Re: [Patch] Fortran: Accept -std=f2023, update line-length for Fortran 2023

2023-11-17 Thread Mikael Morin 

Le 17/11/2023 à 12:38, Tobias Burnus a écrit :


Unless there are follow up comments, I will commit it later today.


I skimmed quickly through the patch, and noticed one typo to fix:

> diff --git a/gcc/fortran/invoke.texi b/gcc/fortran/invoke.texi
> index 10387e39501..5f87b330a22 100644
> --- a/gcc/fortran/invoke.texi
> +++ b/gcc/fortran/invoke.texi
> @@ -958,8 +959,8 @@ The following example will trigger the warning.
>  @item -Wampersand
>  Warn about missing ampersand in continued character constants. The
>  warning is given with @option{-Wampersand}, @option{-pedantic},
> -@option{-std=f95}, @option{-std=f2003}, @option{-std=f2008} and
> -@option{-std=f2018}. Note: With no ampersand given in a continued
> +@option{-std=f95}, @option{-std=f2003}, @option{-std=f2008}, 
@option{-std=f2018}

> +and @option{-std=f203}. Note: With no ampersand given in a continued
>  character constant, GNU Fortran assumes continuation at the first
>  non-comment, non-whitespace character after the ampersand that
>  initiated the continuation.

s/f203/f2023/

Re: [Patch] Fortran: Accept -std=f2023, update line-length for Fortran 2023

2023-11-17 Thread Harald Anlauf 

Hi Tobias,

On 11/17/23 12:38, Tobias Burnus wrote:

Hi Harald, hi all,

On 16.11.23 20:30, Harald Anlauf wrote:

According to the standard one can have 99 lines with only
"&" and then an ";", but then only 100 lines with 1 characters.


I believe a single '&' is not valid, you either need '&&' or something
else + '&'; thus, you can have only half a million lines + 1.


after looking at the F2023 standard again I wonder why
they did such a disservice to compiler developers...

You are right: a single '&' is not valid.

6.3.2.4 also has:

"When used for continuation, the “&” is not part of the statement"

And 6.3.2.5 (also 6.3.3.4): "The “;” is not part of the statement".

So a million "&"-continued lines is possible in free form.

For fixed form, 6.3.3.1 has: "If a source line contains only characters
of default kind, it shall contain exactly 72 characters; otherwise, its
maximum number of characters is processor dependent."

I wonder what I should make out of this...


In the code, I still use 1,000,000 but now with a comment.


Yeah, for the time being this is the most reasonable solution.
Let's claim that the 10^6 line limit is the new GNU standard ;-)

Cheers,
Harald