Re: [IPFIX] RFC 5101bis

Hi, Andrew,

Good points...

On Apr 10, 2012, at 10:12 PM, Andrew Feren wrote:

> I noticed today that section 6.2 is named "Reduced Size Encoding of Integer and Float Types", but then in the first sentence  also lists "string" and "octetArray" types.
> 
> "Information Elements containing integer, string, float, and
>    octetArray types in the information model MAY be encoded using fewer
>    octets than those implied by their type in the information model"
> 
> I'm not sure how a type of string or octetArray implies a length.

This is a terminology question, I think; strings and octetArrays can be any length, so it's not really "reduced length encoding".

> It seems that the section should be renamed to be more generic (remove "Integer and Float Types") 

Agreed.

> In which case I assume rather than a size "implied by their type" the first paragraph should read something like.
> 
> "Information Elements containing integer, string, float, and
>    octetArray types in the information model MAY be encoded using fewer
>    octets than the specified maximum encoding size"

This would be one good way to handle it.

> Or is reduced encoding intended only for integer types?

The fact that an Information Element inherently has a type which inherently has a length (except for strings and octetArrays), which may be different than the length for the IE specifier in a given template record (subject to certain restrictions: you can have a 3-byte unsigned64 but not a 3-byte float64, for instance) means no; however, there is a question as to what of this is reduced-length encoding and what is just "using templates".

Cheers,

Brian
_______________________________________________
IPFIX mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/ipfix