Re: [Openhpi-devel] SaHpiControlGet: spec issue

[Renier M. <re...@us...>]

Ok, thanks. I stand corrected.

Regards,

Renier Morales
IBM Linux Technology Center
OpenHPI Team
http://openhpi.sf.net
(845) 435-2003 (TL 295)

ope...@li... wrote on 29/11/2005 08:54:07 AM:

> I have to respectfully disagree.  In the call to saHpiControlGet(),=20
> it is legal for CtrlState->Type to
> be undefined.  Returning NOT=5FPRESENT is incorrect since the control=20
> can be present.  It's presence
> is determined by the [SessionId, ResourceId, CtrlNum], not the Type.
> If I am correct, an HPI
> implementation must first look at the Type for the actual control,=20
> not the Type passed in via the call.
> If the control is a Text control, then the implementation looks at=20
> the Line number to determine what
> to return to the caller.=20
>=20
> Don Barre
> WG Review team member
>=20
> Renier Morales wrote:=20
>=20
> I think we should validate the CtrlType and return NOT=5FPRESENT if=20
> invalid, since if invalid, the control will not be present. Also,=20
> NOT=5FPRESENT is one of the listed error codes for this function.=20
> Don't think the confomance testers will complain at all if we do this.=20
>=20
> Regards,
>=20
> Renier Morales
> IBM Linux Technology Center
> OpenHPI Team
> http://openhpi.sf.net
> (845) 435-2003 (TL 295)=20
>=20
> ope...@li... wrote on 29/11/2005 05:08:38=20
AM:
>=20
> >=20
> > Spec says: "For text controls, the line number to read is passed in=20
via
> > CtrlState->StateUnion.Text.Line" (page 94, description of CtrlState
> > argument).
> > But spec doesn't require CtrlState->Type must be set properly.
> > What should we check in SaHpiControlGet() implementation for CtrlState
> > argument?
> > Your opinion?
> >=20
> >=20
> > -------------------------------------------------------
> > This SF.net email is sponsored by: Splunk Inc. Do you grep throughlog=20
files
> > for problems?  Stop!  Download the new AJAX search engine that makes
> > searching your log files as easy as surfing the  web.  DOWNLOAD=20
SPLUNK!
> > http://ads.osdn.com/?ad=5Fidv37&alloc=5Fid=16865&op=3Dclick
> > =5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=
=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F
> > Openhpi-devel mailing list
> > Ope...@li...
> > https://lists.sourceforge.net/lists/listinfo/openhpi-devel