[GRASS-dev] Re: [GRASS GIS] #867: WinGrass64-compiling:
`libintl_printf' is an unrecognized format function type'
GRASS GIS
trac at osgeo.org
Tue Jan 12 20:23:19 EST 2010
#867: WinGrass64-compiling: `libintl_printf' is an unrecognized format function
type'
------------------------------+---------------------------------------------
Reporter: hellik | Owner: grass-dev at lists.osgeo.org
Type: defect | Status: new
Priority: minor | Milestone: 6.4.0
Component: Packaging | Version: svn-releasebranch64
Resolution: | Keywords:
Platform: MSWindows Vista | Cpu: x86-32
------------------------------+---------------------------------------------
Changes (by glynn):
* priority: normal => minor
Comment:
Replying to [ticket:867 hellik]:
> unfortunately I wasn't all the building time in front of my laptop and
couldn't catch all the messages,
That's why you should redirect output to a file, e.g.
{{{
make &> build.log
}}}
> but I've seen in a few-modules (raster and vector etc.; i.e. nviz,
v.what, v.out.pov, v.support, v.transform, v.to.db ) following warnings:
>
> `libintl_printf' is an unrecognized format function type'
I count 77 distinct occurrences (each one will generate multiple warnings,
one for each occurrence of `__attribute__((format(printf,...)))` in
gisdefs.h or G3d.h.
Downgrading to "minor", as these are only warnings.
They could be silenced by re-defining __attribute__ to an empty macro (as
we do if __GNUC__ isn't defined).
We could eliminate the redirection to libintl_printf, as none of the
translations are currently using the %1$ syntax (which the MSVCRT printf()
doesn't support, hence the redirection). However, chances are we'll forget
to re-instate it if we ever need it.
--
Ticket URL: <https://trac.osgeo.org/grass/ticket/867#comment:1>
GRASS GIS <http://grass.osgeo.org>
More information about the grass-dev
mailing list