Simplify DCEL label management #469
Merged
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
In the DCEL data structure, there are labels on the vertex records, (half) edge records, and face records. These labels keep track information such as which input geometry the record originates from and whether it will appear in the output geometry. The original approach to labelling was to use the same data structure for all record types. The label had two booleans per operand. The first boolean determined whether the record would appear in the final output, and the second boolean would show whether the first boolean was populated or not. This was essentially tri-state logic. The trouble with this approach was that the DCEL algorithm doesn't know whether each record should appear in the output geometry right up until the last stage. The labels are used in the earlier stages of the algorithm to record other facts, such as whether or not the records appear in the input geometry. The semantics of the fields subtly change over the course of the algorithm, making the code hard to understand and maintain. The new approach (implemented by this PR) changes how label management is performed. Instead of having a single labels data structure, the fields are slightly different for vertex, edge, and face records. This is to account for the differences in information available when processing vertexes vs. edges vs. faces. The number of fields is also expanded, so that concepts like "is this record explicitly present in the input geometry's control points" are separate from concepts like "is this record present in the input geometry but not explicitly in its control points". The populated fields are also removed, and tracked separately in each algorithm that modifies the labels.
albertteoh
approved these changes
Nov 11, 2022
| @@ -13,7 +13,7 @@ func Union(a, b Geometry) (Geometry, error) { | |||
| if b.IsEmpty() { | |||
| return a, nil | |||
| } | |||
| g, err := setOp(a, b, selectUnion) | |||
| g, err := setOp(a, or, b) | |||
There was a problem hiding this comment.
This reads much more nicely 👍🏼
Yes, that's right. The original implementation reused the labels struct or each of the 3 record types which saved a bit of code but was a fairly inaccurate/confusing modelling. |
|
Thanks for the review! 😁 |
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.
Description
commit 45ccc4b
Author: Peter Stace peterstace@gmail.com
Date: Mon Nov 7 20:19:20 2022 +1100
Check List
Have you:
Added unit tests? No, relies on existing.
Add cmprefimpl tests? (if appropriate?) No, relies on existing.
Updated release notes? (if appropriate?) No, relies on existing.
Related Issue
Benchmark Results
Click to expand