Default encoding can't be set to 'auto-detect'

[willamowius]

The 'Default file encoding' in the preferences doesn't have 'auto-detect' as a choice.

[tinymins]

+1 I have lua file in GBK encoding and py file in UTF-8 encoding in the same project. Every time I switch files, I need to change the encoding by ctrl+shift+u and select Auto Detect. Why not auto run encoding detect when open file?

[ghost]

detectEncoding such as the one in encoding-selector atom/text-buffer#86

[ujay68]

When editing well-formed XML files, a tool should definitely detect and also enforce the encoding. Not doing that brings a very high risk of producing a non-wellformed XML file in an edit. If you regard an XML editor like Atom as an XML processor, it even violates the XML specification. See also atom/atom#10052 (closed, sadly). IMO, if Atom forces the user to manually select the encoding of every XML file you edit (and every time you open it?), that makes Atom close to unusable for XML editing in many project settings.

[natinusala]

+1, this is very annoying and breaks all special characters when I save without paying attention

[mconrado]

I try to encourage all my workmates to use Atom but problems with encoding avoid them. Vide too: #8

[tinymins]

atom/atom#12127 #34

[bohning]

My vote for this issue as well, in my eyes it is essential and I have to use a different editor until this is fixed. Manually selecting the non-default enconding is tiresome. I literally have destroyed dozens, if not hundreds of files due to this.

It would be really great to see this feature soon.

[grivasgervilla]

Add vote for this feature. When I open a iso-8859-1 Atom reads it like an UTF-8 file, but if a I choose auto-detect encoding then it reads it find. So if we can set auto-detect as default encoding can be more confortable.

[winstliu]

I am working on this.

[pmgouveia]

any news or commit on this? any workaround with a plugin? I tried https://atom.io/packages/default-encoding but I can't find how to create the config file

[winstliu]

The PRs are open (atom/atom#13943, atom/node-pathwatcher#120, atom/text-buffer#220, #42), but haven't been reviewed yet. Most of the Atom team is very busy working on performance improvements or other initiatives right now though, so I'm content to wait for those improvements to ship first before getting my PRs reviewed.

[tinymins]

So, anywhere to download and install available dev builds? I want to have a try.

[lipeng1667]

This package saved me. https://atom.io/packages/auto-encoding

[burtonsys]

You're more patient that I was, irudoy. I also switched to VS Code. The encoding-detection / file-trashing problem was the fatal flaw, for me.

VS Code is not perfect. The 32-bit version crashed about once a day, which was disconcerting, though it usually recovered nicely, and I rarely lost more than a few characters of typing. But it is getting better. I installed the 64-bit version about 2.5 weeks ago, and it hasn't crashed since then.

[grodrigo]

wow, this is really annoying, this problem exists since 2014! and nothing changed.
SOLUTION:
apm install auto-encoding

I have a legacy system, with windows1252 coding, I plan to migrate it to utf-8, but if the editor changes the coding it would be good to let me know ... especially if it is not going to change correctly.

Unfortunately, some users have switched to another editor, why is not autodetect default behavior? It does not make sense and it made me break files and waste time until I realized that it was a my editor.

The autocoding package detects the charset correctly when the file is opened, really thank you very much!!!! but many users will continue to have problems, please make the default behavior.

[BitGrub]

Incredible that Atom is still lacking this basic functionality after so many years