com.jaeksoft.searchlib.url_config.xml Maven / Gradle / Ivy

Go to download

Show more of this group Show more artifacts with this name
Show all versions of opensearchserver Show documentation

OpenSearchServer is a powerful, enterprise-class, search engine program. Using the web user interface, the crawlers (web, file, database, ...) and the REST/RESTFul API you will be able to integrate quickly and easily advanced full-text search capabilities in your application. OpenSearchServer runs on Windows and Linux/Unix/BSD.

The newest version!

<?xml version="1.0" encoding="UTF-8"?>
<!-- License Agreement for OpenSearchServer -->
<!-- Copyright (C) 2008-2016 Emmanuel Keller / Jaeksoft -->
<!-- http://www.open-search-server.com -->
<!-- This file is part of OpenSearchServer. -->
<!-- OpenSearchServer is free software: you can redistribute it and/or modify 
	it under the terms of the GNU General Public License as published by the 
	Free Software Foundation, either version 3 of the License, or (at your option) 
	any later version. OpenSearchServer is distributed in the hope that it will 
	be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of 
	MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General 
	Public License for more details. You should have received a copy of the GNU 
	General Public License along with Jaeksoft Open Search Server. If not, see 
	<http://www.gnu.org/licenses/>. -->
<configuration>
    <indices>
        <index name="url" searchCache="0" filterCache="0" fieldCache="0"/>
    </indices>
    <schema>
        <fields default="url" unique="url">
            <field name="lang" indexed="yes" stored="no" termVector="no"/>
            <field name="langMethod" indexed="yes" stored="no" termVector="no"/>
            <field name="contentBaseType" indexed="yes" stored="no"
                   termVector="no"/>
            <field name="contentTypeCharset" indexed="yes" stored="no"
                   termVector="no"/>
            <field name="contentEncoding" indexed="yes" stored="no"
                   termVector="no"/>
            <field name="contentLength" indexed="yes" stored="no"
                   termVector="no"/>
            <field name="url" indexed="yes" stored="no" termVector="no"/>
            <field name="outlink" indexed="yes" stored="no" termVector="no"/>
            <field name="inlink" indexed="yes" stored="no" termVector="no"/>
            <field name="host" indexed="yes" stored="no" termVector="no"/>
            <field name="subhost" indexed="yes" stored="no" termVector="no"/>
            <field name="when" indexed="yes" stored="no" termVector="no"/>
            <field name="responseCode" indexed="yes" stored="no"
                   termVector="no"/>
            <field name="robotsTxtStatus" indexed="yes" stored="no"
                   termVector="no"/>
            <field name="parserStatus" indexed="yes" stored="no"
                   termVector="no"/>
            <field name="fetchStatus" indexed="yes" stored="no" termVector="no"/>
            <field name="indexStatus" indexed="yes" stored="no" termVector="no"/>
            <field name="md5size" indexed="yes" stored="no" termVector="no"/>
            <field name="lastModifiedDate" indexed="yes" stored="no"
                   termVector="no"/>
            <field name="contentUpdateDate" indexed="yes" stored="no"
                   termVector="no"/>
            <field name="parentUrl" indexed="yes" stored="no" termVector="no"/>
            <field name="redirectionUrl" indexed="yes" stored="no"
                   termVector="no"/>
            <field name="origin" indexed="yes" stored="no" termVector="no"/>
            <field name="headers" indexed="yes" stored="no" termVector="no"/>
            <field name="backlinkCount" indexed="yes" stored="no"
                   termVector="no"/>
            <field name="depth" indexed="yes" stored="no" termVector="no"/>
            <field name="urlWhen" indexed="yes" stored="no" termVector="no"/>
        </fields>
    </schema>
    <requests>
        <request name="urlSearch" defaultOperator="and" start="0"
                 rows="0" allowLeadingWildcard="yes">
            <query>$$</query>
            <returnFields>
                url,host,contentBaseType,contentTypeCharset,contentEncoding,contentLength,
                lang,langMethod,when,responseCode,
                robotsTxtStatus,parserStatus,fetchStatus,indexStatus,
                md5size,lastModifiedDate,parentUrl,origin,headers,backlinkCount,depth
            </returnFields>
        </request>
        <request name="urlExport" defaultOperator="and" start="0"
                 rows="0" allowLeadingWildcard="yes">
            <query>$$</query>
            <returnFields>url,lastModifiedDate</returnFields>
        </request>
        <request name="hostFacet" defaultOperator="or" start="0" rows="0">
            <query>$$</query>
            <facetFields>
                <facetField name="host" minCount="1"/>
            </facetFields>
        </request>
    </requests>
</configuration>