> Indeed, this issue of linewidth zero is just a dead mashed horse. I made the > deliberate choice to explicitely never issue a linewidth zero to any output > device whatsoever; I issue a very very thin (but constant and mastered) > width instead. The rationale is in the source, changelogs and in the > ML archives; I see zero reason to change this behaviour.
While a dont have any problem with this behaviour i suspect the specifics of the implementation it might have something to do with an unwanted side effect i have been suffering when attempting to create shapes with just a fill but no outline. I think i may have mentioned this before but not explained it particularly well. Later Alan
<<attachment: Dia090win32_screenshot_misbehaviour.png>>
