[GH-ISSUE #968] MD048: Make it automatically fixable #575

New issue

Open

opened 2026-03-03 01:28:08 +03:00 by kerem · 3 comments

kerem commented

2026-03-03 01:28:08 +03:00

Owner

Originally created by @skwde on GitHub (Sep 13, 2023).
Original GitHub issue: https://github.com/DavidAnson/markdownlint/issues/968

The description of rules at https://github.com/DavidAnson/vscode-markdownlint#rules is followed by the list of fixable rules.

MD048 seems fairly easy to fix (I also haven't found info on why it is not fixable).

So can it be done or is there a good reason why it is not?

Originally created by @skwde on GitHub (Sep 13, 2023). Original GitHub issue: https://github.com/DavidAnson/markdownlint/issues/968 The description of rules at <https://github.com/DavidAnson/vscode-markdownlint#rules> is followed by the list of fixable rules. MD048 seems fairly easy to fix (I also haven't found info on why it is not fixable). So can it be done or is there a good reason why it is not?

kerem added the

enhancement

label

2026-03-03 01:28:08 +03:00

kerem commented

2026-03-03 01:28:10 +03:00

Author

Owner

@DavidAnson commented on GitHub (Sep 13, 2023):

This looks like an oversight. I'll implement this soon.

@DavidAnson commented on GitHub (Sep 13, 2023): This looks like an oversight. I'll implement this soon.

kerem commented

2026-03-03 01:28:10 +03:00

Author

Owner

@DavidAnson commented on GitHub (Sep 16, 2023):

This isn't implemented because it requires a single fix that touches both the open and close of the fence (different lines) and that is not easily possible today. It will be possible someday and this behavior should be straightforward to implement then.

@DavidAnson commented on GitHub (Sep 16, 2023): This isn't implemented because it requires a single fix that touches both the open and close of the fence (different lines) and that is not easily possible today. It will be possible someday and this behavior should be straightforward to implement then.

kerem commented

2026-03-03 01:28:10 +03:00

Author

Owner

@wisnij commented on GitHub (Feb 21, 2026):

I was taking a stab at this one recently and only belatedly found this issue. The roadblock I ran into is that this rule has an interaction with MD031/blanks-around-fences – if both rules are enabled and the opening fence is not preceded by a blank line, they both need to emit a fix at the same location, but only one of them can actually be applied (if I understand the logic in applyFixes properly, anyway).

It looks like the options here are:

Make it possible to apply more than one fix at the same location
Allow one rule to detect whether another is enabled, to modify its own behavior accordingly
Just leave this one unfixable

Option 1 would obviously be ideal, but it might not be possible in the general case if there are genuine conflicts. As for option 2, I certainly understand the desire to keep rules independent, but FWIW I would submit that these two rules are already coupled due to triggering on the same lines. Introducing a direct reference between them would not be adding new coupling but converting the existing implicit coupling into an explicit one, which seems less bad, at least.

Here's as far as I've gotten so far, in the hope that it helps:

https://github.com/DavidAnson/markdownlint/compare/main...wisnij:markdownlint:wisnij/20260220-fix-md048

@wisnij commented on GitHub (Feb 21, 2026): I was taking a stab at this one recently and only belatedly found this issue. The roadblock I ran into is that this rule has an interaction with `MD031/blanks-around-fences` – if both rules are enabled and the opening fence is not preceded by a blank line, they both need to emit a fix at the same location, but only one of them can actually be applied (if I understand the logic in `applyFixes` properly, anyway). It looks like the options here are: 1. Make it possible to apply more than one fix at the same location 2. Allow one rule to detect whether another is enabled, to modify its own behavior accordingly 3. Just leave this one unfixable Option 1 would obviously be ideal, but it might not be possible in the general case if there are genuine conflicts. As for option 2, I certainly understand the desire to keep rules independent, but FWIW I would submit that these two rules are _already_ coupled due to triggering on the same lines. Introducing a direct reference between them would not be adding new coupling but converting the existing implicit coupling into an explicit one, which seems less bad, at least. Here's as far as I've gotten so far, in the hope that it helps: https://github.com/DavidAnson/markdownlint/compare/main...wisnij:markdownlint:wisnij/20260220-fix-md048

kerem referenced this issue

2026-03-03 01:30:42 +03:00

[PR #575] [MERGED] Bump JustinBeckwith/linkinator-action from 1.10.2 to 1.10.3 #902

kerem referenced this issue

2026-03-07 20:10:23 +03:00

[PR #575] [MERGED] Bump JustinBeckwith/linkinator-action from 1.10.2 to 1.10.3 #2748