Experimental Physics and
Industrial Control System

Eric Norum <[email protected]> · Thu, 19 Aug 2010 21:03:41 -0700

On Aug 19, 2010, at 8:38 PM, Angus Gratton wrote:

> Hi Eric,
> 
> On Thu, 2010-08-19 at 19:13 -0700, Eric Norum wrote:
>> The 1<<N form seems reasonable, but I'm not 100% that the bitXX names are a good idea.   My concern stems from the fact that some architectures number bits from the right (which is what I presume this patch does) while others number them from the left.  Thus 'bit14' is somewhat ambiguous.
> 
> That seems fair enough. I hadn't considered different bit numbering
> conventions. FWIW, the documentation section describes 'bitX' as
> equivalent to '1<<X', but I can still see your point.
> 
> I have no objection to swapping the string "bit" for "1<<" in the patch.
> 
> The reason why I didn't go with '1<<' up front is that this implies
> there may be a more complete set of expressions available to the user.
> Which could also be nice to have, but I think would require a more
> complex parser for @asynMask fields. Unless there's one elsewhere in
> EPICS (CALC?) that could readily be plugged in?
> 

Sure.
Have a look at 'postfix()' and 'calcPerform()' in the appDevGuide.

-- 
Eric Norum
[email protected]

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System