Add bipedal-locomotion-framework #526

GiulioRomualdi · 2020-11-22T14:25:24Z

This is the first test for having bipedal-locomotion-framework in the superbuild.
I'm pretty sure I missed something

traversaro · 2020-11-22T14:28:42Z

Thanks! At a first glance it seem nice, let's see what is missing. For adding a new components the documentation as outline by the steps in https://github.com/robotology/robotology-superbuild/blob/master/doc/developers-faqs.md#how-to-add-a-new-profile, but let's first see if CI passes, and then we can go forward.

traversaro · 2020-11-22T14:30:37Z

cmake/Buildbipedal-locomotion-framework.cmake

+              TAG master
+              COMPONENT dynamics
+              FOLDER robotology
+              CMAKE_ARGS -DBUILD_TESTING:BOOL=OFF


Could it make sense to pass the options on enabling/disabling the dependencies explicitly?

do you mean passing FRAMEWORK_USE_manif etc? https://github.com/dic-iit/bipedal-locomotion-framework/blob/master/cmake/BipedalLocomotionFrameworkFindDependencies.cmake#L179

traversaro · 2020-11-22T14:30:41Z

cmake/Buildbipedal-locomotion-framework.cmake

+
+include(YCMEPHelper)
+
+set(bipedal-locomotion-framework_DEPENDS "")


Probably in the CMake world we can stick to the CMake package name for consistency, so:

Suggested change

set(bipedal-locomotion-framework_DEPENDS "")

set(BipedalLocomotionFramework_DEPENDS "")

traversaro · 2020-11-22T14:33:12Z

CMakeLists.txt

@@ -161,9 +162,17 @@ if(ROBOTOLOGY_ENABLE_DYNAMICS)
  find_or_build_package(walking-controllers)
  find_or_build_package(whole-body-estimators)
  find_or_build_package(whole-body-controllers)
+  find_or_build_package(bipedal-locomotion-framework)


The name here should match the one of the Build<package>.cmake and the one passed to ycm_ep_helper . Even if not strictly necessary, it is a good idea to just use the cmake package name (so in this case BipedalLocomotionFramework), because find_or_build_package mirrors find_package, so if we have:

find_package(BipedalLocomotionFramework)

we should also have:

find_or_build_package(BipedalLocomotionFramework)

traversaro · 2020-11-22T14:41:27Z

The casadi error on Debian Buster are one of those that make you feel alive:

2020-11-22T14:35:49.3238803Z CMake Deprecation Warning at CMakeLists.txt:31 (cmake_policy):
2020-11-22T14:35:49.3239537Z   The OLD behavior for policy CMP0011 will be removed from a future version
2020-11-22T14:35:49.3240053Z   of CMake.
2020-11-22T14:35:49.3240274Z 
2020-11-22T14:35:49.3240935Z   The cmake-policies(7) manual explains that the OLD behaviors of all
2020-11-22T14:35:49.3241680Z   policies are deprecated and that a policy should be set to OLD only under
2020-11-22T14:35:49.3242608Z   specific short-term circumstances.  Projects should be ported to the NEW
2020-11-22T14:35:49.3243301Z   behavior and not rely on setting a policy to OLD.
2020-11-22T14:35:49.3243625Z 
2020-11-22T14:35:49.3243818Z 
2020-11-22T14:35:49.3244321Z CMake Deprecation Warning at CMakeLists.txt:35 (cmake_policy):
2020-11-22T14:35:49.3245034Z   The OLD behavior for policy CMP0005 will be removed from a future version
2020-11-22T14:35:49.3245541Z   of CMake.
2020-11-22T14:35:49.3245767Z 
2020-11-22T14:35:49.3246437Z   The cmake-policies(7) manual explains that the OLD behaviors of all
2020-11-22T14:35:49.3247173Z   policies are deprecated and that a policy should be set to OLD only under
2020-11-22T14:35:49.3248110Z   specific short-term circumstances.  Projects should be ported to the NEW
2020-11-22T14:35:49.3248791Z   behavior and not rely on setting a policy to OLD.
2020-11-22T14:35:49.3249126Z 
2020-11-22T14:35:49.3249316Z 
2020-11-22T14:35:49.3249752Z CMake Error at CMakeLists.txt:99 (enable_language):
2020-11-22T14:35:49.3250410Z   The Ninja generator does not support Fortran using Ninja version
2020-11-22T14:35:49.3250813Z 
2020-11-22T14:35:49.3251081Z     1.8.2
2020-11-22T14:35:49.3251261Z 
2020-11-22T14:35:49.3251785Z   due to lack of required features.  Kitware has implemented the required
2020-11-22T14:35:49.3252538Z   features and they have been merged to upstream ninja for inclusion in Ninja
2020-11-22T14:35:49.3253426Z   1.10 and higher.  As of this version of CMake, Ninja 1.10 has not been
2020-11-22T14:35:49.3254170Z   released.  Meanwhile, Kitware maintains a branch of Ninja at:
2020-11-22T14:35:49.3254575Z 
2020-11-22T14:35:49.3255358Z     https://github.com/Kitware/ninja/tree/features-for-fortran#readme
2020-11-22T14:35:49.3255884Z 
2020-11-22T14:35:49.3256526Z   with the required features.  One may build ninja from that branch to get
2020-11-22T14:35:49.3257083Z   support for Fortran.
2020-11-22T14:35:49.3257347Z 
2020-11-22T14:35:49.3257521Z 
2020-11-22T14:35:49.3258115Z -- Configuring incomplete, errors occurred!
2020-11-22T14:35:49.3259290Z See also "/__w/robotology-superbuild/robotology-superbuild/build/external/casadi/CMakeFiles/CMakeOutput.log".

However, Debian Buster is one of those platforms for which we keep compatibility mostly for the Core software to run on the pc104 of old robots, probably we do not need for now to support ROBOTOLOGY_ENABLE_DYNAMICS_FULL_DEPS on it.

traversaro · 2020-11-22T14:45:56Z

However, Debian Buster is one of those platforms for which we keep compatibility mostly for the Core software to run on the pc104 of old robots, probably we do not need for now to support ROBOTOLOGY_ENABLE_DYNAMICS_FULL_DEPS on it.

Anyhow, I just check and this is sweet: Debian Buster backports has indeed Ninja 1.10 (see https://packages.debian.org/buster-backports/ninja-build), so we just need to install it from the backports as we already to for CMake in https://packages.debian.org/buster-backports/ninja-build . However, if more problem emerge we can just avoid to test it on Debian Buster, even because now that we support conda we can easily compile our code on arbitrary old Linux distributions through it.

GiulioRomualdi · 2020-11-22T15:22:03Z

I think we need to change the idyntree tag

traversaro · 2020-11-22T16:33:05Z

I think we need to change the idyntree tag

I will release iDynTree 2.0 soon in the next days, this should handle that. Instead we need to have a way to exclude some components from the release.

GiulioRomualdi · 2020-11-22T16:46:16Z

Instead we need to have a way to exclude some components from the release.

I didn't get it

traversaro · 2020-11-22T17:33:12Z

Instead we need to have a way to exclude some components from the release.

I didn't get it

The release of the robotology-superbuild is coherent group of tags for all the subprojects, that are continuously tested together. As BFL still do not have a release for the 2020.11 release of the superbuild, I would like to exclude it from the release 2020.11, while including in the "rolling release" version of the superbuild that we typically used in the lab. However it was not a request for a PR, more a note for me.

traversaro · 2020-11-22T18:14:58Z

I think we need to change the idyntree tag

Once this PR is approved: robotology/idyntree#778 , I will release iDynTree 2.0.0 and then we will have a bfl-compatible iDynTree on both 2020.11 release, Stable and Unstable branches.

…ndencies - manif - qhull - cppad - casadi

GiulioRomualdi · 2020-11-24T17:34:41Z

As discussed in #517 (comment) some dependencies can be installed with brew/vcpkg. We now should decide what to do for them.

For instance, qhull can be installed with brew. Should the superbuild fail if qhull is missing in macOS while it will compile it in Linux?

GiulioRomualdi · 2020-11-24T21:57:46Z

On the conda profile casadi is failing because it's not able to find ipopt. However, ipopt is installed

robotology-superbuild/.github/workflows/ci.yml

Line 55 in 1a9cd37

    
                   mamba install -c conda-forge ace asio boost eigen gazebo glew glfw gsl ipopt libjpeg-turbo libmatio libode libxml2 opencv pkg-config portaudio qt sdl sdl2 sqlite tinyxml

on debian-sid

2020-11-24T17:56:27.5679257Z FAILED: bin/multiple_shooting_from_scratch 
2020-11-24T17:56:27.5681527Z : && /usr/bin/c++ -fPIC -g -rdynamic 
docs/examples/cplusplus/CMakeFiles/multiple_shooting_from_scratch.dir/multiple_shooting_from_scratch.cpp.o -o 
bin/multiple_shooting_from_scratch -L/usr/lib/gcc/lib -Wl,-rpath,/usr/lib/gcc/lib:/__w/robotology-superbuild/robotology-superbuild/build/external/casadi/lib  lib/libcasadi.so.3.6  -ldl && :
2020-11-24T17:56:27.5686120Z /usr/bin/ld: docs/examples/cplusplus/CMakeFiles/multiple_shooting_from_scratch.dir/multiple_shooting_from_scratch.cpp.o: in function `casadi::GenericMatrix<casadi::Matrix<casadi::SXElem> >::sparsity() const':
2020-11-24T17:56:27.5689123Z /__w/robotology-superbuild/robotology-superbuild/external/casadi/./casadi/core/generic_matrix.hpp:860: undefined reference to  `casadi::Matrix<casadi::SXElem>::sparsity() const'
2020-11-24T17:56:27.5690345Z collect2: error: ld returned 1 exit status
2020-11-24T17:56:27.5690961Z ninja: build stopped: subcommand failed.

traversaro · 2020-11-24T22:04:38Z

I would disable for now the ROBOTOLOGY_ENABLE_DYNAMICS_FULL_DEPS in conda and Debian sid. Both are still highly experimental environments that are definitely not our first target.

traversaro · 2020-11-24T22:05:30Z

As discussed in #517 (comment) some dependencies can be installed with brew/vcpkg. We now should decide what to do for them.

For instance, qhull can be installed with brew. Should the superbuild fail if qhull is missing in macOS while it will compile it in Linux?

At the moment I would for the simplest solution, i.e. compile everywhere, we can complexify it once the first PR is landed.

GiulioRomualdi · 2020-11-24T22:06:09Z

on Tags:Release202011@ubuntu-18.04@Release is failing with

BipedalLocomotion::Planners::TimeVaryingDCMPlanner::Impl::computeConstraintElementsECMP(const 
BipedalLocomotion::Planners::ContactPhase&, casadi::DM&, casadi::DM&)’:
2020-11-24T18:11:03.4963635Z /home/runner/work/robotology-superbuild/robotology-superbuild/robotology/BipedalLocomotionFramework/src/Planners/src/TimeVaryingDCMPlanner.cpp:312:53: error: unused variable ‘contactName’ [-Werror=unused-variable]
2020-11-24T18:11:03.4965142Z          for (const auto& [contactName, activeContact] : contactPhase.activeContacts)
2020-11-24T18:11:03.4971726Z                                                      ^
2020-11-24T18:11:07.1228673Z cc1plus: some warnings being treated as errors
2020-11-24T18:11:07.1376820Z src/Planners/CMakeFiles/Planners.dir/build.make:105: recipe for target 'src/Planners/CMakeFiles/Planners.dir/src/TimeVaryingDCMPlanner.cpp.o' failed
2020-11-24T18:11:07.1378922Z CMakeFiles/Makefile2:980: recipe for target 'src/Planners/CMakeFiles/Planners.dir/all' failed
2020-11-24T18:11:07.1383038Z make[5]: *** [src/Planners/CMakeFiles/Planners.dir/src/TimeVaryingDCMPlanner.cpp.o] Error 1
2020-11-24T18:11:07.1384252Z Makefile:146: recipe for target 'all' failed
2020-11-24T18:11:07.1385871Z make[4]: *** [src/Planners/CMakeFiles/Planners.dir/all] Error 2
2020-11-24T18:11:07.1389338Z CMakeFiles/BipedalLocomotionFramework.dir/build.make:134: recipe for target 'robotology/BipedalLocomotionFramework/CMakeFiles/YCMStamp/BipedalLocomotionFramework-build' failed
2020-11-24T18:11:07.1390862Z make[3]: *** [all] Error 2
2020-11-24T18:11:07.1392036Z make[2]: *** [robotology/BipedalLocomotionFramework/CMakeFiles/YCMStamp/BipedalLocomotionFramework-build] Error 2
2020-11-24T18:11:07.1393701Z make[1]: *** [CMakeFiles/BipedalLocomotionFramework.dir/all] Error 2
2020-11-24T18:11:07.1395211Z CMakeFiles/Makefile2:547: recipe for target 'CMakeFiles/BipedalLocomotionFramework.dir/all' failed
2020-11-24T18:11:07.1396710Z Makefile:111: recipe for target 'all' failed
2020-11-24T18:11:07.1397361Z make: *** [all] Error 2
2020-11-24T18:11:07.1416062Z ##[error]Process completed with exit code 2.

This happens because the master version of manif is used. (I think something is missing in the release yml file). This problem has been solved in Fix undesired compile options when gcc is used artivis/manif#157 (manif devel)

traversaro · 2020-11-24T22:07:59Z

I would disable for now the ROBOTOLOGY_ENABLE_DYNAMICS_FULL_DEPS in conda and Debian sid. Both are still highly experimental environments that are definitely not our first target.

As the situation is quickly growing complex for the relative compatibility of options and platform support, I am preparing a matrix to clarify which profiles are a supported on which platform, I am adding it after #535 lands.

traversaro · 2020-11-24T22:16:42Z

* on `Tags:Release202011@ubuntu-18.04@Release` is failing with
  ```
  BipedalLocomotion::Planners::TimeVaryingDCMPlanner::Impl::computeConstraintElementsECMP(const 
  BipedalLocomotion::Planners::ContactPhase&, casadi::DM&, casadi::DM&)’:
  2020-11-24T18:11:03.4963635Z /home/runner/work/robotology-superbuild/robotology-superbuild/robotology/BipedalLocomotionFramework/src/Planners/src/TimeVaryingDCMPlanner.cpp:312:53: error: unused variable ‘contactName’ [-Werror=unused-variable]
  2020-11-24T18:11:03.4965142Z          for (const auto& [contactName, activeContact] : contactPhase.activeContacts)
  2020-11-24T18:11:03.4971726Z                                                      ^
  2020-11-24T18:11:07.1228673Z cc1plus: some warnings being treated as errors
  2020-11-24T18:11:07.1376820Z src/Planners/CMakeFiles/Planners.dir/build.make:105: recipe for target 'src/Planners/CMakeFiles/Planners.dir/src/TimeVaryingDCMPlanner.cpp.o' failed
  2020-11-24T18:11:07.1378922Z CMakeFiles/Makefile2:980: recipe for target 'src/Planners/CMakeFiles/Planners.dir/all' failed
  2020-11-24T18:11:07.1383038Z make[5]: *** [src/Planners/CMakeFiles/Planners.dir/src/TimeVaryingDCMPlanner.cpp.o] Error 1
  2020-11-24T18:11:07.1384252Z Makefile:146: recipe for target 'all' failed
  2020-11-24T18:11:07.1385871Z make[4]: *** [src/Planners/CMakeFiles/Planners.dir/all] Error 2
  2020-11-24T18:11:07.1389338Z CMakeFiles/BipedalLocomotionFramework.dir/build.make:134: recipe for target 'robotology/BipedalLocomotionFramework/CMakeFiles/YCMStamp/BipedalLocomotionFramework-build' failed
  2020-11-24T18:11:07.1390862Z make[3]: *** [all] Error 2
  2020-11-24T18:11:07.1392036Z make[2]: *** [robotology/BipedalLocomotionFramework/CMakeFiles/YCMStamp/BipedalLocomotionFramework-build] Error 2
  2020-11-24T18:11:07.1393701Z make[1]: *** [CMakeFiles/BipedalLocomotionFramework.dir/all] Error 2
  2020-11-24T18:11:07.1395211Z CMakeFiles/Makefile2:547: recipe for target 'CMakeFiles/BipedalLocomotionFramework.dir/all' failed
  2020-11-24T18:11:07.1396710Z Makefile:111: recipe for target 'all' failed
  2020-11-24T18:11:07.1397361Z make: *** [all] Error 2
  2020-11-24T18:11:07.1416062Z ##[error]Process completed with exit code 2.
  ```
  
  
  This happens because the master version of `manif` is used. (I think something is missing in the release `yml` file). This problem has been solved in [artivis/manif#157](https://github.com/artivis/manif/pull/157) (`manif devel`)

As in any case we are not going to add tags for bfl in 2020.11, I think we can skip that profile for now in release CIs, for example adding after all the configure steps:

    - name: Disable profiles that do not support releases for now
      if: contains(matrix.project_tags, 'Release')
      run: | 
        cd build
        cmake -DROBOTOLOGY_ENABLE_DYNAMICS_FULL_DEPS:BOOL=OFF .

GiulioRomualdi · 2020-11-24T22:19:07Z

As in any case we are not going to add tags for bfl in 2020.11,

ops I did it in 082ec05
I remove it now

traversaro · 2020-11-24T22:20:23Z

This happens because the master version of manif is used. (I think something is missing in the release yml file). This problem has been solved in artivis/manif#157 (manif devel)

As in any case we are not going to add tags for bfl in 2020.11,

ops I did it in 082ec05
I remove it now

Ah cool, that's fine then, you can leave them, I did not think we wanted to do that for the moment, but if you did the effort we can leave them.

…work

GiulioRomualdi · 2020-11-25T21:48:53Z

I also added matio-cpp that it's a dependency of blf. See 7e6f604
cc @S-Dafarra

traversaro · 2020-11-25T22:12:57Z

I also added matio-cpp that it's a dependency of blf. See 7e6f604
cc @S-Dafarra

I think we need to add matio to the apt and brew dependencies in docs and CI, in vcpkg binary archive it is already available.

GiulioRomualdi · 2020-11-26T10:06:20Z

I realized that ROBOTOLOGY_ENABLE_DYNAMICS_FULL_DEPS was disabled in Windows CI. b62342b fixes the problem
Furthermore a97dab5 forces casadi to be installed following "normal" folder structure. This is required in windows. Please check here

traversaro · 2020-11-26T11:23:11Z

Casadi failures on Linux:

2020-11-26T11:01:14.0058780Z -- Install configuration: "Debug"
2020-11-26T11:01:14.0059403Z CMake Error at cmake_install.cmake:49 (file):
2020-11-26T11:01:14.0060296Z   file called with unknown argument "lib/cmake/casadi""".

GiulioRomualdi · 2020-11-26T11:55:16Z

@traversaro

Casadi failures on Linux:

2020-11-26T11:01:14.0058780Z -- Install configuration: "Debug"
2020-11-26T11:01:14.0059403Z CMake Error at cmake_install.cmake:49 (file):
2020-11-26T11:01:14.0060296Z   file called with unknown argument "lib/cmake/casadi""".

It should be fixed in 91377a0

GiulioRomualdi · 2020-11-26T15:54:52Z

On windows is failing but I'm not able to find it. @traversaro could you help me?

GiulioRomualdi · 2020-11-26T20:08:35Z

Everything is compiling!!!

traversaro · 2020-11-26T20:11:05Z

Thanks @GiulioRomualdi

GiulioRomualdi mentioned this pull request Nov 22, 2020

Porting bipedal-locomotion-framework in the superbuild #517

Closed

traversaro reviewed Nov 22, 2020

View reviewed changes

GiulioRomualdi added 4 commits November 24, 2020 17:28

Add ROBOTOLOGY_ENABLE_DYNAMICS_FULL_DEPS compile option and some depe…

5621b59

…ndencies - manif - qhull - cppad - casadi

Implement Buildbipedal-locomotion-framework.cmake file

7d6db2c

Enable the compilation of bipedal-locomotion-framework in the superbuild

1c7aabb

Update the CI to enable the compilation of bipedal-locomotion-framework

9503fa9

GiulioRomualdi force-pushed the add_blf branch from 9a798ea to 9503fa9 Compare November 24, 2020 16:28

GiulioRomualdi added 2 commits November 24, 2020 18:29

Update Buildcasadi.cmake

b6186b5

Update Buildqhull.cmake

19fd7ab

GiulioRomualdi linked an issue Nov 24, 2020 that may be closed by this pull request

Porting bipedal-locomotion-framework in the superbuild #517

Closed

Update 2020.11.yaml

082ec05

GiulioRomualdi force-pushed the add_blf branch from 24fb79b to d0056b0 Compare November 25, 2020 11:43

GiulioRomualdi added 6 commits November 25, 2020 22:37

Rename BipedalLocomotionFramework into bipedal-locomotion-framework

3f1107d

Pass FRAMEWORK_USE_* options while compiling bipedal-locomotion-frame…

ab99357

…work

Implement Buildmatio-cpp.cmake file

7e6f604

Enable the compilation of matio-cpp

422a9c2

Fix missing dependencies in bipedal-locomotion-framework.cmake

9f5e551

Add matio-cpp in 2020.11.yaml file

369e137

GiulioRomualdi self-assigned this Nov 25, 2020

GiulioRomualdi added 5 commits November 26, 2020 10:40

Install libmatio-dev in Linux CI

6130a33

Merge remote-tracking branch 'origin/master' into add_blf

320ff40

Update matio installation guide in doc/profiles.md

520d3e1

Make Buildcasadi.cmake compatible with Windows

a97dab5

Enable ROBOTOLOGY_ENABLE_DYNAMICS_FULL_DEPS in Windows CI

b62342b

Fix typos in Buildcasadi.cmake

91377a0

GiulioRomualdi mentioned this pull request Nov 26, 2020

Add cppad to the compiled ports robotology/robotology-superbuild-dependencies-vcpkg#37

Merged

Disable the compilation of CppAD on windows

8ddb11c

traversaro approved these changes Nov 26, 2020

View reviewed changes

GiulioRomualdi closed this Nov 26, 2020

GiulioRomualdi reopened this Nov 26, 2020

traversaro merged commit b3e3833 into robotology:master Nov 26, 2020

GiulioRomualdi deleted the add_blf branch November 26, 2020 21:19

diegoferigo mentioned this pull request Feb 8, 2021

Remove qpOASES from the Dynamics component and the superbuild? #239

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add bipedal-locomotion-framework #526

Add bipedal-locomotion-framework #526

GiulioRomualdi commented Nov 22, 2020

traversaro commented Nov 22, 2020

traversaro Nov 22, 2020

GiulioRomualdi Nov 25, 2020

GiulioRomualdi Nov 25, 2020

traversaro Nov 22, 2020

traversaro Nov 22, 2020

traversaro commented Nov 22, 2020

traversaro commented Nov 22, 2020

GiulioRomualdi commented Nov 22, 2020

traversaro commented Nov 22, 2020

GiulioRomualdi commented Nov 22, 2020

traversaro commented Nov 22, 2020

traversaro commented Nov 22, 2020

GiulioRomualdi commented Nov 24, 2020

GiulioRomualdi commented Nov 24, 2020 •

edited

Loading

traversaro commented Nov 24, 2020

traversaro commented Nov 24, 2020

GiulioRomualdi commented Nov 24, 2020

traversaro commented Nov 24, 2020

traversaro commented Nov 24, 2020

GiulioRomualdi commented Nov 24, 2020 •

edited

Loading

traversaro commented Nov 24, 2020

GiulioRomualdi commented Nov 25, 2020

traversaro commented Nov 25, 2020

GiulioRomualdi commented Nov 26, 2020

traversaro commented Nov 26, 2020

GiulioRomualdi commented Nov 26, 2020

GiulioRomualdi commented Nov 26, 2020

GiulioRomualdi commented Nov 26, 2020

traversaro commented Nov 26, 2020


		include(YCMEPHelper)

		set(bipedal-locomotion-framework_DEPENDS "")

	set(bipedal-locomotion-framework_DEPENDS "")
	set(BipedalLocomotionFramework_DEPENDS "")

Add bipedal-locomotion-framework #526

Add bipedal-locomotion-framework #526

Conversation

GiulioRomualdi commented Nov 22, 2020

traversaro commented Nov 22, 2020

traversaro Nov 22, 2020

Choose a reason for hiding this comment

GiulioRomualdi Nov 25, 2020

Choose a reason for hiding this comment

GiulioRomualdi Nov 25, 2020

Choose a reason for hiding this comment

traversaro Nov 22, 2020

Choose a reason for hiding this comment

traversaro Nov 22, 2020

Choose a reason for hiding this comment

traversaro commented Nov 22, 2020

traversaro commented Nov 22, 2020

GiulioRomualdi commented Nov 22, 2020

traversaro commented Nov 22, 2020

GiulioRomualdi commented Nov 22, 2020

traversaro commented Nov 22, 2020

traversaro commented Nov 22, 2020

GiulioRomualdi commented Nov 24, 2020

GiulioRomualdi commented Nov 24, 2020 • edited Loading

traversaro commented Nov 24, 2020

traversaro commented Nov 24, 2020

GiulioRomualdi commented Nov 24, 2020

traversaro commented Nov 24, 2020

traversaro commented Nov 24, 2020

GiulioRomualdi commented Nov 24, 2020 • edited Loading

traversaro commented Nov 24, 2020

GiulioRomualdi commented Nov 25, 2020

traversaro commented Nov 25, 2020

GiulioRomualdi commented Nov 26, 2020

traversaro commented Nov 26, 2020

GiulioRomualdi commented Nov 26, 2020

GiulioRomualdi commented Nov 26, 2020

GiulioRomualdi commented Nov 26, 2020

traversaro commented Nov 26, 2020

GiulioRomualdi commented Nov 24, 2020 •

edited

Loading

GiulioRomualdi commented Nov 24, 2020 •

edited

Loading