Key-based merge

The key-based merge is used to merge events with identical merge keys. The first merge event imported is deleted and the number of identical merge keys is reduced. The merge process is repeated until no more identical merge keys are found within the current process instance.

Merge key rules are defined in the XML file KeyRules.xml. This is done by specifying the attributes of a fragment event involved in the creation of the merge key. The merge key itself is created by combining the specified attribute values.

Example

The file extract below defines a merge key rule, which uses the internal object name of the event (AT_OBJNAME_INTERN attribute type) to create the merge key for the start and end event in a process instance fragment. As several fragments with the same fragment definition can occur in a process instance, the merge key is extended to include the value of the AT_END_TIME attribute.

<?xml version="1.0" encoding="ISO-8859-1"?>

<!DOCTYPE keyrules SYSTEM "keyrules.dtd">

<keyrules>

...

<processkeyrule>

...

</processkeyrule>

...

<mergekeyrule>

<refobjects>

<refobject objecttype="OT_EVT">

<objectname name="EVT_START"/>

<objectname name="EVT_END"/>

</refobject>

</refobjects>

<keyparts>

<keypart attributetype="AT_OBJNAME_INTERN"/>

<keypart attributetype="AT_END_TIME"/>

</keyparts>

</mergekeyrule>

...

</keyrules>