Re: colored tables work... almost

[Hans Hagen <pragma@wxs.nl>]

W.H. Dekker wrote:

> So I replaced raster with color because the raster thing did not work,
> possible because I translated the keywords (background,
> backgroundraster, raster) wrongly. By the way: do you have a list of

                    screen

> english vs. dutch keywords somewhere?)

Look into file mult-con/mult-com or get the qrc docs from the ntg site,
these are cross linked (click on dutch and get english etc). 

It should work because I use it daily. Keep in mind that some
programs/printers (indeed also GS) just don't show light shades of gray. 

> - in your solution empty fields are considered 0, while they actually
>   stand for non-measured data and may not be replaced with 0. I tried to
>   solve this by testing if the cell has a single 0 (\ifx#1\leeg) but
>   this does not work, the single 0's are not recognized.

I suggest you take a look at syst-gen.tex (I can send you a typeset
documented listing if needed). You can use (and there is a lot of this
stuff there):

\doifelse{#1}{}
  {..}
  {..}

But: the check macro needs a value, so you can best put a * in each
column that does not have a value, and say

\doifelse{#1}{*}
  {..} {..}

An alternative solution would be:

\ifx#1\VL
  etc etc 

but I didn't test this. 

> - The framed numbers (both colored and white) are all shifted up with
>   the top of the frame against the top of the table-cell. So the first
>   column does not align with the number columns and the table becomes
>   longer than without coloring. I have shown this more explicitly by
>   adding some extra \HL's.

First you have to use \inframed instead of \framed (that one normalizes
the framed data to the lineheight. You also have some \.R wrong. 

> - The left vertical line of the cells containing "method" is broken in
>   the middle.

Indeed. I'll look into it. Seperate mail. 

> - Perhaps not important, but intriguing: I don't understand the
>   f{\Check}. According to the TaBlE manual this looks like a font
>   redefinition, but is it?

It sort of is, but such a font definition ends up in front of the data,
and what I do is picking up that data (space terminated). You could also
have let your perl script put 
\Check{..} into each column. 

> I hope you another few of those nice suggestions!

Well, just let me know. When you look in syst-gen, syst-ext, supp-box
you already see a lot of trickery. 

Hans 

-----------------------------------------------------------------
                                              Hans Hagen | PRAGMA
              Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
    tel: 038 477 53 69 | fax: 038 477 53 74 | mail: pragma@wxs.nl
-----------------------------------------------------------------