Paul Pluzhnikov <[EMAIL PROTECTED]> writes:

> You apparently do not yet comprehend amount of work that is sometimes
> required maintaining large bodies of code with numerous external
> dependencies, especially if said code must integrate with other
> products, some of which may only be available in binary form.

You have almost exectly described my work as a programmer in the last
36 years (I have started in 1969) -- apart from the fact that I don't
have to depend from "other products, some of which may only be
available in binary form".

My work is feasible because, e.g. at FNAL, we use strict ANSI Standard
C++.  In the dark days of egcs and g++ 2.95, this meant not to use gcc
at all; we were linked to Kuck & Associates, Inc. (KAI) KCC.  After
the first release of gcc 3, we started to investigate the
compatibility of our code with that compiler; most of the
incompatibilites were due to bad code.  And, for the others, we had to
cooperate strictly with the gcc developement team; after that these
issues (the most of them regarding templates) have been solved, we
switched back from KCC to g++.

Nowadays nobody of us fears a new version of the compiler.  Not only;
from what we see testing the snapshots of g++ version 4, we WOULD LIKE
to have the new compiler soon available in a stable version.

I don't mind what other people think about me; I said that the best
solution for the OP is to upgrade to an up-to-date g++.  If his code
is not Standard, or if he depends from non-Standard code distributed
in a binary form, well, he has done a very bad choice and he in on its
own.

-- 
Maurizio Loreti                         http://www.pd.infn.it/~loreti/mlo.html
Dept. of Physics, Univ. of Padova, Italy              ROT13: [EMAIL PROTECTED]
_______________________________________________
Help-gplusplus mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/help-gplusplus

Reply via email to