(j3.2006) (SC22WG5.5439) RE: Proposal of a new requirement : CMDMSG optional argument in 3 subroutines

[Rafik Zurob]

Hello

I'm not against adding a message argument in principle.

I'm not sure I fully appreciate how it would be useful for these 
intrinsics, however.  For example, in the case of 
get_environment_variable, the standard is explicit about the values 
assigned to the status argument:

?1 if the VALUE argument is present and has a length less than the 
significant length of the environment variable.
 1 if the specified environment variable does not exist
 2 if the processor does not support environment variables
 >2 Processor-dependent values greater than 2 may be assigned for other 
error conditions.

I assume that a message argument would only be useful for status values > 
2 since they are processor-dependent.  So the question (IMHO) is, are 
there actually many status values out there that are > 2?

In the case of XL Fortran, we set status to 3 if we're out of memory.  In 
such a case, I doubt we'd have enough memory to open the message catalogue 
and retrieve a localized message to store into the message argument.  We 
don't have any other status values > 2.  I'm not sure about all the other 
vendors, but I checked the manuals of 2 other compilers and didn't see a 
definition of values > 2.  I also checked the POSIX spec page for getenv() 
to see if it defines any errors -- it does not.  (I'm not saying everyone 
on UNIX/Linux/BSD calls getenv().  I was just thinking that if getenv() 
defined errors, some vendors might have defined status > 2 values to 
correspond to those errors.)  The Windows GetEnvironmentVariable function 
also does not appear to define errors not covered by status values -1 to 
1.

Regards

Rafik


j3-bounces at mailman.j3-fortran.org wrote on 27/01/2015 12:35:21 PM:

> From: "Craig Dedo CAPM" <craig at ctdedo.com>
> To: "WG5 Mailing List" <sc22wg5 at open-std.org>
> Date: 27/01/2015 12:52 PM
> Subject: (j3.2006) (SC22WG5.5438) RE: Proposal of a new requirement 
> : CMDMSG optional argument in 3 subroutines
> Sent by: j3-bounces at mailman.j3-fortran.org
> 
> Everyone:
>             Bernard, thank you very much for bringing this idea to 
> everyone on the WG5 Mailing List.
> 
>             I think this is a great idea.  It would be very helpful 
> to have the command message text available in these three 
> subroutines, just as it is available in Execute_Command_Line.  I?m 
> surprised someone didn?t bring this up earlier.  Is it possible to 
> fit this into the development schedule for Fortran 2015?
> 
>             What do other think about:
>             (1) The merit of the idea.
>             (2) The ability to fit this into the development 
> schedule of Fortran 2015.
> 
>             Please feel free to contact me at any time with any 
> questions or concerns that you may have.  I am looking forward to 
> hearing from you soon.
> 
> Sincerely,
> Craig T. Dedo, CAPM
> 17130 W. Burleigh Place
> P. O. Box 423                         Mobile Phone:  (414) 412-5869
> Brookfield, WI   53008-0423    E-mail:  <craig at ctdedo.com>
> USA
> Linked-In:  http://www.linkedin.com/in/craigdedo
> > -----Original Message-----
> > From: j3-bounces at mailman.j3-fortran.org [mailto:j3-
> bounces at mailman.j3-fortran.org]
> > On Behalf Of Bernard PICHON
> > Sent: Tuesday, January 27, 2015 06:58
> > To: sc22wg5 at open-std.org
> > Subject: (j3.2006) (SC22WG5.5437) Proposal of a new requirement : 
> CMDMSG optional
> > argument in 3 subroutines
> > 
> > Dear all members of WG5
> > 
> > In the past I suggested to introduce in Fortran some means to 
> retrieve ?error
> > message? from the processor. This feature is now standardized with
> keywords IOMSG=
> > for I/O routines, keyword ERRMSG= for memory and image control 
> routines and the
> > optional argument CMDMSG is defined in EXECUTE_COMMAND_LINE 
subroutine.
> > 
> > I encounter these days some difficulties with the subroutine
> > GET_ENVIRONMENT_VARIABLE and discovered that this subroutine 
> doesn?t implement any
> > means to retrieve ?error message? from the host processor although
> the optional
> > argument STATUS is defined.
> > 
> > So, I put to WG5 the following requirement written by analogy with
> > EXECUTE_COMMAND_LINE subroutine :
> > 
> > Note that you may receive this mail twice due to a change in my 
> mail address ;
> > sorry for that !!
> > 
> > ==================================================
> > ===================================
> > 
> > Add the optional argument CMDMSG in subroutines GET_COMMAND , 
> GET_COMMAND_ARGUMENT
> > and GET_ENVIRONMENT_VARIABLE with this edit :
> > 
> > CMDMSG (optional) shall be a default character scalar. It is an 
> INTENT (INOUT)
> > argument. If an error condition occurs, it is assigned a 
processor-dependent
> > explanatory message.
> > Otherwise, it is unchanged.
> > 
> > ==================================================
> > ====================================
> > 
> > Dr. Bernard PICHON (from France)
> > CNRS ? Observatoire de la C?te d?Azur (OCA)
> > 
> > 
> > _______________________________________________
> > J3 mailing list
> > J3 at mailman.j3-fortran.org
> > http://mailman.j3-fortran.org/mailman/listinfo/j3
>  _______________________________________________
> J3 mailing list
> J3 at mailman.j3-fortran.org
> http://mailman.j3-fortran.org/mailman/listinfo/j3