mergiraf/mergiraf
mergiraf/mergiraf
9
213
Fork 45
Code Issues 51 Pull requests 7 Releases 19 Activity Actions

fix: Ignore empty nodes in AstNode::isomorphic_to_source #480

Merged
wetneb merged 3 commits from wetneb/fix_isomorphism into main 2025-07-07 22:43:24 +02:00 AGit
Conversation 5 Commits 3 Files changed 7 +39 -2
wetneb commented 2025-07-07 19:33:05 +02:00
Owner
Copy link

When checking that our merged output is isomorphic to our internal tree (as a consistency check), it can be that isomorphism fails because one of the sides contains extra empty nodes.

I think it makes sense to filter out those nodes to avoid that.

WIP because it depends of #478.

When checking that our merged output is isomorphic to our internal tree (as a consistency check), it can be that isomorphism fails because one of the sides contains extra empty nodes. I think it makes sense to filter out those nodes to avoid that. <s>WIP because it depends of #478.</s>
chore(go.mod): Update parser to 0.4.0
All checks were successful
/ test (pull_request) Successful in 1m52s
Details
f7ad7f9f3b 
This adds support for `ignore` and `godebug` directives,
bringing the parser on-par to the specs (as far as I can tell).
wetneb changed title from fix: Ignore empty nodes in AstNode::isomorphic_to_source to WIP: fix: Ignore empty nodes in AstNode::isomorphic_to_source 2025-07-07 19:33:27 +02:00
ada4a commented 2025-07-07 19:48:49 +02:00
Owner
Copy link

WIP because it depends of #478.

Not logically at least, I don't think? Unless it's the go.mod tests that revealed this issue, but still I think it would be to nice a have separate test for this..

> WIP because it depends of #478. Not logically at least, I don't think? Unless it's the go.mod tests that revealed this issue, but still I think it would be to nice a have separate test for this..
wetneb commented 2025-07-07 19:52:01 +02:00
Author
Owner
Copy link

Yes, the test I have included in 4c56d94338 is specific for this issue but uses a feature of go.mod that requires #478 to be parsed in the first place.

Yes, the test I have included in https://codeberg.org/mergiraf/mergiraf/commit/4c56d94338b68cc841442010462f482448004df6 is specific for this issue but uses a feature of `go.mod` that requires #478 to be parsed in the first place.
ada4a commented 2025-07-07 19:58:41 +02:00
Owner
Copy link

the test I have included in 4c56d94338

yes but that's an integration test -- I was wondering whether we could make just a unit test specifically for AstNode::isomorphic_to_source. Let me see if I can come up with something...

> the test I have included in 4c56d94338 yes but that's an integration test -- I was wondering whether we could make just a unit test specifically for `AstNode::isomorphic_to_source`. Let me see if I can come up with something...
wetneb commented 2025-07-07 20:01:07 +02:00
Author
Owner
Copy link

Given that it's a method of MergedTree, it involves creating a MergedTree instance… it would be great to be able to create those more easily for unit tests, but so far we've just being creating them by running the merge process, so it's not very far from an integration test.

Given that it's a method of `MergedTree`, it involves creating a `MergedTree` instance… it would be great to be able to create those more easily for unit tests, but so far we've just being creating them by running the merge process, so it's not very far from an integration test.
ada4a commented 2025-07-07 20:15:05 +02:00
Owner
Copy link

it would be great to be able to create those more easily for unit tests

Fully agree!

Maybe at least add a README.md saying that that integration test also tests the case fixed by this PR, and link the PR -- I'm sure that whoever stumbles upon a regression in this will be very thankful^^

> it would be great to be able to create those more easily for unit tests Fully agree! Maybe at least add a README.md saying that that integration test also tests the case fixed by this PR, and link the PR -- I'm sure that whoever stumbles upon a regression in this will be very thankful^^
ada4a reviewed 2025-07-07 20:18:19 +02:00
src/merged_tree.rs
@ -406,0 +407,4 @@
MergedChild::Original(child) => {
!child.source.is_empty()
},
MergedChild::Merged(MergedTree::ExactTree { node, revisions, .. }) => {
ada4a commented 2025-07-07 20:17:55 +02:00
Owner
Copy link

maybe reorder these two match arms to keep all the MergedChild::Mergeds together? Or you could put the MergedChild::Original case as the very first, since the MergedChild::Merged are roughly in the order of ascending complexity

maybe reorder these two match arms to keep all the `MergedChild::Merged`s together? Or you could put the `MergedChild::Original` case as the very first, since the `MergedChild::Merged` are _roughly_ in the order of ascending complexity
👍 1
ada4a marked this conversation as resolved
wetneb force-pushed wetneb/fix_isomorphism from 4c56d94338 to adefd0dc90 2025-07-07 22:13:15 +02:00 Compare
wetneb changed title from WIP: fix: Ignore empty nodes in AstNode::isomorphic_to_source to fix: Ignore empty nodes in AstNode::isomorphic_to_source 2025-07-07 22:13:38 +02:00
wetneb force-pushed wetneb/fix_isomorphism from adefd0dc90 to bb0e1bacc2 2025-07-07 22:16:52 +02:00 Compare
Move MergedChild::Original case at the top
All checks were successful
/ test (pull_request) Successful in 55s
Details
52c0af226b
Document the relationship between code and test
All checks were successful
/ test (pull_request) Successful in 59s
Details
0fb87d6f6c
ada4a approved these changes 2025-07-07 22:33:17 +02:00
ada4a left a comment
Owner
Copy link

Very nice!

Very nice!
wetneb merged commit d18cf1c489 into main 2025-07-07 22:43:24 +02:00
Sign in to join this conversation.
Reviewers
No reviewers
ada4a
Labels
Clear labels   
Compat/Breaking

Breaking change that won't be backward compatible

  
Kind
Bad merge
Examples where the merging algorithm doesn't work as well as it should

  
Kind
Bug
Something is not working

  
Kind
Documentation
Documentation changes

  
Kind
Enhancement
Improve existing functionality

  
Kind
Feature
New functionality

  
Kind
New language

  
Kind
Security
This is security issue

  
Kind
Testing
Issue or pull request related to testing

  
Priority
Critical
The priority is critical

  
Priority
High
The priority is high

  
Priority
Low
The priority is low

  
Priority
Medium
The priority is medium

  
Reviewed
Confirmed
Issue has been confirmed

  
Reviewed
Duplicate
This issue or pull request already exists

  
Reviewed
Invalid
Invalid issue

  
Reviewed
Won't Fix
This issue won't be fixed

  
Status
Abandoned
Somebody has started to work on this but abandoned work

  
Status
Blocked
Something is blocking this issue or pull request

  
Status
Need More Info
Feedback is required to reproduce issue or to continue work

No labels
Compat/Breaking
Kind
Bad merge
Kind
Bug
Kind
Documentation
Kind
Enhancement
Kind
Feature
Kind
New language
Kind
Security
Kind
Testing
Priority
Critical
Priority
High
Priority
Low
Priority
Medium
Reviewed
Confirmed
Reviewed
Duplicate
Reviewed
Invalid
Reviewed
Won't Fix
Status
Abandoned
Status
Blocked
Status
Need More Info
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
2 participants
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference: mergiraf/mergiraf#480
No description provided.
Delete branch "wetneb/fix_isomorphism"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?