Replies: 1 comment
|
looked at difftastic before, the model is interesting but doesn't fit revdiff's design. revdiff is built around unified-diff line format end to end. a few other practical issues:
so not planning to pursue this. If you have a concrete scenario where the current diff (with word-diff on) fails badly, drop a sample, that's more useful than the general direction. |
Uh oh!
There was an error while loading. Please reload this page.
Is it worthy exploring https://github.com/Wilfred/difftastic style differ to get higher SNR diffs?
They seem to provide machine readable output format
All reactions