Hi Konstantin et al, On Wed, 28 Feb 2001 19:52:35 -0500 Konstantin Olchanski <olchansk@ux1.phy.bnl.gov> wrote concerning ": Re: BRAT update": > On Wed, Feb 28, 2001 at 09:23:32PM +0100, Christian Holm Christensen wrote: > > Hi all, > > > > I made some very minor changes to BRAT, so that it would compile with > > ROOT 3.00/05 on a Redhat 7 machine. Basically it consisted of > > typecasting various Float_t's to Int_t's which made the build fail > > because of our policy not to allow warnings! > > > > > So warnings are evil but casts are okey? > > Or was it the other way round? No, casts are evil. However, I did a very VERY quick and dirty hack, since I believe that the next generation GCC 3.0 will not have some of this rather annoying behaviour. That's what you get for using Redhat - they'll happily compromise everything by taking a development snapshot rather then waiting for the next stable release. I guess the reason is that 2.96 sounds better then 2.95.2 when you want to _sell_ your distribution. I was compiling BRAT on Peter's Laptop and ran into this pile of problems. I'm not sure if it's because the method signatures (or prototypes) has changed in ROOT 3.00/05 or if it's some feature of "GCC 2.96" that it's more stringent then it's predecessors. Anyway, the point is that it should be fixed properly soon and is not something that anyone should be inspirred by! Yours, Christian ----------------------------------------------------------- Holm Christensen Phone: (+45) 35 35 96 91 Sankt Hansgade 23, 1. th. Office: (+45) 353 25 305 DK-2200 Copenhagen N Web: www.nbi.dk/~cholm Denmark Email: cholm@nbi.dk
This archive was generated by hypermail 2b29 : Thu Mar 01 2001 - 10:45:38 EST