[All] issofa_bugfix: cleans and fixes #218

guparan · 2017-03-23T15:56:10Z

This PR will be further discussed during STC#3.

issofa_bugfix

Cleans

BaseObject: Output message in serr for required datas. Error word should be reserved for messages that will make the application fail.
RayTriangleVisitor and SceneLoaderFactory: clean warnings
ParticlesRepulsionForceField: Add empty implementation of addKToMatrix to get rid of console warnings.

Bugfixes

AttachConstraint: fix crash when indices are resized
BTDLinearSolver: fix constant log + operator new [] vs operator delete mismatch
CatmullRomSplineMapping: fix applyJ compilation
ConstantForceField: fix avoid crashes in draw when trying to apply a constantforcefield to a non existing point
DistanceGrid: fix incorrect use of serr
GeneralConstraintSolver: fix incorrect parent class in SOFA_CLASS macro
IndicesFromValues: fix use getTemplateName to avoid problems of links with other data
EulerImplicitSolver: fix Disable calls to solveConstraint from EulerImplicitSolver by default as it is not needed unless an very specific solver is added and it currently crashes in other cases + wrong argument order in calls to AdvancedTimer::stepNext()
FixedConstraint and PartialFixedConstraint: fix "fixed FixedConstraint & PartialFixedConstraint for size-varying mechanical objects".
MechanicalObject: fix crash in debug with null pointer
Mass: fix remove error logging in Mass method that are inherited from Forcefield API and that are not relevant for Mass
MechanicalMatrixVisitor and MechanicalOperations: fix if using a Linear Solver, projective constraints were wrongly applied when a force field is in a child node
Mesh2PointMechanicalMapping: fix constraints ApplyJT
Metis(gk_arch.h): fix compilation with Visual Studio 2015
ParticlesRepulsionForceField and RepulsiveSpringForceField: fix avoid division by 0 on repulsion force fields
PersistentContactBarycentricMapping: fix init variables in constructor
RestShapeSpringsForceField: fix Runtime stiffness tunning was not working + optimisation and cleaning
SkinningMapping: fix compilation of SofaRigid
SofaViewer: fix crashes when current camera of pick-handler is NULL
SurfacePressureForceField: fix volume computation formula
TaitSurfacePressureForceField: fix contribution to the stiffness matrix for the second component df = P+dN in TaitSurfacePressureForceField
TopologicalMapping: fix display error messages when a TopologicalMapping failed to be created
TriangularFEMForceFieldOptim: fix principal values ordering with input matrix is diagonal + uninitialized value warning
VisualModelImpl: fix wrong object (triangles) called when adding quads in handleTopologyChange()
VoxelGridLoader: std::unique result was not used

This PR:

builds with SUCCESS for all platforms on the CI.
does not generate new warnings nor unit test failures.
does not break existing scenes.
does not break API compatibility.
has been reviewed and agreed to be transitional.
is more than 1 week old.

Reviewers will merge only if all these checks are true.

…ntainer according to std spec. git-svn-id: https://code.insimo.fr/svn/IS/ISSofa/branches/ISSofa@10404 f8a8aa39-ebdb-4476-bae3-08f188f1b544

git-svn-id: https://code.insimo.fr/svn/IS/ISSofa/branches/ISSofa@10439 f8a8aa39-ebdb-4476-bae3-08f188f1b544

git-svn-id: https://code.insimo.fr/svn/IS/ISSofa/branches/ISSofa@10467 f8a8aa39-ebdb-4476-bae3-08f188f1b544

ADD : Optimizations & cleaning.

git-svn-id: https://code.insimo.fr/svn/IS/ISSofa/branches/ISSofa@10494 f8a8aa39-ebdb-4476-bae3-08f188f1b544

…logyChange() git-svn-id: https://code.insimo.fr/svn/IS/ISSofa/branches/ISSofa@10500 f8a8aa39-ebdb-4476-bae3-08f188f1b544

git-svn-id: https://code.insimo.fr/svn/IS/ISSofa/branches/ISSofa@10512 f8a8aa39-ebdb-4476-bae3-08f188f1b544

git-svn-id: https://code.insimo.fr/svn/IS/ISSofa/branches/ISSofa@10515 f8a8aa39-ebdb-4476-bae3-08f188f1b544

This reverts commit 7507f05d07b5932974a68b121d64101f0f92a678. Conflicts: framework/sofa/core/ObjectFactory.cpp NB FJ: worth investigating later, but right now it just generates tons of warnings and does not seem to comply to the promises the initial commit made

…be reserved for messages that will make the application fail.

…as it is not needed unless an very specific solver is added and it currently crashes in other cases git-svn-id: https://code.insimo.fr/svn/IS/ISSofa/branches/ISSofa@10544 f8a8aa39-ebdb-4476-bae3-08f188f1b544

git-svn-id: https://code.insimo.fr/svn/IS/ISSofa/branches/ISSofa@10545 f8a8aa39-ebdb-4476-bae3-08f188f1b544

… warnings

…acro

…ialized value warning

…andle multiple points generation according to only one source primitive

…plied when a force field is in a child node

…to a non existing point

…ying mechanical objects". Not the best way to handle this, probably bugged, and in all cases it broke projectJacobianMatrix in FixedConstraint, at least with the implementation of constraint Jacobians using sofa::defaulttype::MapMapSparseMatrix. In this implementation row.size() is not equal to the size of the mechanicalObject, but to the number of dofs that are affected by a given constraint equation. This partially reverts commit c4aa645fd431369db795e317cd05df6f86460515

…KToMatrix to get rid of console warnings.

… = P+dN in TaitSurfacePressureForceField (Ref T1019, T1010)

…ssofa_bugfix

review taken into account

hugtalbot · 2019-01-23T09:17:48Z

@fjourdes and @ChristianDuriez I think this comes from the collaboration between you guys.
Last moment for a feedback before merge, thx

ChristianDuriez · 2019-01-25T14:14:33Z

SofaKernel/framework/sofa/simulation/MechanicalMatrixVisitor.h

@@ -223,17 +223,6 @@ class SOFA_SIMULATION_CORE_API MechanicalAddMBK_ToMatrixVisitor : public Mechani
        return RESULT_CONTINUE;
    }

-    //Masses are now added in the addMBKToMatrix call for all ForceFields
-
-    virtual Result fwdProjectiveConstraintSet(simulation::Node* /*node*/, core::behavior::BaseProjectiveConstraintSet* c)


Why this has been removed ?
Is it replaced by MechanicalApplyProjectiveConstraint_ToMatrixVisitor ?

Yes, it's related to

MechanicalMatrixVisitor and MechanicalOperations: fix if using a Linear Solver, projective constraints were wrongly applied when a force field is in a child node

When you have a MechanicalObject and a projective constraint in a (parent) node and a ForceField in a child node, which you have to do if for instance you want to apply it to the DOFs but with a different topology for the elements (a subset, or converting hexas to tetras for example), then the matrix was projected before the forcefield would add its contributions.
In order to fix it in all cases (allowing for both forcefield and projective constraints in child nodes) the solution is now to do it with two visitors.

damienmarchal · 2019-01-25T14:15:17Z

modules/SofaSparseSolver/src/SofaSparseSolver/SparseLDLSolver.h

-    void solve (Matrix& M, Vector& x, Vector& b) override ;
-    void invert(Matrix& M) override;
-    bool addJMInvJtLocal(TMatrix * M, ResMatrixType * result,const JMatrixType * J, double fact) override;
+    virtual void solve (Matrix& M, Vector& x, Vector& b) override;


virtual not needed.

damienmarchal · 2019-01-25T14:15:29Z

modules/SofaMiscMapping/CatmullRomSplineMapping.inl

@@ -178,7 +178,7 @@ void CatmullRomSplineMapping<TIn, TOut>::apply( const sofa::core::MechanicalPara


 template <class TIn, class TOut>
-void CatmullRomSplineMapping<TIn, TOut>::applyJ( const sofa::core::MechanicalParams* mparams, OutDataVecDeriv& outData, const InDataVecDeriv& inData)
+void CatmullRomSplineMapping<TIn, TOut>::applyJ( const sofa::core::MechanicalParams* mparams /* PARAMS FIRST */, OutDataVecDeriv& outData, const InDataVecDeriv& inData)


PARAMS_FIRST is to removed.

ChristianDuriez · 2019-01-25T14:16:09Z

For me, it's ok.
I just had a question about the applyconstraint that has been removed in the visitor..

…issofa_bugfix

Due to missing override

hugtalbot · 2020-08-24T14:04:50Z

Looks fine to me now

hugtalbot · 2020-08-26T08:24:05Z

CentOS: 15 scenes time out
MacOS: 87 crashes (time out as well) ..

Let's :

[ci-build][with-all-tests][force-full-build]

[all] issofa_bugfix: cleans and fixes

thomasjundinsimo and others added 30 commits October 12, 2016 16:26

FIX: std::unique result wasn't used : unknown elements remained in co…

91d6ece

…ntainer according to std spec. git-svn-id: https://code.insimo.fr/svn/IS/ISSofa/branches/ISSofa@10404 f8a8aa39-ebdb-4476-bae3-08f188f1b544

CLEAN: var not init in constructor

390084d

git-svn-id: https://code.insimo.fr/svn/IS/ISSofa/branches/ISSofa@10439 f8a8aa39-ebdb-4476-bae3-08f188f1b544

FIX: crashes ni SofaViewer when current camera or pick handler is NULL

95b5949

git-svn-id: https://code.insimo.fr/svn/IS/ISSofa/branches/ISSofa@10467 f8a8aa39-ebdb-4476-bae3-08f188f1b544

FIX : Runtime stiffness tunning was not working.

b4f147a

ADD : Optimizations & cleaning.

FIX: crash in debug with null pointer

ebe4ae7

git-svn-id: https://code.insimo.fr/svn/IS/ISSofa/branches/ISSofa@10494 f8a8aa39-ebdb-4476-bae3-08f188f1b544

FIX : wrong object (triangles) called when adding quads in handleTopo…

7299302

…logyChange() git-svn-id: https://code.insimo.fr/svn/IS/ISSofa/branches/ISSofa@10500 f8a8aa39-ebdb-4476-bae3-08f188f1b544

FIX: crash when indices are resized

8eb0b66

git-svn-id: https://code.insimo.fr/svn/IS/ISSofa/branches/ISSofa@10512 f8a8aa39-ebdb-4476-bae3-08f188f1b544

FIX: avoid division by 0 on repulsion force fields

1e97dbf

git-svn-id: https://code.insimo.fr/svn/IS/ISSofa/branches/ISSofa@10515 f8a8aa39-ebdb-4476-bae3-08f188f1b544

FIX: compilation of SofaRigid

9359e97

FIX: compilation of SofaMiscMapping

7e89c58

FIX: warnings

7e97ad2

CHANGE: Output message in serr for required datas. Error word should …

959a943

…be reserved for messages that will make the application fail.

Disable calls to solveConstraint from EulerImplicitSolver by default …

81d1e4d

…as it is not needed unless an very specific solver is added and it currently crashes in other cases git-svn-id: https://code.insimo.fr/svn/IS/ISSofa/branches/ISSofa@10544 f8a8aa39-ebdb-4476-bae3-08f188f1b544

FIX: wrong argument order in calls to AdvancedTimer::stepNext()

98b9468

git-svn-id: https://code.insimo.fr/svn/IS/ISSofa/branches/ISSofa@10545 f8a8aa39-ebdb-4476-bae3-08f188f1b544

FIX: constant log in BTD, incorrect use of serr in DistanceGrid, misc…

831d521

… warnings

CLEAN: incorrect parent class in GenericConstraintSolver SOFA_CLASS m…

c753bc7

…acro

UPDATE: use getTemplateName to avoid problems of links with other data

869c89d

FIX: principal values ordering with input matrix is diagonal + uninit…

0cb6f4e

…ialized value warning

FIX: constraints ApplyJT in Mesh2PointMechanicalMapping in order to h…

bd1c313

…andle multiple points generation according to only one source primitive

FIX: volume computation formula in SurfacePressureForceField

285bec5

FIX: if using a Linear Solver, projective constraints were wrongly ap…

13551e2

…plied when a force field is in a child node

FIX: operator new [] vs operator delete mismatch

a5ae657

FIX: avoid crashes in draw when trying to apply a constantforcefield …

6f1b939

…to a non existing point

UPDATE: ParticlesRepulsionForceField. Add empty implementation of add…

6946f6e

…KToMatrix to get rid of console warnings.

FIX: contribution to the stiffness matrix for the second component df…

700d0a9

… = P+dN in TaitSurfacePressureForceField (Ref T1019, T1010)

Merge branch 'reset_issofa_integration' into issofa_bugfix

2ff3586

Merge remote-tracking branch 'github/reset_issofa_integration' into i…

450cee7

…ssofa_bugfix

[SofaKernel] FIX Windows dllimport macro

adb5a93

hugtalbot added pr: status to review To notify reviewers to review this pull-request and removed pr: status wip Development in the pull-request is still in progress labels Jan 13, 2019

make PR cleaner and use msg

4a55108

ChristianDuriez reviewed Jan 25, 2019

View reviewed changes

damienmarchal reviewed Jan 25, 2019

View reviewed changes

guparan added pr: status wip Development in the pull-request is still in progress and removed pr: status to review To notify reviewers to review this pull-request labels Jan 30, 2019

epernod added the STC#8 Simple tasks for STC#8 coding sprint label Nov 15, 2019

hugtalbot added STC#9 Simple tasks for STC#5 coding sprint and removed STC#5 STC#8 Simple tasks for STC#8 coding sprint labels Feb 27, 2020

Merge branch 'master' of https://github.com/sofa-framework/sofa into …

6549465

…issofa_bugfix

fredroy added pr: status to review To notify reviewers to review this pull-request and removed pr: status wip Development in the pull-request is still in progress labels Aug 14, 2020

hugtalbot added 2 commits August 19, 2020 12:42

Fix warning in MechanicalMatrixVisitor.h

a978186

Due to missing override

Update MechanicalMatrixVisitor.h

3b78cc8

hugtalbot added pr: status ready Approved a pull-request, ready to be squashed and removed pr: status to review To notify reviewers to review this pull-request labels Sep 9, 2020

Update SparseLDLSolver.h

3d5cb02

epernod merged commit 114cd81 into master Sep 11, 2020

guparan deleted the issofa_bugfix branch September 18, 2020 12:18

guparan added this to the v20.12 milestone Nov 17, 2020

guparan changed the title ~~[all] issofa_bugfix: cleans and fixes~~ [All] issofa_bugfix: cleans and fixes Jan 18, 2021

bakpaul pushed a commit to bakpaul/sofa that referenced this pull request Feb 23, 2024

Merge pull request sofa-framework#218 from sofa-framework/issofa_bugfix

0ea613b

[all] issofa_bugfix: cleans and fixes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[All] issofa_bugfix: cleans and fixes #218

[All] issofa_bugfix: cleans and fixes #218

guparan commented Mar 23, 2017 •

edited

Loading

hugtalbot commented Jan 23, 2019

ChristianDuriez Jan 25, 2019

JeremieA Jan 31, 2019

damienmarchal Jan 25, 2019

damienmarchal Jan 25, 2019

ChristianDuriez commented Jan 25, 2019

hugtalbot commented Aug 24, 2020

hugtalbot commented Aug 26, 2020

[All] issofa_bugfix: cleans and fixes #218

[All] issofa_bugfix: cleans and fixes #218

Conversation

guparan commented Mar 23, 2017 • edited Loading

issofa_bugfix

Cleans

Bugfixes

hugtalbot commented Jan 23, 2019

ChristianDuriez Jan 25, 2019

Choose a reason for hiding this comment

JeremieA Jan 31, 2019

Choose a reason for hiding this comment

damienmarchal Jan 25, 2019

Choose a reason for hiding this comment

damienmarchal Jan 25, 2019

Choose a reason for hiding this comment

ChristianDuriez commented Jan 25, 2019

hugtalbot commented Aug 24, 2020

hugtalbot commented Aug 26, 2020

guparan commented Mar 23, 2017 •

edited

Loading