Auto-detect encoding on an already open file does not properly detect in all cases

[alexandernst]

Maybe related to atom/atom#4160 , the "auto-detect encoding" in this package is completely broken.

Simplest test: create a new UTF16 encoded file and write a few "weird" characters (蔡葕ቅ for example) and save the file. Now open it with Atom and try to auto-detect the encoding.

[lee-dohm]

Per @kevinsawicki here, could you open this on Atom Core?

If auto-detect on open is added, it will have to be done in core since that is where the buffer's are built.

https://github.com/atom/atom/issues/new

[lee-dohm]

Ah sorry, I misinterpreted what you were trying to say. Rewording the title to make it more clear.

[renkun-ken]

It's frustrating to re-select Auto Detect for each non-utf-8 encoded file every time I open them especially when I need to view quite a number of files located in a folder. Really appreciate it if someone could fix this.

[jamadam]

Wired behavior reproduces even without reopening the file.

Simplest test: create a new UTF16 encoded file and write a few "weird" characters (蔡葕ቅ for example) and save the file. Click the project pane and then click the file tab again.

atom: 1.32.2
encoding-selector: 0.23.9

[aledmb]

here's a text file created with notepad on windows 10.

all editors open it using the correct encoding, except atom.

atom could do the same thing phpstorm does, which is tell the user what's happening.

test.txt