enabled support for DAQmxSelfCal

[Sov-trotter]

• This contribution adheres to CONTRIBUTING.md.

What does this Pull Request accomplish?

This PR enables the support for DAQmxSelfCal hence enabling self calibration of the supported DAQ devices.

Why should this Pull Request be merged?

Self Calibration is critical functionality for many devices that is encouraged to be run regularly. This also a simple and straight-forward addition that doesn't require tackling the complexity of #22 and #23

What testing has been done?

I have manually tested this on Ni-6133 and Ni-4309 DAQ devices.
cc: @neutronstriker

This is my first time contributing to this project. Please let me know if I have missed something. :)

[bkeryan]

Hi @Sov-trotter, thank you for contributing.

source/codegen/metadata/functions.py is auto-generated by a build process located in an NI internal repo. Someone at NI will need to update that build process to produce the updated SelfCal function definition, or else this change will get reverted the next time we update the API metadata.

[bkeryan]

Will close #22 and #23

This is a step in the right direction but we shouldn't close either issue yet.

#22 includes the external calibration session class and external calibration adjustment functions, which is a sizeable effort.

#23 includes the calibration info properties, which is a smaller effort.

[zhindes]

Hi @Sov-trotter, thank you for contributing.

source/codegen/metadata/functions.py is auto-generated by a build process located in an NI internal repo. Someone at NI will need to update that build process to produce the updated SelfCal function definition, or else this change will get reverted the next time we update the API metadata.

I am willing to take on that effort for this contribution!

[zhindes]

Will close #22 and #23

This is a step in the right direction but we shouldn't close either issue yet.

#22 includes the external calibration session class and external calibration adjustment functions, which is a sizeable effort.

#23 includes the calibration info properties, which is a smaller effort.

I agree. I'll clean up this PR description a bit.

[bkeryan]

Hi @Sov-trotter, thank you for contributing.
source/codegen/metadata/functions.py is auto-generated by a build process located in an NI internal repo. Someone at NI will need to update that build process to produce the updated SelfCal function definition, or else this change will get reverted the next time we update the API metadata.

I am willing to take on that effort for this contribution!

I forgot: we can put customizations in https://github.com/ni/nidaqmx-python/blob/master/src/codegen/metadata/functions_addon.py so that they are not overwritten.

However, updating the build process that generates functions.py is better in the long term.

[Sov-trotter]

Thanks for the feedback @bkeryan. I moved the SelfCal customization to functions_addon.py . This doesn't quite seem to work though.

[flef]

Can you please add support for LibraryInterpreter ? selfCal has the same C prototype than Selftest.

The body of the Selftest function can be almost copy paste.

Regards

[bkeryan]

Can you please add support for LibraryInterpreter ? selfCal has the same C prototype than Selftest.

The body of the Selftest function can be almost copy paste.

The interpreter classes are auto-generated and LibraryInterpreter was supported in the first commit, before @Sov-trotter tried my suggestion of using functions_addon.py (which didn't work and I need to look into).

[Sov-trotter]

Hi @Sov-trotter, thank you for contributing.
source/codegen/metadata/functions.py is auto-generated by a build process located in an NI internal repo. Someone at NI will need to update that build process to produce the updated SelfCal function definition, or else this change will get reverted the next time we update the API metadata.

I am willing to take on that effort for this contribution!

Hey @zhindes @bkeryan let me know if guys are able to update the said build process.

Also it would be really great if you can tag a release as soon as possible. :)

[zhindes]

Hi @Sov-trotter, thank you for contributing.
source/codegen/metadata/functions.py is auto-generated by a build process located in an NI internal repo. Someone at NI will need to update that build process to produce the updated SelfCal function definition, or else this change will get reverted the next time we update the API metadata.

I am willing to take on that effort for this contribution!

Hey @zhindes @bkeryan let me know if guys are able to update the said build process.

Also it would be really great if you can tag a release as soon as possible. :)

This week has been hell, but next week will be much more calm. I'll dig in then, and I also want to make a release soon as well.

[bkeryan]

Thanks for the feedback @bkeryan. I moved the SelfCal customization to functions_addon.py . This doesn't quite seem to work though.

I figured out what's going on with functions_addon.py. In your first commit, you removed 'python_codegen_method': 'no',. However, src/codegen/metadata/__init__.py merges the original function metadata with the addon metadata, so the override in your second commit leaves 'python_codegen_method': 'no', in place.

Adding 'python_codegen_method': 'yes', or 'python_codegen_method': 'None', to the override doesn't entirely work, because some of the codegen assumes that the presence of python_codegen_method means that the function is either hand-written or unsupported.

It's possible to work around this by adding del all_functions["SelfCal"]["python_codegen_method"] to src/codegen/metadata/__init__.py, but it would be better to change src/codegen/utilities/function_helpers.py to do if function_data.get("python_codegen_method") or function_name in EXCLUDED_FUNCTIONS: instead of if "python_codegen_method" in function_data or function_name in EXCLUDED_FUNCTIONS: and set 'python_codegen_method': 'None', in the override.

Since @zhindes is planning to update the metadata next week, this is kind of a moot point, but I wanted to understand why the override didn't work.

[bkeryan]

it would be better to change src/codegen/utilities/function_helpers.py to do if function_data.get("python_codegen_method") or function_name in EXCLUDED_FUNCTIONS: instead of if "python_codegen_method" in function_data or function_name in EXCLUDED_FUNCTIONS:

BTW, I made this change in #462

[zhindes]

Hi @Sov-trotter, thank you for contributing.
source/codegen/metadata/functions.py is auto-generated by a build process located in an NI internal repo. Someone at NI will need to update that build process to produce the updated SelfCal function definition, or else this change will get reverted the next time we update the API metadata.

I am willing to take on that effort for this contribution!

@Sov-trotter - You can update your metadata from users/zhindes/self_cal_metadata. The SelfCal changes look identical to me, but some other metadata was updated in the process that has no functional impact on the generated Python API (it is used by other tooling throughout our stack).

[zhindes]

As mentioned in other places: add a test and update to the latest official metadata

[zhindes]

One more issue (I think), but then we'll be good. Since we're here, can you also add this to the changelog under Merged Pull Requests? A link like this: 460: enabled support for DAQmxSelfCal.

I can make a release ASAP once this is in

[zhindes]

@Sov-trotter - looks great. I'll get a release out ASAP.

[zhindes]

@Sov-trotter - bah, one last issue, I think.

./tests/component/system/test_device.py:33:1: BLK100 Black would make changes.

You can run poetry run black tests/component/system/test_device.py, and it should fix it for you.

[zhindes]

@Sov-trotter - bah, one last issue, I think.

./tests/component/system/test_device.py:33:1: BLK100 Black would make changes.

You can run poetry run black tests/component/system/test_device.py, and it should fix it for you.

We're good now. I'll babysit it from here.

[zhindes]

@Sov-trotter its in! I should be able to make a release this afternoon, 🤞

[zhindes]

@Sov-trotter its in! I should be able to make a release this afternoon, 🤞

0.9.0 is out

[Sov-trotter]

Thanks for this @zhindes @bkeryan

[ClaesFredo]

@zhindes @bkeryan

Just upgraded to 0.9.0 - thx. :)

I find the ai_dev_scaling_coeff but not much else.

Dumb questions?
a) Has the device calibration date been added? (I assume no & yes, self calibration is more important)
b) will the calibration info of #22 be added? It is great info for QA purposes.

[zhindes]

@zhindes @bkeryan

Just upgraded to 0.9.0 - thx. :)

I find the ai_dev_scaling_coeff but not much else.

Dumb questions? a) Has the device calibration date been added? (I assume no & yes, self calibration is more important) b) will the calibration info of #22 be added? It is great info for QA purposes.

a) No, we did not add that. However, nisyscfg (pypi, github) is high quality and should have support for that.

b) We do not have any current plans to add broader calibration support. Reflection of calibration metadata should already be covered by nisyscfg, and I don't think calibration adjustment is a feature we will prioritize highly, likely ever. Self Calibration is a small-impact API that is expected to be run regularly for the health of a DAQ board, so that was something that I prioritized adding based on the high level of interest shown in this PR.

[ClaesFredo]

Hi

Thx for clarifying matters and for the nisycfg info. :)

For what it is worth and, perhaps, as food 4 thought - the device calibration_date and its calibration_due_date is useful as Quality information data, ie to assess the device status.

[alra0351]

Hi,
thank you for the new release: I am trying to use the self_cal() function on the NI 9205, but i am getting an error... whtas could be wrong here?

for device in nidaqmx.system.System.local().devices: if device.product_type == 'NI 9205': device.self_cal()

Output:

raise DaqError(extended_error_info, error_code)
nidaqmx.errors.DaqError: Some or all of the samples requested have not yet been acquired.
To wait for the samples to become available use a longer read timeout or read later in your program. To make the samples available sooner, increase the sample rate. If your task uses a start trigger,  make sure that your start trigger is configured correctly. It is also possible that you configured the task for external timing, and no clock was supplied. If this is the case, supply an external clock.
Property: DAQmx_Read_RelativeTo
Corresponding Value: DAQmx_Val_CurrReadPos
Property: DAQmx_Read_Offset
Corresponding Value: 0

Status Code: -200284

[zhindes]

@alra0351 That error being returned from self_cal is definitely unexpected, and I doubt it is related to the Python API. Can you try hitting the "Self Calibrate" button in MAX? I expect it will return the same error.

[alra0351]

@zhindes thank you for your response. "Self Calibration" from MAX works without any problems. Any further ideas what I can try?
The Python API runs under Linux. MAX runs under Windows. That is the only difference.

[zhindes]

@alra0351 Can you create a new discussion and share the full code?