ead2dcterms_config.xml

<?xml version="1.0" encoding="UTF-8"?>
<?xml-stylesheet type="text/xsl" href="XmlMapper/view/xml_mapper_table.xsl"?>
<!--
Convert a finding aid in EAD to a list of records in Dublin Core Metadata Terms.

For main description, see readme.md [https://github.com/Daniel-KM/Ead2DCterms].

For technical notes, see xml_mapper_config.xml.

This file allows to set the configuration for the process, in particular to
define the set of records and the type of identifier and relations.

For more info about the mapping between EAD and Dublin Core Terms, see the
file of mappings.

See other notes in default xml_mapper_config.xml and xml_mapper_mappings.xml.

Choice of type of records
=========================

* Define the main records

The finding aid itself can be converted into one, two or three records:
- mixed "eadheader" and "frontmatter", so each "frontmatter" part ("titlepage"
  and "div") is added as a description of a single record created from the
  "eadheader".
- separated "eadheader" and "frontmatter", so two records are built.
- mixed or separated archival description.

The recommended option is to use one main record (ead header and front matter)
and another separated record for the archival description, but it depends on
original source.

* Digital objects as records or references

Furthermore, two default profiles of records are prepared according to files and
digital objects:
- separated records: each file or digital object is a record with its own
  metadata. Relation with main part or component is made with the DC terms
  "Is Part Of" / "Has Part". Note: this choice implies to define an "identifier"
  for each target.
- mixed records: each file or digital object is a simple metadata of a record,
  included with the DC term "Has Format".

A third profile is added to list other prepared records, that separate
everything (header, front matter, archival description, components and files).

Simply set true (default) or false to select and process the needed records.

The "use" main mapping allows to set the base of the record, that may define a
base node for the other mappings.

Identifiers
===========

The format EAD doesn't require an identifier for each part and component of the
xml. It may be the id or the unit id, but they are not always set, the unit
title, but there is a risk of duplicates, the position in the source tree, but
it's not stable, the xpath, but it's not clear for people, the list of levels
and types, but this can be too long, an url or an urn, but it should be defined.
It may be a combination of them too and there may be a public one and a internal
one. It may be anything else, like an ark.

By default (see the mapping), the base identifier will be the one defined in the
mapping (set in the source, or the default). This base is followed by the xpath
to the element, for example "http://localhost/id/ead/archdesc/c01[6]/c02[3]".

For digital objects, the identifier is the "href", the "entityref", the "target"
or any other unique location or reference.

Options
=======

Options used to create each type of records:
- "process": extract metadata for this type.
- "element_root": define the name of the root, else use the default;
- "element_record: define the name of the element, else use the default;
- "record_name": add the mapping name as the attribute "type" of each record.
- "normalize_space": spaces of values can be normalized.
- "deduplicate": process can create duplicates, for example for the title.
- "remove_empty": process can create empty elements.
- "order": order the result according to the mapping specified in the mapping
  file (attribute for "records" only).

Options can be set at the level record, else at the records level, else it will
be "true".

TODO
====

- Add an option to output records nested.

@version: 20150824
@see https://www.loc.gov/ead
@see http://dublincore.org/specifications
@copyright Daniel Berthereau, 2015
@license CeCILL v2.1 http://www.cecill.info/licences/Licence_CeCILL_V2.1-en.html
@link https://github.com/Daniel-KM/Ead2DCterms
-->
<config>
    <!-- These infos are the same as in mappings, and used only for the display. -->
    <from name="EAD 2002" prefix="ead" namespace="http://www.loc.gov/ead" />
    <to name="DCMI Metadata Terms" prefix="dcterms" namespace="http://purl.org/dc/terms/" />

    <!-- Mapping between EAD and Dublin Core Metadata Terms. -->
    <option name="mappings" value="ead2dcterms_mappings.xml" />

    <!-- Stylesheet used for rules that are not managed by the mapper currently.
    NOTE This stylesheet should be imported by the mapper manually, or a
    specific mapper should be done. -->
    <option name="rules" value="ead2dcterms_rules.xsl" />

    <!-- Default element name of the root and each record. -->
    <option name="default_element_root" value="records" />
    <option name="default_element_record" value="record" />

    <!-- Prepend the prefix of the namespace to each element output. -->
    <option name="output_prefix" value="true" />

    <!-- Possibilities to add other namespaces. -->
    <!-- <namespace prefix="ead" namespace="http://www.loc.gov/ead" /> -->

    <!-- This base id will be used to create the unique identifier of each
    record, that may be used to create relations between records. If no
    attribute "from" is used, the one set in the mapping will be used.
    By default, records are the finding aid, the archival description and each
    components. Digital archival objects (files) are records too, but they use
    their own id.
    Note: "document-uri(/)" doesn't work with xslt 1.
    -->
    <baseid from="document-uri(/)" default="http://localhost/id" />

    <!-- This default conversion sets a main record for the Finding Aid, with
    data from the Front Matter, another record for the Archival Description, one
    record for each component and one record for each digital archival object.
    -->
    <records process="true"
             element_root="records"
             element_record="record"
             record_name="true"
             normalize_space="true"
             deduplicate="true"
             remove_empty="true"
             order="true">
        <!-- One record for the main informations (Header and Front Matter). -->
        <record element="record" name="Finding Aid" process="true">
            <use mapping="Finding Aid" type="base" />
            <use mapping="Identifier: baseid/xpath" node="/ead/eadheader" process="true" />
            <use mapping="Has part: Archival Description" process="true" />
            <use mapping="Has part: Digital objects as records" node="/ead/eadheader|frontmatter" process="true" />
        </record>

        <!-- One record for the Archival Description (with the main "dsc"). -->
        <record name="Archival Description" process="true">
            <use mapping="Archival Description + Description of Subordinate Components" type="base" />
            <use mapping="Identifier: baseid/xpath" node="/ead/archdesc" />
            <!-- This node is used because the mapping is looking for the first
            ancestor of the node, that is the "eadheader". -->
            <use mapping="Is part of" node="/ead/eadheader/eadid" />
            <use mapping="Has part: Components" node="/ead/archdesc/dsc" />
            <use mapping="Has part: Digital objects as records" node="/ead/archdesc/mapper:copy-except('dsc', .)" />
            <use mapping="Has part: Digital objects as records" node="/ead/archdesc/dsc/mapper:copy-except('c|c01', .)" />
        </record>

        <!-- One record for each component (cXX), at any level. -->
        <record name="Component" process="true">
            <use mapping="Component" type="base" />
            <use mapping="Identifier: baseid/xpath" />
            <use mapping="Is part of" />
            <use mapping="Has part: Components" />
            <use mapping="Has part: Digital objects as records" node="self::c/mapper:copy-except('c', .)" />
            <use mapping="Has part: Digital objects as records" node="self::c01/mapper:copy-except('c02', .)" />
            <use mapping="Has part: Digital objects as records" node="self::c02/mapper:copy-except('c03', .)" />
            <use mapping="Has part: Digital objects as records" node="self::c03/mapper:copy-except('c04', .)" />
            <use mapping="Has part: Digital objects as records" node="self::c04/mapper:copy-except('c05', .)" />
            <use mapping="Has part: Digital objects as records" node="self::c05/mapper:copy-except('c06', .)" />
            <use mapping="Has part: Digital objects as records" node="self::c06/mapper:copy-except('c07', .)" />
            <use mapping="Has part: Digital objects as records" node="self::c07/mapper:copy-except('c08', .)" />
            <use mapping="Has part: Digital objects as records" node="self::c08/mapper:copy-except('c09', .)" />
            <use mapping="Has part: Digital objects as records" node="self::c09/mapper:copy-except('c10', .)" />
            <use mapping="Has part: Digital objects as records" node="self::c10/mapper:copy-except('c11', .)" />
            <use mapping="Has part: Digital objects as records" node="self::c11/mapper:copy-except('c12', .)" />
        </record>

        <!-- One related record (is part of) for each digital archival object
        ("dao"). -->
        <record name="Digital Archival Object" process="true">
            <use mapping="Digital Archival Object" type="base" />
            <use mapping="Is part of" />
        </record>
        <!-- One related record for (is part of) each digital archival object
        pointer ("daogrp" locations and references). -->
        <record name="Digital Archival Object" process="true">
            <use mapping="Digital Archival Object Group pointers" type="base" />
            <use mapping="Digital Archival Object Group common" />
            <use mapping="Is part of" />
        </record>
    </records>

    <!-- This configuration builds same records as the previous one, but without
    separated digital objects. So it creates a main record for the Finding Aid,
    with data from the Front Matter, another record for the Archival Description
    and records for each component. Each digital archival object is referenced
    by its parent.
    -->
    <records process="false">
        <!-- One record for the main informations (Header and Front Matter). -->
        <record name="Finding Aid">
            <use mapping="Finding Aid" type="base" />
            <use mapping="Identifier: baseid/xpath" node="/ead/eadheader" />
            <use mapping="Has part: Archival Description" />
            <use mapping="Has format: Digital objects as links" node="/ead/eadheader|frontmatter" />
        </record>

        <!-- One record for the Archival Description (with the main "dsc"). -->
        <record name="Archival Description">
            <use mapping="Archival Description + Description of Subordinate Components" type="base" />
            <use mapping="Identifier: baseid/xpath" node="/ead/archdesc" />
            <use mapping="Is part of" node="/ead/eadheader/eadid" />
            <use mapping="Has part: Components" node="/ead/archdesc/dsc" />
            <use mapping="Has format: Digital objects as links" node="/ead/archdesc/mapper:copy-except('dsc', .)" />
            <use mapping="Has format: Digital objects as links" node="/ead/archdesc/dsc/mapper:copy-except('c|c01', .)" />
        </record>

        <!-- One record for each component (cXX), at any level. -->
        <record name="Component">
            <use mapping="Component" type="base" />
            <use mapping="Identifier: baseid/xpath" />
            <use mapping="Is part of" />
            <use mapping="Has part: Components" />
            <use mapping="Has format: Digital objects as links" node="self::c/mapper:copy-except('c', .)" />
            <use mapping="Has format: Digital objects as links" node="self::c01/mapper:copy-except('c02', .)" />
            <use mapping="Has format: Digital objects as links" node="self::c02/mapper:copy-except('c03', .)" />
            <use mapping="Has format: Digital objects as links" node="self::c03/mapper:copy-except('c04', .)" />
            <use mapping="Has format: Digital objects as links" node="self::c04/mapper:copy-except('c05', .)" />
            <use mapping="Has format: Digital objects as links" node="self::c05/mapper:copy-except('c06', .)" />
            <use mapping="Has format: Digital objects as links" node="self::c06/mapper:copy-except('c07', .)" />
            <use mapping="Has format: Digital objects as links" node="self::c07/mapper:copy-except('c08', .)" />
            <use mapping="Has format: Digital objects as links" node="self::c08/mapper:copy-except('c09', .)" />
            <use mapping="Has format: Digital objects as links" node="self::c09/mapper:copy-except('c10', .)" />
            <use mapping="Has format: Digital objects as links" node="self::c10/mapper:copy-except('c11', .)" />
            <use mapping="Has format: Digital objects as links" node="self::c11/mapper:copy-except('c12', .)" />
        </record>
    </records>

    <!-- Other prepared mappings, that are not used in previous profiles, set
    here to remember. It can be completed with previous records.
    Notes:
    - Mappings for included "use" may need to be defined.
    - Functions inside the file used for special rules may need to be updated.
    -->
    <records process="false">
        <!-- The mapping "main all" is prepared to aggregate "Finding Aid" and
        "Archival Description" together. -->
        <record name="Finding Aid">
            <use mapping="Finding Aid + Archival Description" />
        </record>
        <!-- Separated records for the main informations (header, front matter
        and archival description). -->
        <record name="Finding Aid">
            <use mapping="Header" />
        </record>
        <record name="Front Matter">
            <use mapping="Front Matter" />
        </record>
        <record name="Archival Description">
            <use mapping="Archival Description" />
        </record>
        <record name="Description of Subordinate Components">
            <use mapping="Description of Subordinate Components" />
        </record>
        <!-- One related record (is part of) for each digital archival object
        group ("daogrp"). -->
        <record name="Digital Archival Object">
            <use mapping="Digital Archival Object Group" />
        </record>
        <record name="Digital Archival Object Group locations">
            <use mapping="daogrp locations" />
        </record>
        <record name="Digital Archival Object Group references">
            <use mapping="daogrp references" />
        </record>
        <!-- Next "Has part" allow to distinct archival objects (dao) and
        pointers (daoloc...). -->
        <record name="Has part: Digital archival objects as records">
            <use mapping="Has part: Digital archival objects as records" />
        </record>
        <record name="Has part: Digital objects pointers as records">
            <use mapping="Has part: Digital objects pointers as records" />
        </record>
        <record name="Has part: Digital archival objects as links">
            <use mapping="Has part: Digital archival objects as links" />
        </record>
        <record name="Has part: Digital objects pointers as links">
            <use mapping="Has part: Digital objects pointers as links" />
        </record>
        <!-- Partial mapping for the identifier, with the ead id followed by the
        xpath. -->
        <record>
            <use mapping="Identifier: eadid xpath" />
        </record>
        <!-- The simple, unique and absolute xpath to the current node. -->
        <record>
            <use mapping="Identifier: xpath" />
        </record>
    </records>
</config>