[metapost] Bug: wrong bounding box with mptopdf?

[Oliver Buerschaper]

Dear MetaPost wizards,

there seems to be a bug with the bounding box handling in mptopdf …  
see this test case:

---
beginfig(1);
     draw fullsquare scaled 3cm;
endfig;

end
---


Log output:

---
MPtoPDF 1.3.2 : running 'texexec --mptex '
TeXExec | processing graphic 'boundingbox.mp'
This is MetaPost, Version 0.993 (Web2C 7.5.6)
(/usr/local/texlive/2007/texmf/web2c/natural.tcx)
(boundingbox.mp [1]
metafun 2007-12-1 11:43
  )
1 output file written: boundingbox.1
Transcript written on boundingbox.log.
TeXExec | runtime: 0.061077
This is pdfTeXk, Version 3.141592-1.40.3 (Web2C 7.5.6)
  %&-line parsing enabled.
[MP to PDF] (./boundingbox.1) [1]
Output written on boundingbox.pdf (1 page, 1330 bytes).
Transcript written on boundingbox.log.

MPtoPDF 1.3.2 : boundingbox is converted to boundingbox-1.pdf
---


PDF output:

-------------- next part --------------
A non-text attachment was scrubbed...
Name: boundingbox-1.pdf
Type: application/pdf
Size: 1330 bytes
Desc: not available
Url : http://tug.org/pipermail/metapost/attachments/20080813/e84bd58b/attachment.pdf 
-------------- next part --------------




Note how the top bit is chipped off … compare this problem with the  
one described in http://www.tug.org/pipermail/metapost/2007-June/000964.html 
  where TeX labels are cut off.

Best wishes,
Oliver