Fix HierarchicalPathFinder considering some unreachable cells as reachable #20239
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
RoosterDragon
force-pushed
the
fix-hpf-domains
branch
from
b791564
to
355535b
Compare
|
Need a rebase |
…hable. When using the internal AbstractCellForLocalCell method to check if a local cell is reachable, this should return null when the cell is unreachable. If multiple abstract cells were required for that grid, this worked as intended. Only reachable cells are stored in the localCellToAbstractCell mapping. For a grid that required only a single abstract cell, which is the common case, we optimize this to store only the single abstract cell rather than the whole mapping for potentially 100 cells in that grid. However this makes no distinction between the reachable and unreachable cells, so when we check later we get incorrect results. If a cell is unreachable but belongs to the same grid as a single group of reachable cells then we incorrectly report it as reachable. The easiest way to see this incorrect behaviour is when the PathExists is called and can sometimes indicate a path exists when it does not. To fix this, we now ensure we perform a check to see if the cell is reachable in this single layer case, this allows us to retain the optimization where we don't need to store the whole mapping, but allows us to correctly indicate when cells are unreachable.
RoosterDragon
force-pushed
the
fix-hpf-domains
branch
from
355535b
to
b543c85
Compare
|
Rebased. |
PunkPun
approved these changes
Sep 2, 2022
|
My test was the following:
|
|
Sounds like this should be marked as depending on #20227. |
Mailaender
reviewed
Sep 3, 2022
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
When using the internal AbstractCellForLocalCell method to check if a local cell is reachable, this should return null when the cell is unreachable. If multiple abstract cells were required for that grid, this worked as intended. Only reachable cells are stored in the localCellToAbstractCell mapping. For a grid that required only a single abstract cell, which is the common case, we optimize this to store only the single abstract cell rather than the whole mapping for potentially 100 cells in that grid. However this makes no distinction between the reachable and unreachable cells, so when we check later we get incorrect results. If a cell is unreachable but belongs to the same grid as a single group of reachable cells then we incorrectly report it as reachable. The easiest way to see this incorrect behaviour is when the PathExists is called and can sometimes indicate a path exists when it does not.
To fix this, we now ensure we perform a check to see if the cell is reachable in this single layer case, this allows us to retain the optimization where we don't need to store the whole mapping, but allows us to correctly indicate when cells are unreachable.
I have confirmed this resolves the suspect behaviour noted in #20227 (comment).