NavigationDataPlayer readings from file differ from NavigationDataRecorder outputs retrieved from the according NavigationDataPlayer. For details refer to Bug-11635 where a testing takes place that reveals this bug.
Description
Description
| Status | Assigned | Task | ||
|---|---|---|---|---|
| Duplicate | None | T11971 NavigationDataPlayer readings differ from NavigationDataRecorder outputs | ||
| Duplicate | None | T11635 Improve test of class NavigationDataRecorder |
Event Timeline
Comment Actions
It appears that the Datarecorder is the source of error. Preliminary testing suggests faithful reproduction of recorded data by SequentialDataPlayer.
Comment Actions
Turns out that this is a bug that only shows up specifically in the situation that a recorder is connected to a sequential player recording multiple tools.
What happens is that the recorder calls Update() on each of it's input. The sequential Player is happy to oblige and updates each input eacht time it recieves an update call. That means that when recording 2 tools, only half of the ouput is generated
One should discuss whether this behaviour is acceptable or not. It will however be highly difficult to fix. Assigning back to Chief of module.
Additional Notes:
- Current implementation is a little unclean regarding pipeline behaviour. One might consider redisign (GenerateData Method is empty, but Update has proprietary Implementation.
- Implementation seems to round values pretty randomly. This makes it practically unsuable for scientific purposes.