Solve #600 (ExtrapolationException & its error_code) #601
Conversation
|
This patch would be important to include in Iron's release |
|
Thank you both for this! I'll fire up CI on this, and assuming that it passes I'll merge it in and then backport to Iron. CI:
|
|
@clalancette CI looks green! 🐢 |
|
@Mergifyio backport iron |
✅ Backports have been created
|
|
@roncapat Thanks again for the fix! Since this was a pretty urgent fix, I went ahead and merged it as-is, and also backported it to |
|
@clalancette it was a pleasure for me (and also an headache, I'm very sorry for having introduced this nasty bug). I was thinking exactly the same... we need a test for it. I lack time now for work reason, but if you want assign to me an issue as a reminder, I will make sure to think about it as soon as I can. |
* error_code shall follow same path of its error string * Lint (cherry picked from commit ac2479e) Co-authored-by: Patrick Roncagliolo <[email protected]>
in interim until next rolling sync - ros2/geometry2#601
Thank you, that is appreciated. I've now opened up #603 and assigned it to you. |
in interim until next rolling sync - ros2/geometry2#601
in interim until next rolling sync - ros2/geometry2#601
in interim until next rolling sync - ros2/geometry2#601 Signed-off-by: enricosutera <[email protected]>
This solves #600. There was an oversight in how the code, before introduction of
error_code, handled theerror_string. In particular, one code path saves the error in a "side" (so not directly inerror_stringbut inextrapolation_error_string) and this HAS to be done also with the newerror_codestuff.