[metapost] Re: pdf output and extensions

Larry Siebenmann laurent at math.toronto.edu
Wed Mar 16 05:48:55 CET 2005



Hi Jacko,

 > One ``yes'' (which then?) or two ``yeses''?
 > (Hint: the answer ``yes'' is OK ;-)

If you are on the verge of explaining the current symantics
of "inner" and "outer" in mp, I'm listening.

 > Anyway, I would not agree that there is something like inner 
 > or outer envelope -- think, e.g., of 8-shaped paths.

Yes "innerenvelope" and "outerenvelope"  are reasonable
terms only for cyclic paths with *no* self-intersections.
The Jordan curve theorem tells us that there is an inside and
an outside.

You suggest  "leftenvelope" and "rightenvelope" instead.

 > Anyway, I would not agree that there is something like inner
 > or outer envelope -- think, e.g., of 8-shaped paths. One can
 > only speak about leftside and rightside envelope with
 > respect to the direction of the pen movement -- these are
 > well-defined entities.

Alternatives are  "leftoffsetpath" and "rightoffsetpath",
terms I have spotted in the literature.

As for "well-defined", there is a quandry at cusps on the
path (as at the top of handwritten dotless i). 

Consequently, there is no connected path definition of (say)
"rightenvelope" that varies continuously with the path.  On
the other hand the (possibly disconnected) "perimeter" (=
"frontier") of the inked region does vary continuously with
the path, and that is all that meets the eye of the reader.
So the "well-definition" difficulties here are artifacts of
the path technology.

Cheers

Laurent S.



More information about the metapost mailing list