Re: section 2.5 Transfer Encodings of HTTP adaption draft
This is still ugly --
the callout service has no way to know whether the OPES processor
removed encodings (and the Transfer-Encoding header) or did not
remove the encodings (but removed the header), etc.
What problem do you see here?
We defined:
In the absence of explicit transfer-encoding negotiations, an OCP
agent MUST NOT send transfer-encoded application messages.
Informally, this means that the agent or its environment have to make
sure that all transfer encodings are stripped before an application
message enters OCP scope.
So, an OPES processor must assume that there is no encoding.
Of course, it gets a problem if the processor violates this rule.
How to detect? Very hard. But it is the same problem as applying a
Transfer-Encoding in HTTP without setting the header.
Or do you think it is to weak because it does not foresee what will
happen if explicit transfer-encoding negotiation will be introduced
in future?
Shall we introduce a Transfer-Encoding parameter to AMS that MUST be
set if a transfer encoding is (still) applied to the content?
Implementations could be better prepared in this case and can at least
terminate the transaction in this case.
Regards
Martin