Re: [RFC] Reducing the use of current_language - some patches

[Joel Brobecker <brobecker@adacore.com>]

Hey Ulrich,

> So I was wondering if we shouldn't add just a "language" parameter
> to the parsing routines, but something more general, like a "parse
> context" structure.  This could encapsulate all that global state,
> and would be easily extensible in the future ...

Thanks for your feedback. Indeed, I had completely forgotten about
the input radix, and it should indeed be passed. Having a structure
is a really good idea.

On the other hand, I'm not sure we're ready to have the parse context
contain the gdbarch.  For instance, the breakpoint structure doesn't
save it, so we don't have access to it when we want to re_set the
breakpoint.

I propose we start with just the language and the input radix for
this task, and then we can think about incorporating the gdbarch.


> Maybe (if this is a good idea), it could be done after your current
> set of patches; on the other hand, as you're touching all those files
> anyway, maybe it would be simpler to do it in one go.

It would actually be simpler for me to do the work in two phases,
since I'm almost done with the first phase. Now, I have to redo
probably most of it. However, I will do it in one go as it will
create less work for the insight guys, as they will need to update
their sources only once.

Thanks again.

-- 
Joel