Skip to content

Conversation

@mnot
Copy link
Member

@mnot mnot commented Feb 24, 2021

Fixes #683.

@mnot mnot requested review from reschke and royfielding February 24, 2021 04:21
Copy link
Contributor

@reschke reschke left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I though the proposal was to go back to what RFC723* said - CTL characters are invalid, but do not require them to be rejected or converted.

@mnot
Copy link
Member Author

mnot commented Feb 24, 2021

That's what this does; the text above already has that effect.

It does raise the question of whether we should add CTL to obs-text, however. Thoughts?

@reschke
Copy link
Contributor

reschke commented Feb 25, 2021

That's what this does; the text above already has that effect.

I believe the proposed text now is somewhat misleading, as it does not mention that CTLs are (and always have been invalid.

It does raise the question of whether we should add CTL to obs-text, however. Thoughts?

No, that would be a breaking change.

Copy link
Member

@royfielding royfielding left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

How about this variation?

Co-authored-by: Roy T. Fielding <fielding@gbiv.com>
@mnot mnot requested review from reschke and royfielding March 1, 2021 03:50
@reschke
Copy link
Contributor

reschke commented Mar 1, 2021

We still need to rewrite or remove

Clarify that control characters in field values are to be rejected or mapped to SP. (Section 5.5)

@reschke reschke changed the title Relax invalid characters in field values Relax recipient requirements for invalid characters in field values Mar 1, 2021
@reschke reschke changed the title Relax recipient requirements for invalid characters in field values Restore recipient requirements for invalid characters in field values Mar 1, 2021
@royfielding royfielding merged commit 94e465f into master Mar 4, 2021
@royfielding royfielding deleted the mnot-683 branch March 4, 2021 04:41
triple-underscore added a commit to triple-underscore/triple-underscore.github.io that referenced this pull request Mar 6, 2021
Network congestion or server load httpwg/http-core#724 Remove TRACE requirement on message/http media type (last normative ref to HTTP/1.1) httpwg/http-core#790 Remove unneeded normative references to MESSAGING httpwg/http-core#796 Remove mid-stream trailers httpwg/http-core#792 Move statement about relation between Content-Length and Transfer-Encoding into Messaging editorial h1-messaging semantics httpwg/http-core#747 Restore recipient requirements for invalid characters in field values httpwg/http-core#783
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

4 participants