discuss: simplify function names?

[stevevanhooser]

readchannels_epochsamples => readsamples read_epochevents => readevents?

or just read('samples', ...) read('events',...)

For discussion...

[stevevanhooser]

Look at all of the functions in ndr.reader

and see if there are simpler names

[Sophie-Greer]

Here are the new function names I propose:

base:

• ndr_reader_base_obj --> base
• epochclock --> clock
• t0_t1 --> t0_t1_base
• getchannelsepoch --> channels_base
• readchannels_epochsamples --> readsamples_base
• readevents_epochsamples --> readsamples_base
• readevents_epochsamples_native --> readnative_samples_base
• samplerate --> samplerate_base
• mfdaq_channeltypes --> mfdaq_channels
• mfdaq_prefix can stay the same
• mfdaq_type can stay the same

ced_smr

• ced_smr --> ced_smr_new
• getchannelsepoch --> channels_cedsmr
• readchannels_epochsamples --> readsamples_ced
• t0_t1 --> t0_t1_ced
• readevents_epochsamples_native --> readnative_samples_ced
• samplerate --> samplerate_cedsmr
• cedsmrfile --> identify_smr
• cedsmrheader2reacherchanneltype --> intan_to_reader

intan_rhd

• intan_rhd_obj --> intan_new
• t0_t1 --> t0_t1_intan
• getchannelsepoch --> channels_intan
• readevents_epochsamples --> readsamples_intan
• samplerate --> samplerate_intan
• filenamefromepochfiles --> filename_intan
• mfdaqchanneltype2intanheadertype --> mfdaq_to_intan
• intanheadertype2mfdaqchanneltype --> intan_to_mfdaq
• mfdaqchanneltype2intanchanneltype --> mfdaq_to_intan_channels
• intanname2mfdaqname --> intan_to_mfdaq_channels
• mfdaqchanneltype2intanfreqheader --> header_freq

somecompany_someformat

• somecompany_someformat --> new_generic_base

spikegadgets_rec

• spikegadgets_rec --> spike_new
• getchannelsepoch --> channels_spike
• samplerate --> samplerate_spike
• t0_t1 --> t0_t1_spike
• getepochprobemap --> probe
• readchannels_epochsamples --> readsamples_spike
• filenamefromepochfiles --> filename

[Sophie-Greer]

Ok, I've taken another look at this and here are the functions that should keep their names across all classes (thanks, Steve, for the help): t0_t1 getchannelsepoch readchannels_epochsamples readevents_epochsamples readevents_epochsamples_native samplerate filenamefromepochfiles

[stevevanhooser]

read(channelType, channelNames_Numbers, 't0', -Inf, 't1', Inf,'useTime',1 ) % %. Step 1: decode the channel names and channel types %. Step 2: if it is something that needs to be read with readchannels_epochsamples, then it calls that %. if it is something that needs to be read with readevents_epochsamples, then it calls that

I will write a function prototype, ask NDR developers to write the channel decoders for their own reader

[lakesare]

Method names have been pretty confusing for me too. When I was implementing their Python analogues I initially renamed them to more understandable names (and then went back to NDR names for consistency).

E.g.:

ndr_reader_base_obj --> self // It can be short, people know what `self`/`obj` means when it goes as the first argument in a method
t0_t1 --> get_t0_t1_of_epoch()
getchannelsepoch --> get_channels_of_epoch()
readchannels_epochsamples --> read_samples()
readevents_epochsamples --> read_events()
samplerate --> get_sample_rate()

daqchannels2internalchannels was also confusing, it's not immediately clear what's a daq channel name and what's an internal channel name, especially since some readers don't directly follow the distinction. convert_channels_from_device_to_ndr_format()?

I would probably also rename some arguments:

channelnumber --> channels // And require it to consistently be a cell array?
epochfiles --> files // Sometimes there is more than one epoch in a file

Not urgent, just a couple of thoughts.