Re: hdotsfor?

I understand that you want lots of dots.  U+2026 may not be the ideal
character to say that a stretchy version has as many dots as needed to fill
the area; perhaps there is a better Unicode character.  I think the idea of
using a stretchy char though is one that fits with the MathML spec as it
stands now.

Neil Soiffer
Senior Scientist
Design Science, Inc.
www.dessci.com
~ Makers of MathType, MathFlow, MathPlayer, MathDaisy, Equation Editor ~




On Sun, Oct 3, 2010 at 4:13 PM, Jacques Distler <distler@golem.ph.utexas.edu
> wrote:

>
> On Oct 3, 2010, at 4:46 PM, Neil Soiffer wrote:
>
> > I don't believe it is possible to do this in presentation MathML with
> current renderers.
> >
> > Your suggestion of
> >    <mtd columnspan="...">
> > would not fit in with the spec because columnspan is used to indicate the
> number of columns spanned, so some other method would have to be used.
>
> Sorry. I didn't mean that the ellipsis was the value of the columnspan
> attribute.
>
> The value of the columnspan attribute was supposed to be the mandatory
> argument of \hdotsfor[]{}.
>
> > One way to do this (but I very much doubt it works in any MathML
> implementation) would be something like
> >       <mtd columnspan="3"> <mo stretchy="true">&#x2026; <!-- horiz
> ellipsis --> </mo> </mtd>
> >
> > The idea here is to make use of the property of stretchy chars and
> rows/cols that span.  This idea would work for both horizontal and vertical
> dots in a table/matrix.
> >
> > MathML's operator dictionary [1] provides some hints to developers as to
> what chars to stretch (at least that should stretch by default).  Ellipsis
> is not on that list.  I think it would be useful to hear from users as to
> what characters they would like to see stretchable, even if they don't
> stretch by default.
>
> It should be understood that what's wanted here is not a mere
> stretching-out of the three dots in U+2026, but an arbitrary length row of
> dots (something along the lines of using U+2026 as an 'extender character').
>
> But, yes, that would be the idea.
>
> Jacques