|
FAST Protocol
< Previous Next >
Re: swx MDI decoding issue
Michael Wong / Chi-X Global Technology
13 May 2009 11:29PM ETRolf,
Oh I made a mistake that I think mantissa is optional. so the value is decreased by 1.
I want to know how to handle if a field is absent.
Michael
> > Rolf,
> >
> > I am sorry I got confused again. For tag 271, the pmap bit is 0. so
> > the default value 0 is used for exponent. Can we say that the exponent
> > has a value so that we need to decode mantissa? If so, the encoded
> > value is 0x82 for the mantissa. The decoded value should be 1 since it
> > is delta operator, Right?
>
> Yes, 0x82 is for the mantissa. No, the decoded mantissa value is 2. Why
> do you think the value should be 1?
>
> > BTW, if the length field of a sequence field is greater than 1, does
> > the sequence pmap appear each iteration?
>
> yes, as noted in my mail below
>
> (FAST 1.1 spec section 6.2.5 says: "If any instruction of the group
> needs to allocate a bit in a presence map, each element is represented
> as a segment in the transfer encoding." "element" in this context is
> synonymous with my use of "iteration" in my mail below)
>
> /Rolf
>
> ... snip ...
>
> >> Michael,
> >>
> >> The pmap entry for the length field is located in the _message_ level
> >> pmap (bit 7 in this case). The sequence level pmap occurs for every
> >> iteration (repetition) of the sequence and provides presence info for
> >> the fields in the current iteration.
Re: swx MDI decoding issue
Michael Wong / Chi-X Global Technology 13 May 2009 11:29PM ET
|
