Write attributes paths are not validated unlike read

[fessehaeve]

Problem

./chip-tool any read-by-id 0x11110001 0x1111f001 1 1 fails with INVALID_ACTION but
./chip-tool any write-by-id 0x11110001 0x1111f001 "abc" 1 1 succeeds.

This is concerning the valid range for cluster id and attribute id.

Proposed Solution

Do we need to have this check?

[mrjerryjohns]

On master as of today, the write does not actually succeed - it returns back an UNSUPPORTED_CLUSTER error.

This however isn't presumably the right error. The WriteHandler logic today does not actually validate that the Attribute and Cluster IDs are indeed well formed. This should be added.

[CamWms]

Working on PR now. Should the action abort, or generate a new error: INVALID_PATH?
For now the IM aborts, just as it would for a illegal wildcards, group or missing elements in the path.
https://github.com/CHIP-Specifications/connectedhomeip-spec/pull/5425

[mrjerryjohns]

@bzbarsky-apple I'm going to recommend we not fix this for v1.0, and defer this, since we don't actually have an issue with spec compliance.

[CamWms]

I have a PR #5425 to check if paths are well-formed, but IDs should not be checked for in-range by the SDK code except if the range is within the MS range. The valid range is to direct new ID allocation, not for operational checks. We may expand or reduce the range in the future. If the SDK does not recognize an ID (in or outside the current range), then that is enough.

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs.

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs.

[stale]

This stale issue has been automatically closed. Thank you for your contributions.

[bzbarsky-apple]

Fixed by #29461