[reportlab-users] About a bug in barcodes

Germán M. Bravo german.mb at gmail.com
Fri Sep 3 08:16:51 EDT 2010


On Sep 3, 2010, at 3:53 AM, Andy Robinson <andy at reportlab.com> wrote:

> On 3 September 2010 09:48, Robin Becker <robin at reportlab.com> wrote:

>> strokeWidth==0 is not the same as no stroke being drawn. In all rational

>> worlds it would be, but according to some person in postscript land zero

>> means the smallest width possible.

>

> I think there was a CAD person involved in the spec at some point in

> the past. I used to work with CAD systems and lines have different

> meanings. If it's supposed to be the 'edge' of a real world object,

> then it should still be drawn on the monitor with a one pixel line,

> however much you zoom in. If it's a shape with a finite width,

> you'll get what you expect.


I understand about the "smallest width" and the "edge" of the object... but other than in monitors, for some printers, the "smallest width" is just a stroke too wide, and the barcodes get all screwed up. The border of the bars should *not* exist at all, never, no matter what... so I think even taking out the barStrokeWidth and barStrokeColor attributes is not at all a completely wild thought.

Anyway, I'm not sure, 'cause the patch I sent you only covered EAN barcodes, but I suppose it's a good time for a check to see if the problem exists in other barcodes as well.

Best regards,
German.


More information about the reportlab-users mailing list