MantisBT
Mantis Bug Tracker Workflow

View Issue Details Jump to Notes ] Issue History ] Print ]
IDProjectCategoryView StatusDate SubmittedLast Update
0029893Open CASCADE[OCCT] OCCT:Modeling Algorithmspublic2018-06-22 13:542018-06-25 16:14
Reporternbv 
Assigned Tomsv 
PrioritynormalSeverityminor 
StatusnewResolutionopen 
PlatformOSOS Version
Product Version 
Target Version[OCCT] 7.4.0*Fixed in Version 
Summary0029893: IntPatch_TheSOnBounds cannot find precise solution
DescriptionClass IntPatch_TheSOnBounds is used for computation IntPatch_Point(s) and putting them to IntPatch_Aline or IntPatch_GLine (and IntPatch_RLine but another initialization algorithm is used).

Indeed the found solution must lie on the both surfaces strictly. However, the returned solution is in the distance 2.5177674590093085e-006 from the surface. It leads to wrong IntPatch_Point computation. After that, this bad IntPatch_Point is put to the intersection curve and bounds it incorrectly (intersection curve has become not closed).

As we can see in Steps To Reproduce, two solution is found. The 1st one is wrong the 2nd one is OK. If we try to use extrema then we will obtain the good IntPatch_Point but only one (2nd solution cannot be found by extrema).

Please note that the result of IntPatch_TheSOnBounds algorithm does not depend on the input tolerance.
Steps To Reproduce# Test bugs modalg_7 bug29807_i5001
# Test parasolid doc_3 A4

binrestore [locate_data_file bug29807_f1.bin] f1
binrestore [locate_data_file bug29807_f2.bin] f2

mksurface s1 f1
mksurface s2 f2

trim s1 s1
trim s2 s2

# Take U-isoline of the cylindrical surface (s2)
uiso cc s2 0
# Its 2D-representation
line liso 0 0 0 1

# New DRAW-command (from the file). s1 - quadric surface; s2 - 2nd surface, which liso is chosen on.
# Later first and last parameters of iso-line and tolerance for computation are set.

sonbou s1 s2 liso -92588.227836497696 101953.32390724987 1.0e-4
# D1 = -2.5177674590093085e-006; P(120.7, 4.795054507370125, 5.6500000000000057) ==> WRONG
# D2 = 0; P(120.7, 8726.5714710291286, 5.6500000000000057) ==> OK

# Compute with another tolerance

sonbou s1 s2 liso -92588.227836497696 101953.32390724987 1.0e-7
# The same result

# Try to compute with extrema

extrema cc s1
# Only one extrema is found, which really precise the result found by "sonbou" command.
# But extrema cannot find 2nd solution which is found by "sonbou" command precisely

length ext_1
# 1.664854851506445e-012 ==> OK

Additional information
and documentation updates
Please use new DRAW-command from the file QABugs_InterSOnB.cxx to reproduce this problem in DRAW-application.
TagsNo tags attached.
Test case number
Attached Filescxx file icon QABugs_InterSOnB.cxx (3,074 bytes) 2018-06-22 14:00

- Relationships
related to 0029807verifiedbugmaster [Regression to 7.0.0] Impossible to cut cone from prism 

-  Notes
There are no notes attached to this issue.

- Issue History
Date Modified Username Field Change
2018-06-22 13:54 nbv New Issue
2018-06-22 13:54 nbv Assigned To => msv
2018-06-22 14:00 nbv File Added: QABugs_InterSOnB.cxx
2018-06-22 14:24 nbv Relationship added related to 0029807
2018-06-22 15:02 nbv Steps to Reproduce Updated View Revisions
2018-06-25 16:14 nbv Steps to Reproduce Updated View Revisions


Copyright © 2000 - 2018 MantisBT Team
Powered by Mantis Bugtracker