process-tpstream.md

Processing TP Streams

process_tpstream.cxx (and the application trgtools_process_tpstream) processes a timeslice HDF5 that contains TriggerPrimitives and creates a new HDF5 that also includes TriggerActivities and TriggerCandidates. The primary use of this is to test TA algorithms, TC algorithms, and their configurations, with output diagnostics available from ta_dump.py and tc_dump.py. The application also outputs the latencies per-tp for TA emulation, and per-TA for TC emulation, into a CSV file, if --latencies option is added. The CSV is the format: row as a TP (TA) for TA(TC) emulation, with time_start, adc_integral, processing time, and whether it was TP that closed TA (1) or not (0).

Example

trgtools_process_tpstream -i input_file.hdf5 -o output_file.hdf5 -j algo_config.json -m VDColdboxChannelMap --quiet

trgtools_process_tpstream --latencies -i input_file.hdf5 -o output_file.hdf5 -j algo_config.json

In the second case, the default map will be VDColdboxChannelMap. The DUNE-DAQ/detchannelmaps repository has a table of available channel maps.

Configuration

The algo_config.json configuration mirrors the format that is used in daqconf. An example is shown below.

{
	"trigger_activity_config": [
	    {
		"adc_threshold": 10000,
		"adj_tolerance": 4,
		"adjacency_threshold": 6,
		"n_channels_threshold": 8,
		"prescale": 100,
		"print_tp_info": false,
		"trigger_on_adc": false,
		"trigger_on_adjacency": true,
		"trigger_on_n_channels": false,
		"window_length": 10000,
		"max_time_over_threshold": 10000
	    }
	],
	"trigger_activity_plugin": [
	    "TAMakerPrescaleAlgorithm"
	],
	"trigger_candidate_config": [
	    {
		"adc_threshold": 10000,
		"adj_tolerance": 4,
		"adjacency_threshold": 6,
		"n_channels_threshold": 8,
		"prescale": 100,
		"print_tp_info": false,
		"trigger_on_adc": false,
		"trigger_on_adjacency": true,
		"trigger_on_n_channels": false,
		"window_length": 10000
	    }
	],
	"trigger_candidate_plugin": [
	    "TCMakerPrescaleAlgorithm"
	]
}

When developing new algorithms, it is sufficient to change the plugin name and insert the new configurable parameters.