Replace custom validators with attrs.validators

[RHammond2]

Replace custom validators with attrs.validators

This PR replaces the use of custom validators contained within h2integrate/core/validators.py with functionality provided by attrs.validators to reduce code maintenance by taking advantage of existing solutions.

The only validator that was not replaced was not_equal. There are two potential approaches we could take to ensure a complete replacement:

1. Use a single-valued in_ validator:

   val = field(default=2022, validator=validators.in_([2022]))

   UPDATE: this approach was applied.

2. Wait for this PR to be merged and released to allow for:

val = field(default=2022, validator=validators.eq(2022))

I would personally opt for option 2 as it doesn't allow for the addition of newly added values in the future, maintaining the spirit and implementation of the must_equal. attrs prefers the use of in_ to cover equality.

Section 1: Type of Contribution

• Feature Enhancement
  • Framework
  • New Model
  • Updated Model
  • Tools/Utilities
  • Other (please describe):
• Bug Fix
• Documentation Update
• CI Changes
• Other (please describe): Reducing duplicate/workaround code with existing solutions.

Section 2: Draft PR Checklist

N/A

Section 3: General PR Checklist

• PR description thoroughly describes the new feature, bug fix, etc.
• Added tests for new functionality or bug fixes
• Tests pass (If not, and this is expected, please elaborate in the Section 6: Test Results)
• Documentation
  • Docstrings are up-to-date
  • Related docs/ files are up-to-date, or added when necessary
  • Documentation has been rebuilt successfully
  • Examples have been updated (if applicable)
• CHANGELOG.md
  • At least one complete sentence has been provided to describe the changes made in this PR
  • After the above, a hyperlink has been provided to the PR using the following format:
    "A complete thought. [PR XYZ]((https://github.com/NatLabRockies/H2Integrate/pull/XYZ)", where
    XYZ should be replaced with the actual number.

Section 4: Related Issues

Section 5: Impacted Areas of the Software

Section 5.1: New Files

N/A

Section 5.2: Modified Files

Too many files have been modified to reasonably list them all, so the replacement validators will be described below.

must_equal(val)

value = field(validator=validators.in_([val]))

gt_zero

num = field(validator=validators.gt(0))

ge_zero

num = field(validator=validators.ge(0))

range_val(min_val, max_val)

num = field(validator=(validators.ge(min_val), validators.le(max_val)))

range_val_or_none(min_val, max_val)

num = field(validator=validators.optional((validators.ge(min_val), validators.le(max_val))))

contains([val1, val2])

val = field(validator=validators.in_((val1, val2))

has_required_keys([key1, key2])

This validator's replacement is a little more involved. After discussion with @elenya-grant, it appears the checks in the original were incomplete, so we have to do a slightly roundabout validation to ensure only key1 and key2 are keys in the dictionary as was originally intended for the validator. First we use the key_validator to validate the key names are valid. Then we use the mapping_validator to enforce an equally sized min and max length of the of the dictionary to ensure we have a dictionary with exactly key1 and key2 keys.

dct = field(
    validator=validators.deep_mapping(
        key_validator=validators.in_([key1, key2]),
        mapping_validator=(
            validators.instance_of(dict), validators.min_len(2), validators.max_len(2)
        ),
    )
)

Section 6: Additional Supporting Information

Section 7: Test Results, if applicable

Tests pass.

Section 8 (Optional): New Model Checklist

N/A

[johnjasa]

Thanks for these changes, Rob! I support this effort.

My only initial hang-up is how verbose some of the new lines become, especially the dict ones, e.g.:

    peak_range: dict = field(
        validator=validators.deep_mapping(
            key_validator=validators.in_(["start", "end"]),
            mapping_validator=(
                validators.instance_of(dict),
                validators.min_len(2),
                validators.max_len(2),
            ),
        )
    )

That being said, being more strictly correct (I saw your note in the PR that we weren't fully checking everything for those cases) and also not wrapping our own validators seems worthwhile.

I like that you deleted all the now-unused functions in validators.py! Good to see.

[RHammond2]

Thanks for these changes, Rob! I support this effort.

My only initial hang-up is how verbose some of the new lines become, especially the dict ones, e.g.:

    peak_range: dict = field(
        validator=validators.deep_mapping(
            key_validator=validators.in_(["start", "end"]),
            mapping_validator=(
                validators.instance_of(dict),
                validators.min_len(2),
                validators.max_len(2),
            ),
        )
    )

That being said, being more strictly correct (I saw your note in the PR that we weren't fully checking everything for those cases) and also not wrapping our own validators seems worthwhile.

I like that you deleted all the now-unused functions in validators.py! Good to see.

This was the only validator I could see keeping, but only if we corrected it to apply the correct checks. This is also more of a use case for having a dataclass object represent the dictionary, so I would probably be in favor of folks taking the following approach for more involved checks as it's far cleaner.

@define
class PeakRange(BaseConfig):
    start: int = field(validator=validators.instance_of(int))
    end: int = field(validator=validators.instance_of(int))

@define
class PeakLoadManagementHeuristicOpenLoopStorageControllerConfig:
    demand_profile_upstream: int | float | list | None = field()
    peak_range: PeakRange | dict = field(converter=PeakRange.from_dict)

[johnjasa]

Thanks for these changes, Rob! I support this effort.
My only initial hang-up is how verbose some of the new lines become, especially the dict ones, e.g.:

    peak_range: dict = field(
        validator=validators.deep_mapping(
            key_validator=validators.in_(["start", "end"]),
            mapping_validator=(
                validators.instance_of(dict),
                validators.min_len(2),
                validators.max_len(2),
            ),
        )
    )

That being said, being more strictly correct (I saw your note in the PR that we weren't fully checking everything for those cases) and also not wrapping our own validators seems worthwhile.
I like that you deleted all the now-unused functions in validators.py! Good to see.

This was the only validator I could see keeping, but only if we corrected it to apply the correct checks. This is also more of a use case for having a dataclass object represent the dictionary, so I would probably be in favor of folks taking the following approach for more involved checks as it's far cleaner.

@define
class PeakRange(BaseConfig):
    start: int = field(validator=validators.instance_of(int))
    end: int = field(validator=validators.instance_of(int))

@define
class PeakLoadManagementHeuristicOpenLoopStorageControllerConfig:
    demand_profile_upstream: int | float | list | None = field()
    peak_range: PeakRange | dict = field(converter=PeakRange.from_dict)

Thanks for this suggestion, Rob! I dig it. Sometimes it's tough to know when to use a certain object, but this validation is so much cleaner looking.