Most likely, the biggest win by far will be to implement draw_markers.
You will see the start of an attmept in draw_markers_OLD. Not having
this function means symbol plots must make poosibly thousands of
separate calls to draw_polygon and friends, each with their own set of
gc calls. backend_ps is probably the best place to look for
inspiration. If you search the devel archives for draw_markers,
you'll find lots of discussion ....
JDH
···
On 7/12/07, Carl Worth <cworth@...528...> wrote:
Anyway, I'll try to get setup to profile the cairo run above and see
if any obvious things stick out that could be easily fixed.
Carl Worth wrote:
Carl Worth wrote:
[...]
Later, you said...
Aha! Cairo ps output is full of "initclip" commands, so it can't be
converted to eps without substantial reworking.
Oh, actually that should be really easy to get rid of. The cairo
semantics match PostScript's very closely:
cairo_clip -> clip
cairo_reset_clip -> initclip
So, the only problematic case is if the application is actually
calling cairo_reset_clip, (which could simply be made into an error
when targeting EPS). If the application isn't, and cairo is still
generating initclip in the PostScript output then it should be very
simple to fix that.
I have looked at the mpl cairo backend code and at the source code for the pycairo package that mpl uses, and I have not figured out where the initclip call is coming from; it does not appear to be anything explicit. reset_clip is wrapped by pycairo, but an example snippet that generates ps with an initclip does not include any call to reset_clip. Nor does the mpl backend. So I conclude that it is being generated implicitly. I have not checked for any of the other postscript commands that are not allowed in eps, and I really can't pursue this any farther. But thanks for the ideas and encouragement; and in turn, if you are interested, I certainly encourage you to pursue the better utilization of cairo in mpl. The author of the backend, as well as of pycairo, is Steve Chaplin.
Eric
···
On Thu, 12 Jul 2007 11:35:25 -1000, Eric Firing wrote: