{{Quickfixn}} Possible bug in QuickFIX/n?

[Manuel Lopez]

>"Perhaps the most key question now is whether the default config should be
to reject or not reject."

I'd vote the default config should be NOT to reject.  A standard is its
implementation, and here we have one rather big counterparty (Barclays)
that interprets the standard as not requiring that
item (OrigSendingTime).  A default of not requiring would not cause
problems for counterparties that do require it; it means only that they
would be missing a validation check until the default setting is changed;
but a default of requiring would cause outright failure against
counterparties that take it as not required.  (And apart from that, I don't
see a good reason why it's required.)

--I assume there are only a handful of quirks like this (I mean where the
interpretations of defined (non-custom) FIX messages differ).  But if there
are more than a few, then ideally the config file would have a setting to
select from a list of major counterparties, though I imagine that would be
a lot of work.  Or there could be a choice of three settings: NoValidation,
RelaxedValidation, and StrictValidation.  RelaxedValidation would skip
checks where the interpretation is unclear (as implemented by major
counterparties).

(@Thomas Fleming, thanks for reporting the issue and the workaround.)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quickfixn.com/pipermail/quickfixn-quickfixn.com/attachments/20120906/142f3156/attachment-0002.htm>