You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, if you pass a patch as a string to applyPatch and it has incorrect line counts in the hunk headers, we'll fail to apply it but won't throw an error to tell you that the patch itself was invalid. That seems bad; I spent a long time struggling to figure out why a patch wasn't applying in #238 because of this.
I don't really see any reason for strict mode not to be on by default, and suggest we make it so in the next major version.
The text was updated successfully, but these errors were encountered:
While we're at it, we should probably also fix applyPatch to be able to apply a patch where the only thing wrong is the line counts. Either that or we should rip the strict option out of jsdiff entirely and make what we currently call "strict" mode be the only mode. It's misleading to have a non-strict mode if patches we parse in non-strict mode aren't actually usable by jsdiff.
Oh, man, when I make strict mode be always-on I see thirty test failures when running yarn test. I guess I need to figure out which of those (if any) are valid patches that strict mode was wrongly rejecting before, and which of them (if any) are invalid patches we were trying to parse or apply in our tests.
Currently, if you pass a patch as a string to
applyPatch
and it has incorrect line counts in the hunk headers, we'll fail to apply it but won't throw an error to tell you that the patch itself was invalid. That seems bad; I spent a long time struggling to figure out why a patch wasn't applying in #238 because of this.I don't really see any reason for strict mode not to be on by default, and suggest we make it so in the next major version.
The text was updated successfully, but these errors were encountered: