Re: [Rlib-devel] RLIB-1.3.2 questions, problems and an ODBC fix

[Zoltan B. <zb...@du...>]

Bob Doan =C3=ADrta:
> Thank you!  Thank you!  Thank you!
>=20
> Have you actually produced a report in windows? ;)

Not yet. And there is a bug still in my RLIB work.
When "make" is finished, "ldd libr.dll" produces the library list
I expect. When I run "make install", libr.dll is relinked
and after that, the DLL file is not what ldd expects.
I am trying to do something with it.

> I'm going to assume=20
>=20
> DEF_FILE=3Dlibrpdf.def
> if WIN32
> def_ldflags=3D-Wl,--output-def,$(DEF_FILE) -export-dynamic -avoid-versi=
on
> endif
>=20
> Is the magic required to make the windows dll's.

Almost. The "-no-undefined" linker option and
the occurence of AC_LIBTOOL_WIN32_DLL before
AC_PROG_LIBTOOL does most of the work.
The -export-dynamic is something I am not sure is needed
but the -avoid-version is because I wanted to too libr.dll
instead of libr-1.dll or libr-1-3-3.dll.

> What is the license on the nl_langinfo stuff?=20

GPL

> We should do something nice about strfmon.  One would think it's lgpl'd
> code and if it is we might want to include it.
>=20
> Out of curiousity do you have any experiences packaging WIN32 programs?

Very little.

> I think it would be in RLIB's best interest to have a nice installable
> package (Most windows users probably won't want to go through what you
> went through in order to get this going ;))
>=20
> I see you fixed my automake flaws.  Were you planning on making a RLIB
> RPM?

That was the first thing I did on Linux. :-) I will send my RPM
specfile, too.

> Thanks for the strings.  I'll test w/ them tomorrow.

You're welcome.

> The bottom line is the patch looks very nice.  When do you think you
> will find yourself near a printer to print/sign the (c) thingie??? :)

Today. I am going back to my primary work.

Best regards,
Zolt=C3=A1n B=C3=B6sz=C3=B6rm=C3=A9nyi