Closed tmcw closed 2 years ago
Yes, the file-save
legacy handling I wrote incorrectly. I'll make a fix in the coming days.
Thanks, @dwelle! By the way, I noticed that the current Excalidraw workaround doesn't trigger in realtime for Firefox when I played with it again now. It's similarly ignored as was my earlier attempt based on similar ideas (https://github.com/GoogleChromeLabs/browser-fs-access/commit/9340b5810e4a311e3715a63121f765eecf6f829b and even earlier). It seems to trigger upon saving: open a file dialog, cancel it, then save, and only then you can see the "file open was canceled" log message trigger.
Thanks, @dwelle! By the way, I noticed that the current Excalidraw workaround doesn't trigger in realtime for Firefox when I played with it again now. It's similarly ignored as was my earlier attempt based on similar ideas (9340b58 and even earlier). It seems to trigger upon saving: open a file dialog, cancel it, then save, and only then you can see the "file open was canceled" log message trigger.
Mhm, when did it regress, because I'm sure it worked around the time I was writing it ðŸ˜
Wait, nvm, it still works for me. I've had warnings
disabled in FF. Don't ever recall turning it off, can you check if that's the case for you as well?
It works somewhat, but I don't get the logic by which it works:
https://user-images.githubusercontent.com/145676/165718054-165f4f51-31ae-4b99-acee-052b2e7c9e3b.mp4
Ok, I was checking on Windows. I can repro the same issue on MacOS as you. Will take a look later.
https://github.com/GoogleChromeLabs/browser-fs-access/blob/b3be8c6bd93d7be3388aed3ca77da258283494ba/src/legacy/file-save.mjs#L38
There's no
reject
variable in scope here. It looks like this bug has existed since the code was introduced:https://github.com/GoogleChromeLabs/browser-fs-access/commit/8ed86033972796fc1531de91d13b86bc632571ea#diff-0a1bdb6eb4ee8c5dd800ceba3ea9beda6618f51150ef0152f552454c6cbfd92bR31