[tex-k] Bug#364505: dvips: the -T options doesn't always act as asked

Frank Küster frank at debian.org
Mon Apr 24 11:15:02 CEST 2006


Dear TeX-k people,

here's a bugreport against dvips, with a suggested fix.  I'd be thankful
if you could comment on this.  As you can see, the codebase is from
teTeX-3.0, or dvips(k) 5.95a, and it seems to be still the same in 5.95b
in TeXLive 2005.  But I haven't checked whether it's still that way in
any more current code.

Thanks in advance, Fullquote follows,

Frank

Vincent Fourmond <vincent.fourmond at 9online.fr> wrote:

> Package: tetex-bin
> Version: 3.0-16
> Severity: normal
>
>   Hello !
>
>   I work with a publishing house, and they use dvips to produce postscript versions of their books. 
> They did find recently that sometimes, providing a paper size with the -T option doesn't produce the 
> exact size asked.
>
>   For instance, the option -T17cm,25cm does actually produces something way closer to -T17.6cm,25cm. 
> This is coming directly from the following code in texk/dvipsk/output.c (findpapersize, around line 
> 1000)
>
>          for (ps=papsizes; ps; ps = ps->next) {
>             if (indelta(ps->xsize-hpapersize) &&
>                 indelta(ps->ysize-vpapersize)) {
>                landscape = 0 ;
>                break ;
>             }
>          }
>
>   Commenting out the break does solve the problem. Even though I can understand that it is comfortable 
> to specify an approximate paper size and having it automatically stick to the closest predefined size 
> available, it is way better to use it's name, and not to mess the things for people who really need 
> precise paper size, such as publishers. 
>
>   Here is what I propose:
>
>   * unless this feature has some important reasons to be, could it simply be removed ? (it wasn't 
>     present, as far as I know, in TeTex 2 series) 
>   * in any other case, would it be possible to provide a switch to turn it on or off at will ?
>
>   Thanks for considering this report !
>
> 	Vincent Fourmond
>
>
> -- System Information:
> (unnecessary)

-- 
Frank Küster
Single Molecule Spectroscopy, Protein Folding @ Inst. f. Biochemie, Univ. Zürich
Debian Developer (teTeX)



More information about the tex-k mailing list