Checking for paths.length after filtering #43
Conversation
xs.map(x => String(x)) is the same as xs.map(String)
| @@ -1,3 +1,3 @@ | |||
| * text=auto | |||
| * text eol=lf | |||
There was a problem hiding this comment.
This doesn't work like you think. It's super flawed and really means "convert everything to text and with lf newlines" instead of meaning "convert text to use lf newlines". So it will ruin anything binary.
There was a problem hiding this comment.
What about changing it to
*.js test eol=lf
There was a problem hiding this comment.
Partly because of the .editorconfig and mostly because when I cloned and ran the test, I got an error for every line telling me it encountered a CRLF but expected a LF. I did not have an editor config plugin installed at first.
Probably an oversight when globbing was added. It should definitely check the final paths array. |
| @@ -1,3 +1,3 @@ | |||
| * text=auto | |||
There was a problem hiding this comment.
This should still be preserved.
|
Thank you! Merry xmas 🎄 |
|
Haha thanks you! Happy hollidays! |
Hey @sindresorhus,
I'm not sure if the paths.length === 0 check is before the filter(fileExist) check on purpose but to me it makes sense to put it after the filter.
If I want a file to be removed but it doesn't exist in the first place then that is perfectly fine with me. There shouldn't be an error (which the windows binary yields).
Also I set the line endings to LF in the .gitattributes file and made some code simplifications (at least I think so).
Let me know what you think :)