On 24/7/11 11:27, ~suv wrote:

On 23/7/11 23:20, Ivan Louette wrote:

...

Rev 10495 on kubuntu

When page shadow is disabled the horizontal line at the bottom of the page border disappears at launching.

If you disable and reactivate page border it comes back.

But if you scroll this line ouside the document window (up or down) it disappears again when you scroll back.

On Mac OS X 10.5.8 (i386) with both backends (GTK+/X11 2.24.4, cairo 1.10.2; GT+/Quartz 2.24.4, cairo 1.11.2):

• not reproduced with r10493
• reproduced with r10495

Additional artifacts/misrendering: The page border (without shadow) sometimes gets only partially redrawn when dragging an object near or across the page border at the right side or at the bottom:

http://imgh.us/page-border-incompletely-redrawn-r10497-osx.png

~suv

2011/7/24 SorinN <nemes.sorin@...400...>:

Confirmed with Ubuntu 11.0 ( Classic Desktop ) Inkscape r10495

This is caused by Johan's NRRectL -> Geom::IntRect rewrites, because IntRect contained a serious bug which I have recently discovered in my GSoC branch. Basically, an IntRect would not be contained in itself, and as a result neither the intersect() or contains() functions would work correctly. I committed an update of 2Geom that should fix this.

Regards, Krzysztof

Krzysztof,

Does your bug take into account his commit to fix the issue? (if you're not subscribed to commits for trunk, you may want to entertain the idea)

Cheers, Josh

2011/7/24 Krzysztof Kosiński <tweenk.pl@...400...>:

2011/7/24 SorinN <nemes.sorin@...400...>:

...

Confirmed with Ubuntu 11.0 ( Classic Desktop ) Inkscape r10495

This is caused by Johan's NRRectL -> Geom::IntRect rewrites, because IntRect contained a serious bug which I have recently discovered in my GSoC branch. Basically, an IntRect would not be contained in itself, and as a result neither the intersect() or contains() functions would work correctly. I committed an update of 2Geom that should fix this.

Regards, Krzysztof

---

Magic Quadrant for Content-Aware Data Loss Prevention Research study explores the data loss prevention market. Includes in-depth analysis on the changes within the DLP market, and the criteria used to evaluate the strengths and weaknesses of these DLP solutions. http://www.accelacomm.com/jaw/sfnl/114/51385063/ _______________________________________________ Inkscape-devel mailing list Inkscape-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-devel

On Jul 24, 2011, at 4:33 PM, Krzysztof Kosiński wrote:

This is caused by Johan's NRRectL -> Geom::IntRect rewrites, because IntRect contained a serious bug which I have recently discovered in my GSoC branch. Basically, an IntRect would not be contained in itself, and as a result neither the intersect() or contains() functions would work correctly. I committed an update of 2Geom that should fix this.

Cool!

Did you get unit tests set up on this? Aside from preventing regression, they do serve nicely as documentation for people looking at both the implementing and the using code.