[This message was posted by Naresh Maharaj of Instinet Europe plc 
<[email protected]> to the "FAST Protocol" discussion forum at 
http://fixprotocol.org/discuss/46. You can reply to it on-line at 
http://fixprotocol.org/discuss/read/2830f020 - PLEASE DO NOT REPLY BY MAIL.]

David 

I am clearing out the dictionary after every single message (block) read. 

If you are not able to do this then what would happen if a message was missed 
or dropped.

Surely after each block you have to clear the dictionary.


Naresh


> Naresh -
> 
> There is no such default length. The semantics of an operator on the
> sequence length preamble is no different than the semantics you'd have
> if you applied the same operator to a normal field of type uInt32.
> 
> You say that you cut out a single message for playback. This is
> generally not possible with FAST since you lose the previous
> value history.
> 
> /David
> 
> >
> >
> > Hi Dale
> >
> > Thank you for the response and the behaviour you mentioned below is
> > being used and is inherent in the copy operator I use as below:
> >
> > long segmentLength = copyOperatorForInt(buffer, TypeDefs.NoMDEntries,
> > pmapBits[pmapCounter--]);
> >
> > This same line is working on several other messages.
> >
> > The message in question is a single message that has been cut out of
> > the stream for playback.
> >
> > When I play the single message back via openfast I get the result
> > perfectly.
> >
> > 8=FIXT.1.1,35=W,1128=7,49=SWXGROUP-MDI.P01,56=SWXGROUP- 
> > MDC,34=19341,97=N,52=20091028-
> > 05:11:05.849,1021=1,48=CZ0001000749,22=4,268=1
> >
> > Am confident that I am reading the bytes correctly in my app and have
> > gone through every bit..
> >
> > As it is a single message there is no prev value in the dictionary.
> > Hence I end up getting the pmap bit zero for the length and when going
> > to check the dictionary it is undefined and no initial value hence ERR
> > D5 (InitialValue NULL and Mandatory)
> >
> > It leads me to suggest that if the pmap bit concludes that the segment
> > length is not in the stream that it should default to a minimum of 1.
> >
> > This is my own interpretation which I prefer not to use but I cant see
> > any other option here.
> >
> > Naresh
> >
> > > And of course when I said "segment" in my message below, I meant
> > > "Sequence".
> > >
> > > Dale
> > >
> > > > Hi Naresh,
> > > >
> > > > > > Is it possible to have a sequence as below where the pmap bit
> > > > > > is set to false for the length of the sequence.
> > > > > >
> > > > > > [sequence name="MDEntries"] [typeRef name="MDIncGrp"/] [length
> > > > > > name="NoMDEntries" id="268"] [copy/] [/length] [string
> > > > > > name="MDUpdateAction" id="279"] [copy value="0"/] [/string]
> > > > > > [!--
> > > > > > : New --] ....
> > > > > >
> > > > > > The field [length name="NoMDEntries" id="268"] [copy/]
> > > > > > [/length]
> > > > > >
> > > > > > has a false in the pmap bit
> > > > > >
> > > > > > Therefore am not sure who many times to iterate through the
> > > > > > sequence
> > > > > >
> > > >
> > > > The pmap bit associated with a segment applies to the length field
> > > > of the segment. This pmap bit tells you whether the length appears
> > > > on-the- wire. What it means when the length does not appear on-the-
> > > > wire is determined by the operator associated with the length
> > > > field.
> > > >
> > > > In your case the length field of the sequence has a copy operator.
> > > > The zero pmap bit is telling you that the sequence length does not
> > > > appear on-the-wire. Instead you should re-use the previous length
> > > > value from the appropriate dictionary.
> > > >
> > > > Dale
> > > > --
> > > > Dale Wilson Principal Software Engineer Object Computing, Inc.
> > > > (www.ociweb.com) Lead developer for QuickFAST
> > > > (http:www.quickfast.org)


[You can unsubscribe from this discussion group by sending a message to 
mailto:[email protected]]

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"Financial Information eXchange" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/fix-protocol?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to