Conversation
|
Ok, sure, why not |
|
Ready for review |
|
Other than missing implementation for minimal diff, that's all I can say. |
|
@Therzok I've been refraining from implementing the Any idea? |
|
Only seems to do things for big diffs. I just did: Returned same results until around ~600 commits, and I tracked down what changed the stuff. It seems to be XML that's the culprit of making minimal better, but only in a big diff context. Here's the diff between the two. https://gist.github.com/Therzok/0f1f127ab2c9bb95681b They make chrome choke when trying to gist them: Differences also happen in libgit2sharp repo with ~300 commits from HEAD. |
|
In short it only affects huge diffs. This would be useful for someone diffing HEAD to 1 year prior, but not many people do this. So it would be best to keep it. Short things short: minimal is just myers with better diffs for huge stuff. Edit: The time difference is actually not noticeable. |
Expose a more generic DiffAlgorithm property in CompareOptions
|
Published as NuGet pre-release package |
Following #1039 and in order to avoid future breaking changes maybe would it be wise to expose a property accepting an enum of Diff algorithms (Meyer, Minimal, ...).
/cc @dmalikov