The system relies on abstract indexes that are mapped to corresponding XPath.
These indexes are mainly used in the fcs
module which is the currently the core module for searching in the data.
The index mappings are defined in
/mets:mets/mets:amdSec/mets:techMD[@ID="projectMappings"]
There is a schema for this map (mappings.xsd), but it is outdated (#9).
A sample could look like:
<mdWrap MDTYPE="OTHER">
<xmlData>
<map xmlns="">
<namespaces>
<ns prefix="tei" uri="http://www.tei-c.org/ns/1.0"/>
<ns prefix="mets" uri="http://www.loc.gov/METS/"/>
<ns prefix="fcs" uri="http://clarin.eu/fcs/1.0"/>
<ns prefix="cr" uri="http://aac.ac.at/content_repository"/>
</namespaces>
<map key="" path="" title="">
<index key="rf">
<path match="@xml:id" label="@n">pb</path>
</index>
<index key="fcs.toc" type="default">
<path match="@cr:id">div[@type='chapter']</path>
<path match="@cr:id">front</path>
<path match="@cr:id">back</path>
</index>
<index key="cql.serverChoice" type="ft">
<path>p</path>
<path>l</path>
<path>head</path>
<path>table</path>
<path>titlePage</path>
</index>
<index key="fcs.resource" type="ft" on-data="project">
<path match="@ID" label="@LABEL">mets:div</path>
</index>
<index key="fcs.rf" type="ft">
<path match="@resourcefragment-pid" label="@rf-label">fcs:resourceFragment</path>
</index>
<index key="w" label="Wortform" type="ft" scan="true">
<path>tei:w</path>
</index>
<index key="lemma" label="Lemma" type="ft" scan="true">
<path match="@lemma" label="@lemma">tei:w[@lemma]</path>
</index>
<index key="persName" label="Personennamen" type="default" facet="persType" scan="true">
<path match="@key" label="./descendant::tei:w/@lemma">persName</path>
<path match="@key" label="./descendant::tei:w/@lemma">tei:persName</path>
</index>
<index key="persType" type="default">
<path match="@type">tei:persName</path>
</index>
</map>
</map>
</xmlData>
</mdWrap>
Some of the indexes have a special meaning:
rf
defines the xpath delimiting the basic resource fragments (usually the pages, marked in TEI by empty element <pb/>
fcs.toc
index defining the higher-level structures of the resource, meant for generating a table of contents. The paths should be adapted to match the structures in the resources
fcs.resource
internally used index to identify a resource. The xpath relates to the internal data and shouldn't be adapted
fcs.rf
internally used index to identify a resource fragment. The xpath relates to the internal data and shouldn't be adapted
facs
defines the xpath to the filename/url of a facsimile/image pertaining to a given resource fragment
cql.serverChoice this represents the default index, i.e. where should the system search, if the user searches just with a term (unqualified by any specific index)