The XML data import format for process instance-independent measures is preset by the following document type definitions:
DTD pikidata.dtd (referencing of data series as import data)
DTD _tabledata.dtd (table format for importing process instance-independent data):
Thus, import data of process instance-independent measures in XML format have the following general structure:
<?xml version="1.0" encoding="ISO-8859-1"?>
<!DOCTYPE pikidata SYSTEM "pikidata.dtd">
<pikidata>
<pikicube name="...">
<datacols>
<datacol name="..."/>
...
</datacols>
<datarow>
<value relevance="...">...</value>
...
</datarow>
...
</pikicube>
...
</pikidata>
ELEMENT and ATTLIST pikidata |
Description |
---|---|
pikidata |
List of import data on any number of process instance-independent data series (PIKI cube) |
pikicube |
Data series into which the specified data is to be imported |
name |
Data series identifier. Must match the name of the PIKI cube (pikicube name XML attribute) specified in the measure configuration. |
ELEMENT and ATTLIST datacols |
Description |
---|---|
datacols |
Import data structure (table columns of the data series) |
datacol |
Specification of a table column. A column is specified for each process instance-independent measure and each relevant referenced dimension. |
name |
Table column identifier. Must match the names of the process instance-independent measures (pikidef name) specified in the measure configuration and the names of the referenced dimensions (refdim name). |
format |
Optional import format for column values. Supported formats are floating point numbers, time and time of day values. |
datarow |
Specific import data for the data series in the form of a data row. The sequence of import values (value elements) must match the column sequence (datacols) so that each criterion of the data series (process instance-independent measure or referenced dimension) can be assigned a unique value. |
value |
Data row column value to be imported (value of a referenced dimension or a process instance-independent measure) |
relevance |
Relevance value referring to the corresponding value of a process instance-independent measure |
Example 1
<?xml version="1.0" encoding="ISO-8859-1"?>
<!DOCTYPE pikidata SYSTEM "pikidata.dtd">
<pikidata>
<pikicube name="PIKICUBE_COSTS">
<datacols>
<datacol name="OVERHEAD_COSTS"/>
<datacol name="PROCESSTYPE"/>
<datacol name="TIME" format="MMM yyyy"/>
<datacol name="MATERIAL"/>
</datacols>
<datarow>
<value relevance="100">1000 EUR</value>
<value>Order processing\Car industry</value>
<value>June 2000</value>
<value>NaviSet B6{Navigation system}</value>
</datarow>
<datarow>
...
</datarow>
...
</pikicube>
...
</pikidata>
Specification of the import data structure (XML elements datacols and datacol)
The XML-Element datacols specifies the identifiers of the process instance-independent measures and referenced dimensions (datacol name="...") that the data import refers to for the relevant data row. The import data structure is specified in a simple table format in which each criterion of the data series (process instance-independent measure or referenced dimension) represents a table column.
Data import
The actual import values are extracted from the datarow XML elements. Each of these elements represents a data row that is imported. Each specified value element corresponds to the matching datacol element of the specified import data structure.
The sequence of the values (value XML element) in a data row (datarow XML element) must match the sequence of columns (in the datacols XML element) in the import data structure.
In the XML attribute format="MMM yyyy", the data format for the value to be imported for the referenced dimension TIME is specified. Format information is optional.
Specifying a descriptive text in curly brackets { } means that the values for the one-level dimension MATERIAL are expected in the form <Identifier>{<Descriptive text>}. Specifying a descriptive text is optional, as only the identifier is crucial for the dimension reference.
The example data set has a relevance value for the process instance-independent measure OVERHEAD_COSTS. If you do not specify a relevance value, relevance="1" is the default value.
If no unit is specified for a data value, by default the base unit for the attribute data type is used on which the process instance-independent measure or referenced dimension is based.
Example 2
There are two different formats for importing values of two-level dimensions:
The values are separated by a back slash (\).
...
<datacols>
...
<datacol name="PRINCIPAL"/>
...
</datacols>
...
<datarow>
...
<value>DE{Germany}\0000000001{Becker}</value>
...
</datarow>
...
The values are specified in two XML elements that directly follow one another.
...
<datacols>
...
<datacol name="PRINCIPAL"/>
<datacol name="PRINCIPAL"/>
...
</datacols>
<datarow>
...
<value>DE{Germany}</value>
<value>0000000001{Becker}</value>
...
</datarow>
...