Re: [sip-clf] draft-ietf-sipclf-format-01, TLV Start Pointer

Works for me

Peter Musgrave
(as individual)

On 2011-03-15, at 3:15 PM, Gonzalo Salgueiro wrote:

> 
> If this simplifies getting the length of the final mandatory field (i.e Client-Txn), then I'm not opposed to having "TLV Start Pointer" point to the terminating line feed, rather than being set to all zeroes in the case of no optional fields.
> 
> Are others OK with this logic?
> 
> Gonzalo
> 
> 
> 
> On Mar 15, 2011, at 2:02 PM, Anders Nygren wrote:
> 
>> Hi
>> When extracting the mandatory fields the length of one parameter can normally be
>> calculated by
>> Length = <Pointer To Next Parameter> - <Pointer To This Parameter> -1
>> 
>> However if there are no optional fields
>> "If the OPTIONAL TLV groups are not implemented, then the "TLV Start Pointer"
>> field MUST be set to zero (0x0000)."
>> 
>> But to calculate the length of <Client-Txn> You first have to check if
>> <TLV Start Pointer> = 0x0000 in which case the <Client-Txn> length
>> is
>> Length = <Record Length> - <Pointer To Client-Txn> -1
>> 
>> Similarly when extracting the TLVs it is necessary to do an extra test if
>> the <TLV Start Pointer> = 0x0000.
>> 
>> I would like to propose that in the case when there are no optional fields that
>> the <TLV Start Pointer> points to the 0x0A at the end of the record.
>> 
>> It is not a big complication but it will remove a couple of tests in a parser.
>> 
>> /Anders
>> _______________________________________________
>> sip-clf mailing list
>> sip-clf@ietf.org
>> https://www.ietf.org/mailman/listinfo/sip-clf
> 
> _______________________________________________
> sip-clf mailing list
> sip-clf@ietf.org
> https://www.ietf.org/mailman/listinfo/sip-clf