Conversation
|
This PR should be reviews used changes in ros-perception/image_common#84, right? |
|
Correct, it depends on the image_transport port to ROS2. |
| { | ||
|
|
||
| void CompressedPublisher::advertiseImpl(ros::NodeHandle &nh, const std::string &base_topic, uint32_t queue_size, | ||
| const image_transport::SubscriberStatusCallback &user_connect_cb, |
There was a problem hiding this comment.
We are not porting the SubscriberStatusCallback to ROS2. Looking for the similiar feature I found a answer.ros.org by William explaining in March that the feature is not implemented yet. I assume that it is still unimplemented.
There was a problem hiding this comment.
Correct, it remains unimplemented.
| float cRatio = (float)(cv_ptr->image.rows * cv_ptr->image.cols * cv_ptr->image.elemSize()) | ||
| / (float)compressed.data.size(); | ||
| ROS_DEBUG("Compressed Image Transport - Codec: jpg, Compression Ratio: 1:%.2f (%lu bytes)", cRatio, compressed.data.size()); | ||
| //ROS_DEBUG("Compressed Image Transport - Codec: jpg, Compression Ratio: 1:%.2f (%lu bytes)", cRatio, compressed.data.size()); |
There was a problem hiding this comment.
RCUTILS_LOG_DEBUG here and in the rest of removals?
| else | ||
| { | ||
| ROS_ERROR("cv::imencode (png) failed on input image"); | ||
| //ROS_ERROR("cv::imencode (png) failed on input image"); |
There was a problem hiding this comment.
RCUTILS_LOG_ERROR here and in the rest of removals?
| void CompressedPublisher::advertiseImpl(ros::NodeHandle &nh, const std::string &base_topic, uint32_t queue_size, | ||
| const image_transport::SubscriberStatusCallback &user_connect_cb, | ||
| const image_transport::SubscriberStatusCallback &user_disconnect_cb, | ||
| const ros::VoidPtr &tracked_object, bool latch) |
There was a problem hiding this comment.
tracked_object was removed from the arguments, just checking that it was on propose.
There was a problem hiding this comment.
Correct, it was removed in the image_transport as well.
| } | ||
|
|
||
|
|
||
| void CompressedSubscriber::configCb(Config& config, uint32_t level) |
There was a problem hiding this comment.
not sure if it could help the future implementation of parameters to leave the configCb code as a comment
There was a problem hiding this comment.
I made a config struct that mirrors the config, I figure we can always jump back in the git log to get this.
|
Latest changes look good to me. |
Basic port
TODO: