Expose cdr #267
Expose cdr #267
Conversation
test_communication/CMakeLists.txt
Outdated
| macro(serialize) | ||
| set(serialize_target_name "test_serialize${target}${target_suffix}") | ||
| ament_add_gtest( | ||
| ${serialize_target_name} test/test_message_serialization.cpp |
There was a problem hiding this comment.
recommend indenting the arguments to ament_add_gtest()
| TEST_F(CLASSNAME(TestMessageSerialization, RMW_IMPLEMENTATION), raw_callback) { | ||
| size_t counter = 0; | ||
|
|
||
| auto raw_callback = [&counter](const std::shared_ptr<rmw_message_raw_t> raw_msg) { |
There was a problem hiding this comment.
It looks like this test will pass if raw_callback is never called. Maybe add EXPECT_GT(counter, 0u) after the loop at the bottom?
There was a problem hiding this comment.
done in f6400e0
|
|
||
| auto ret = RMW_RET_OK; | ||
| ret = | ||
| rmw_serialize(&bounded_array_nested_msg_c, message_c_typesupport, &raw_message_c); |
There was a problem hiding this comment.
It looks like this call would fit on the line above
| EXPECT_EQ(RMW_RET_OK, ret); | ||
|
|
||
| print_raw_buffer(raw_message_c, "raw message c"); | ||
| print_raw_buffer(raw_message_cpp, "raw message cpp"); |
There was a problem hiding this comment.
Are raw_message_c.buffer and raw_message_cpp.buffer expected to be identical?
There was a problem hiding this comment.
I believe this might be RMW implementation dependent.
* raw->serialized * use size_t
|
This change might be the reason for the recently failing build: https://ci.ros2.org/job/ci_osx/3897/ |
|
but the static typesuppport branch has conflicts. Most likely due to my merge. |
Connects to ros2/demos#185