After posting this problem I did probe a lot further in this and a similar problem. As I reported shortly, the problem goes away when solely replacing in supp-mps.tex the line \ifx\TEXEXECcommand \undefined \def\TEXEXECcommand{texmfstart texexec} \fi by \ifx\TEXEXECcommand \undefined \def\TEXEXECcommand{texexec} \fi This gives me the impression it might not be in context itself, especially since to my (inexperienced) eye there are no or few changes between the working and nonworking versions that might explain the problem. It might be the change from the perl to the ruby scripts with a problem in one of the latter. I expect Hans or Taco to react soon with there usual speed and will await for their comments first. Hans van der Meer On Jun 15, 2006, at 17:52, Mojca Miklavec wrote:
Replying off-list since I have no place to test anything and no time righ now:
I know that Hans has played with transparency at the beginning of May in order to enable it in XeTeX (but that should happen/break already earlier then), but I can't help you any further. Normar "with color transparent" worked a couple of days ago as far as I can remember, but I might have had an older version installed (from May).
However: I would like to ask you to commit such examples (you have much more bug reports everywhere) to "contexttest" suite, see
I usually do this when in doubt if my code is right. But in this case formerly perfectly working code is suddenly broken. And of course I need to run my production and development runs on my own machine, so I have to solve it for that one.
http://wiki.contextgarden.net/Test.
Mojca
On 6/13/06, Hans van der Meer wrote:
I used to draw in transparent color in MetaFun with:
newinternal tfill_mode; tfill_mode := 1; % transparency mode for inside newinternal tfill_fact; tfill_fact := 1; % transparency factor for inside def withFillColor = withcolor transparent(tfill_mode, tfill_fact, fillcolor_) enddef; fill somepath withFillColor;
Suddenly (since the june update I guess) this fails. Because
fill somepath withcolor fillcolor_;
is still working I wonder if something has done to this transparency business. If there was a change, why? It brings me trouble.
Hans van der Meer