This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH, Fortran] PR 33141: Better diagnostics and handling of intrinsics with -std=* settings


Daniel Kraft wrote:
FYI, that's the updated patch for the changes described below.

Committed as rev 138122. Will update the wiki now.


Thanks,
Daniel

Daniel

Daniel Kraft wrote:
Tobias Burnus wrote:
"Warning: The intrinsic 'abort' at (1) is not included in the selected
 standard (GNU Fortran extension) and 'abort' will be treated as if
 declared EXTERNAL.  Use an appropriate -std=* option or define
 -fall-intrinsics to allow this intrinsic."

"in the selected standard (GNU Fortran exension)" sounds as if one used
-std=gnu instead of, e.g., -std=f95. Additionally, warning/error
messages in gfortran do not end with a ".".  Maybe better:

"Warning: 'abort' at (1) is not an intrinsic in the selected standard
 but a GNU extension. It will be treated as if declared EXTERNAL.  Use
 an appropriate -std=* option or define -fall-intrinsics to allow this
 intrinsic"

But maybe someone else comes up with an even better wording.

I tried to get something easy to translate and language-independent (because those "GNU Fortran extension", "new in Fortran 2003" and such are substituted), but you're right. I suggest to change those standard-names to


"a GNU Fortran extension", "new in XXX", "for backwards compatibility"...

so that we can do use your suggested message:

is not an intrinsic in...but %s. It will...

Is this ok, even for a translator's point of view? At least for German, where I can tell, this can be translated fairly straight:

"eine GNU Fortran Erweiterung", "neu in XXX", "für Legacy-Code gedacht"...
ist nicht im gewählten Standard enthalten, sondern %s.


which works out well.

+/* Check if the passed name is name of an intrinsic (taking into accout
the

Spelling: account

Oops, will be changed of course.


The patch was regression tested and bootstrapped on GNU/Linux-x86-32
without any failures.  Ok to commit?
OK. But maybe wait one day or two to give other a chance to comment.

I'll check it in tomorrow evening (europe) if no updates.


Can you additionally update
http://gcc.gnu.org/wiki/GFortran#news
(At some point we need to convert those items into a patch for
 http://gcc.gnu.org/gcc-4.4/changes.html )

Ok, I'll update the wiki with the new/removed flag(s).


Thanks for the review,
Daniel





--
Done:     Arc-Bar-Sam-Val-Wiz, Dwa-Elf-Gno-Hum-Orc, Law-Neu-Cha, Fem-Mal
Underway: Cav-Dwa-Law-Fem
To go:    Cav-Hea-Kni-Mon-Pri-Ran-Rog-Tou


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]