You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
After playing around with creating dynamic classes and monkeypatching them, I think the way forward is to add a processing pipeline to DataDevice. The device will do it's own processing first, then push the data through each stage in the pipeline before returning.
In the first instance, the pipeline can just be a list of functions that take one argument, data, and return that argument. Any object that wants to add to the pipeline can do so by adding a function to the list.
There is still the issue of the clarity_processor modules
(https://github.com/pptman/clarity_processor). Because they are not
packaged it is difficult to use them. I see several options going
forward:
Create a PR to add setup.py and whatever else is necessary to make
it a proper package.
Create a fork in MicronOxford and modify it however we want.
Add a new config option for the Clarity device that specifies a
path to clarity_process.py, then load the module dynamically with
importlib.util.
Don't do anything. Leave it to the the user to setup the
clarity_process module.
Do you have any preferences? I like the first two options the best and
I am happy to talk to Mark Neil about it.
Can we please solve this issue and get the working version of the clarity processing into the main branch.
Activity
mickp commentedon Nov 27, 2019
Work in progress
iandobbie commentedon Nov 6, 2020
Danny worked on integrating this properly on the engineering system. His current version is at https://github.com/dstoychev/microscope/tree/aurox
There is however an outstanding issue:
Can we please solve this issue and get the working version of the clarity processing into the main branch.