Remove HTTP_PUSH_ALREADY_IN_CACHE error #2813
Conversation
draft-ietf-quic-http.md
Outdated
| {{errors}}). This asks the server not to transfer additional data and indicates | ||
| that it will be discarded upon receipt. | ||
| CANCEL_PUSH frame, a QUIC STOP_SENDING frame with an error code of | ||
| HTTP_PUSH_REFUSED. This asks the server not to transfer additional data and |
There was a problem hiding this comment.
This sentence is missing a verb.
|
My intention was to keep PRs somewhat atomic and fragment the error code space in each PR that separately removes or consolidates an error code. Once complete, I would then defrag it on a PR that let's people weigh in on which error code should have which number. (It also simplifies dependencies which having branches off branches) Happy to do something different if you want. |
| @@ -785,10 +785,9 @@ amount of data a server may commit to the pushed stream. | |||
|
|
|||
| If a promised server push is not needed by the client, the client SHOULD send a | |||
| CANCEL_PUSH frame. If the push stream is already open or opens after sending the | |||
| CANCEL_PUSH frame, a QUIC STOP_SENDING frame with an appropriate error code can | |||
There was a problem hiding this comment.
To me the subject and the verb being omitted in this sentence was a bit surprising, but that might be due to me not being a native speaker.
There was a problem hiding this comment.
It is an error, thanks for picking it up. I would convert this to
"a QUIC STOP_SENDING frame with an error code of HTTP_REQUEST_CANCELLED can be used."
Alternatively, elsewhere we talk about "triggering QUIC STOP_SENDING" which is under review as part of @MikeBishop #2805
| @@ -1903,9 +1901,9 @@ The entries in the following table are registered by this document. | |||
| | ----------------------------------- | ---------- | ---------------------------------------- | ---------------------- | | |||
| | HTTP_NO_ERROR | 0x0000 | No error | {{http-error-codes}} | | |||
| | HTTP_WRONG_SETTING_DIRECTION | 0x0001 | Setting sent in wrong direction | {{http-error-codes}} | | |||
| | HTTP_PUSH_REFUSED | 0x0002 | Client refused pushed content | {{http-error-codes}} | | |||
| | Reserved | 0x0002 | N/A | N/A | | |||
There was a problem hiding this comment.
My sense is that we want to do a final cleanup for reserved error codes. We might reserve some for keeping the relationship with RFC 7540 clean, but I don't think that a renumbering is a bad idea after all this work on fixing errors completes.
There was a problem hiding this comment.
Definitely. Not many good reasons exist to reserve codes for the shipped RFC.
Fixes #2812