is
            easy. You need to perform the above alteration, and also to
            add data to the <filename>mimeconf</filename> file
            (typically in <filename>~/.recoll/mimeconf</filename>):</para>
          <itemizedlist>
            <listitem><para>Under the <literal>[index]</literal>
		section, add the following line (more about the
		<replaceable>rclblob</replaceable> indexing script
		later):<programlisting>
application/x-blobapp = exec rclblob
</programlisting></para>
            </listitem>
            <listitem><para>Under the <literal>[icons]</literal>
		section, you should choose an icon to be displayed for the
		files inside the result lists. Icons are normally 64x64
		pixels PNG files which live in
		<filename>/usr/[local/]share/recoll/images</filename>.</para>
            </listitem>
            <listitem><para>Under the <literal>[categories]</literal>
		section, you should add the mime type where it makes sense
		(you can also create a category). Categories may be used
		for filtering in advanced search.</para>
            </listitem>
          </itemizedlist>

          <para>The <replaceable>rclblob</replaceable> filter should
            be an executable program or script which exists inside
            <filename>/usr/[local/]share/recoll/filters</filename>. It
            will be given a file name as argument and should output the
            text or html contents on the standard output.</para>

          <para>The <link linkend="rcl.program.filters">filter
              programming</link> section describes in more detail how
              to write a filter.</para> 

        </sect3>

      </sect2>

    </sect1>

  </chapter>

</book>