Add an io_value type for integers

[Steve Mullerworth (stevemullerworth)]

PR Summary

Sci/Tech Reviewer: allynt
Code Reviewer: Pierre Siddall (@Pierre-siddall)

Currently, support for storing real arrays in the checkpoint dump is provided by the io_value_type and associated support routines in the LFRic-XIOS component. This change adds support for storing and checkpointing integer arrays.

As it changes the existing API, a linked MetOffice/lfric_apps#427 PR must be committed alongside this one.

As XIOS does not have an API for writing integer data, integer values are converted to dp_xios which would normally be a real64 value. This is accurate for all int32 numbers, but would not be should dp_xios ever be real32 (precision is insufficient for integer values greater than or equal to 2 to the power 24). The unit tests were written to check for this possibility.

Note on implementation 1: Currently, io_value_type is hardwired to r_def which can be real64 or real32 depending on the model configuration. Therefore, r_def is explicitly stated in the updated read/write API in the LFRic-XIOS component. Like with integers, there is a conversion to dp_xios which would have no impact on values whether r_def is 32-bit or 64-bit.

Note on implementation 2: The writing/reading of fields of different types/kinds is done through a single generic routine, such as write_field_generic, which includes a select type to differentiate the types/kinds. This setup is possible because the write_interface API in the common field_parent_type relates to the field_parent_proxy rather than the field_parent_type. As the IO value types do not have an equivalent proxy, the analogous write interface must refer to the specific object itself. In other words, the integer and r_def read and write interfaces take the actual integer_io_value_type or io_value_type as arguments, requiring the code to define specific subroutines for reading/writing.

As the read and write APIs should have different intent (the write interface should have intent(in)), the existing abstract interface that was shared has been split into two.

As indicated above, the change also adds some unit tests for the LFRic-XIOS capabilities for both real and integer arrays as previously no io_value testing existed. During this work, it was noted that the test data was not being allocated correctly, so this was tidied.

Code Quality Checklist

• I have performed a self-review of my own code
• My code follows the project's style guidelines
• Comments have been included that aid understanding and enhance the readability of the code
• My changes generate no new warnings
• All automated checks in the CI pipeline have completed successfully

Testing

• I have tested this change locally, using the LFRic Core rose-stem suite
• If required (e.g. API changes) I have also run the LFRic Apps test suite using this branch
• If any tests fail (rose-stem or CI) the reason is understood and acceptable (e.g. kgo changes)
• I have added tests to cover new functionality as appropriate (e.g. system tests, unit tests, etc.)
• Any new tests have been assigned an appropriate amount of compute resource and have been allocated to an appropriate testing group (i.e. the developer tests are for jobs which use a small amount of compute resource and complete in a matter of minutes)

trac.log

Test Suite Results - lfric_core - core_chkpt_seed/run2

Suite Information

Item	Value
Suite Name	core_chkpt_seed/run2
Suite User	steve.mullerworth
Workflow Start	2026-04-10T13:48:02
Groups Run	developer

Dependency	Reference	Main Like
lfric_core	stevemullerworth/lfric_core@checkpoint_seed	False
SimSys_Scripts	MetOffice/SimSys_Scripts@2025.12.1	True

Task Information

✅ succeeded tasks - 382

Security Considerations

• I have reviewed my changes for potential security issues
• Sensitive data is properly handled (if applicable)
• Authentication and authorisation are properly implemented (if applicable)

Performance Impact

• Performance of the code has been considered and, if applicable, suitable performance measurements have been conducted

AI Assistance and Attribution

• Some of the content of this change has been produced with the assistance of Generative AI tool name (e.g., Met Office Github Copilot Enterprise, Github Copilot Personal, ChatGPT GPT-4, etc) and I have followed the Simulation Systems AI policy (including attribution labels)

Documentation

• Where appropriate I have updated documentation related to this change and confirmed that it builds correctly

PSyclone Approval

• If you have edited any PSyclone-related code (e.g. PSyKAl-lite, Kernel interface, optimisation scripts, LFRic data structure code) then please contact the TCD Team

Sci/Tech Review

• I understand this area of code and the changes being added
• The proposed changes correspond to the pull request description
• Documentation is sufficient (do documentation papers need updating)
• Sufficient testing has been completed

(Please alert the code reviewer via a tag when you have approved the SR)

Code Review

• All dependencies have been resolved
• Related Issues have been properly linked and addressed
• CLA compliance has been confirmed
• Code quality standards have been met
• Tests are adequate and have passed
• Documentation is complete and accurate
• Security considerations have been addressed
• Performance impact is acceptable

[allynt]

This PR correctly extends io_value_type to store integers as well as reals (r_def). It does this by creating a new type just for integers, integer_io_value_type, and then adding interfaces where distinguishing between reals and integers is required.

LGTM.

[Pierre Siddall (Pierre-siddall)]

Thanks Steve Mullerworth (@stevemullerworth), this mostly looks ready to head into testing but I have one small suggestion regarding a code block which is repeated throughout. I'm hoping I've understood the technical aspect of this correctly, but let me know if I have missed context.

[Pierre Siddall (Pierre-siddall)]

Given that this is a repetition of the XIOS real conversion above, would it make more sense to move this into a small helper subroutine to improve readability and future maintenance ?

[Steve Mullerworth (stevemullerworth)]

The io_value object is an abstract type. The select type disambiguates between either io_value_type or integer_io_value_type so that within the first block io_value behaves as an io_value_type but within the second it behaves as an integer_io_value_type. So two helper subroutines would be needed to satisfy the two types!

[Pierre Siddall (Pierre-siddall)]

Thanks Steve Mullerworth (@stevemullerworth), this is super helpful, I may need to pick your brains some more to understand this further but for now I'm happy for this to head into testing and then onto main :) .

[Pierre Siddall (Pierre-siddall)]

Similarly to the last comment the moving of this block to a subroutine would benefit the code here.

[Steve Mullerworth (stevemullerworth)]

Same reply as above

[Pierre Siddall (Pierre-siddall)]

Replacement also needs to happen here.

[Pierre Siddall (Pierre-siddall)]

New context about the repeated code blocks has now been clarified by the developer, and therefore I'm happy to put this into testing.