[metapost] problem with drawoptions and fake Euro symbol
Stephan Hennig
mailing_list at arcor.de
Thu Nov 15 16:49:08 CET 2007
Taco Hoekwater schrieb:
> Stephan Hennig wrote:
>>
>> Would it be difficult to mark paths originating from makempx (if that's
>> correct?) as immune to withpen or dashed drawoptions?
>
> An internal all-or-nothing switch would be possible, the commands that
> come from mpx files are pretty easy to isolate. But ...
That means in the first case in a file
prologues := 3;
beginfig(1);
picture pic;
pic := btex X\vrule width20pt height.5pt X etex;
addto pic doublepath fullcircle scaled 100 withpen currentpen;
drawoptions(withpen pencircle scaled 5bp);
draw pic;
endfig;
beginfig(2);
picture pic;
pic := image(draw fullcircle scaled 100);
addto pic also btex X\vrule width20pt height.5pt X etex;
drawoptions(withpen pencircle scaled 5bp);
draw pic;
endfig;
end
the circle wouldn't obey drawoptions as well, since the pic originated
from an mpx file? One could easily work around that, so I think it were ok.
What about the second (reverse) case?
> ... it is not feasible to split 'rules from \rules' from 'rules from vf
> characters', so I doubt it would help much.
No problem,
* John Hobby talks of 'labels' in "For instance, a dashed or withpen
option will affect all the lines in the picture but none of the labels."
On the other hand, if one assumes John Hobby only spoke about string
labels, it seems a natural generalization to handle TeX labels
uniformly, i.e., not distinguish between virtual fonts and plain rules
in TeX material.
* I find it intuitive not to distinguish rules in TeX material.
* Dan already presented a use-case where by the old/current behaviour
math fraction bars come out dotted, which is hardly the intended
behaviour.
> If you don't want to use drawoptions for labels, why not redefine the
> label() macro?
* I've just yesterday learnt about that solution.
* It is not easy to use that in pre-generated code.
* One can argue, the current behaviour is a bug.
* I'm using it now (but would like to consider it a workaround).
Best regards,
Stephan Hennig
More information about the metapost
mailing list