View Issue Details

IDProjectCategoryView StatusLast Update
0024065Open CASCADEOCCT:Modeling Algorithmspublic2013-12-19 13:57
ReporteremvAssigned Tobugmaster  
PrioritynormalSeverityminor 
Status closedResolutionfixed 
Product Version6.6.0 
Target Version6.7.0Fixed in Version6.7.0 
Summary0024065: Wrong intersection point
DescriptionIntersection point between attached curve and surface is wrong.
The problem has been detected in the test case bugs modalg_2 bug239_2.
Steps To Reproducerestore f.brep f
restore e.brep e
mksurface s f
mkcurve c e
intersect i c s
projponf f i_2
#proj dist = 6.0298589392945979e-006
xdist c s 0 1 100
cvalue c 0.30303 X Y Z
point p X Y Z
projponf f p
#proj dist = 2.0458970352054304e-008
#in the result should point p instead of i_2
TagsNo tags attached.
Test case numberbugs moddata_3(013) bug24065

Attached Files

  • f.brep (70,797 bytes)
  • e.brep (24,716 bytes)

Activities

emv

2013-07-11 15:37

developer  

f.brep (70,797 bytes)

emv

2013-07-11 15:37

developer  

e.brep (24,716 bytes)

ifv

2013-08-13 18:48

developer   ~0025295

Branch CR24065 is created for rew.

ifv

2013-08-13 18:48

developer   ~0025296

Ok

mkv

2013-08-15 09:21

tester   ~0025310

Dear BugMaster,

Branch CR24065 (and products from GIT master) was compiled on Linux and Windows platforms and tested.
SHA-1: f43ffa6ee81287ee55da8c40c865358cc5e8920e

Number of compiler warnings:

occt component :
Linux: 916 (917 on master)
Windows: 855 (855 on master)

products component :
Linux: 188 (188 on master)
Windows: 287 (287 on master)

Regressions:
http://occt-tests/CR24065-master-occt/Debian60-64/bugs/modalg_1/bug10232.html
http://occt-tests/CR24065-master-occt/Windows-32-VC9/bugs/modalg_1/bug10232.html
bugs modalg_1 bug10232 - new length of result shape

http://occt-tests/CR24065-master-occt/Debian60-64/bugs/modalg_2/bug239_2.html
http://occt-tests/CR24065-master-occt/Windows-32-VC9/bugs/modalg_2/bug239_2.html
bugs modalg_2 bug239_2 - new length of result shape

Improvements:
No improvements

Testing cases:
bugs moddata_3(013) bug24065 - OK.

Testing on Linux:
Total MEMORY difference: 363497700 / 363214852
Total CPU difference: 58780.270000000586 / 49125.790000000496

Testing on Windows:
Total MEMORY difference: 424267232 / 424364848
Total CPU difference: 40447.03125 / 32226.875

There are not differences in images found by testdiff.
http://occt-tests/CR24065-master-occt/Debian60-64/diff-Debian60-64.html
http://occt-tests/CR24065-master-occt/Windows-32-VC9/diff-Windows-32-VC9.html
IMAGE bugs modalg_2 bug239_1: bug239_1.png differs
IMAGE bugs modalg_2 bug23: bug239_2.png differs

ifv

2013-08-15 15:40

developer   ~0025322

Dear EMV, as far as I understand, regression bugs modalg_2 bug239_2 is really improvement - it is the same case which you reported in bug. Now result is shorter because of more correct cutting initial wire by solid.
What about regression bugs modalg_1 bug10232? Since nobody knows what is correct result, give please you expert conclusion about this case.
I think that we must concider new results as correct ones and change values used in tests for checking results.
What is your opinion?

emv

2013-08-15 18:13

developer   ~0025324

Last edited: 2013-08-15 18:17

Dear colleagues,
I have analyzed these two cases and my opinion coincides with IFV's opinion.

bugs modalg_2 bug239_2 - now result is shorter because of more correct cutting.
bugs modalg_1 bug10232 - now result is longer because there are some new section edges created which current version of OCCT is unable to create. For example, try in DRAW:
restore [locate_data_file OCC10232_ms1.brep] b1
restore [locate_data_file OCC10232_ms2.brep] b2
explode b1 f; copy b1_109 b1
explode b2 f; copy b2_109 b2
bop b1 b2
bopsection r

Thus, dear MKV please change values used in these test scripts for checking results.

mkv

2013-08-16 12:09

tester   ~0025328

Dear emv,
There are different results on Linux and Windows platforms in bugs modalg_1 bug10232
http://occt-tests/CR24065-master-occt/Debian60-64/bugs/modalg_1/bug10232.html
http://occt-tests/CR24065-master-occt/Windows-32-VC9/bugs/modalg_1/bug10232.html

ifv

2013-08-27 10:52

developer   ~0025387

I suggest to integrate this fix in spite of result of bugs modalg_1 bug10232
for following reasons:
1. this test was wrong in muster and probably remains wrong after fix because nobody knows correct result for this test.
2. Fixing this test will take rather much time (if success can be reached at all)
So I suggest to use different values of length for Windows and Linux (it is better to remove this test at all) - fix for bug is really useful and more important then results of useless test.

abv

2013-08-27 16:57

manager   ~0025390

Eugene, can you explain why results are different for Linux and Windows (if I understand correctly, it was not like that before the fix)?

emv

2013-08-27 17:55

developer   ~0025392

Last edited: 2013-08-27 18:26

Actually, it was. Here are the results of tests from IR-2013-08-22-master-occt:
1. Debian60-64 - length is 3200.06 ( http://occt-tests/IR-2013-08-22-master-occt/Debian60-64/bugs/modalg_1/bug10232.html );
2. Windows-32-VC9 - length is 3202.9 ( http://occt-tests/IR-2013-08-22-master-occt/Windows-32-VC9/bugs/modalg_1/bug10232.html ).

mkv

2013-08-27 20:28

tester   ~0025397

Dear BugMaster,
Test cases bugs/modalg_1(006)/bug10232 and bugs/modalg_2(007)/bug239_2 was corrected and pushed to git repository in branch CR24065.

Related Changesets

occt: master 4355f260

2013-08-29 09:30:41

emv


Committer: bugmaster Details Diff
0024065: Wrong intersection point

Adding test case for issue CR24065

Small corrections of test cases
Affected Issues
0024065
mod - src/IntAna/IntAna_QuadQuadGeo.cxx Diff File
mod - tests/bugs/modalg_1/bug10232 Diff File
mod - tests/bugs/modalg_2/bug239_2 Diff File
add - tests/bugs/moddata_3/bug24065 Diff File

Issue History

Date Modified Username Field Change
2013-07-11 15:37 emv New Issue
2013-07-11 15:37 emv Assigned To => ifv
2013-07-11 15:37 emv File Added: f.brep
2013-07-11 15:37 emv File Added: e.brep
2013-08-13 18:48 ifv Note Added: 0025295
2013-08-13 18:48 ifv Status new => resolved
2013-08-13 18:48 ifv Note Added: 0025296
2013-08-13 18:48 ifv Status resolved => reviewed
2013-08-13 19:06 mkv Assigned To ifv => mkv
2013-08-15 09:21 mkv Note Added: 0025310
2013-08-15 09:22 mkv Test case number => bugs moddata_3(013) bug24065
2013-08-15 09:22 mkv Assigned To mkv => ifv
2013-08-15 09:22 mkv Status reviewed => assigned
2013-08-15 15:40 ifv Note Added: 0025322
2013-08-15 15:40 ifv Assigned To ifv => emv
2013-08-15 15:40 ifv Status assigned => feedback
2013-08-15 18:13 emv Note Added: 0025324
2013-08-15 18:13 emv Assigned To emv => mkv
2013-08-15 18:14 emv Note Edited: 0025324
2013-08-15 18:17 emv Note Edited: 0025324
2013-08-16 12:09 mkv Note Added: 0025328
2013-08-16 12:09 mkv Assigned To mkv => emv
2013-08-26 10:47 mkv Assigned To emv => ifv
2013-08-27 10:52 ifv Note Added: 0025387
2013-08-27 10:52 ifv Status feedback => reviewed
2013-08-27 16:57 abv Note Added: 0025390
2013-08-27 16:57 abv Assigned To ifv => emv
2013-08-27 16:57 abv Status reviewed => feedback
2013-08-27 17:54 emv Assigned To emv => mkv
2013-08-27 17:55 emv Note Added: 0025392
2013-08-27 18:26 emv Note Edited: 0025392
2013-08-27 20:28 mkv Note Added: 0025397
2013-08-27 20:28 mkv Assigned To mkv => bugmaster
2013-08-27 20:28 mkv Status feedback => tested
2013-08-28 14:33 bugmaster Target Version => 6.7.0
2013-09-03 12:22 bugmaster Changeset attached => occt master 4355f260
2013-09-03 12:22 bugmaster Status tested => verified
2013-09-03 12:22 bugmaster Resolution open => fixed
2013-12-19 13:52 bugmaster Status verified => closed
2013-12-19 13:57 bugmaster Fixed in Version => 6.7.0