LATEX-L Archives

Mailing list for the LaTeX3 project

LATEX-L@LISTSERV.UNI-HEIDELBERG.DE

Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Condense Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Sender:
Mailing list for the LaTeX3 project <[log in to unmask]>
Subject:
From:
Frank Mittelbach <[log in to unmask]>
Date:
Sun, 14 Nov 1999 10:24:36 +0100
In-Reply-To:
Reply-To:
Mailing list for the LaTeX3 project <[log in to unmask]>
Parts/Attachments:
text/plain (117 lines)
James wrote:

> > my questions to this list would be now:
> >
> >  c) can you define the layout you are looking for?
> >     if not:
> >     - are there keys missing for the current templates?
> >     - are there other templates missing that do radically different things?
> >     - are the keys perhaps the wrong ones?
>
> One more key : number-indent =l [0pt] \parindent
> Perhaps it should be `first-line-indent'.  Used below.

yes, probably. and probably all keys should be a little bit more
uniform in names and bahviour, eg if we have right-margin-sep why
don't we have left-margin-sep?

> For the contents I am trying to simulate I seem to need both
> chapter and un-numbered-chapter instances. That would have implications
> for the templates which write the table of contents file. (I also need
> a demi-bold font, but thats by-the-by)

i strongly believe that we do something seriously wrong if something
like the design of the TOC has implications for the templates that
generate the toc entries. if we have such a dependency then i think we
need to consider modifying the contentsobject templates instead. after
all, i as a designer, want to change the layout of the TOC by changing
the instances for contentsobject but i don't want to be forced to
change the instances of `headings' at the same time. the main reason
for having a proper interface that passes along enough data was to
avoid such dependencies.

so the conclusion for me is that James' simulation of the Ktheory TOC
suggests that we do need a different type of contentsobject template.

since my feeling is that the current templates do nicely for a large
number of toc layouts i would not try to extend them directly to cater
for the case that the unnumbered object requires a very different
layout in comparison to the numbered object. Instead my proposal is to
add a `combination' template which allows that specification;
something like this:


 \DeclareTemplate{contentsobject}{combined}{6}{
   numbered-setup   =i {contentsobject} \contents@numbered,
   unnumbered-setup =i {contentsobject} \contents@unnumbered,
 }
 {
     \DoParameterAssignments
     \IfNoValueTF{#3}
        { \contents@unnumbered }
        { \contents@numbered }
     {#1}{#2}{#3}{#4}{#5}{#6}
 }


This template does nothing but decides (depending on whether or not a
number is present) which real template (or rather which instance) to
use. --- the name `combined' is very much rubbish: anybody having a
good idea how to call such types of templates; i'm sure that one will
see them more often in the future.

having this template the instances for James' simulation would then
look like this:

\DeclareCollectionInstance{Ktheory}{contentsobject}{chapter}{combined}
 {
   numbered-setup   = \UseTemplate{contentsobject}{JK}
        {pre-v-action  = \DelayEvaluation{
                          \addpenalty{10000}\addvspace{0.5ex plus 1pt}},
         number-format = \textbf{\hfil Chapter~#1.\nobreakspace},
         title-format  = \textbf{#1},
         pnum-format   = \textbf{\hfil#1},
         leaders-sep   = 100cm,
         object-indent = 2cm,
         number-indent = -2cm,
         number-width  = 2cm},
   unnumbered-setup = \UseTemplate{contentsobject}{JK}
         {pre-v-action = \DelayEvaluation{
                          \addpenalty{10000}\addvspace{1ex plus 1pt}},
          title-format = \textbf{#1},
          pnum-format  = \textbf{\hfil#1},
          leaders-sep  = 100cm},
 }

or assuming that the indivdual instances have been named:

\DeclareCollectionInstance{Ktheory}{contentsobject}{chapter}{combined}
 {
   numbered-setup   = numberedchapter,
   unnumbered-setup = unnumberedchapter,
 }


> >  e) how difficult is it to implement a given design?
>
> It's straightforward but fiddley,

is it? if so why? because things could be improved or because the
situation is complex and nothing could be done about it?

i would like to repeat my questions from the previous post (with one
additional question at the end):

 - can you define the layout you are looking for?
    if not:
    - are there keys missing for the current templates?
    - are there other templates missing that do radically different things?
    - are the keys perhaps the wrong ones?

 - share your thoughts and sample trials with us

 - do you have a good suggestion for the name of the above template
   `combined'?

frank

ATOM RSS1 RSS2