Inreasingly experimental setups are made up of independent detector subsystems. The NSCLDAQ event builder is then used to stitch together coincident event fragments into full experiment events.
It is convenient for the developer of each detector subsystem to indpendently build a software ecosystem to support their subsystem. It is equally convenient for the experiment integrator to be able to leverage parts of that ecosystem. Specifically, each detector subsystem may have a well developed set of event processors for SpecTcl, and it would be very good to be able to drop those into a SpecTcl built to analyze the event built data.
The evpunpack command leverages event processing pipline support and internal SpecTcl classes to help do just that. Specifically, you can create event processors programmatically from registered event processing pipelines that are capable of unpacking data from the event builder.
URLs of the following form,
Create a new unpacker for event builder data. The following are all mandatory query parameters:
Name of the new event unpacker. This will be registered with the pipeline manager and, therefore, can participate in pipelines created by that command/REST interface.
The frequency of the event builder timestamp in MHz. This is used to provide an elapsed run time that can be used in creating diagnostic spectra from the parameters the processor automatically creates.
The base name of the set of parameters the processor makes. For each event, the event processor fills in several diagnostic parameters in addition to any the pipelines attached to it produce:
basename.event_noThe event number.
basename.run_timeElapsed run time based on the timestamp and frequency, in seconds.
basename.sourcesNumber of fragments in an event.
basename.unrecognized_sourceNumber of sources in the event for which no event processing pipeline was registered.
basename.n_presentFlag to indicate source n was present.
The intent of these parameters is to provide support for several standard event builder diagnostic spectra.
URLs of the following form:
Registers an event processing pipeline (either programmatically created or created using the pman command/REST interface) to process the event fragments from a specific event source. The data presented to the event processor is the same as if it were seeing data from a system that only took data from that subsystem.
The mandatory query parameter name is the name of the event processor being created. the mandatory query parameter source is the source id of the data that we are teaching it to process and pipe specifies the name of an event processing pipeline that will be invoked to process the data from fragments from the source.
URLs of the following form:
List the names of event processors that have been created via this command. This is returned in the detail attribute of the returned object as an array of strings.