[
https://issues.apache.org/jira/browse/PDFBOX-5258?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17399168#comment-17399168
]
Tilman Hausherr edited comment on PDFBOX-5258 at 8/14/21, 12:35 PM:
--------------------------------------------------------------------
Not part of the upcoming commit but worth to mention; this moves the
intersectioning to the graphics device (which sadly makes it slower (tested on
my 1000 files), but makes the files from PDFBOX-4718 and PDFBOX-4743 faster)
{code}
protected void transferClip(PDGraphicsState graphicsState, Graphics2D
graphics)
{
List<Path2D.Double> paths = graphicsState.getCurrentClippingPaths();
boolean allDone = true;
for (Path2D.Double path : paths)
{
if (!path.getPathIterator(null).isDone())
{
allDone = false;
break;
}
}
if (allDone)
{
// PDFBOX-4821: avoid bug with java printing that empty clipping
path is ignored by
// replacing with empty rectangle, works because this is not an
empty path
graphics.setClip(new Rectangle());
return;
}
Path2D.Double first = paths.get(0);
graphics.setClip(first);
for (int i = 1; i < paths.size(); i++)
{
graphics.clip(paths.get(i));
}
}
{code}
was (Author: tilman):
Not part of the commit but worth to mention; this moves the intersectioning to
the graphics device (which sadly makes it slower (tested on my 1000 files), but
makes the files from PDFBOX-4718 and PDFBOX-4743 faster)
{code}
protected void transferClip(PDGraphicsState graphicsState, Graphics2D
graphics)
{
List<Path2D.Double> paths = graphicsState.getCurrentClippingPaths();
boolean allDone = true;
for (Path2D.Double path : paths)
{
if (!path.getPathIterator(null).isDone())
{
allDone = false;
break;
}
}
if (allDone)
{
// PDFBOX-4821: avoid bug with java printing that empty clipping
path is ignored by
// replacing with empty rectangle, works because this is not an
empty path
graphics.setClip(new Rectangle());
return;
}
Path2D.Double first = paths.get(0);
graphics.setClip(first);
for (int i = 1; i < paths.size(); i++)
{
graphics.clip(paths.get(i));
}
}
{code}
> Lazier clipping
> ---------------
>
> Key: PDFBOX-5258
> URL: https://issues.apache.org/jira/browse/PDFBOX-5258
> Project: PDFBox
> Issue Type: Improvement
> Components: Rendering
> Affects Versions: 2.0.24, 3.0.0 PDFBox
> Reporter: Tilman Hausherr
> Priority: Major
> Fix For: 2.0.25, 3.0.0 PDFBox
>
>
> From Stephan Schwiebert in linked PR:
> {quote}
> Calculating the intersection of two Area can take a lot of time. However,
> depending on the Graphics2D that is used for rendering, it may not be
> necessary to actually perform this operation.
> For instance, when generating an SVG, the individual clipping paths can be
> serialized individually, and the intersection is then calculated at runtime,
> when the SVG file is rendered.
> The idea of this PR is to replace PDGraphicsState.clippingPath with a list of
> GeneralPaths, which is lazily evaluated, truncated & cached when
> getCurrentClippingPath() is called (effectively leaving the current behaviour
> of PdfBox unchanged, and it should also not have any significant impact on
> the performance).
> Additionally, a new method protected void transferClip(PDGraphicsState
> graphicsState, Graphics2D graphics) is added to PageDrawer. By default, this
> method makes use of getCurrentClippingPath() to call graphics.setClip(...),
> which again is what PdfBox currently does.
> However, classes that extend PageDrawer can override this method, and
> directly access the individual clipping paths, without any need to calculate
> their intersection.
> In some cases (shading fills & transparency groups), it is still necessary to
> calculate the intersection.
> {quote}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
