Neo testing utilities for neuralynx raw io

[Peter N. Steinmetz]

I am now developing some new classes and programs to help producing the plain_data files needed for testing which make no assumptions about the existing file header structure. 

This has been a sort of catch-22 when people provide new testing data with a slightly different header structure. Those won't load with the existing NeuralynxRawIO because the headers don't work. At the same time, plain testing data helps complete the test suite before the changes to read the header are committed. In the past I have used an old java program I had to produce the plain text dump, but I don't feel like modifying that further for new header structures since I mostly use Neo now. 

My question is whether the main developers think this is appropriate to add to python-neo itself? Presently I have a local package that contains these. If it is appropriate, where in the package structure should they go. They are specific to Neuralynx. 

[Samuel Garcia]

Hi Peter,
I not sure to understand fully what you have in mind and your goal.
You want to make a generator to create some fake files for testing ?
If yes, my feeling is that getting real files (generated by the recording) is better for testing.
Relying on something generated by us is not a proof that the reader code correct.
For write/read round trip this make sens but I am not sure if this is what you have in mind.

Lets discuss this during a call maybe but I am pretty busy for the next 2 weeks. Sorry.

Have a nice day.

Samuel

--
You received this message because you are subscribed to the Google Groups "Neural Ensemble" group.
To unsubscribe from this group and stop receiving emails from it, send an email to neuralensembl...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neuralensemble/d722f9b7-dc77-478f-8346-6b1dbf883529n%40googlegroups.com.