[Patch/fortran] Add documentation section about implemented F2003 features

[Brooks Moses]

Janne Blomqvist wrote:
> Here is an updated patch. It adds back the mention of eventual Fortran
> 90 support and contains some minor formatting improvements as well.
[...]
> Index: gfortran.texi
> ===================================================================
> RCS file: /cvsroot/gcc/gcc/gcc/fortran/gfortran.texi,v
> retrieving revision 1.20
> diff -r1.20 gfortran.texi
> 955,957c955,986
> < In the future it may also support other variants and extensions to the Fortran
> < language.  This includes ANSI Fortran 77, Fortran 90, Fortran 2000 (not yet
> < finalized), and OpenMP.
> ---
> 
>>In the future it may also support other variants and extensions to the
>>Fortran language.  This includes ANSI Fortran 77, ISO Fortran 90, ISO
>>Fortran 2003 and OpenMP.

That should be "These include", not "This includes".

I think it should, pedantically speaking, also be "support other 
variants of and extensions to", with an "of" added after "variants". 
(Otherwise, the parallelism expands to "other variants to Fortran," 
which is wrong.)

I also wonder if it would be better to phrase this a bit stronger: "In 
the future it will also support....  These may include...."

>>
>>@menu
>>* Fortran 2003 status::
>>@end menu
>>
>>@node Fortran 2003 status
>>@section Fortran 2003 status
>>
>>Although @command{gfortran} for the time being focuses on implementing
>>the Fortran 95 standard, a few Fortran 2003 features are also
>>implemented.

I think this should be worded as: "Although @command{gfortran} focuses 
on implementing the Fortran 95 standard for the time being, a few...."

Also, I think it should be "a few Fortran 2003 features have been 
implemented."  Or, if you prefer the present tense, "... are also 
available" or "... are also supported."

Actually, I think I prefer "... are currently supported," really.

>>@itemize
>>@item 
>>Intrinsics command_argument_count, get_command, get_command_argument,
>>get_environment_variable.

Should that be ", and get_environment_variable"?  (I don't know what 
house style is on this.)

>>
>>@item 
>>[...] style array constructors

This seems to me to be potentially ambiguous: I mostly see "[...]" used 
as an editorial notation that some material has been omitted.  Maybe 
something like the following:

"Array constructors using square brackets.  That is, [...] rather than 
(/.../)"

I have also seen this sort of thing written as "[... , ... , ...]"; this 
would be clearer as well, but it's typographically a bit uglier.

>>
>>@item 
>>FLUSH statement. Note that the IOMSG feature of the I/O statements is
>>not yet implemented.
>>@end itemize

Perhaps "Note that this is not a complete implementation of the Fortran 
2003 FLUSH statement, as the IOMSG feature...."

(I mean, I know why the comment about IOMSG is in there, because I saw 
the conversation here.  But without knowing that, it would seem to be a 
bit of a non sequitor.)

Otherwise, it looks good to me.

- Brooks