4.3.2. Configuration

The PICA+ configuration takes place within the enclosing <PicaPlus> element, which is directly subordinate to the <Formats> element. Within the <PicaPlus> element, the type-specific elements <Person>, <Metadata> and <DocStruct> are used, each of which maps an internal data type. As with the RDF/XML format, a 1:1 mapping is always performed here.

The characteristics of the PICA+ structure are used for the mapping. Within these type-specific elements, the following sub-elements are permitted:

Tag	Description
<picaMainTag>	The PICA+ field number. This element must exist exactly once.
<picaSubTag>	The PICA+ field separator within the field specified with <picaMainTag>. The content of this subfield corresponds to the value of the metadata specified in the <name> element. Within the <Metadata> and <DocStruct> element, this field must exist exactly once. Different rules apply to the <Person> element (see below).
<valueCondition>	Conditional mapping: By specifying a regular expression (Perl5 syntax), a condition can be specified here. Only if this condition applies to the content of the PICA+ field defined above, the value is assigned to the internal metadata. This field is optional, see example below.
<name>	The internal metadata name. This element must also exist exactly once.
<valueRegExp>	Subsequent editing of PICA+ field values: By specifying a regular expression (Perl5 syntax), the value of the metadata can be manipulated. This field is optional, see example below.

Example: Mapping of PICA+ fields

<Metadata>
      <picaMainTag>021A</picaMainTag>
      <picaSubTag>a</picaSubTag>
      <Name>TitleDocMain</Name>
</Metadata>

Example: Use of conditional assignments

<Metadata>
    <picaMainTag>007S</picaMainTag>
    <picaSubTag>0</picaSubTag>
    <ValueCondition>/^VD17/</ValueCondition>
    <Name>CatalogFieldVDseventeen</Name>
    <ValueRegExp>s/^VD17\s(.*)/$1/</ValueRegExp>
</Metadata>
<Metadata>
    <picaMainTag>007S</picaMainTag>
    <picaSubTag>0</picaSubTag>
    <ValueCondition>/^VD18/</ValueCondition>
    <Name>CatalogFieldVDeighteen</Name>
    <ValueRegExp>s/^VD18\s(.*)/$1/</ValueRegExp>
</Metadata>

The value of picaMainTag 007s and picaSubTag 0 is assigned to the internal metadata CatalogFieldVDseventeen only if its content starts with DVD17. If the content of the PICA+ field starts with Vd18, it will be assigned to the internal metadata CatalogFieldVDeighteen.

If one of the conditions is met, the VD17 or the VD18 in front of the actual VD17 or VD18 number (including the space) is also removed.

Example: Manipulation of metadata during import

<Metadata>
    <picaMainTag>003@</picaMainTag>
    <picaSubTag>0</picaSubTag>
    <Name>CatalogIDDigital</Name>
    <ValueRegExp>s/(.*)/PPN$1/</ValueRegExp>
</Metadata>

This regular expression adds the string PN to the value of PICA+ (PPN).

Since persons have other characteristics and are not just a simple type-value pair, a <picaSubTag> can exist within the <person> element for each of these characteristics. The corresponding attribute type specifies the corresponding characteristic. The following values are valid for this attribute:

Attribut	Description
firstname	First name of the person
lastname	Last name of the person
identifier	Identifier of the person (e.g. from an authority file)
expansion	First and last name can also be extracted from the pica field "Expansion of the attachment form".

Example: Mapping of persons

<Person>
<picaMainTag>028A</picaMainTag> <Name>Author</Name>
<picaSubTag type="firstname">d</picaSubTag> <picaSubTag type="lastname">a</picaSubTag> <picaSubTag type="identifier">9</picaSubTag> <picaSubTag type="expansion">8</picaSubTag>
</Person>

Corporations also consist of several fields. They are defined within a <Corporate> element. Here, similar to the persons, the type attribute for <picaSubTag> is available to define into which field a value is to be imported. The following values are possible:

Attribute	Description
mainName	This field contains the code of the <subfield> element from which the main name of the corporation is to be imported. The field is repeatable to allow different code to be specified. If more than one field is found, the first value will be imported.
subName	This field contains the code from which further name details of the corporation are to be imported. The field is repeatable to allow different code to be specified. If more than one field is found, each value is imported separately.
partName	This field contains the code from which to import censuses, locations or dates for the corporation. The field is repeatable to allow different codes to be specified. If more than one field is found, all values are imported into a common field, the individual entries are separated by the value configured in <separator>.
identifier	This field contains the code of the <subfield> element in which a standard data identifier of the entity has been defined.

Beispiel: Mapping von Körperschaften

<Corporate>
    <Name>Corporation</Name>
    <picaMainTag>029F</picaMainTag>
    <picaSubTag type="mainName">a</picaSubTag>
    <picaSubTag type="subName">b</picaSubTag>
    <picaSubTag type="partName">c</picaSubTag>
    <picaSubTag type="partName">d</picaSubTag>
    <picaSubTag type="partName">n</picaSubTag>
    <picaSubTag type="identifier">7</picaSubTag>
</Corporate>

Since the type of the respective document structure is dependent on the value of a field, there is another mandatory element within the element <DocStrct>. Only if the element specified with <picaMainTag> and <picaSubTag> has the content defined in <picaContent>, a corresponding document structure is created and the metadata is assigned to it. The type of the document structure is specified in the element <name>.

This can also result in an n:1 mapping, i.e. several types from the pica system can be assigned to an internal type. This is necessary because the letter combination of the PICA+ format contains not only the bibliographic type but also the appearance (printed, digital, microform). Accordingly, a mapping must be made for each letter combination.

Example: Mapping of a MultiVolume work

<DocStruct>
      <picaMainTag>002@</picaMainTag>
      <picaSubTag>0</picaSubTag>
      <picaContent>Oc</picaContent>
      <Name>MultivolumeWork</Name>
</DocStruct>
<DocStruct>
      <picaMainTag>002@</picaMainTag>
      <picaSubTag>0</picaSubTag>
      <picaContent>Ac</picaContent>
      <Name>MultivolumeWork</Name>
</DocStruct>

For the mapping of document structures, it should be noted that only the first letters defined in the rule set are compared. This is virtually equivalent to a truncated search. Since in the example above only two letters are specified in the field <picaContent>, only the first two letters in the PICA+ file are taken into account. The comparison is also case-sensitive.