Difference difference

C R

23 Apr 23 Apr

6:05 p.m.

Could not duplicate the error, however if you have a stroke applied to your shapes that is the same colour as your fill, that may explain what we are seeing here. :)

Boolean shape ops are done to fills only, and do not take into account line thickness on the resulting paths (this is intentional).

-C

On Sat, Apr 23, 2016 at 6:21 PM, Brynn <brynn@...3089...> wrote:

...

Hi Friends, It seems like I should know the answer to this. But I don't recall hearing about it. Starting with these 2 rings, which have fill and no stroke: http://inkscapecommunity.com/jd/ss/diff1.png I duplicate both, and then Difference. At 100% zoom they look ok. But zoom in (in this case, all the way, for best view) and you can see that the Difference wasn't correct. http://inkscapecommunity.com/jd/ss/diff2.png It's not a huge amount, but I would expect there shouldn't be any difference at all. This on Windows 7, 64-bit, sp1, Inkscape 0.91, 32-bit, exe installer. Is that a bug? Or is there some reason it's expected not to be precise?

Thanks for your help :-) brynn

Find and fix application performance issues faster with Applications Manager Applications Manager provides deep performance insights into multiple tiers of your business applications. It resolves application problems quickly and reduces your MTTR. Get your free trial! https://ad.doubleclick.net/ddm/clk/302982198;130105516;z _______________________________________________ Inkscape-user mailing list Inkscape-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-user

Reply

ugajin

3:09 p.m.

There are a couple of things.

All the paths are drawn clockwise. If you run Union on each ring before executing Difference you will get a different result. Not necessrilt better!

If however you break apart the each ring and reverse the direction of one component path from each pair, and then combine, before running Difference you will I think get the result you are looking for.

-u

---- On Sun, 24 Apr 2016 14:19:27 +0000 Brynn <brynn@...3089...> wrote ----

Thanks C R.

That was the first thing I thought too, at first. The bottom object must have a stroke. But as I said, they are both filled with no stroke.

Here's link to SVG file: http://inkscapecommunity.com/jd/imgs/diff-diff.svg . It has the rings already duplicated. So to reproduce, select the top green and top gold rings, and do Difference.

Thanks again :-) brynn

___________________________________ From: C R Sent: Saturday, April 23, 2016 12:05 PM To: Inkscape User Community Subject: Re: [Inkscape-user] Difference difference

Could not duplicate the error, however if you have a stroke applied to your shapes that is the same colour as your fill, that may explain what we are seeing here. :)

Boolean shape ops are done to fills only, and do not take into account line thickness on the resulting paths (this is intentional).

-C

On Sat, Apr 23, 2016 at 6:21 PM, Brynn <brynn@...3089...> wrote:

Hi Friends, It seems like I should know the answer to this. But I don't recall hearing about it. Starting with these 2 rings, which have fill and no stroke: http://inkscapecommunity.com/jd/ss/diff1.png I duplicate both, and then Difference. At 100% zoom they look ok. But zoom in (in this case, all the way, for best view) and you can see that the Difference wasn't correct. http://inkscapecommunity.com/jd/ss/diff2.png It's not a huge amount, but I would expect there shouldn't be any difference at all. This on Windows 7, 64-bit, sp1, Inkscape 0.91, 32-bit, exe installer. Is that a bug? Or is there some reason it's expected not to be precise?

Thanks for your help :-) brynn

------------------------------------------------------------------------------ Find and fix application performance issues faster with Applications Manager Applications Manager provides deep performance insights into multiple tiers of your business applications. It resolves application problems quickly and reduces your MTTR. Get your free trial! https://ad.doubleclick.net/ddm/clk/302982198;130105516;z _______________________________________________ Inkscape-user mailing list Inkscape-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-user

------------------------------------------------------------------------------ Find and fix application performance issues faster with Applications Manager Applications Manager provides deep performance insights into multiple tiers of your business applications. It resolves application problems quickly and reduces your MTTR. Get your free trial! https://ad.doubleclick.net/ddm/clk/302982198;130105516;z

_______________________________________________ Inkscape-user mailing list Inkscape-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-user

------------------------------------------------------------------------------ Find and fix application performance issues faster with Applications Manager Applications Manager provides deep performance insights into multiple tiers of your business applications. It resolves application problems quickly and reduces your MTTR. Get your free trial! https://ad.doubleclick.net/ddm/clk/302982198;130105516;z _______________________________________________ Inkscape-user mailing list Inkscape-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-user

Reply

C R

3:27 p.m.

Heh, interesting! I'm getting a new and different offset error in inkscape trunk. It does not appear to be a problem with the difference boolean op, (fortunately) but rather the position of the result is off by a little.

I'm not sure what causes this, but I suspect it's a roundoff error to do with the precision. To fix, simply erase the shapes directly behind your result, or use snapping to snap the

It is a bug though. Whether it's one that can be easily fixed, I have no idea. :)

I say file a bug report for it.

My 2p. -C

On Sun, Apr 24, 2016 at 3:19 PM, Brynn <brynn@...3089...> wrote:

...

Thanks C R.

That was the first thing I thought too, at first. The bottom object must have a stroke. But as I said, they are both filled with no stroke.

Here's link to SVG file: http://inkscapecommunity.com/jd/imgs/diff-diff.svg . It has the rings already duplicated. So to reproduce, select the top green and top gold rings, and do Difference.

Thanks again :-) brynn

From: C R Sent: Saturday, April 23, 2016 12:05 PM To: Inkscape User Community Subject: Re: [Inkscape-user] Difference difference

Could not duplicate the error, however if you have a stroke applied to your shapes that is the same colour as your fill, that may explain what we are seeing here. :)

Boolean shape ops are done to fills only, and do not take into account line thickness on the resulting paths (this is intentional).

-C

On Sat, Apr 23, 2016 at 6:21 PM, Brynn <brynn@...3089...> wrote:

Hi Friends, It seems like I should know the answer to this. But I don't recall hearing about it. Starting with these 2 rings, which have fill and no stroke: http://inkscapecommunity.com/jd/ss/diff1.png I duplicate both, and then Difference. At 100% zoom they look ok. But zoom in (in this case, all the way, for best view) and you can see that the Difference wasn't correct. http://inkscapecommunity.com/jd/ss/diff2.png It's not a huge amount, but I would expect there shouldn't be any difference at all. This on Windows 7, 64-bit, sp1, Inkscape 0.91, 32-bit, exe installer. Is that a bug? Or is there some reason it's expected not to be precise?

Thanks for your help :-) brynn

Find and fix application performance issues faster with Applications Manager Applications Manager provides deep performance insights into multiple tiers of your business applications. It resolves application problems quickly and reduces your MTTR. Get your free trial! https://ad.doubleclick.net/ddm/clk/302982198;130105516;z _______________________________________________ Inkscape-user mailing list Inkscape-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-user

Find and fix application performance issues faster with Applications Manager Applications Manager provides deep performance insights into multiple tiers of your business applications. It resolves application problems quickly and reduces your MTTR. Get your free trial! https://ad.doubleclick.net/ddm/clk/302982198;130105516;z

Inkscape-user mailing list Inkscape-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-user

Find and fix application performance issues faster with Applications Manager Applications Manager provides deep performance insights into multiple tiers of your business applications. It resolves application problems quickly and reduces your MTTR. Get your free trial! https://ad.doubleclick.net/ddm/clk/302982198;130105516;z _______________________________________________ Inkscape-user mailing list Inkscape-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-user

Reply

Brynn

5:58 p.m.

Hi C R, I just realized something. I use a custom default doc with landscape orientation. Could this be related to the transform attribute issues? I was one second away from posting the bug report when I thought of it. Do you think it should still be posted, given the new info about custom default doc? Or would that explain it?

Btw, ugajin is right about reversing the subpaths. It improves the situation, resulting in less overlap. But there's still a tiny bit of overlap.

(And C R, in my previous message about positioning, I think I probably misunderstood you. Both rings still sit at the same x and y coordinates as before the Difference. So nevermind about that comment :-))

Thanks, brynn _____________________________ From: C R Sent: Sunday, April 24, 2016 9:27 AM To: Inkscape User Community Subject: Re: [Inkscape-user] Difference difference

Heh, interesting! I'm getting a new and different offset error in inkscape trunk. It does not appear to be a problem with the difference boolean op, (fortunately) but rather the position of the result is off by a little.

I'm not sure what causes this, but I suspect it's a roundoff error to do with the precision. To fix, simply erase the shapes directly behind your result, or use snapping to snap the

It is a bug though. Whether it's one that can be easily fixed, I have no idea. :)

I say file a bug report for it.

My 2p. -C

On Sun, Apr 24, 2016 at 3:19 PM, Brynn <brynn@...3089...> wrote:

Thanks C R.

That was the first thing I thought too, at first. The bottom object must have a stroke. But as I said, they are both filled with no stroke.

Here's link to SVG file: http://inkscapecommunity.com/jd/imgs/diff-diff.svg . It has the rings already duplicated. So to reproduce, select the top green and top gold rings, and do Difference.

Thanks again :-) brynn

___________________________________ From: C R Sent: Saturday, April 23, 2016 12:05 PM To: Inkscape User Community Subject: Re: [Inkscape-user] Difference difference

Could not duplicate the error, however if you have a stroke applied to your shapes that is the same colour as your fill, that may explain what we are seeing here. :)

Boolean shape ops are done to fills only, and do not take into account line thickness on the resulting paths (this is intentional).

-C

On Sat, Apr 23, 2016 at 6:21 PM, Brynn <brynn@...3089...> wrote:

Hi Friends, It seems like I should know the answer to this. But I don't recall hearing about it. Starting with these 2 rings, which have fill and no stroke: http://inkscapecommunity.com/jd/ss/diff1.png I duplicate both, and then Difference. At 100% zoom they look ok. But zoom in (in this case, all the way, for best view) and you can see that the Difference wasn't correct. http://inkscapecommunity.com/jd/ss/diff2.png It's not a huge amount, but I would expect there shouldn't be any difference at all. This on Windows 7, 64-bit, sp1, Inkscape 0.91, 32-bit, exe installer. Is that a bug? Or is there some reason it's expected not to be precise?

Thanks for your help :-) brynn

------------------------------------------------------------------------------ Find and fix application performance issues faster with Applications Manager Applications Manager provides deep performance insights into multiple tiers of your business applications. It resolves application problems quickly and reduces your MTTR. Get your free trial! https://ad.doubleclick.net/ddm/clk/302982198;130105516;z _______________________________________________ Inkscape-user mailing list Inkscape-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-user

------------------------------------------------------------------------------ Find and fix application performance issues faster with Applications Manager Applications Manager provides deep performance insights into multiple tiers of your business applications. It resolves application problems quickly and reduces your MTTR. Get your free trial! https://ad.doubleclick.net/ddm/clk/302982198;130105516;z

_______________________________________________ Inkscape-user mailing list Inkscape-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-user

------------------------------------------------------------------------------ Find and fix application performance issues faster with Applications Manager Applications Manager provides deep performance insights into multiple tiers of your business applications. It resolves application problems quickly and reduces your MTTR. Get your free trial! https://ad.doubleclick.net/ddm/clk/302982198;130105516;z _______________________________________________ Inkscape-user mailing list Inkscape-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-user

------------------------------------------------------------------------------ Find and fix application performance issues faster with Applications Manager Applications Manager provides deep performance insights into multiple tiers of your business applications. It resolves application problems quickly and reduces your MTTR. Get your free trial! https://ad.doubleclick.net/ddm/clk/302982198;130105516;z

_______________________________________________ Inkscape-user mailing list Inkscape-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-user

Reply

ugajin

4 May 4 May

10:53 p.m.

Finally found an answer to this. Do you still need a solution?

-u

---- On Sun, 24 Apr 2016 20:33:31 +0000 ugajin <ugajin@...3218...> wrote ----

I hadn't noticed the small remaining overlap as it was hidden by the stacking order.

-u

---- On Sun, 24 Apr 2016 17:58:39 +0000 Brynn<brynn@...3089...> wrote ----

Hi C R, I just realized something. I use a custom default doc with landscape orientation. Could this be related to the transform attribute issues? I was one second away from posting the bug report when I thought of it. Do you think it should still be posted, given the new info about custom default doc? Or would that explain it?

Btw, ugajin is right about reversing the subpaths. It improves the situation, resulting in less overlap. But there's still a tiny bit of overlap.

(And C R, in my previous message about positioning, I think I probably misunderstood you. Both rings still sit at the same x and y coordinates as before the Difference. So nevermind about that comment :-))

Thanks, brynn _____________________________ From: C R Sent: Sunday, April 24, 2016 9:27 AM To: Inkscape User Community Subject: Re: [Inkscape-user] Difference difference

Heh, interesting! I'm getting a new and different offset error in inkscape trunk. It does not appear to be a problem with the difference boolean op, (fortunately) but rather the position of the result is off by a little.

I'm not sure what causes this, but I suspect it's a roundoff error to do with the precision. To fix, simply erase the shapes directly behind your result, or use snapping to snap the

It is a bug though. Whether it's one that can be easily fixed, I have no idea. :)

I say file a bug report for it.

My 2p. -C

On Sun, Apr 24, 2016 at 3:19 PM, Brynn <brynn@...3089...> wrote:

Thanks C R.

That was the first thing I thought too, at first. The bottom object must have a stroke. But as I said, they are both filled with no stroke.

Here's link to SVG file: http://inkscapecommunity.com/jd/imgs/diff-diff.svg . It has the rings already duplicated. So to reproduce, select the top green and top gold rings, and do Difference.

Thanks again :-) brynn

___________________________________ From: C R Sent: Saturday, April 23, 2016 12:05 PM To: Inkscape User Community Subject: Re: [Inkscape-user] Difference difference

Could not duplicate the error, however if you have a stroke applied to your shapes that is the same colour as your fill, that may explain what we are seeing here. :)

Boolean shape ops are done to fills only, and do not take into account line thickness on the resulting paths (this is intentional).

-C

On Sat, Apr 23, 2016 at 6:21 PM, Brynn <brynn@...3089...> wrote:

Hi Friends, It seems like I should know the answer to this. But I don't recall hearing about it. Starting with these 2 rings, which have fill and no stroke: http://inkscapecommunity.com/jd/ss/diff1.png I duplicate both, and then Difference. At 100% zoom they look ok. But zoom in (in this case, all the way, for best view) and you can see that the Difference wasn't correct. http://inkscapecommunity.com/jd/ss/diff2.png It's not a huge amount, but I would expect there shouldn't be any difference at all. This on Windows 7, 64-bit, sp1, Inkscape 0.91, 32-bit, exe installer. Is that a bug? Or is there some reason it's expected not to be precise?

Thanks for your help :-) brynn

------------------------------------------------------------------------------ Find and fix application performance issues faster with Applications Manager Applications Manager provides deep performance insights into multiple tiers of your business applications. It resolves application problems quickly and reduces your MTTR. Get your free trial! https://ad.doubleclick.net/ddm/clk/302982198;130105516;z _______________________________________________ Inkscape-user mailing list Inkscape-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-user

------------------------------------------------------------------------------ Find and fix application performance issues faster with Applications Manager Applications Manager provides deep performance insights into multiple tiers of your business applications. It resolves application problems quickly and reduces your MTTR. Get your free trial! https://ad.doubleclick.net/ddm/clk/302982198;130105516;z

_______________________________________________ Inkscape-user mailing list Inkscape-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-user

------------------------------------------------------------------------------ Find and fix application performance issues faster with Applications Manager Applications Manager provides deep performance insights into multiple tiers of your business applications. It resolves application problems quickly and reduces your MTTR. Get your free trial! https://ad.doubleclick.net/ddm/clk/302982198;130105516;z _______________________________________________ Inkscape-user mailing list Inkscape-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-user

------------------------------------------------------------------------------ Find and fix application performance issues faster with Applications Manager Applications Manager provides deep performance insights into multiple tiers of your business applications. It resolves application problems quickly and reduces your MTTR. Get your free trial! https://ad.doubleclick.net/ddm/clk/302982198;130105516;z

_______________________________________________ Inkscape-user mailing list Inkscape-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-user

------------------------------------------------------------------------------ Find and fix application performance issues faster with Applications Manager Applications Manager provides deep performance insights into multiple tiers of your business applications. It resolves application problems quickly and reduces your MTTR. Get your free trial! https://ad.doubleclick.net/ddm/clk/302982198;130105516;z _______________________________________________ Inkscape-user mailing list Inkscape-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-user

------------------------------------------------------------------------------ Find and fix application performance issues faster with Applications Manager Applications Manager provides deep performance insights into multiple tiers of your business applications. It resolves application problems quickly and reduces your MTTR. Get your free trial! https://ad.doubleclick.net/ddm/clk/302982198;130105516;z____________________... Inkscape-user mailing list Inkscape-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-user

Reply

ugajin

10:29 a.m.

I am surprised, you were satisfied with the work around. The 'answer' turns out to be one those things that seem obvious after the fact. If you double the number of node you get a better result! Stricly speaking you only need to double the node count for the intersecting segments. Also, it may or may not not surprise you, but you need to add nodes to both objects.

See grab, as proof of concept.

-u

---- On Sat, 07 May 2016 04:07:19 +0000 Brynn <brynn@...3089...> wrote ----

Hi u, No, I don't need a solution for the Difference problem. I already made it work using your first idea. But, I'm always interested to learn new things! What did you find?

brynn _________________________ From: ugajin Sent: Wednesday, May 04, 2016 4:53 PM To: Inkscape User Community Cc: Inkscape User Community Subject: Re: [Inkscape-user] Difference difference