Conflicting instructions in sections for a given file

[mipo256]

Consider that we have file, located in /some/dir/MyFile.java

And we have the following .editorconfig file:

root = true

[/some/dir/*.java]
end_of_line = cr

[/**/MyFile.{java,js}]
end_of_line = lf

This is a very simple example to demonstrate the problem. The problem essentially is that any file X, such as /some/dir/MyFile.java in our example, could be potentially matched by multiple sections. And those sections can (the spec correctly does not forbid it) have conflicting instructions, such as an example above.

What we need to do:

The spec must clarify what we require the implementation to do in such case.

[mipo256]

CC: @xuhdev @cxw42

What are your thoughts on this?

[SunsetTechuila]

I wasn't asked, but I think this doesn't have to be a special case, it's enough to specify the order in which the sections are combined, as they can already be, and to state that a that a section can override options from the previous one.

[mipo256]

@SunsetTechuila Yeah, this is the very first thought that comes to mind, but I doubt it is extensible. We could've had 2 files, and both of them match to both sections. The problem is that we might want for one file to take settings from section A, and for the second file take settings from section B.

So although this solution is very simple, I do not think it is ideal.

[xuhdev]

I wasn't asked, but I think this doesn't have to be a special case, it's enough to specify the order in which the sections are combined, as they can already be, and to state that a that a section can override options from the previous one.

I agree this is the intended behavior. The spec says:

Files are read top to bottom and the most recent rules found take precedence. If multiple EditorConfig files have matching sections, the rules from the closer EditorConfig file are read last, so pairs in closer files take precedence.

I think we just need to replace the word "rule" with "pair", which is defined in the spec.

[mipo256]

Okay, if we have stated this already in the doc, then, I guess, the solution proposed by @SunsetTechuila comes naturally

[SunsetTechuila]

It's not a solution proposed by me, but by the spec) I read it once more after seeing your issue, but somehow missed that exact paragraph.

[mipo256]

@SunsetTechuila I agree, that is just a question of interpretation. In any case, thank you for pointing this out, I appreciate that 🤝