groff
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [Groff] Typesetting sqrt with geqn


From: Ted Harding
Subject: RE: [Groff] Typesetting sqrt with geqn
Date: Tue, 12 Feb 2002 08:34:10 -0000 (GMT)

On 11-Feb-02 Gary E. RAFE, PhD wrote:
> Using v 1.17.2, consider the following:
> 
> .EQ
> sqrt { x }
> .EN
> 
> I'm finding that the "radicalex" character is offset to the right
> of the radical sign by the width of the "radicalex" character when
> no size changes are made to the radical sign.
> 
> Setting this "by-hand" seems to produce a correct result:
> 
> \(sr\h^G-\w'\[radicalex]'u^G\l'XXu\[radicalex]'
> 
> where XX is the length of the expression under the radical bar.
> 
> Can this be confirmed ?
> Is there any way to get geqn to typeset $sqrt { x }$ correctly ?!

The following may have nothing to do with the specific problem
you are describing, but it is worth bearing in mind.

Which groff device are you using (ps, dvi, ...?) for output? And
what are you using to physically print/display it?

There's a curious feature to the way eqn/groff does this, namely
that the horizontal extender is printed "absolutely" and not
relative to the radical sign itself. I.e. the absolute position
of the top of the radical sign is computed; this is then used
in a "moveto" instruction to position the start of the horizontal
extender (and the extender is actually printed before the
radical sign!). Provided the font metrics used by the physical
display or the physical printer match those used by groff in
computing this position, it will be OK. But if they are different,
the extender will be offset relative to the radical sign (because
the radical sign will be in the wrong place).

I used to get this (ages ago) with old versions of ghostscript,
though I have not seen it happen for recent ones.

If you are using ghostscript as a component of your print
spool filter (i, order to print groff's PostScript output
to a non-PS printer), and your ghostscript has this font
metrics problem, then this could to happen.

Ted.

--------------------------------------------------------------------
E-Mail: (Ted Harding) <address@hidden>
Fax-to-email: +44 (0)870 167 1972
Date: 12-Feb-02                                       Time: 08:34:10
------------------------------ XFMail ------------------------------

reply via email to

[Prev in Thread] Current Thread [Next in Thread]