[XeTeX] line spacing

[Ross Moore]

On 08/03/2005, at 4:02 AM, Yves Codet wrote:

> Yes, it would be handy indeed for fonts such as Zapfino.
>
> The following remark is for Ross. Jonathan advised me to write in 
> preambles:
>
> \catcode"00A0=\active
> \def^^^^00a0{\char32\relax}
>
> so that "no-break space" appears in the output. Could it be taken care 
> of by "xunicode"?

OK, nice idea.
But maybe not so easy to do this "properly & robustly"


There is a problem with just declaring this \catcode change globally.
Not all fonts (used with LaTeX) have a space in the \char32 .


You can do something like:

\def\activatenonbreakspace{\catcode"00A0=\active}
\def\deactivatenonbreakspace{\catcode"00A0=12\relax}
{\activatenonbreakspace\gdef^^^^00a0{\char32\relax}}

  ... which gives switches:
   \activatenonbreakspace
   \deactivatenonbreakspace
to switch on/off this feature.


Now you want this turned on for most Apple & Adobe fonts,
but turned off for Computer Modern, any Math fonts, Xy-pic fonts
and other traditional TeX fonts.

You also want it to happen automatically for the "right" fonts,
rather than putting it explicitly in every document.
In a sense this is an encoding issue.


Here's a first (very simple) attempt at getting the logic right:

If the font has 'unicode' encoding  (so 'U' at present)
then
   use  \activatenonbreakspace

otherwise  (e.g.  `T1', `OT1', `LY1', ... )
   use  \deactivatenonbreakspace .


I could try to hack at the \fontencoding command to implement
this. But firstly I'd need to know:

   What other fonts use `U' encoding ?
   What other encodings have the \char32  as a space ?



There is another way to tackle this, which is to keep
    \catcode^^^^00a0=\active
always, but expanding into a "robust" macro
   \nonbreakingspace , say.

Then alter the expansion of \nonbreakingspace
according to the \fontencoding , as discussed above.
(Essentially to  either \char32  when appropriate,
  or to  ~  when not.)


Indeed, I think it has to be done this way,
else you will not be able to use ^^^^00a0
within section headings where:

   a.  the document body uses Fancy fonts
         or some titles need them; but...
   b.  the Table-of-Contents  uses Computer Modern.

Similarly for any situation where LaTeX writes out
into a .aux, .toc, .idx, .lof, etc.  file,
expecting to reuse the character string within a
different context to how it was input.



Alternatively, why not use ~  in your documents,
the way it has always been done traditionally in TeX ?

The expansion of ~ should be linked to the encoding
(in the same way that xunicode does for other characters)
so as to produce the \char32 character as appropriate.


Yes, I think this is the *correct* LaTeX way to proceed:

  (i)  Examine how ~ currently works in the T1 encoding.
(ii)  mimic this for a robust macro \nonbreakingspace .
(iii) extend this action for `U' encoding => \char32
(iv)  activate ^^^^00a0
(v)   \let ~ \nonbreakingspace   (if advantageous)
(vi)  \let ^^^00a0\nonbreakingspace

That should result in a fully backwards-compatible
definition of ^^^^00a0 that will do the right thing
for auxiliary files, TofCs, indexes, etc.
(It also will ensure that ~ has this latter property too!)


Comments or improvements requested please,
before I try to implement this scheme.


>
> Thanks to all for your help.


Hope this helps,

	Ross


>
> Best wishes,
>
> Yves
>
> _______________________________________________
> XeTeX mailing list
> postmaster at tug.org
> http://tug.org/mailman/listinfo/xetex
>
>
------------------------------------------------------------------------
Ross Moore                                         ross at maths.mq.edu.au
Mathematics Department                             office: E7A-419
Macquarie University                               tel: +61 +2 9850 8955
Sydney, Australia                                  fax: +61 +2 9850 8114
------------------------------------------------------------------------