[Xy-pic] Fw: Xy-pic and Qcircuit

[Kristoffer H Rose]

Dear Daniel,

> is it always the same macro and/or trigonometric calculation which 
> produces the errors in Bryan's examples with elliptic shapes? If the 
> overflow errors are caused by a single macro which does a certain 
> numeric calculation, could we reimplement this with \dimexpr?

The trigonometry is in xy.doc (note the ".doc" which is the literate 
programming source for xy.tex). There are two trigonometric calculation 
types: approximate, based on the xydash font (in the "Agorithms / 
Directions" section), and calculated (in the "angle in current base" note) 
- the latter is used by \circularEdge at .

> As you are well aware of, doing calculations in TeX's fixed point 
> arithmetic is a pain. But one thing has improved in the 12 years since 
> Xy-pic was written: e-TeX has the \dimexpr macro which makes 
> multiplication and division a lot easier than in the days when 
> everything had to be done with the \multiply and \divide primitives.

First, not all users are "with the times" and I am sure that some still 
run TeX installations that predate e-TeX, so whatever we do has to be 
guarded with fallbacks to the old way.

> Xy-pic's engine is a complicated system, but is it possible to identify 
> the macros which generated errors with arrows and frames in connection 
> with very small objects? If the problem is somewhat limited to the 
> arithmetic at a few spots, I could offer my help with that.

...which would be most welcome, certainly :-)

Did you progress with the last xypdf changes?

Best,
        Kris

__
Kristoffer H Rose, ph.d. <krisrose at us.ibm.com> +1(914)784-7642 (fax -6324)
IBM T.J.Watson Research Center, PO Box 704, Yorktown Heights, NY 10598
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://tug.org/pipermail/xy-pic/attachments/20110320/676c41a1/attachment.html>