The join
accessor is useful when developing in DPML as it allows you get
the results from asynchronously issue requests using the
async accessor.
If you are developing an accessor or using a scriping
language you may find it easier to use the underlying asynchronous request
methods in the NKF API.
If an exception is thrown in the subrequest then the join will propagate
that exception.
Conversely, if you do not join, an exception or any other result
from an asynchronous request will be lost.
Example
Here is an example showing asynchronous execution of a DPML sub-process with
join used to obtain the result of the sub-process.
<idoc>
<seq>
<comment>Fork off a DPML process</comment>
<instr>
<type>async</type>
<uri>active:dpml</uri>
<operand>another-script.idoc</operand>
<param>this:param:param</param>
<target>var:handle</target>
</instr>
...
...
Do some local processing in parallel
...
...
<instr>
<type>join</type>
<operand>var:handle</operand>
<target>var:result-from-asynch-fork</target>
</instr>
...
...
Do some more processing
...
...
</seq>
</idoc>
Preface
Accessor Listing
ASpell Spell Checker
