Hi,

I think that - even though it’s probably slower - it’s fine to keep the null check separate. The reason for me is that, it’s probably less error prone this way (and we can’t forget to add it to one of the sub-switches) and it
clearly communicates intent in the code.
A hybrid solution could be to add the tid2 null check before and add the
tid1 check in the top-level switch. But that would need to be nicely
documented as the asymmetry could be confusing otherwise.

My 2c,
Till

On 25 Jul 2016, at 10:07, Riyafa Abdul Hameed wrote:

Hi,

I added the tests for all comparison checks (eq, ne, lt, gt, ge, le) as suggested[1]. Here I have used an if statement in the FunctionHelper class instead of checking for null tag inside switch statements because it would be tedious to add the null tag to all the switch statements for the tid2 tag. Shall I proceed to attempt to add the null tag into all the switch
statements or leave it as a if check?

[1] https://github.com/apache/vxquery/pull/109

Thank you.

Yours sincerely,
Riyafa


On 19 July 2016 at 16:34, Riyafa Abdul Hameed <[email protected]>
wrote:

Hi,

I have made Changes to value comparison semantics for JSONiq[1].

[1] https://github.com/apache/vxquery/pull/109

Thank you.

Yours sincerely,
Riyafa

--
Riyafa Abdul Hameed
Undergraduate, University of Moratuwa

Email: [email protected]
Website: https://riyafa.wordpress.com/ <http://riyafa.wordpress.com/>
<http://facebook.com/riyafa.ahf>  <http://lk.linkedin.com/in/riyafa>
<http://twitter.com/Riyafa1>




--
Riyafa Abdul Hameed
Undergraduate, University of Moratuwa

Email: [email protected]
Website: https://riyafa.wordpress.com/ <http://riyafa.wordpress.com/>
<http://facebook.com/riyafa.ahf>  <http://lk.linkedin.com/in/riyafa>
<http://twitter.com/Riyafa1>

Reply via email to