Hi all,

According to is a git issue raised here [1]
<https://github.com/wso2/product-is/issues/4352>, x5t generated manually
and the one we send in the id_token are different.

When I debug the identity-inbound-oauth component, Thumbprint generated
manually and generated in the id_token are the same.
Thumbprint generated manually : 501fc1432d87155dc431382aeb843ed558ad61b1
Thumbprint generated in id_token : 501fc1432d87155dc431382aeb843ed558ad61b1

But the issue is in base64 encoding. x5t generated manually after base64
encoding is different form the x5t generated in id_token.
x5t generated manually :
*NTAxZmMxNDMyZDg3MTU1ZGM0MzEzODJhZWI4NDNlZDU1OGFkNjFiMQ==*
x5t generated in the id_token :
*NTAxZmMxNDMyZDg3MTU1ZGM0MzEzODJhZWI4NDNlZDU1OGFkNjFiMQ*

In identity-inbound-oauth, We use org.apache.commons.codec.binary.Base64
library for base64 encoding during x5t generation. In the documentation of
this library, it is mentioned that  *no padding is added when using the
URL-safe alphabet *[2]
<https://commons.apache.org/proper/commons-codec/apidocs/org/apache/commons/codec/binary/Base64.html>*.
*In id_token building process, we pass urlSafe = true. So this will remove
the trailing "=" during base64 encoding.

So is this a proper way to remove trailing "=" during base64 encoding in
x5t generation?

[1]https://github.com/wso2/product-is/issues/4352
[2]
https://commons.apache.org/proper/commons-codec/apidocs/org/apache/commons/codec/binary/Base64.html

Thanks and Regards,
Piraveena
*Piraveena Paralogarajah*
Software Engineer | WSO2 Inc.
*(m)* +94776099594 | *(e)* [email protected]
_______________________________________________
Dev mailing list
[email protected]
http://wso2.org/cgi-bin/mailman/listinfo/dev

Reply via email to