Re: [RFC] libstdc++/5816

[Paolo Carlini <pcarlini at unitus dot it>]

Nathan Myers wrote:

> > I think Peter expected a succesful parsing of doubles in de_DE when
> > no grouping chars are found. Some different implementations (Dinkum)
> > behave this way. This is what v3 does for integers.
> >
> > I'm confused as of what the standard prescribes.
>
> There is some dispute about what should happen with doubles for the
> case of formatting, because of a one-word typo which Dinkumware chose
> to implement (apparently just because it was less work), and which the
> LWG has failed thus far to fix just because Dinkumware has threatened
> to refuse to implement it.  The final resolution will almost certainly
> permit applying grouping on doubles, although, if the LWG is particularly
> craven that day, it might not require it.

Ah, thanks. Perhaps you already kindly explained something about this, but I did
not remember the details.

> No such ambiguity about parsing doubles has been reported.  The standard
> is clear that parsing must read grouping characters and check grouping,
> and that if no grouping characters are seen, no grouping check is applied.

Thanks for this clarification, in particular: "if no grouping characters are
seen, no grouping check is applied".
I was really confused about it!

Therefore, tomorrow I will refine, test and complete with a testcase a patch
along the lines I had already tentatively envisaged earlier today.

Ciao, Paolo.