Closed spaulaus closed 6 years ago
(1) I was wanting to suggest you provide some "standard" templates as examples for users, such as TemplateExpProcessor. I used your initial Pr270Processor in your pr270 branch as template for my adapted version. Good place to start. (2) Users should still have option to change their ROOT TTree, e.g. rename the tree, add branches etc. as I'm doing in the experiment processor.
The name of the tree will be defined by the user, ditto for branches. RootHandler will only keep track of that information and ensure that duplicates are not entered.
This turns out to be much more complicated than initially thought. You can find the full details in this thread : https://root-forum.cern.ch/t/disable-ttree-autosave-when-calling-ttree-fill-to-write-tree-in-parallel/28337
I'm going to investigate a workaround here. If we define two TFiles one for histograms and one for trees, we may be able to circumvent this issue.
Expected Behavior
RootHandler should handle Trees for us as well as histograms. This prevents users from needing specialized root files and writing that may overlap with the "official one"
Current Behavior
The guts for TTrees exists, but isn't used.
Context
We have having an issue in @jjvz's code where we're unable to view the online histograms due to the tree constantly writing out of sync with the histogramming.
Possible Solution (optional)
Acceptance Criteria
TwoChanTimingProcessor
to provide users concrete exampleTemplateExpProcessor
to provide users concrete example