Re: wishes for lout

[P.J. Kersten]

Valeriy E. Ushakov wrote:
> 
> On Thu, Jan 15, 1998 at 09:52:30AM +0100, P.J. Kersten wrote:
> 
> > Mentioning whishes, is it possible to include 'table awareness' in lout?
> > This should include repeating headers, colored cells and random row and
> > column straddling.
> 
> Repeating headers are tough in general as object has no way to notice
> it was split into components.
> 

I am aware of that. That's why it's a wish. More to the point: the
algorithm which splits the objects _has_ notice. This is the point where
such a change could be adopted. I am also aware of the risk of damaging
lout's processing model. Perhaps it is time to add 'trigger procedures'
as n/t/groff has. In this environments one can express:
  at pagebottom -1 inch start macro ...
Ofcourse one has to agree about what a pagebottom actualy is in lout.
Using louts object model, it could be made possible to add
(standard)methods to objects which are triggered in a standard way, say
header(), footer() or something similar. If such methods are
overrideable one could add many features to lout compareable with the
best DTP-packages around.

> Adding colored cells to @Tab (or random cell beckground - to be more
> generic) should be possible.
> 
> Random spanning columns is tough with @Tab.  Spanning rows is not
> supported by @Tab (see recent discussion on the list).

So I noticed. Still this is the way most (DTP)tables are built (I hereby
deliberately ignore the 'normal' text-processing tasks). Lout is built
on top of PostScript and there are plenty of commercial DTP-packages
which can accomplish the straddling task with ease using PostScript.
Even is it is hard to accomplish, it would be an enrichment of lout if
such a job could be done. That's is why I mentioned it 'table
awareness'. Implementing a table as a special(=internal) object, one can
express it in (opt header, footer,) columns and rows. Lout can then
calculate and serve straddle-requests based on rows and columns instead
of being stuck with a almost non-related sequence of random objects
which can only be split and aligned.

> 
> If your Lout code is machine generated you might consider the
> possibility to abandon @Tab and use some custom replacement.  Note
> well that @Tab is written to provide users with convenient syntax.
> Not all things that are possible with Lout can be packaged to use some
> nice user friendly syntax, but machine generated Lout code is not
> constrained with user friendliness.  At the extreme, you can directly
> produce a sequence of concatenation operators.
> 

Don't forget that machine-generated code has to be (hand)written too
(just kidding). But seriously, you get stuck with the same lack of
splitting notice. Ofcourse one could make an image of a page in the
program itself and use lout as a form of high-level PostScript,
signaling table splitting at the source, but that would bypass similar
problems in handwritten documents. This can therefore only be used as a
work-around.

> > Is it possible to make a full-page background for ie. forms without
> > using EPS-graphics?
> 
> Doesn't @PageBackground option to @DocumentLayout do what you need?
> 

Perhaps. I will investigate it. Thanks.

> Hope it helps.
> 
> SY, Uwe
> --
> address@hidden                         |       Zu Grunde kommen
> http://www.ptc.spbu.ru/~uwe/            |       Ist zu Grunde gehen

-- 
Pieter J. Kersten
Erasmus Universiteit Rotterdam
dir. I&A/I&S, WH2-10
Postbus 1738
3000 DR  Rotterdam
Nederland
Tel +31-10-4082599
Fax +31-10-4527236