Support setting encoding per file

[awvalenti]

files.encoding allows setting encoding user-wide or workspace-wide. Sometimes we need to set different encodings for specific files. I'd like something like:

{
  "files.encoding": "iso88591",
  "file-specific": {
    "/src/my-file.md": {
      "files.encoding": "utf8"
    }
  }
}

(EDITED: adding the missing second files.encoding part)

Actual situation: my project uses iso88591. I'm using marp and need to save only that markdown file in utf8.

[bpasero]

I think there should rather be general support to scope a setting to a resource. Related #35350

[vs-code-engineering]

This feature request is now a candidate for our backlog. The community has 60 days to upvote the issue. If it receives 20 upvotes we will move it to our backlog. If not, we will close it. To learn more about how we handle feature requests, please see our documentation.

Happy Coding!

[ped7g]

OP: I assume you are aware you can reopen the file with different encoding in the VSC and then it somehow does remember the specific encoding for that particular file as far as I can tell. But I'm unable to tell where that setting is stored, doesn't modify any of the settings.json files, so it must be elsewhere? So your issue is asking to move this into settings? (asking just to clarify it to myself - confused user of VSC, nothing more significant, sry :) )

I do see some value in having this pre-set in git repo file, so any newly cloned repo will behave correctly without manually reopening the file.

[awvalenti]

It remembers that choice only while the editor is open. If you close it and reopen, it resorts again to workspace setting. It's problably on some very temporary config file, i.e., too fragile.

[yhatt]

files.encoding is just setting for the default character set. An actually selected encoding by UI may become different from that setting. It has stored to the memory of each text editing area, and currently it seems to be able to access that only by VS Code internally.

vscode/src/vs/workbench/browser/parts/editor/editorStatus.ts

Line 82 in 11b2f94

function toEditorWithEncodingSupport(input: EditorInput): IEncodingSupport | null {

Actual situation: my project uses iso88591. I'm using marp and need to save only that markdown file in utf8.

I'm the author of Marp it mentioned at here. The original issue was recieved at marp-team/marp#558, as an possibility issue report for the Marp for VS Code extension.

If VS Code extensions can access to the current encoding for each text editor, the original issue by the reporter would be solved because the extension can do proper conditional branching by the selected character set. So if provided a solution against the long-standing issue #824, this proposal might not be needed from the standpoint of reporter. (e.g. PR #177434)

Our extension might implement heuristic based on the current files.encoding setting and might avoid the problem due to uncommon encoding, but it is not accurate because the actually preferred character set in the document may not same as files.encoding.

Similar?: #209503, microsoft/vscode-cpptools#13047

[vs-code-engineering]

This feature request has not yet received the 20 community upvotes it takes to make to our backlog. 10 days to go. To learn more about how we handle feature requests, please see our documentation.

Happy Coding!

[ped7g]

I think there should rather be general support to scope a setting to a resource. Related #35350

That would be nice, but doesn't sound like it would allow to change encoding for specific files. (you can have project where for example most of the .cpp files are utf-8, but for various technical reasons you have couple of .cpp files with cp1252 encoding (for example with German comments) and keeping them in original form is important enough to prevent just converting them to utf8.
(yes, lot of those "technical reasons" are more like legacy/organisation rules issues than truly technical, but the end result is same, you are stuck with file having different encoding in your project repo).

[vs-code-engineering]

🙁 In the last 60 days, this feature request has received less than 20 community upvotes and we closed it. Still a big Thank You to you for taking the time to create this issue! To learn more about how we handle feature requests, please see our documentation.

Happy Coding!

[artemgur]

I also need to set separate encoding for some files. Because on Windows bat and ps1 scripts with Cyrillic symbols need very specific encodings to work properly (OEM866 and UTF-8 with BOM correspondingly), otherwise Cyrillic symbols become broken and scripts don't work