[ 
https://issues.apache.org/jira/browse/PDFBOX-2081?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14000744#comment-14000744
 ] 

Tilman Hausherr edited comment on PDFBOX-2081 at 5/17/14 11:28 AM:
-------------------------------------------------------------------

I have doubts whether my theory is the cause of the problem; I looked into the 
stream and there is "W n" many times. "W" defines which intersection method 
with the old clipping path is to be used, "n" is a dummy operator that is used 
to set the clipping region without painting/filling/stroking anything.

What I also noticed is that at 300dpi, the lines are rendered. The two files I 
mentioned at the beginning were rendered as part of my "diff tests" at 96dpi.


was (Author: tilman):
I have doubts whether my theory is the cause of the problem; I looked into the 
stream and there is "W n" many times. "n" is a dummy operator that is used to 
set the clipping region without painting/filling/stroking anything.

What I also noticed is that at 300dpi, the lines are rendered. The two files I 
mentioned at the beginning were rendered as part of my "diff tests" at 96dpi.

> Lines that exceeds clipping area are not drawn
> ----------------------------------------------
>
>                 Key: PDFBOX-2081
>                 URL: https://issues.apache.org/jira/browse/PDFBOX-2081
>             Project: PDFBox
>          Issue Type: Bug
>          Components: Rendering
>    Affects Versions: 2.0.0
>            Reporter: Juraj Lonc
>         Attachments: Obyčajné zásielky.pdf, rendered_(missing_lines).png, 
> rendered_(with_null_clipping).png
>
>
> PDF contains shapes that are partly on the paper and partly outside (shape 
> overflows paper borders).
> Those shapes are not rendered to image.
> It is caused by clipping area.
> When I replace line in PDFDrawer.strokePath()
> {noformat}
> graphics.setClip(getGraphicsState().getCurrentClippingPath());
> {noformat}
> to
> {noformat}
> graphics.setClip(null);
> {noformat}
> then everything is rendered correctly.
> Possibly bug in Java?



--
This message was sent by Atlassian JIRA
(v6.2#6252)

Reply via email to