Re: [ptp-dev] Need new attribute type

["Randy M. Roberts" <rsqrd@xxxxxxxx>]

Greg,

Do you want me to do it, or do you want to do it?

R^2

On Thu, 2007-06-21 at 12:12 -0600, Greg Watson wrote:
> I'll add a BigIntegerAttribute.
> 
> Greg
> 
> On Jun 21, 2007, at 11:31 AM, Dave Wootton wrote:
> 
> > Randy
> > I'm not quite sure I follow your suggestion. What I have is a  
> > single PE
> > option which has the format mmm,nnn. This appears in the GUI as a  
> > single
> > Text widget which the user fills in in this form. (I suppose I  
> > could also
> > have two separate Text fields to contain the two parts of the  
> > field).  It
> > also gets passed back to the proxy in this form.
> >
> > Is what you are suggesting is that the proxy create two attribute
> > definitions, one for each part, named say, ATTR_PART_A and  
> > ATTR_PART_B,
> > then use those two attribute definitions to perform validations for  
> > each
> > of the components of the option?
> >
> > I think that would work for me, and also solves the second part of my
> > validation problem where an option can have a value that can either  
> > be one
> > of an enumerated set or an integer with range limits.
> >
> > In that case, in order for my validation to be correct, I would have a
> > definite need for a LONG attribute definition event or a BIGINTEGER
> > attribute definition event.
> >
> > Dave
> >
> >
> >
> > "Randy M. Roberts" <rsqrd@xxxxxxxx>
> > Sent by: ptp-dev-bounces@xxxxxxxxxxx
> > 06/21/2007 12:00 PM
> > Please respond to
> > Parallel Tools Platform general developers <ptp-dev@xxxxxxxxxxx>
> >
> >
> > To
> > ptp-dev <ptp-dev@xxxxxxxxxxx>
> > cc
> >
> > Subject
> > Re: [ptp-dev] Need new attribute type
> >
> >
> >
> >
> >
> >
> > Dave,
> >
> > Since the field is nnn,mmm, couldn't you present
> > it as two fields, nnn and mmm, with a "," label between
> > them?  Then you could verify the nnn and the mmm separately.
> >
> > R^2
> >
> > On Thu, 2007-06-21 at 09:09 -0400, Dave Wootton wrote:
> >> Greg
> >> At the moment, I don't have a specific need for > 32 bit integer
> > support.
> >> The one attribute that can be > 32 bit integer has the format nnn,mmm
> >> where mmm can be > 32 bits. So it doesn't fit the model of a simple
> >> integer attribute, and I have to figure out how to validate it
> > otherwise.
> >> At the moment, this is one case where I have to hard code the  
> >> validation
> >
> >> ranges in the gui code.
> >>
> >> We may need larger integers for LL, but we aren't sure yet. Our  
> >> specific
> >
> >> need is for 32 bit and 64 bit integers but have no objection to a  
> >> more
> >> general BigInteger implementation.
> >>
> >> Dave
> >>
> >>
> >>
> >> Greg Watson <g.watson@xxxxxxxxxxxx>
> >> Sent by: ptp-dev-bounces@xxxxxxxxxxx
> >> 06/20/2007 06:15 PM
> >> Please respond to
> >> Parallel Tools Platform general developers <ptp-dev@xxxxxxxxxxx>
> >>
> >>
> >> To
> >> Parallel Tools Platform general developers <ptp-dev@xxxxxxxxxxx>
> >> cc
> >>
> >> Subject
> >> Re: [ptp-dev] Need new attribute type
> >>
> >>
> >>
> >>
> >>
> >>
> >> I could implement this as a BigInteger (i.e. arbitrary precision
> >> integer). When the attribute definition is created, you would specify
> >> the precision which would be fixed from then on. That might help with
> >> the validation?
> >>
> >> Greg
> >>
> >> On Jun 20, 2007, at 2:48 PM, Dave Wootton wrote:
> >>
> >>> On second thought, I may not really need this. The 64 bit value  
> >>> is an
> >>> optional part of an attribute which I am treating as a string
> >>> attribute
> >>> (since it's in the form ATTR=nnn,mmm). As long as I treat it as a
> >>> string,
> >>> with the possible exception of some validation code, I think I
> >>> don't need
> >>> the additional attribute types,
> >>> Dave
> >>>
> >>>
> >>>
> >>> Dave Wootton/Poughkeepsie/IBM@IBMUS
> >>> Sent by: ptp-dev-bounces@xxxxxxxxxxx
> >>> 06/20/2007 04:33 PM
> >>> Please respond to
> >>> Parallel Tools Platform general developers <ptp-dev@xxxxxxxxxxx>
> >>>
> >>>
> >>> To
> >>> ptp-dev@xxxxxxxxxxx
> >>> cc
> >>>
> >>> Subject
> >>> [ptp-dev] Need new attribute type
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>> I found out that I have one case where I have an attribute that
> >>> should be
> >>> defined as a 64 bit integer. Is it possible to define a
> >>> LongIntegerAttribute and LongIntegerAttributeDefinition, along with
> >>> supporting events, etc?
> >>>
> >>> Thanks
> >>> Dave
> >>> _______________________________________________
> >>> ptp-dev mailing list
> >>> ptp-dev@xxxxxxxxxxx
> >>> https://dev.eclipse.org/mailman/listinfo/ptp-dev
> >>>
> >>>
> >>> _______________________________________________
> >>> ptp-dev mailing list
> >>> ptp-dev@xxxxxxxxxxx
> >>> https://dev.eclipse.org/mailman/listinfo/ptp-dev
> >>>
> >>
> >> _______________________________________________
> >> ptp-dev mailing list
> >> ptp-dev@xxxxxxxxxxx
> >> https://dev.eclipse.org/mailman/listinfo/ptp-dev
> >>
> >>
> >> _______________________________________________
> >> ptp-dev mailing list
> >> ptp-dev@xxxxxxxxxxx
> >> https://dev.eclipse.org/mailman/listinfo/ptp-dev
> >
> > _______________________________________________
> > ptp-dev mailing list
> > ptp-dev@xxxxxxxxxxx
> > https://dev.eclipse.org/mailman/listinfo/ptp-dev
> >
> >
> > _______________________________________________
> > ptp-dev mailing list
> > ptp-dev@xxxxxxxxxxx
> > https://dev.eclipse.org/mailman/listinfo/ptp-dev
> >
> 
> _______________________________________________
> ptp-dev mailing list
> ptp-dev@xxxxxxxxxxx
> https://dev.eclipse.org/mailman/listinfo/ptp-dev