Use ROS isolated version of tests #133
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: Pete Baughman <pete.baughman@apex.ai>
pbaughman
force-pushed
the
test_isolation
branch
from
84b05a1
to
a1a457a
Compare
|
edit also test_play in rosbag2_transport |
thomas-moulard
pushed a commit
to thomas-moulard/rosbag2
that referenced
this pull request
Oct 22, 2019
* ros2GH-138 Move calculation of bag size - previously in rosbag2::Info - now in storage plugin * ros2GH-130 Add rosbag2_bag_v2_plugins package -This package will contain storage and converter plugins * ros2GH-131 don't build plugins on Windows * ros2GH-129 Add function to be generated - massive if/else between all message types - will be generated similar to ros1_bridge plugin * ros2GH-138 Write storage plugin for rosbag v2 bags * ros2GH-138 Make sure that no attempt to create a converter is made when trying to read a rsbag v2 bag file * ros2GH-138 Add play end-to-end test for rosbag v2 plugin * ros2GH-138 Use cmake files to find ros1 packages - Use files from ros1_bridge via PkgConfig * ros2GH-138 Add generator code * ros2GH-141 Add initial version of vendor package * ros2GH-141 Improve vendor package to build on Mac * ros2GH-138 Cleanup CMakeLists * ros2GH-141 Use unmanaged Instance of class-loader - managed instance somehow isn't available for gcc 6.3 * ros2GH-141 Reduce patch and copy new toplevle CMakeLists by hand * ros2GH-141 Fix Shared Instance usage * ros2GH-141 Improve maintainability of vendor package - Document what patches do and why changes are necessary - Load ros1 packages through cmake macro - Do not export ros1 packages via ament - use commit hash of current master which is more stable than using melodic-devel * ros2GH-138 Link against rclcpp - necessary for ros1_bridge * ros2GH-138 Avoid crash when trying to play v2 bags which contain unknown message types * ros2GH-138 Add CLI -s <storage_id> option to ros2 bag info and use it in rosbag2::info - this allows ros2 bag info to work also when the yaml metadata file does not exsist - this is always the case for rosbag1 bagfiles - it could also happen for sqlite or other storage based bagfiles * ros2GH-138 Add end-to-end info test for rosbag v2 files * ros2GH-138 Add unit tests to rosbag_storage * ros2GH-138 Add method to extract filename from path to FilesystemHelpers * ros2GH-138 Add proper logging for topics which cannot be converted * ros2GH-138 Improve finding dependencies of ros1 * ros2GH-141 Explicitly import transitive dependencies of vendor package * ros2GH-138 Skip tests via ament if ros1 is not available * ros2GH-133 First split of plugins * ros2GH-133 Write serialized rosbag message * ros2GH-133 Improve converter plugin - move generation templates outside of plugin folders as both plugins need it - use ros::serialization routines to deserialize the ros message * ros2GH-133 Add plugin to be found by pluginlib * ros2GH-133 Remove empty check in converter - With the rosbag_v2_converter_plugin, we don't need to treat rosbag_v2 storage any different * ros2GH-133 Assert serialization format in unit tests for storage * ros2GH-133 Delete superfluous include folder - Only needed if we want to link against the library * ros2GH-133 get_all_topics_and_types returns only valid ros2 types - This is necessary as the information is used by rosbag2_transport - ros2 bag info still shows all topics and types - rosbag::View::getConnections() can return multiple connections corresponding to the same topic * ros2GH-133 Improve end to end test - use a bagfile with messages not known to ros2 * ros2GH-133 Reformulate info message in case of missing ros1-ros2 mapping for a topic * ros2GH-14 Find messages first * Explicitly print message when on Windows Co-Authored-By: Martin-Idel-SI <external.Martin.Idel@bosch-si.com> * ros2GH-14 Refactor rosbag_storage vendor package - Improve toplevel CMakeLists - Put all patches into a resource subfolder * ros2GH-14 Reflect renames of converter interfaces * ros2GH-156 Workaround for path problems * ros2GH-156 Add documentation for plugin * ros2GH-156 Fix the pluginlib version to greater 2 * ros2GH-156 Prohibit CMake from declaring paths as system paths This switches the order of ros2 and ros1 directories resulting in build failures * ros2GH-156 Prohibit system include paths for rosbag plugins This can lead to switching ros1 and ros2 include paths resulting in missing symbols as the wrong pluginlib gets included * ros2GH-14 Split patches * make README more verbose * add plugin specific readme * more readme for bag_v2 plugin
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This is my suggested fix for rosbag2 tests failing when run in parallel with other tests.
I've checked this by running the rosbag2 tests in a loop while I run
colcon test --packages-up-to rcl rclcpp rclpy
In another terminal window. Prior to this change, I would see test failures every time.
I didn't change all of the tests to use the ros_isolated version - only the ones that were failing because of interference.