MantisBT - Community
View Issue Details
0031131Community[OCCT] OCCT:Meshpublic2019-11-05 11:432020-12-02 17:12
kgv 
bugmaster 
normalminor 
closedfixed 
[OCCT] 7.4.0 
[OCCT] 7.5.0[OCCT] 7.5.0 
bugs/mesh/bug31131
0031131: [REGRESSION] Mesh - location of Wire affects Poly_Polygon3D
The following scripts creates a compound of unlocated Wire and translated Wire.
Poly_Polygon3D generated by TKMesh are unexpectedly misplaced.
This is a regression compared to OCCT 7.3.0.

Reference:
https://www.opencascade.com/content/confusing-occt-74-mesh-results-compound [^]
pload MODELING VISUALIZATION
vertex v1  0  0 0
vertex v2 10  0 0
vertex v3 10 10 0
edge e12 v1 v2
edge e23 v2 v3
wire w e12 e23
tcopy w w0
copy w w11
copy w w12
copy w w13
ttranslate w12 0 0 10
ttranslate w13 0 0 20
compound w11 w12 w13 c
incmesh c 1
vclear
vinit View1
vdefaults -autoTriang 0
explode w11 v
#vdisplay w11 w0 w11_1 w11_2 w11_3
vdisplay w11 w11_1 w11_2 w11_3
vfit

bounding w11 -save xmin ymin zmin xmax ymax zmax
checkpoint Min [list [dval xmin] [dval ymin] [dval zmin]] { -1 -1 -1 } 0.0001
checkpoint Max [list [dval xmax] [dval ymax] [dval zmax]] { 11 11  1 } 0.0001

Result: wire presentation doesn't match its vertices.

It seems that -extToler and -noTriangulation parameters of "bounding" command have no effect for computed box (in case of Wire), while should be.
No tags attached.
png wire_730_OK.png (1,317) 2019-11-05 11:45
https://tracker.dev.opencascade.org/
png wire_740_KO.png (1,041) 2019-11-05 11:45
https://tracker.dev.opencascade.org/
Issue History
2019-11-05 11:43kgvNew Issue
2019-11-05 11:43kgvAssigned To => oan
2019-11-05 11:44kgvDescription Updatedbug_revision_view_page.php?rev_id=22151#r22151
2019-11-05 11:45kgvFile Added: wire_730_OK.png
2019-11-05 11:45kgvFile Added: wire_740_KO.png
2019-11-05 14:34kgvSteps to Reproduce Updatedbug_revision_view_page.php?rev_id=22153#r22153
2019-11-05 16:18gitNote Added: 0088732
2019-11-05 16:21gitNote Added: 0088733
2019-11-05 18:24oanNote Added: 0088736
2019-11-05 18:24oanAssigned Tooan => msv
2019-11-05 18:24oanStatusnew => resolved
2019-11-06 09:48msvNote Added: 0088742
2019-11-06 09:48msvAssigned Tomsv => bugmaster
2019-11-06 09:48msvStatusresolved => reviewed
2019-11-06 09:55msvAssigned Tobugmaster => oan
2019-11-06 09:55msvStatusreviewed => assigned
2019-11-06 09:55msvNote Added: 0088743
2019-11-06 10:42kgvRelationship addedrelated to 0031077
2019-11-06 10:44kgvNote Added: 0088744
2019-11-06 10:45kgvNote Edited: 0088744bug_revision_view_page.php?bugnote_id=88744#r22159
2019-11-06 10:45kgvNote Edited: 0088744bug_revision_view_page.php?bugnote_id=88744#r22160
2019-11-06 10:51oanAssigned Tooan => msv
2019-11-06 10:51oanStatusassigned => resolved
2019-11-06 10:59msvNote Added: 0088746
2019-11-06 10:59msvAssigned Tomsv => bugmaster
2019-11-06 10:59msvStatusresolved => reviewed
2019-11-07 10:46bugmasterNote Added: 0088767
2019-11-07 10:46bugmasterStatusreviewed => tested
2019-11-07 10:47bugmasterTest case number => bugs/mesh/bug31131
2019-11-10 11:18bugmasterChangeset attached => occt master e9d05765
2019-11-10 11:18bugmasterStatustested => verified
2019-11-10 11:18bugmasterResolutionopen => fixed
2019-11-10 11:41gitNote Added: 0088836
2020-12-02 16:40emoFixed in Version => 7.5.0
2020-12-02 17:12emoStatusverified => closed

Notes
(0088732)
git   
2019-11-05 16:18   
Branch CR31131 has been created by oan.

SHA-1: 084f7642d9f14abf095be5129c21fa66457bb901


Detailed log of new commits:

Author: oan
Date: Tue Nov 5 16:13:38 2019 +0300

    0031131: [REGRESSION] Mesh - location of Wire affects Poly_Polygon3D
    
    Use TopoDS_Edge without location in case of free edges to produce Poly_Polygon3D without offset.
(0088733)
git   
2019-11-05 16:21   
Branch CR31131 has been updated forcibly by oan.

SHA-1: 038e2e7922c9884ebda8be9973e6040c0273234e
(0088736)
oan   
2019-11-05 18:24   
Please review the patch.

http://jenkins-test-12.nnov.opencascade.com/view/CR31131-master-OAN/view/COMPARE/ [^]
(0088742)
msv   
2019-11-06 09:48   
Reviewed.
(0088743)
msv   
2019-11-06 09:55   
Please explain differences in images in products tests "dxf read *".
(0088744)
kgv   
2019-11-06 10:44   
(edited on: 2019-11-06 10:45)
Before #0031077 most DXF tests have no screenshot dumps in 3D Viewer, so that regression was unnoticed.

I have not analyzed DXF models, but to me all changes look like exactly fixing the bug.

(0088746)
msv   
2019-11-06 10:59   
I suspected that it is improvement. Just wanted some confirmation of this.
(0088767)
bugmaster   
2019-11-07 10:46   
Tested in framework of WEEK-45

Combination -
OCCT branch : WEEK-45
master SHA - a1a9b249523438cf9b3d510fb4ff74bd4a5675b6
5f5b1aed1c6e139bbd34314eca77ae7abcd8895c
Products branch : WEEK-45 SHA - f88bc93e3637c4c42f3c154c0b0e167d4e297ae6
was compiled on Linux, MacOS and Windows platforms and tested in optimize mode.

Number of compiler warnings:
No new/fixed warnings

Regressions/Differences/Improvements:
No regressions/differences

CPU differences:
Debian80-64:
OCCT
Total CPU difference: 16844.70000000007 / 16848.430000000117 [-0.02%]
Products
Total CPU difference: 10837.970000000065 / 10873.690000000062 [-0.33%]
Windows-64-VC14:
OCCT
Total CPU difference: 18366.671875 / 18319.515625 [+0.26%]
Products
Total CPU difference: 12848.578125 / 12848.859375 [-0.00%]


Image differences :
No differences that require special attention

Memory differences :
No differences that require special attention
(0088836)
git   
2019-11-10 11:41   
Branch CR31131 has been deleted by inv.

SHA-1: 038e2e7922c9884ebda8be9973e6040c0273234e