FAB Token Driver: Optimizing Identity Deserialization Performance

[AkramBitar]

Problem

Optimize identity desterilization in FAB Token drive

For more information see comment #1410 (comment) in PR: #1410

Solution

Review the current identity desterilization in FAB Token drive and tries to optimize it,(e.g., cache the desterilized identities in case of same identity desterilized more than one time).

[Rama542]

Hi @adecaro, I'm a newcomer looking to contribute to fabric-token-sdk and I'm preparing for the LFX Summer 2026 mentorship program.
I'd like to work on this issue. I've looked at the codebase and found where IdentityType is currently defined as a string, and I've also checked how the token type already uses int32 as the pattern to follow.
Is this issue available for an external contributor to pick up, or is it already being worked on? Happy to start immediately if it's free.

[adecaro]

Hi @Rama542 , I think @AkramBitar already prepared something on this. Not sure.
@AkramBitar ?

[AkramBitar]

Hi @Rama542 , I think @AkramBitar already prepared something on this. Not sure. @AkramBitar ?

Hi, @Rama542 I will be very happy if you can pick this one.
Please let me know if you need any additional clarifications on it.
Good luck.

[Rama542]

Hi @AkramBitar! Thank you so much for the warm welcome and for assigning this to me.
I'll go through the issue carefully and reach out if I need any clarifications.
Really excited to work on this one! 😊

[Rama542]

Hi @adecaro &@AkramBitar! I've opened a PR for this issue.

The fix adds a sync.Map cache to TypedVerifierDeserializerMultiplex so
that repeated calls with the same identity bytes skip deserialization entirely
and return the cached verifier directly.

PR: #<1589>

Happy to make any changes based on your feedback 🙏