Chris,
Apologies. This is a bit of loose code that should have been completely
removed. On earlier versions of the compiler it apparently doesn't
complain when const class members are left uninitialised. So you can
either uncomment the initialisation (as you did) or comment out the data
member declarations of ntapsi and ntpsi later in the class. These data
members are unused in the code.
Cheers, Mark.
Mark Woolrich.
Oxford University Centre for Functional MRI of the Brain (FMRIB),
John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK.
Tel: (+44)1865-222782 Homepage: http://www.fmrib.ox.ac.uk/~woolrich
On Mon, 20 Dec 2004, Christopher Bailey wrote:
>
> Dear list,
>
> I'm still running into a (minor!) build problem I posted earlier:
>
> http://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind0404&L=fsl&P=R2768&I=-1
>
> Attached is a snippet of the build.log after an otherwise successfull
> build. Essentially, gsmanager.h contains two commented lines
>
> 36: /* ntapsi(16), */
> 37: /* ntpsi(23), */
>
> which to me seem to be the issue. Once uncommented, flame can be built
> without error. I suppose adding a -Wsomething-or-other flag to g++ call
> might turn the e~rror into a warning too.
>
> So I guess what I'm after is, why are those two lines commented out, and
> am I the only one getting this error? I have no clue what ntapsi(16) or
> ntpsi(23) actually initialise, so perhaps it's a bad idea to uncomment
> those lines?
>
> Best regards,
>
> Chris Bailey
>
> P.S. Thank you for advice on my previous film-related post: I couldn't
> imagince I would be hovering so close to the process memory limit with
> "only" 330 volumes. avwroi did the trick.
>
> --
> Christopher Bailey <[log in to unmask]>
> PET Centre and
> Center for Functionally Integrative Neuroscience
> Aarhus University Hospital, Denmark
> http://www.cfin.au.dk/
>
|