Pierre-Alain Loizeau authored 1 year ago and Volker Friese committed 1 year ago

DISCLAIMERS:
1) emission happens on every iteration of the main loop over TS!
=> No memory between iterations to accumulate statistics
=> No limitation (time interval or nb iterations interval) on emission, lots of small TS will lead to lots of messages emission
2) ZMQ addressing hardcoded in header, would need to be added to binary arguments and maybe also to YAML config!

Pierre-Alain Loizeau authored 1 year ago and Volker Friese committed 1 year ago

- Add to DigiEventQaConfig a static method to get the name of digi time histos (avoid copies w/ risk of divergence)
- Add in DigiEventQaConfig a method to get Config strings for all histos (ensure same output folder)
- Add in DigiEventQaConfig a method to get Config strings for all canvases + prepare it for Digi Time in Event canvas
- Change storage for Histo1D in DigiEventQaData to a vector + modify map to provide reference to entry in it (preparation for addition of other histograms)
- Add GetConfig accessor to DigiEventQa

Pierre-Alain Loizeau authored 1 year ago and Volker Friese committed 1 year ago

- Screenshot
- Protocol description
- example commands
- roadmap update

Pierre-Alain Loizeau authored 1 year ago and Volker Friese committed 1 year ago

Runs and pop the server but:
- Input interface not defined! Something need to be invented to replace FairMQ
- Commands do not work as the static boolean accessed by CLING do not seem to be share to the main threads (worked in FairTask world)
- Due to this, no mechanism to stop the main loop apart from Ctrl+C
- Not tested with histos, neither with canvases

Pierre-Alain Loizeau authored 1 year ago and Volker Friese committed 1 year ago

=> protect temporary histogram in CbmTaskDigiEventQa::ToTH1D from ROOT name registration clash

Pierre-Alain Loizeau authored 1 year ago and Volker Friese committed 1 year ago

- Also fix boundary bug in Histo1D content access

* Add `--dump-setup` to dump unpacker setups to file.
* Add TRD and TRD2D setup files converted to YAML
* Enable TRD and TRD2D Unpacker in cbmreco
* Make TRD and TRD2D Unpacker parallel

- Small refactoring
- Add support for std::map
- Add some unittest coverage
- Add CBM_FORMAT to specify how structs should formatted in yaml

CA [Bugfix]: Turing the old performance off, if the point or hit match branches are not presented in the cbmsim tree