Padding matters

[ccoVeille]

Describe the bug

Padding can be used in markdown files.

Config and version

Any

To Reproduce

- one item

    Some note about this item. Here is an example

   ```go
   var expired bool
   ```

- a second one

   Some comment

It gets rendered like that

• one item

  Some note about this item. Here is an example

  var expired bool

• a second one

  Some comment

Expected behavior

The tool reports an issue with the formatting, while the code is fine

It wants this

- one item

    Some note about this item. Here is an example

```go
var expired bool
```

- a second one

   Some comment

But then, it's not rendered the same way

• one item

  Some note about this item. Here is an example

var expired bool

• a second one

  Some comment

Screenshots
If applicable, add screenshots to help explain your problem.

Additional context
Add any other context about the problem here.

[ccoVeille]

This can break formatting with code embedded within <details/> node that are padded.

I can provide an example if you are interested.

[hougesen]

If I understand correctly this is a bug in #616 / #628?

I am unable to reproduce, could you share the expected input/output as a raw text file?

This is formatted

- one item

    Some note about this item. Here is an example

   ```go
   var        expired     bool
   ```

- a second one

   Some comment

into this :

- one item

    Some note about this item. Here is an example

   ```go
   var expired bool
   ```

- a second one

   Some comment

using the default config @ v0.9.5

{
  "$schema": "https://raw.githubusercontent.com/hougesen/mdsf/main/schemas/v0.9.5/mdsf.schema.json",
  "languages": {
    "c": "clang-format",
    "cpp": "clang-format",
    "csharp": "csharpier",
    "go": [
      ["gci", "goimports-reviser", "goimports"],
      ["gofumpt", "gofmt"]
    ],
    "java": "google-java-format",
    "javascript": "prettier",
    "json": "prettier",
    "python": [
      ["usort", "isort"],
      ["ruff:format", "blue", "black", "yapf", "autopep8", "pyink"]
    ],
    "ruby": [["rubocop", "rufo", "rubyfmt", "standardrb"]],
    "rust": "rustfmt",
    "shell": [["shfmt", "beautysh"]],
    "sql": [["sql-formatter", "sqlfluff:format", "sqlfmt"]],
    "toml": "taplo",
    "typescript": "prettier",
    "yaml": "prettier"
  }
}

[ccoVeille]

Here it is

https://github.com/mgechev/revive/actions/runs/15257512277/job/42908541751#step:10:1

[ccoVeille]

I found it back

https://github.com/ccoveille-forks/mgechev-revive/blob/fffc1ad831dac9e1b5c8352c55e60d6e8adb457e/RULES_DESCRIPTIONS.md?plain=1#L1085

Maybe I'm totally wrong, that's also possible 😅

[hougesen]

I am still a bit confused 😅

This is the diff I get when I format using this config https://github.com/mgechev/revive/blob/6becd540e4f864330381c0f2cd0cf05089aa8aa3/mdsf.json

Could your issue be caused by the file missing an empty line at the end? IIRC mdsf inserts one at the end.

Also shouldn't this be a </details>?

https://github.com/ccoveille-forks/mgechev-revive/blob/fffc1ad831dac9e1b5c8352c55e60d6e8adb457e/RULES_DESCRIPTIONS.md?plain=1#L1133

[hougesen]

The first 2 chars of each function argument line is a space, followed by a tab ( \t). Could that be related?

[ccoVeille]

I am still a bit confused 😅

Me too

This is the diff I get when I format using this config https://github.com/mgechev/revive/blob/6becd540e4f864330381c0f2cd0cf05089aa8aa3/mdsf.json

When I read it here, I feel like the mdsf formatting might be good.

So maybe the issue I had, was with the way the mdsf action displayed them? Just a idea. Like a tab being rendered as a single space? Making me think mdsf wanted me to remove some padding spaces in my block.

Could your issue be caused by the file missing an empty line at the end? IIRC mdsf inserts one at the end.

Dunno

Also shouldn't this be a </details>?

Yes, I fixed it in a later commit.

[hougesen]

I think that is caused by https://github.com/ccoveille-forks/mgechev-revive/blob/fffc1ad831dac9e1b5c8352c55e60d6e8adb457e/.github/workflows/lint.yaml#L94-L98

But I do think that the actual logs made by mdsf could be simplified.

When I look at 100 lines of uncolored text like this I get a bit overwhelmed

./RULES_DESCRIPTIONS.md:1202 to :1211 toml block using taplo
INFO taplo:format_files:collect_files: found files total=1 excluded=0 files=["/home/runner/.cache/mdsf/mdsfFpOsXv98wN04.toml"] cwd="/home/runner/work/revive/revive"
./RULES_DESCRIPTIONS.md:1243 to :1245 go block using goimports
./RULES_DESCRIPTIONS.md:1247 to :1250 toml block using taplo
INFO taplo:format_files:collect_files: found files total=1 excluded=0 files=["/home/runner/.cache/mdsf/mdsfOsOEmfUaNoLQ.toml"] cwd="/home/runner/work/revive/revive"
./RULES_DESCRIPTIONS.md:1252 to :1255 toml block using taplo
INFO taplo:format_files:collect_files: found files total=1 excluded=0 files=["/home/runner/.cache/mdsf/mdsfP4kt8bwrr7f0.toml"] cwd="/home/runner/work/revive/revive"
./RULES_DESCRIPTIONS.md:1268 to :1270 go block using goimports

I am not sure how it should look though.

[ccoVeille]

I do think that the actual logs made by mdsf could be simplified.

They do.

I was also surprised how verbose was the mdsf-action