[gfortran,patch] Internationalisation of the Fortran front-end

[Tobias Schlüter]

Gabriel Dos Reis wrote:
> Steven Bosscher <stevenb@suse.de> writes:
> 
> | On Thursday 01 September 2005 16:38, Gabriel Dos Reis wrote:
> | > François-Xavier Coudert <fxcoudert@gmail.com> writes:
> | > | gfortran doesn't use locus in the same sense that other front-end
> | > | routines.
> | >
> | > that is a bug.
> | 
> | No, it is just different historically.  Gfortran has its own diagnostics
> | reporting system, and at some point, when the generic system used by the
> | C family front ends finally gets support for carets, it should be changed
> | to use that instead.  Until then, this is not a bug in gfortran, but a
> | shortcoming of the generic system.
> 
> No.  The current scheme used in gfortran finds its place in the
> current diagnostic machinery: Have BEGIN_DIAGNOSTIC output the carret,
> and let the general machinery handle the rest. 

What rest?  Printing a textual message?  Or is there something else that needs
to be done once the location's been output?

What exactly are the advantages of using the common machinery if the most
complicated part of the code (preparing the lines for caret printing and
printing them) will still have to be done in the frontend (and probably
reimplemented, as gfortran's code will probably not drop into place)?

(Maybe we should have this dicussion when somebody actually embarks on doing
it.  For the record, I do agree with Steven, if that wasn't clear.)

- Tobi