Re: [Quickfixj-users] Should fields throw IncorrectDataFormat for bad data?

[Christoph J. <chr...@ma...>]

Hi Philip,

I think it might be sensible to extend the validation by this. It could keep user's application code 
cleaner.
However, I think this validation should be made configurable via a configuration option, e.g. 
ValidateDatatypes or similar. Of course one could still turn off complete validation but in my 
opinion a finer grained control of the various validation options is better.

Cheers,
Chris.


On 27.09.19 17:53, Philip Whitehouse wrote:
> QuickFIX/J Documentation: http://www.quickfixj.org/documentation/
> QuickFIX/J Support: http://www.quickfixj.org/support/
>
>
>
> Hi all,
>
> The FIX spec for Qty states:
>
> float field capable of storing either a whole number (no decimal places) of "shares" (securities 
> denominated in whole units) or a decimal value containing decimal places for non-share quantity 
> asset classes (securities denominated in fractional units).
>
> 	
>
> There’s no mention of negative values (whereas there is for price).
>
> But we parse it as a double and it’s left to applications to validate that it’s positive.
>
> A DAYOFMONTH field is parsed as a string (even though it should be 1-31)
>
> Similar stuff can be said for lots of other fields.
>
> Should we move towards tighter validation of field values here:
> https://github.com/quickfix-j/quickfixj/blob/master/quickfixj-core/src/main/java/quickfix/DataDictionary.java#L706
>
> Thoughts and comments appreciated,
>
> Best regards,
>
> Philip Whitehouse
>
>
>
> _______________________________________________
> Quickfixj-users mailing list
> Qui...@li...
> https://lists.sourceforge.net/lists/listinfo/quickfixj-users

-- 
Christoph John
Software Engineering
T +49 241 557080-28
chr...@ma...

MACD GmbH
Oppenhoffallee 103
52066 Aachen, Germany
www.macd.com

Amtsgericht Aachen: HRB 8151
Ust.-Id: DE 813021663
Geschäftsführer: George Macdonald