Experimental Physics and
Industrial Control System

Andrew Johnson <[email protected]> · Thu, 5 Jul 2012 11:31:09 -0500

Hi,

On 2012-07-04 haquin wrote:
> 
> Will this modification be integrated in the next EPICS release ? when ?

Thanks to your reminder I have implemented this on the 3.15 branch.  However 
the 3.15.0.x releases which we're currently working on are not intended for 
operational use, and at this stage I can't predict when the first operational 
release 3.15.1 will be out -- sorry!

On 2012-07-04 [email protected] wrote:
> 
> I'm pretty sure it's always safe to strtoul(), the behaviour in the
> presence of a minus sign seems well defined in the man page I'm reading
> and I've certainly not encountered problems using it.

You're right, and since the conversation on this topic I read up on this and 
have added a set of fully tested APIs to libCom in the 3.15 branch for parsing 
ASCII numbers into the various epicsInt* and epicsUInt* types, testing for 
overflows in all cases.

Literals that start 0x or 0X will be parsed by epicsCalc as an epicsUInt32.  
Other literals will be parsed using epicsStrtod() as before.  That means it 
will not be possible to use hex floating-point literals in 3.15, although they 
may have worked in earlier releases on platforms where strtod() supports them.

- Andrew
-- 
Never interrupt your enemy when he is making a mistake.
-- Napoleon Bonaparte

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System