[Rivet] [rivet] root!=ansi compatible

[Andy Buckley]

James Monk wrote:
> Hi everyone,
> 
> Has anyone used Rivet with Root on a system with 64bit libs?  Would that
> affect the def'n of long long Vs. long?  If I take out the -pedantic
> flag then it does compile, so it may be that we have to lose that flag
> in order to compile with Root.

I've seen some of these warnings, but this sounds much more serious. I thought
that the use of the -permissive flag (i.e. "Mr G++, I'm trying to compile some
really shoddy code, so please be forgiving") was enough to let ROOT compile,
but by the looks of things this doesn't extend to the 64 bit case.

I'm reluctant to reduce our coding standards just to let ROOT get away with
whatever horrors are going on inside: using the whingy compiler mode helps us
keep our own code portable. It's already bad that we have to add -permissive
because of ROOT: removing -pedantic would be a case of the tail wagging the dog.

Is there any way that we can feed problems like this back to the ROOT
developers? Frankly, anything used so widely through the community *should* be
ANSI compliant! If this can be phrased as a problem for GENSER/ATLAS, do the
ROOT team have any formal support role which means that they have to fix their
code? My suspicion is that there is no such role, and they do what they feel
like, but I'd be interested to know if we have any hope of getting this fixed
upstream. Probably best that I not report it, though: that would be a good way
to guarantee being ignored :)

Andy

-- 
Andy Buckley: CEDAR @ IPPP, Durham
Work: www.cedar.ac.uk
www.insectnation.org