what's the problem with system cairo ?

Enrico Weigelt enrico.weigelt at vnc.biz
Sun Nov 18 08:46:30 PST 2012


> [1] used in:
> - vcl/Library_vclplug_svp : ifeq ($(GUIBASE),unx)
> - vcl/Library_vclplug_gtk : ENABLE_GTK
> - canvas/Library_cairocanvas : ENABLE_CAIRO_CANVAS
> e.g. for Library_cairocanvas there are also WNT... conditionals in
> the
> makefile but configure says:
> if test $_os = Darwin -o $_os = WINNT; then
>     if test "$enable_cairo_canvas" = yes; then
>         AC_MSG_ERROR([The cairo canvas should not be used for this
>         platform])

I'd guess the WINNT conditionals are old leftovers, maybe there
was some unfinished WINNT-cairo code, maybe limitation of cairo
to unix-only was done later.

IMHO, we should drop the WINNT-conditionals in cairo as first step.

By the way: could we use Gtk on WINNT too ?
Not sure about the real implications of that, but from an toplevel
architecture view, I'd really prefer that approach to reduce amount
of code and complexity.

At that point, I'd also raise the question whether we really need
different widget toolkits (even on the same platform). Can't we,
at least on *nix, choose one crossplatform widget toolkit (I'd
personally would prefer Gtk over Qt) and drop all others ?
That, IMHO, should reduce the code (source) size and complexity
and so maintenance and testing efforts.

Mit freundlichen Grüßen / Kind regards 

Enrico Weigelt 
VNC - Virtual Network Consult GmbH 
Head Of Development 

Pariser Platz 4a, D-10117 Berlin
Tel.: +49 (30) 3464615-20
Fax: +49 (30) 3464615-59

enrico.weigelt at vnc.biz; www.vnc.de 

More information about the LibreOffice mailing list