Skip to content

fix wrong edge id in bidir a* expansion#5265

Merged
chrstnbwnkl merged 4 commits intomasterfrom
cb-bidirastar-reverse-expansion
May 27, 2025
Merged

fix wrong edge id in bidir a* expansion#5265
chrstnbwnkl merged 4 commits intomasterfrom
cb-bidirastar-reverse-expansion

Conversation

@chrstnbwnkl
Copy link
Member

Issue

While debugging, I noticed that it was impossible to trace the reverse search expansion, and saw that instead of passing the reverse predecessor, its opposing edge ID was passed (i.e. the one used for traversal only).

@kevinkreiser
Copy link
Member

how is it possible we keep finding these. it feels like after all this time it shouldnt happen anymore especially with the "main" algorithm haha

@chrstnbwnkl
Copy link
Member Author

To be fair, you would only go knees deep into this if you suspect there is something wrong with the algorithm, but bidirectional A* is just flawless at this point 😄

I see the dedupe tests are failing because now there are less features with the same ID. To me this seems expected, if we tracked the wrong direction in reverse, there'd be more duplicates

@kevinkreiser
Copy link
Member

yeah i guess the expansion tracing is "newer" too so its less "settled"

kevinkreiser
kevinkreiser previously approved these changes May 23, 2025
Copy link
Member

@kevinkreiser kevinkreiser left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

preemptive ship for when you get the tests updated

kevinkreiser
kevinkreiser previously approved these changes May 27, 2025
@chrstnbwnkl chrstnbwnkl merged commit 241ebce into master May 27, 2025
9 checks passed
@chrstnbwnkl chrstnbwnkl deleted the cb-bidirastar-reverse-expansion branch May 27, 2025 14:06
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants