[reportlab-users] pb with images

[Tim Roberts]

On Wed, 14 Dec 2005 14:22:32 +0100, Jerome Alet <alet at librelogiciel.com> 
wrote:

>...If I linearize the code using the pycanvas framework included
>in ReportLab, I've got a 54 Mb python program (>41000 lines long)
>which in fact does (for each image) :
>
>--- CUT ---        
># Saves context level 1 {'_mitreLimit': 0, '_leading': 14.4, '_textLineMatrix': (1.0, 0.0, 0.0, 1.0, 0.0, 0.0), '_fontsize': 12, '_horizScale': 100, '_fillMode': 0, '_lineJoin': 0, '_charSpace': 0, '_strokeColorRGB': (0, 0, 0), '_y': 0, '_x': 0, '_dynamicFont': 0, '_textMatrix': (1.0, 0.0, 0.0, 1.0, 0.0, 0.0), '_textRenderMode': 0, '_textMode': 0, '_lineCap': 0, '_lineDash': None, '_currentMatrix': (1.0, 0.0, 0.0, 1.0, 42.519685039370074, 42.519685039370074), '_fontname': 'Times-Roman', '_lineWidth': 0, '_fillColorRGB': (0, 0, 0), '_wordSpace': 0, '_rise': 0}
>c.saveState()
>c.translate(219.40376884083801, 529.37537249388242)
>


I love computers.  Given that the matrix scale is in points, you are 
specifying the translation there with an accuracy down to one 
one-thousandth of the diameter of a proton.  I'm guessing that most of 
today's laser printers don't place toner at the sub-atomic level.

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.