MantisBT - Community
View Issue Details
0025104Community[OCCT] OCCT:Modeling Algorithmspublic2014-07-25 20:182018-06-29 21:19
shoogen 
bugmaster 
normalminor 
closedfixed 
Linux 64-bitDebian7.5
[OCCT] 6.7.1 
[OCCT] 7.3.0[OCCT] 7.3.0 
bugs modalg_7 bug25104
0025104: Prism from BSpline curve can not be chamfered
The shape returned by "prism" fails the bopcheck for an unknown reason.
If i use it for chamfering coord.Modulus() returns 0, which triggers a
"VectorWithNullMagnitude" exception, without a message.
pload ALL
restore whale2-draw-path.brep
renamevar whale2-draw-path path
prism Extrude001 path 0. 0. 5.
donly Extrude001
bopcheck Extrude001
BOPAlgo_OperationAborted
related to 0025879closed bugmaster Community result of blend fails the bopcheck 
related to 0025110new ifv Community Modeling Algorithms - TCollection_IndexedDataMap::FindFromKey error when sweeping circle along BezierCurve 
related to 0025693closed bugmaster Community Wire of BSplines fails bopcheck 
related to 0029316verified bugmaster Open CASCADE Test case bugs mesh bug25806 works unstable 
zip whale2.zip (7,139) 2014-07-25 20:18
https://tracker.dev.opencascade.org/
? pairs.brep (16,679) 2017-09-26 09:03
https://tracker.dev.opencascade.org/
Issue History
2014-07-25 20:18shoogenNew Issue
2014-07-25 20:18shoogenAssigned To => ifv
2014-07-25 20:18shoogenFile Added: whale2.zip
2014-07-31 14:19shoogenNote Added: 0030509
2014-08-12 20:15shoogenNote Added: 0030690
2014-09-11 18:54abvTarget Version => 7.1.0
2015-01-26 11:19shoogenTag Attached: BOPAlgo_OperationAborted
2016-10-28 11:54msvTarget Version7.1.0 => 7.2.0
2017-07-21 11:34msvTarget Version7.2.0 => 7.3.0
2017-08-02 17:20apvTest case number => bugs modalg_7 bug25104
2017-08-02 17:20apvNote Added: 0069060
2017-09-26 09:00emvRelationship addedrelated to 0025693
2017-09-26 09:01emvRelationship addedrelated to 0025879
2017-09-26 09:02emvNote Added: 0070835
2017-09-26 09:03emvNote Edited: 0070835bug_revision_view_page.php?bugnote_id=70835#r17722
2017-09-26 09:03emvFile Added: pairs.brep
2017-11-01 13:45amlNote Added: 0071910
2017-11-01 13:45amlAssigned Toifv => aml
2017-11-01 13:45amlStatusnew => assigned
2017-11-02 12:08gitNote Added: 0071926
2017-11-03 06:51gitNote Added: 0071967
2017-11-03 07:39gitNote Added: 0071968
2017-11-03 10:56amlNote Added: 0071972
2017-11-03 10:56amlAssigned Toaml => msv
2017-11-03 10:56amlStatusassigned => resolved
2017-11-08 17:37msvSeverityjust a question => minor
2017-11-08 17:54msvNote Added: 0072059
2017-11-08 17:54msvAssigned Tomsv => aml
2017-11-08 17:54msvStatusresolved => assigned
2017-11-10 14:08gitNote Added: 0072173
2017-11-11 10:48amlNote Added: 0072193
2017-11-11 10:53amlNote Added: 0072194
2017-11-11 10:53amlNote Edited: 0072193bug_revision_view_page.php?bugnote_id=72193#r18128
2017-11-13 10:26msvNote Added: 0072201
2017-11-13 10:34amlNote Added: 0072202
2017-11-13 10:34amlAssigned Toaml => msv
2017-11-13 10:34amlStatusassigned => resolved
2017-11-13 10:50msvNote Added: 0072203
2017-11-13 10:50msvAssigned Tomsv => bugmaster
2017-11-13 10:50msvStatusresolved => reviewed
2017-11-13 10:57amlRelationship addedrelated to 0029316
2017-11-13 11:48bugmasterNote Added: 0072209
2017-11-13 11:48bugmasterStatusreviewed => tested
2017-11-17 16:10bugmasterChangeset attached => occt master 90f0fdad
2017-11-17 16:10bugmasterStatustested => verified
2017-11-17 16:10bugmasterResolutionopen => fixed
2017-12-07 21:07gitNote Added: 0072780
2018-02-18 00:04abvRelationship addedrelated to 0025110
2018-06-29 21:15aivFixed in Version => 7.3.0
2018-06-29 21:19aivStatusverified => closed

Notes
(0030509)
shoogen   
2014-07-31 14:19   
It's seems to me that there is a general problem with BSplines and the mapping of edges and faces using FindFromKey.
IMHO this is related to 25110
(0030690)
shoogen   
2014-08-12 20:15   
i reproduced the actual filleting with DRAWEXE

> explode Extrude001 E
> blend Fillet Extrude001 1 Extrude001_4 1 Extrude001_7 1 Extrude001_10 1 Extrude001_13 1 Extrude001_16 1 Extrude001_19 1 Extrude001_22 1 Extrude001_25 1 Extrude001_27
returns
An exception was caught 0x7fb5791c5911 : gp_VectorWithNullMagnitude:
** Exception ** 0x7fb5791c5911 : gp_VectorWithNullMagnitude:
(0069060)
apv   
2017-08-02 17:20   
Problem described in issue is reproduced on current state of OCCT.
(0070835)
emv   
2017-09-26 09:02   
(edited on: 2017-09-26 09:03)
The bopcheck fails because intersection of some of the sub-shapes raises exception. Try to intersect the pairs in the attached compound:
restore pairs.brep pairs

foreach c [explode pairs] {
  copy $c pair
  explode pair
  bop pair_1 pair_2
}


(0071910)
aml   
2017-11-01 13:45   
Dear msv,
It seems that this issue has been fixed. I will take care of it.
(0071926)
git   
2017-11-02 12:08   
Branch CR25104 has been created by aml.

SHA-1: 7becd57195395ff0897a831f22c231c0fd81bf75


Detailed log of new commits:

Author: aml
Date: Thu Nov 2 12:08:18 2017 +0300

    0025104: Prism from BSpline curve can not be chamfered
    
    Protection from zero-length vectors has been added.
(0071967)
git   
2017-11-03 06:51   
Branch CR25104 has been updated by aml.

SHA-1: 35c5f26eb60ae554d5b90fb4945571f5c2c72a93


Detailed log of new commits:

Author: aml
Date: Fri Nov 3 06:51:21 2017 +0300

    # Test cases correction.

(0071968)
git   
2017-11-03 07:39   
Branch CR25104 has been updated forcibly by aml.

SHA-1: f4ded11eddd7047da182641c48b3b95360ce9f81
(0071972)
aml   
2017-11-03 10:56   
Dear msv,
Could you please check the current state of the CR25104 branch?
Corresponding Jenkins job is: http://jenkins-test-10.nnov.opencascade.com:8080/view/CR25104-master-AML/ [^]

The following cases are fixed under windows:
bug25104
bug25693_1
bug25693_2
(0072059)
msv   
2017-11-08 17:54   
Many test cases show decreased performance.
Could you analyze the reason on the example of the following tests?

CPU boolean bfuse_2d: 1.3416085999999998 / 0.7956051000000002 [+68.63%]
CPU boolean bopcut_simple: 20.38933070000004 / 17.830914299999993 [+14.35%]
CPU thrusection specific: 5.428834799999996 / 4.461628599999995 [+21.68%]
CPU bugs mesh bug25806: 36.7538356 / 25.4905634 [+44.19%]
(0072173)
git   
2017-11-10 14:08   
Branch CR25104 has been updated forcibly by aml.

SHA-1: ca6b3aec5e5217eb60a88c47cef14d1d4afa9dee
(0072193)
aml   
2017-11-11 10:48   
(edited on: 2017-11-11 10:53)
CR25104-master-AML

testgrid boolean bfuse_2d -parallel 1

master:
Elapsed time: 0 Hours 0 Minutes 7.133023403 Seconds
Elapsed time: 0 Hours 0 Minutes 7.10875429874 Seconds
Elapsed time: 0 Hours 0 Minutes 7.20491770517 Seconds

25104:
Elapsed time: 0 Hours 0 Minutes 7.39321095336 Seconds
Elapsed time: 0 Hours 0 Minutes 7.09034411979 Seconds
Elapsed time: 0 Hours 0 Minutes 7.27562945378 Seconds


testgrid boolean bopcut_simple -parallel 1

master:
Elapsed time: 0 Hours 1 Minutes 45.4793312428 Seconds
Elapsed time: 0 Hours 1 Minutes 43.7380158271 Seconds
Elapsed time: 0 Hours 1 Minutes 43.0334414918 Seconds

25104:
Elapsed time: 0 Hours 1 Minutes 44.3422253874 Seconds
Elapsed time: 0 Hours 1 Minutes 43.5321976635 Seconds
Elapsed time: 0 Hours 1 Minutes 45.3253594649 Seconds


test bugs mesh bug25806 -echo

master:
TOTAL CPU TIME: 26.234375 sec
TOTAL CPU TIME: 28.765625 sec
TOTAL CPU TIME: 28.625 sec
TOTAL CPU TIME: 36.046875 sec
TOTAL CPU TIME: 27.59375 sec

25104:
TOTAL CPU TIME: 26.46875 sec
TOTAL CPU TIME: 36.578125 sec
TOTAL CPU TIME: 33.703125 sec
TOTAL CPU TIME: 29.390625 sec
TOTAL CPU TIME: 27.75 sec


testgrid thrusection specific -parallel 1

master:
Elapsed time: 0 Hours 0 Minutes 31.294814899 Seconds
Elapsed time: 0 Hours 0 Minutes 31.4890486621 Seconds
Elapsed time: 0 Hours 0 Minutes 32.2164092851 Seconds
Elapsed time: 0 Hours 0 Minutes 32.5949582258 Seconds

25104:
Elapsed time: 0 Hours 0 Minutes 32.322451308 Seconds
Elapsed time: 0 Hours 0 Minutes 31.6478428142 Seconds
Elapsed time: 0 Hours 0 Minutes 33.0373280631 Seconds
Elapsed time: 0 Hours 0 Minutes 32.0014748887 Seconds

(0072194)
aml   
2017-11-11 10:53   
So, I cannot see any significant impact on performance. The only thing which should be reported - the unstable behavior of the bug25806 issue.
(0072201)
msv   
2017-11-13 10:26   
I agree.
(0072202)
aml   
2017-11-13 10:34   
Dear msv,
could you please check CR25104?

The corresponding jenkins task: http://jenkins-test-10.nnov.opencascade.com:8080/view/CR25104-master-AML/ [^]
(0072203)
msv   
2017-11-13 10:50   
Reviewed.
(0072209)
bugmaster   
2017-11-13 11:48   
Combination -
OCCT branch: CR25104 SHA: ca6b3aec5e5217eb60a88c47cef14d1d4afa9dee
Products branch: master
was compiled on Linux, MacOS and Windows platforms and tested on optimize mode.

Number of compiler warnings:
No new/fixed warnings.

Regressions/Differences/Improvements:
No regressions/differences.

CPU differences:
OCCT
Debian70-64
Total CPU difference: 20765.97000000055 / 20769.880000000467 [-0.02%]
Windows-64-VC10
Total CPU difference: 18548.28489849851 / 18481.95327329852 [+0.36%]
Products
Debian70-64
Total CPU difference: 7899.500000000094 / 7901.260000000098 [-0.02%]
Windows-64-VC10
Total CPU difference: 7991.182425199982 / 7874.587277799976 [+1.48%]

Image differences :
No differences that require special attention.

Memory differences :
No differences that require special attention.
(0072780)
git   
2017-12-07 21:07   
Branch CR25104 has been deleted by kgv.

SHA-1: ca6b3aec5e5217eb60a88c47cef14d1d4afa9dee