triangle_mesh_create Fail

[kingui1106]

Hi. I meet some error, Is there any solution. I use 'triangle_mesh_create' return TRI_SEG_INTERSECT. I find blow in the code:TODO: Internal error: insertsegment().

[kingui1106]

https://drive.google.com/file/d/1v5fWCiZPEXfIzEeGfcVW_Rgmg1c3IAtV/view?usp=sharing

my input triangle

[wo80]

Please post your input polygon (as text, see below) and the Triangle options used.

Here's what I came up with:

# issue-43.poly
9 2 0 1
0  0.0 0.0  1
1  1.0 1.0  1
2  0.0 2.0  1
3  2.0 0.0  2
4  2.0 2.0  2
5  0.5 1.0  2
6  3.0 0.0  3
7  0.0 3.0  3
8  1.5 1.0  3
# segments
9 1
0  0 1  1
1  1 2  1
2  2 0  1
3  3 4  2
4  4 5  2
5  5 3  2
6  6 7  3
7  7 8  3
8  8 6  3
# holes
0

Triangulating with triangle-cli -pqz issue-43.poly works just fine.

[kingui1106]

Thanks. I've set all the data into triangleio in. I wander how can i write it to poly file

[wo80]

Then just post the code.

Compiling in debug mode (_DEBUG symbol present) might give you a hint whats going wrong.

[wo80]

I wander how can i write it to poly file

There's actually no api method for this. Might be worth adding one.

[wo80]

Just a guess: Are you numbering your nodes starting from 0 and forgot to set behavior->firstnumber = 0?

EDIT: this would probably not produce an error, just a warning and wrong mesh.

[kingui1106]

char* params = (char*)("jpzQYY");
int tristatus;
tristatus = triangle_context_options(ctx, params);

I use the params .It will set behavior->firstnumber = 0.

Just a guess: Are you numbering your nodes starting from 0 and forgot to set behavior->firstnumber = 0?

EDIT: this would probably not produce an error, just a warning and wrong mesh.

[kingui1106]

# issue-43.poly
18 2 1 0
0  0.5 1.0  1
1  2.0 2.0  1
2  2.0 0.0  1
3  1.5 1.0  1
4  3.0 0.0  1
5  0.0 3.0  1
6  0.0 0.0  1
7  1.0 1.0  1
8  0.0 2.0  1
9  1.1666666666666667  1.4444444444444444 1
10 1.3999999999999999 1.6000000000000001 1
11 2.00  0.66666666666666663 1
12 2.0000000000000000  1.0000000000000000  1
13  1.3833333333333333 1.5888888888888890 1
14 1.3499999999999994  1.5666666666666664 1
15 1.4333333333333338  1.5666666666666662 1
16  2.0000000000000000  0.76666666666666727 1
17  2.0000000000000000  1.5666666666666662 1
# segments
9 0
0  0 1 
1  1 2  
2  2 0  
3  3 4  
4  4 5  
5  5 3
6  6 7
7  7 8
8  8 6 
# holes
0

[wo80]

https://www.cs.cmu.edu/~quake/triangle.trouble.html

Strange things can happen if you've taken liberties with your PSLG. Do you have a vertex lying in the middle of a segment? Triangle sometimes copes poorly with that sort of thing. Do you want to lay out a collinear row of evenly spaced, segment-connected vertices? Have you simply defined one long segment connecting the leftmost vertex to the rightmost vertex, and a bunch of vertices lying along it? This method occasionally works, especially with horizontal and vertical lines, but often it doesn't, and you'll have to connect each adjacent pair of vertices with a separate segment. If you don't like it, tough.

[kingui1106]

OK!Thanks!!!

---Original--- From: @.> Date: Wed, Aug 4, 2021 18:21 PM To: @.>; Cc: @.**@.>; Subject: Re: [wo80/Triangle] triangle_mesh_create Fail (#43)

https://www.cs.cmu.edu/~quake/triangle.trouble.html

Strange things can happen if you've taken liberties with your PSLG. Do you have a vertex lying in the middle of a segment? Triangle sometimes copes poorly with that sort of thing. Do you want to lay out a collinear row of evenly spaced, segment-connected vertices? Have you simply defined one long segment connecting the leftmost vertex to the rightmost vertex, and a bunch of vertices lying along it? This method occasionally works, especially with horizontal and vertical lines, but often it doesn't, and you'll have to connect each adjacent pair of vertices with a separate segment. If you don't like it, tough.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.

[wo80]

You're welcome! Hope you find a way to handle the input.

If you think we're done here, please close the issue.

[kingui1106]

I think can add a small (1e-9) to the a vertex lying in the middle of a segment. Or, Break the line first with a vertex . I close the issue by accident. 😭

[kingui1106]

I test add a small (1e-9) to the a vertex lying in the middle of a segment. And then get the result triangles ，judge that three points are collinear by tolerance. It is userful and simple.