BoundingBox in PostScript output for A4 is too large and the co mputation with its comment are bogus

[Reinhard Kotucha]

On 2024-03-10 at 00:47:58 +0100, Zdenek Wagner wrote:

 > Hi,
 > 
 > in fact, the BoundingBox comment is not crucial. It is used if EPS is
 > included as graphics by a program which does not interpret the PS
 > code. If you want to generate a file to be sent to a phototypesetter
 > or to a proffesional printing machine, the boxes should be set in the
 > PS/PDF code and the %%BoundingBox is irrelevant. If you want to print
 > on A4 and the elements (photos, background) should touch the edge of
 > the trimmed page, it should extend outside the final page dimensions
 > because trimming is never pixel precise. You have to set TrimBox to
 > the size of the page after trimming, BleedBox to the size where the
 > material extends so that it goes till the edge of the trimed page, and
 > MediaBox for crop marks and other information for the DTP studio. My
 > zwpagelayout package does that.
  
Hi Zdeněk,
it's true that the BoundingBox is meaningless here except if you want
to include a page in your document (maybe a poster) within a tight
\fbox{}.  eps2pdf is using the BoundingBox it finds in order to
determine an appropriate /MediaBox for the PDF file.

But I think it's worthwhile to look into the code though.  The
BoundingBox and the /PageSize argument of the setpagedevice dictionary
should contain the same values.

  %%BoundingBox: 0 0 596 842
  << /PageSize [595 842] >> setpagedevice 

The /PageSize values are correct.  As far as I remember Tom added some
code a few years ago in order to snap to the nearest known paper size.

I told Karl that according to the PDF-spec the values of the /MediaBox
shouldn't differ by more than 5bp from the standard values so that
office printers can use this information to select an appropriate
paper tray.  Karl forwarded my mail to Tom and Tom's response was
quite positive.  I'm convinced that he implemented it and /page size
is always correct .

But I have one suggestion regarding the BoundingBox:

I think that it's quite useful to provide a HiResBoundingBox commnent
as well.  The latter allows real numbers instead of integers.  The
real numbers are already available before they are rounded.  Hence not
much work involved.  Maybe only an additional printf() call.

Regards,
  Reinhard




-- 
------------------------------------------------------------------
Reinhard Kotucha                            Phone: +49-511-3373112
Marschnerstr. 25
D-30167 Hannover                    mailto:reinhard.kotucha at gmx.de
------------------------------------------------------------------