This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.

Bug 25389 - Define octet string
Summary: Define octet string
Status: RESOLVED FIXED
Alias: None
Product: Web Cryptography
Classification: Unclassified
Component: Web Cryptography API Document (show other bugs)
Version: unspecified
Hardware: PC All
: P2 normal
Target Milestone: ---
Assignee: Mark Watson
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2014-04-18 08:37 UTC by Anne
Modified: 2014-09-24 21:00 UTC (History)
2 users (show)

See Also:

Attachments

Description Anne 2014-04-18 08:37:25 UTC 
It's not really clear what that means.

(Aligning with http://encoding.spec.whatwg.org/#terminology and such would be good here I think.)
Comment 1 Mark Watson 2014-09-22 17:42:02 UTC 
I propose: "An octet string is an ordered sequence of zero or more integers, each in the range 0 to 255 inclusive."
Comment 3 Anne 2014-09-24 20:00:14 UTC 
Why can't we call this a byte sequence as the Encoding standard recommends? Seems weird to use different terminology here.
Comment 4 Mark Watson 2014-09-24 20:15:53 UTC 
(In reply to Anne from comment #3)
> Why can't we call this a byte sequence as the Encoding standard recommends?
> Seems weird to use different terminology here.

I don't see a definition of 'byte sequence' at the link you provided. I see a definition of 'byte' which looks odd to me (I would think a byte is an integer between 0 and 255 inclusive without any implication about how it is 'represented').
Comment 5 Anne 2014-09-24 20:25:20 UTC 
Well, you have to represent it somehow in a document.
Comment 6 Mark Watson 2014-09-24 20:27:16 UTC 
(In reply to Anne from comment #5)
> Well, you have to represent it somehow in a document.

What we need in the WebCrypto spec is an abstract type for describing the processing of information within our procedures and in particular passing information to / from cryptographic algorithms described in other specifications. We don't need to define / assume any particular way to represent members of that type, whether in a document, in memory or elsewhere.
Comment 7 Anne 2014-09-24 20:46:27 UTC 
If you don't represent it you don't have to worry about that part. The larger point here is that we use byte everywhere, not octet. And "ByteString" already has a particular meaning you're not employing here. So byte sequence would be best.
Comment 8 Mark Watson 2014-09-24 20:56:12 UTC 
WebIDL defines both byte and octet, with byte being signed and octet unsigned: http://www.w3.org/TR/WebIDL/#idl-octet.

This is itself an odd definition of byte.

I'm all for consistency / alignment, it's just not clear who we should be consistent / aligned with ;-)
Comment 9 Anne 2014-09-24 21:00:05 UTC 
Fair. I think there's an open bug on IDL.