# ExtendSrf display bug

**URL:** <https://discourse.mcneel.com/t/extendsrf-display-bug/31805>\
**Category:** Serengeti (Rhino BETA)\
**Created:** [April 24, 2016, 10:49pm UTC](https://discourse.mcneel.com/t/extendsrf-display-bug/31805 "2016-04-24T22:49:12Z")\
**Posts on this page:** 15\
**Page:** 1

<div class="post-metadata">

**Author:** ![clement](https://sea3.discourse-cdn.com/mcneel/user_avatar/discourse.mcneel.com/clement/32/759176_2.png) [@clement](https://discourse.mcneel.com/u/clement)\
**Post date:** [April 24, 2016, 10:49pm UTC](https://discourse.mcneel.com/t/extendsrf-display-bug/31805/1 "2016-04-24T22:49:12Z")

</div>

@pascal,

i’ve found one repeatable case where `_ExtendSrf` does not behave. If you extend the edge of the attached surface near point A and move the mouse cursor over point B, some display lines (curves) are shown which stay even when the command is canceled. It looks like this:

 ![](https://global.discourse-cdn.com/mcneel/uploads/default/original/3X/7/a/7ac434dd079847e6131635cafa0183eaef0fafbf.png)

Happens with this file: [ExtendBug.3dm](https://global.discourse-cdn.com/mcneel/uploads/default/original/3X/1/e/1e8f1e68b1d8d426e5c301b6aef7a367e9e4dcf4.3dm) (84.9 KB)

btw. if i `_Undo` it seems to cure it…

- 

c.

---

<div class="post-metadata">

**Author:** ![pascal](https://sea3.discourse-cdn.com/mcneel/user_avatar/discourse.mcneel.com/pascal/32/479395_2.png) [@pascal](https://discourse.mcneel.com/u/pascal)\
**Post date:** [April 25, 2016, 1:42am UTC](https://discourse.mcneel.com/t/extendsrf-display-bug/31805/2 "2016-04-25T01:42:19Z")

</div>

Hi Clement - I see that, thanks.

[http://mcneel.myjetbrains.com/youtrack/issue/RH-33768](http://mcneel.myjetbrains.com/youtrack/issue/RH-33768)

-Pascal

---

<div class="post-metadata">

**Author:** ![davide76](https://sea3.discourse-cdn.com/mcneel/user_avatar/discourse.mcneel.com/davide76/32/502826_2.png) [@davide76](https://discourse.mcneel.com/u/davide76)\
**Post date:** [April 25, 2016, 8:17am UTC](https://discourse.mcneel.com/t/extendsrf-display-bug/31805/3 "2016-04-25T08:17:19Z")

</div>

Pascal why this extension of the edge (a) takes place in this way (see right)?  
  
So it should be corrected:

---

<div class="post-metadata">

**Author:** ![wim](https://sea3.discourse-cdn.com/mcneel/user_avatar/discourse.mcneel.com/wim/32/747482_2.png) [@wim](https://discourse.mcneel.com/u/wim)\
**Post date:** [April 25, 2016, 8:42am UTC](https://discourse.mcneel.com/t/extendsrf-display-bug/31805/4 "2016-04-25T08:42:33Z")

</div>

> [@davide76](#):
>
> So it should be corrected

There is no artificial intelligence in Rhino and it doesn’t read minds. You are asking it to change one edge and so it tries to do so without changing that other edge. In your ‘correct’ solution, you have also changed that other edge.

---

<div class="post-metadata">

**Author:** ![Gijs](https://sea3.discourse-cdn.com/mcneel/user_avatar/discourse.mcneel.com/gijs/32/481553_2.png) [@Gijs](https://discourse.mcneel.com/u/Gijs)\
**Post date:** [April 25, 2016, 8:49am UTC](https://discourse.mcneel.com/t/extendsrf-display-bug/31805/5 "2016-04-25T08:49:06Z")

</div>

@wim I don’t agree here, in his example, the other edge did change as well. It’s just ‘extending’ the wrong direction.

take for example the attached and try to extend one of the rounded edges. Weird things will happen.[extendsrf.3dm](https://global.discourse-cdn.com/mcneel/uploads/default/original/3X/f/9/f985854a558b2d2020a23f352db266c9c1fd69cd.3dm) (319.9 KB)

---

<div class="post-metadata">

**Author:** ![wim](https://sea3.discourse-cdn.com/mcneel/user_avatar/discourse.mcneel.com/wim/32/747482_2.png) [@wim](https://discourse.mcneel.com/u/wim)\
**Post date:** [April 25, 2016, 8:51am UTC](https://discourse.mcneel.com/t/extendsrf-display-bug/31805/6 "2016-04-25T08:51:12Z")

</div>

> [@Gijs](#):
>
> It’s just ‘extending’ the wrong direction.

A bit quick on the trigger here - I stand corrected! 😊  
(thanks!)

---

<div class="post-metadata">

**Author:** ![davide76](https://sea3.discourse-cdn.com/mcneel/user_avatar/discourse.mcneel.com/davide76/32/502826_2.png) [@davide76](https://discourse.mcneel.com/u/davide76)\
**Post date:** [April 25, 2016, 9:17am UTC](https://discourse.mcneel.com/t/extendsrf-display-bug/31805/7 "2016-04-25T09:17:48Z")

</div>

I believe that if Pascal wants, should the developer to show this case, and in case correct it if possible.

---

<div class="post-metadata">

**Author:** ![davide76](https://sea3.discourse-cdn.com/mcneel/user_avatar/discourse.mcneel.com/davide76/32/502826_2.png) [@davide76](https://discourse.mcneel.com/u/davide76)\
**Post date:** [April 25, 2016, 9:30am UTC](https://discourse.mcneel.com/t/extendsrf-display-bug/31805/8 "2016-04-25T09:30:41Z")

</div>

I think the “Extrude srf” command should work properly in this way.  
For more selected edges, not just one at a time (do not know how it behaves in Rhino 6):  
  
The command, to be really convincing, it should work well. Useless to make turns of phrase. Rajaa should work to offer such solutions.  
A command that works in half, is not a good command, did well!

---

<div class="post-metadata">

**Author:** ![davide76](https://sea3.discourse-cdn.com/mcneel/user_avatar/discourse.mcneel.com/davide76/32/502826_2.png) [@davide76](https://discourse.mcneel.com/u/davide76)\
**Post date:** [April 25, 2016, 2:45pm UTC](https://discourse.mcneel.com/t/extendsrf-display-bug/31805/9 "2016-04-25T14:45:12Z")

</div>

Extend a trimmed surface is relatively simple. The problem occurs when the surface is trimmed, as in these simple examples.  
I think that the extension, at the end, is an offset operation.  
In these examples the only problem would be to get to the limit case, that is, up to the limit which is possible to extend a trimmed surface, i believe, until the extension is not self-intersecting.

---

<div class="post-metadata">

**Author:** ![Helvetosaur](https://sea3.discourse-cdn.com/mcneel/user_avatar/discourse.mcneel.com/helvetosaur/32/635974_2.png) [@Helvetosaur](https://discourse.mcneel.com/u/Helvetosaur)\
**Post date:** [April 25, 2016, 2:48pm UTC](https://discourse.mcneel.com/t/extendsrf-display-bug/31805/10 "2016-04-25T14:48:59Z")

</div>

Planar surfaces are always easy. The problem is more extending non-planar surfaces. --Mitch

---

<div class="post-metadata">

**Author:** ![davide76](https://sea3.discourse-cdn.com/mcneel/user_avatar/discourse.mcneel.com/davide76/32/502826_2.png) [@davide76](https://discourse.mcneel.com/u/davide76)\
**Post date:** [April 25, 2016, 3:13pm UTC](https://discourse.mcneel.com/t/extendsrf-display-bug/31805/11 "2016-04-25T15:13:07Z")

</div>

sure, this is true, but if the algorithm is good should work for srf not flat

---

<div class="post-metadata">

**Author:** ![davide76](https://sea3.discourse-cdn.com/mcneel/user_avatar/discourse.mcneel.com/davide76/32/502826_2.png) [@davide76](https://discourse.mcneel.com/u/davide76)\
**Post date:** [April 25, 2016, 3:14pm UTC](https://discourse.mcneel.com/t/extendsrf-display-bug/31805/12 "2016-04-25T15:14:58Z")

</div>

it is also true that at present, does not work so well even on flat surfaces, a trivial example is that I posted above (Rhino 6 I do not know …)

---

<div class="post-metadata">

**Author:** ![davidcockey](https://sea3.discourse-cdn.com/mcneel/user_avatar/discourse.mcneel.com/davidcockey/32/732604_2.png) [@davidcockey](https://discourse.mcneel.com/u/davidcockey)\
**Post date:** [April 26, 2016, 12:45am UTC](https://discourse.mcneel.com/t/extendsrf-display-bug/31805/13 "2016-04-26T00:45:36Z")

</div>

If you want to extend a surface by a set distance try this workflow:  
DupEdge the edge of the surface to be extended.  
ExtendSrf surface further than desired.  
OffsetCrvOnSrf to create an offset curve at the desired location of the new edge.  
Trim the extended surface using the offset curve.  
Delete the offset curve.

---

<div class="post-metadata">

**Author:** ![davide76](https://sea3.discourse-cdn.com/mcneel/user_avatar/discourse.mcneel.com/davide76/32/502826_2.png) [@davide76](https://discourse.mcneel.com/u/davide76)\
**Post date:** [May 15, 2016, 9:20am UTC](https://discourse.mcneel.com/t/extendsrf-display-bug/31805/14 "2016-05-15T09:20:57Z")

</div>

Rhino gives extensions problems in the trimmed surfaces.  
If you extend an edge of a surface with a trimming, it is logical that you change the board near!  
I think the extension code is inaccurate (in v6 do not know): Catia and Siemens Nx extend surfaces trimmed properly (not flat), also multiple edges at once!

---

<div class="post-metadata">

**Author:** ![davide76](https://sea3.discourse-cdn.com/mcneel/user_avatar/discourse.mcneel.com/davide76/32/502826_2.png) [@davide76](https://discourse.mcneel.com/u/davide76)\
**Post date:** [May 15, 2016, 9:23am UTC](https://discourse.mcneel.com/t/extendsrf-display-bug/31805/15 "2016-05-15T09:23:54Z")

</div>

Mitch, what you say is correct, but the problem is that Rhino does not extend properly surfaces trimmed even flat!