I began experimenting with cairo to export PDFs of my GIMP images
several weeks ago. Today I saw the GSoC idea in this direction, and I
thought about mailing my thoughts.
I managed to export text while keeping the same appearance that it had
in GIMP using PangoCairo. Exporting images with cairo was also
possible if I saved the images first as PNGs and then used cairo PNG
surfaces to draw them.
Exporting paths is also possible since cairo has full support for bezier curves.
Cairo has the option to create pages in different sizes, using the api
function cairo_pdf_surface_set_size, which means we can also have
different sizes for each page.
The question is, how to build the interface of the plugin. I have no
expirience in GTK+ programming (except for cairo) which means I doubt
I can build an appropriate user interface... In addition, there is
also the question of how will the plugin work - will it export some
sort of defenitions file (XML for example) and then it will call a
process which handles this file, or whether there will be no
sub-export and everything will be done directly.
There are many more questions about this plugin, which I haven't
raised here. Before I continue, I must know whether my direction
sounds OK. Do we want to use Cairo (PDF and PNG), Pango, gimp's png
plugin and so on, or do we want to do it in some other way.
We probably have the option of getting a pixbuf of GIMP's internal
images (using GDK), but I don't like it - I prefer the PNG export.
I'll do some more advanced experimenting, as soon as I get some
working build environment (and libgimp 2.6 for windows which I can't
on Tor Lillqvist's (tml) page here
Gimp-developer mailing list