Re: [RFA/OBVIOUS?] ui_file_new function missing "extern" in header.

Re: [RFA/OBVIOUS?] ui_file_new function missing "extern" in header.

[Tom Tromey]

>>>>> "Pierre" == Pierre Muller <pierre.muller@ics-cnrs.unistra.fr> writes:

Pierre>    This is both a RFA
Pierre>  and a question about:
Pierre>   should this be considered as an obvious fix?
  
Pierre>   i.e. shouldn't all functions declared in headers
Pierre> have the extern modifier?
Pierre>   Is this mandatory?

It isn't mandatory right now, gdb does it both ways.  I don't really
care about this issue so I just tend to use whatever is used nearby.  I
think Joel at least has argued for making this part of the gdb style.

Tom

[RFA/OBVIOUS?] ui_file_new function missing "extern" in header.

[Pierre Muller]

   This is both a RFA
 and a question about:
  should this be considered as an obvious fix?
  
  i.e. shouldn't all functions declared in headers
have the extern modifier?
  Is this mandatory?
  It doesn't see to change anything in practice, but to me,
it seems like good practice...
... but remember that my C programming knowledge is weak...




Pierre Muller
GDB pascal language maintainer


2012-12-18  Pierre Muller  <muller@sourceware.org>

        * ui-file.h (tee_file_new): Add extern modifier in header
declaration.

Index: ui-file.h
===================================================================
RCS file: /cvs/src/src/gdb/ui-file.h,v
retrieving revision 1.19
diff -u -p -r1.19 ui-file.h
--- ui-file.h   18 Dec 2012 07:01:58 -0000      1.19
+++ ui-file.h   18 Dec 2012 08:31:00 -0000
@@ -135,7 +135,7 @@ extern struct ui_file *gdb_fopen (char *
 /* Create a file which writes to both ONE and TWO.  CLOSE_ONE
    and CLOSE_TWO indicate whether the original files should be
    closed when the new file is closed.  */
-struct ui_file *tee_file_new (struct ui_file *one,
+extern struct ui_file *tee_file_new (struct ui_file *one,
                              int close_one,
                              struct ui_file *two,
                              int close_two);

Re: [RFA/OBVIOUS?] ui_file_new function missing "extern" in header.

[Jan Kratochvil]

On Tue, 18 Dec 2012 09:35:39 +0100, Pierre Muller wrote:
>   i.e. shouldn't all functions declared in headers
> have the extern modifier?
>   Is this mandatory?
>   It doesn't see to change anything in practice, but to me,
> it seems like good practice...

I agree, it does not change anything.  I agree it is probably GNU Coding
Standards do not talk about it so it could be added to gdb/doc/gdbint.texinfo
"Coding Standards".

I am OK with such kind of changes as obvious ones after updating
gdbint.texinfo.


> @@ -135,7 +135,7 @@ extern struct ui_file *gdb_fopen (char *
>  /* Create a file which writes to both ONE and TWO.  CLOSE_ONE
>     and CLOSE_TWO indicate whether the original files should be
>     closed when the new file is closed.  */
> -struct ui_file *tee_file_new (struct ui_file *one,
> +extern struct ui_file *tee_file_new (struct ui_file *one,
>                               int close_one,
>                               struct ui_file *two,
>                               int close_two);

But in this case you need to reformat the parameters "clone_one" and remaining
ones to still align properly under the opening paren.


Thanks,
Jan

RE: [RFA/OBVIOUS?] ui_file_new function missing "extern" in header.

[Pierre Muller]

> -----Message d'origine-----
> De : gdb-patches-owner@sourceware.org [mailto:gdb-patches-
> owner@sourceware.org] De la part de Jan Kratochvil
> Envoyé : mardi 18 décembre 2012 09:47
> À : Pierre Muller
> Cc : gdb-patches@sourceware.org
> Objet : Re: [RFA/OBVIOUS?] ui_file_new function missing "extern" in
header.
> 
> On Tue, 18 Dec 2012 09:35:39 +0100, Pierre Muller wrote:
> >   i.e. shouldn't all functions declared in headers
> > have the extern modifier?
> >   Is this mandatory?
> >   It doesn't see to change anything in practice, but to me,
> > it seems like good practice...
> 
> I agree, it does not change anything.  I agree it is probably GNU Coding
> Standards do not talk about it so it could be added to
> gdb/doc/gdbint.texinfo
> "Coding Standards".
> 
> I am OK with such kind of changes as obvious ones after updating
> gdbint.texinfo.
> 

  OK, thus after change to gdbint.texinfo,
I committed this patch as obvious in:
http://sourceware.org/ml/gdb-cvs/2012-12/msg00175.html

I modified the submitted patch to correctly align the parameters of 
the following lines, as you suggested.

Thanks Jan,

Pierre Muller

Re: [RFA/OBVIOUS?] ui_file_new function missing "extern" in header.

[Yao Qi]

On 12/18/2012 04:35 PM, Pierre Muller wrote:
>    i.e. shouldn't all functions declared in headers
> have the extern modifier?
>    Is this mandatory?

Better to do that...

>    It doesn't see to change anything in practice, but to me,
> it seems like good practice...
> ... but remember that my C programming knowledge is weak...

Tom fixed a bug (gdbserver PR 14917) recently, which is about adding 
'extern' in header.

   http://sourceware.org/ml/gdb-patches/2012-12/msg00072.html

-- 
Yao (齐尧)

Re: [RFA/OBVIOUS?] ui_file_new function missing "extern" in header.

[Jan Kratochvil]

On Tue, 18 Dec 2012 09:53:48 +0100, Yao Qi wrote:
> Tom fixed a bug (gdbserver PR 14917) recently, which is about adding
> 'extern' in header.
> 
>   http://sourceware.org/ml/gdb-patches/2012-12/msg00072.html

I find that irrelevant, that PR is about variables, this thread is about
function declarations.


Jan