Resolve issues with Deserialization of Untyped Types

[WanjohiSammy]

Issues

This pull request fixes

• It doesn't make sense to resolve untyped integer value to Edm.Decimal #2657

Description

The change ensures that untyped values of various primitive types are correctly deserialized and materialized, supporting a broader range of types.

The previous implementation only handled bool, string, and defaulted all other types to decimal. This was insufficient for untyped properties that could be of unresolved types such as int, long, double, and collections.

This change introduces a new boolean flag PreserveUntypedNumericAsDecimal to enhance control over how untyped numeric values are handled during JSON parsing.

• Default behavior: When set to true, untyped numeric values are preserved as decimal to maintain precision.
• Alternative behavior: When set to false, numeric values are inferred and converted to specific types (int, long, double, float) based on their runtime type.

Checklist (Uncheck if it is not completed)

• Test cases added
• Build and test with one-click build and test script passed

Additional work necessary

If documentation update is needed, please add "Docs Needed" label to the issue and provide details about the required document change in the issue.

Repository notes

Team members can start a CI build by adding a comment with the text /AzurePipelines run to a PR. A bot may respond indicating that there is no pipeline associated with the pull request. This can be ignored if the build is triggered.

Team members should not trigger a build this way for pull requests coming from forked repositories. They should instead trigger the build manually by setting the "branch" to refs/pull/{prId}/merge where {prId} is the ID of the PR.

[WanjohiSammy]

/AzurePipelines run

[azure-pipelines]

No pipelines are associated with this pull request.

[WanjohiSammy]

/AzurePipelines run

[azure-pipelines]

No pipelines are associated with this pull request.

[WanjohiSammy]

/AzurePipelines run

[azure-pipelines]

No pipelines are associated with this pull request.

[WanjohiSammy]

/AzurePipelines run

[azure-pipelines]

No pipelines are associated with this pull request.

[xuzhg]

If it's for 9.x, do we need to keep back compatible?

[gathogojr]

@WanjohiSammy I’d like to highlight two specific comments from the thread on the issue you have referenced that are particularly relevant to this PR:

• It doesn't make sense to resolve untyped integer value to Edm.Decimal #2657 (comment)
• It doesn't make sense to resolve untyped integer value to Edm.Decimal #2657 (comment) — this one reflects the discussion we had during the initial scoping of the work.

Regarding backward compatibility, we typically manage such changes using the LibraryCompatibility enum. You can see how this is structured here:
https://github.com/OData/odata.net/blob/main/src/Microsoft.OData.Core/ODataLibraryCompatibility.cs

To align with this approach, I suggest introducing a new enum member, e.g., ReadUntypedNumericAsDecimal, with a value like 1 << 5. This flag can then be included in the Version6 and Version7 enum members, and you can define Version8 = ReadUntypedNumericAsDecimal. Toggling the flag will preserve the old behaviour in 6.x, 7.x, and 8.x.

Since ODataMessageReaderSettings already exposes a LibraryCompatibility property, this approach wouldn’t require introducing a new setting.

Finally, please refer to how compatibility flags are evaluated and toggled across the library to ensure consistency with existing patterns.

[gathogojr]

Given that this method is internal, and we have updated the call points to pass the relevant value, do we need to set a default parameter value?

[WanjohiSammy]

Resolved

[gathogojr]

Just to confirm, if I create a new instance of ODataMessageReaderSettings, without doing much else, will the new behaviour be applied?

[WanjohiSammy]

Just to confirm, if I create a new instance of ODataMessageReaderSettings, without doing much else, will the new behavior be applied?

Yes, this is correct, the new behavior will be applied.

I have added tests: https://github.com/OData/odata.net/pull/3283/files#diff-503ffc0454f3d70fe4153d243d53eb35d79aaddfc1d9dcd09e742f4aa88d9317

[WanjohiSammy]

/AzurePipelines run

[azure-pipelines]

No pipelines are associated with this pull request.