should extract for RheoFreqData take a FreqDataType

[moustachio-belvedere]

At the moment, extract for RheoTimeData types needs a TimeDataType argument which makes sense because you might want to extract time, or time and stress, or time and strain.

However, extract for RheoFreqData types always just extracts the frequency vector as it doesn't make any sense to ever have Gp without Gpp and vice versa.

So when using extract for RheoFreqData, the second argument of FreqDataType is completely redundant.

I can think of a few options:

• We keep behavior as it is

• Remove second argument for RheoFreqData types

• Rename the RheoFreqData method extractfrequency or something more specialised so it is clear from the function name what it does, and remove the redundant second argument.

[akabla]

Hi Louis, I agree that this is odd. We treated the frec data as we did with time data, but they are fundamentally different. We probably still need a function to extract the freq vector as a new RheoFreqData. We may also need a new generator function to create Gp and Gpp from omega. Not sure if this exists. I am happy with renaming the function extractfrequency. Maybe we should then rename the other ones extracttime, extractstrain (t & epsilon) and extractstress (t & sigma).

…

On Sun, 21 Jul 2019 at 17:04, Louis Kaplan ***@***.***> wrote: At the moment, extract for RheoTimeData types needs a TimeDataType argument which makes sense because you might want to extract time, or time and stress, or time and strain. However, extract for RheoFreqData types always just extracts the frequency vector as it doesn't make any sense to ever have Gp without Gpp and vice versa. So the second argument of FreqDataType is completely redundant. I can think of a few options: - We keep behavior as it is, though I think it is a bit strange - Remove second argument for RheoFreqData types, which might also be a little strange - Rename the RheoFreqData method extractfrequency or something more specialised so it is clear from the function name what it does, and remove the redundant second argument. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#23?email_source=notifications&email_token=AA6B24XTZMCVSTBK2WLWH23QASCIVA5CNFSM4IFSY4X2YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HAPZQPQ>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AA6B24Q3WQ4RUR44UR5TRR3QASCIVANCNFSM4IFSY4XQ> .

[akabla]

I'm wondering if it would not be simpler to rename all extract functions for consistency:

extracttime
extractstrain
extractstress
extractfreq

I think it would make the code cleaner and simpler to read for the user, and for us. Most of the code of the extract function is spent working out which parameter to extract.

Note: these functions are currently defined in processing.jl

[moustachio-belvedere]

Sounds good to me!

[akabla]

The functions extract are still there to avoid breaking changes, but now deprecated. To be removed with version 1.0.

[akabla]

Need to include new functions in API