Vit Zyka wrote:
> Firstly, let us show some examples:
> - transparency frames
> ftp://cmp.felk.cvut.cz/pub/cmp/users/zyka/vg-ex/vg05-ex-body1.pdf
> - two columnsets + marginals + figures + balancing
> ftp://cmp.felk.cvut.cz/pub/cmp/users/zyka/vg-ex/vg05-ex-body2.pdf
> - two columnsets + page float
> ftp://cmp.felk.cvut.cz/pub/cmp/users/zyka/vg-ex/vg05-ex-body3.pdf
> - three columnsets + floats
> ftp://cmp.felk.cvut.cz/pub/cmp/users/zyka/vg-ex/vg05-ex-encyk.pdf
> - content
> ftp://cmp.felk.cvut.cz/pub/cmp/users/zyka/vg-ex/vg05-ex-content.pdf
> - index
> ftp://cmp.felk.cvut.cz/pub/cmp/users/zyka/vg-ex/vg05-ex-index.pdf
>
nice; how about an article on such a project for the maps/tugboat
> Comments:
> > grid
> ? if we want to do avoiding orphans and widows automatically,
> we would probably need the glue with DISCRETE STEP of
> stretch/shrink.
> Then we can say e.g.
> \let\bl=\baselineskip
> \setuphead
> [section]
> [before=\vskip1\bl plus1\bl withstep1\bl]
> to get vertical space before section head exactly 1 OR 2
> \baselineskip height.
> Another solution with stepwise box:
> \vbox to2\bl plus2\bl withstep1\bl{\vfil heading \vfil}
>
hm, worth investigating but maybe wait for what pdftex will bring some
day soon
> ! the only problem was in the table of contents + columns;
> from the first line there was a small shift between columns;
> perhaps \splittopskip? but setting it to large enough value
> does not solve it.
>
could be; actually i wish i'd set topskip to strutheight by default
which would make life way easier
>
> after Hans improvement seems that some floats soes not dissappered
> now, but need to be carefull
>
i think that indeed some floats get lost bugs were solved but it's
always a bit tricky with left-overs
> + floats insertion with specifying place
> ? for top float: automatically setting \placeongrid[{broad,high}]
> for bottom float \placeongrid[{broad,low}]
> to be vertically align with text on the rest of column(s)
>
there is some control possible, so remind me when you need it again
(\OTRSETprepareforcolumnslot permits log/middle/high handling but it's
not keyword driven (yet)
> * balancing: only manually by setting:
> \setupcolumnsetlines[columnset_id][columnset_page][column][line]
> ! Not so painful if it not occures next: If there are many
> floats near balancing place they sometimes disappear or cause ignoring
> balancing - even if they can fit on the place.
> ? understand that automatical balancing with floats is extremly
> difficult
>
indeed
> * page floats: big improvements during the work.
> ! running marginal text continue through the page with only float
> = so you have to manually insert blank vertical margin on such place
> * header & footers: Hans did a new column marking:
> \getmarking[marking_id][column][last/first].
> ! There are some problems [perhaps with expansion (different
> counter number and text)] if marking goes between two columnsets (framed
> chapters cross two cols in examples).
>
hm, some day ...
> = found acceptable hack but need more ivestigation on small
> examples for cleaning.
>
ok
> > metapost inclusion
> + the most powerful, smooth and stable part of ConTeXt in my opinion
> (see random frames, circles with transparent graphics, lines with color
> shading)
>
> = if you have many (over 1024!) shadings (-> MP specials) do
> \settrue \manyMPspecials
>
this is the default now (today i played with persistent mp runs and that
will speed up runtime processing enourmously)
>
>> index (register)
>>
> + nice possibility to differentiate several groups (here 8 from which
> 4 have different visual marking of page number, see index example)
> - sorting: until this spring there was no satisfied backround for
> Czech sorting in newtexutil.rb till this spring.
> = I did dirty reimplementation of old texutil.pl for doing so
> * I have to have a look for new sorting now.
> - coupling: page numbers with group preserving (see example)
> = also in texutil.pl reimplementation
>
> > regime
> * my great discovery was that the regime IS
> \enableregime{latin2} % correct
> and NOT
> \enableregime{il2} % incorrect!
>
well, il2 is actually a kind of hack: an incomplete, usable for csr
only, encoding
> > pseudo-small caps
> = http://wiki.contextgarden.net/Pseudo_Small_Caps
>
> > ...
>
> I provide the environment files (design definition) in case of
> somebody's interest. But suddenly they are very very very dirty.
>
ha, look at some of my project files, patch upon patch (esp when
designers start getting inconsistent )
> Partially due to my unsufficient ConTeXt knowledge, partially due to my
> despair during trying many ways to solve some problem, partially due to
> putting pathes to old ConTeXt (it was fresh of course when the work
> begins), ...:
> ftp://cmp.felk.cvut.cz/pub/cmp/users/zyka/vg-ex/vg05-env.tex (core)
> ftp://cmp.felk.cvut.cz/pub/cmp/users/zyka/vg-ex/vg05-mark.tex (markup)
>
well, given that at that time columnsets were even more experimental, you did
quite well
-----------------------------------------------------------------
Hans Hagen | PRAGMA ADE
Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com
| www.pragma-pod.nl
-----------------------------------------------------------------
_______________________________________________
ntg-context mailing list
ntg-context@ntg.nl
http://www.ntg.nl/mailman/listinfo/ntg-context
