Conversation
|
Sorry for the duplicate tag, I saw that this was fixing the same issue as #33395 but not that this was a solution inside the Fwiw, I think having docs and tests would be very helpful here for deciding whether we want to go this route (this would be the first time that |
|
I think it's better to solve this at the |
It’s more of the reverse situation: Windows/NTFS uses UTF-16 for filenames, which we expose to our users as UTF-8, so strings work just fine there. However, while you could call filesystems on other OSes “UTF-8-capable”, what that means in reality is that no specific encoding is enforced at all. That’s why we have (That being said, I don’t have a strong opinion here.) |
Citing the use of this test case fromshackijj
|
I’ve renamed the PR to be a bit more accurate, and so it doesn’t get confused with #33395 |
|
@kkz13250 @addaleax I need buffer support for several of I've started working on an implementation here first before pulling it into Node.js. Would love feedback, specifically I want to make sure we handle some of the concerns around UTF-16. |
5bbd1f7 to
1a626e6
Compare
Internalized the handling of Buffer type in the join function in path.js,