On Tue, 20 May 2008, Alon Bar-Lev wrote:
> It has nothing to do with distcheck.
>
> All it does it put the headers into tarball even if package was
> configured without library.
Hi Alon,
I think the patch looks good but for some reason (which must be my fault)
I couldn't reproduce the problem.
I tried ./configure --disable-library with 'make dist' and 'make distcheck'
but the package always had the headers.
How could I check the failure scenario?
Thanks,
Szaka
> On 5/19/08, Bernhard Kaindl <[EMAIL PROTECTED]> wrote:
> > On Mon, 19 May 2008, Alon Bar-Lev wrote:
> > >
> > > Always distribute headers.
> >
> > That seems to explain what this patch does, which would be very
> > helpful as not everybody knows autoconf by heart.
> >
> > But I am missing *why* you say you want to distribute headers always.
> >
> > I did a quick test:
> >
> > Afterwards, it looks your description is applies to 'make distcheck'
> > and should read somewhat more verbosely like:
> >
> > "Always include ntfs-3g headers with 'make distcheck'"
> >
> > But that is just a pure guess, maybe you can explain what it
> > does more verbosely.
> >
> > From your description, I thought that you want to install the
> > headers even when INSTALL_LIBRARY is not set (triggered by
> > --disable-library)
> >
> > That would not have made sense to me because the point of
> > --disable-library was to build a ntfs-3g binary which has
> > libntfs-3g linked into the binary instead of distributing
> > it separately to make combined binary smaller than the two
> > separate files (ntfs-3g + libntfs-3g.so).
> >
> > As there is no library installed in this case, the header files
> > are quite useless which is why I added this condition of not
> > installing the headers with --disable-library.
> >
> > BTW, I do not use --disable-library anymore but maybe some embedded
> > device makers like it (if they do not include ntfs-3g.probe,
> > they save some space and maybe could even gain some speed,
> > depending on the CPU).
> >
> > Now I tested, but the change does not appear to have an effect
> > even though I verified that it made it into the final Makefile,
> > so the description of the patch does appear wrong to me.
> >
> > But then the question would be what does the patch really change.
> >
> > I guess it is one of the puzzle pieces in order to make
> > 'make distcheck' work and for that it would make sense
> > and looks like a good idea.
> >
> > Bernhard
> >
> >
> > > Index: include/ntfs-3g/Makefile.am
> > > ===================================================================
> > > RCS file: /cvsroot/ntfs-3g/ntfs-3g/include/ntfs-3g/Makefile.am,v
> > > retrieving revision 1.4
> > > diff -u -B -r1.4 Makefile.am
> > > --- include/ntfs-3g/Makefile.am 6 Dec 2007 20:44:03 -0000 1.4
> > > +++ include/ntfs-3g/Makefile.am 19 May 2008 04:55:16 -0000
> > > @@ -3,6 +3,5 @@
> > >
> > > -if INSTALL_LIBRARY
> > > -ntfs3ginclude_HEADERS = \
> > > +headers = \
> > > attrib.h \
> > > attrlist.h \
> > > bitmap.h \
> > > @@ -32,5 +31,10 @@
> > > unistr.h \
> > > version.h \
> > > volume.h
> > > +
> > > +if INSTALL_LIBRARY
> > > +ntfs3ginclude_HEADERS = $(headers)
> > > +else
> > > +noinst_HEADERS = $(headers)
> > > endif
> > >
> >
> >
> > --
> >
> > Bernhard Kaindl
> >
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> ntfs-3g-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/ntfs-3g-devel
>
--
NTFS-3G: http://ntfs-3g.org
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
ntfs-3g-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ntfs-3g-devel
