feat: false positives and negatives for consecutive backslashes

[lumirlumir]

Prerequisites checklist

• I have read the contributing guidelines.

What is the purpose of this pull request?

Which language are you using?

CommonMark and GFM.

What did you do?

I've tried using consecutive backslashes in the no-missing-atx-heading-space, no-missing-label-refs, and no-reversed-media-syntax rules, but they reported false positives and false negatives as shown below:

What did you expect to happen?

no-missing-atx-heading-space

• Only an even number of backslashes should be reported, as \\\\ is not interpreted as an escape backslash character in Markdown. However, it also reports odd numbers of backslashes greater than three.

  

no-missing-label-refs

• When consecutive backslashes appear, it fails to report them, resulting in false negatives:

  

  According to the AST, the foo\\ is a valid label reference.

  

no-reversed-media-syntax

• When consecutive backslashes appear, it fails to report them, resulting in false negatives:

  

What actually happened?

It's also mentioned in the "What did you expect to happen?" section.

Link to minimal reproducible Example

<!-- eslint markdown/no-missing-atx-heading-space: ["error", { checkClosedHeadings: true }] -->

<!-- OK -->
# Heading 1\# 
# Heading 1\\#
# Heading 1\\\\#
# Heading 1\\\\\\#

<!-- NOT OK -->
# Heading 1\\\#
# Heading 1\\\\\#

<!-- eslint markdown/no-missing-label-refs: "error" -->

<!-- OK -->
[foo]
\[foo\]
\\\[foo\\\]

<!-- NOT OK -->
\\[foo\\]
\\\\[foo\\\\]

<!-- eslint markdown/no-reversed-media-syntax: "error" -->

<!-- OK -->
(hi)[hi]
\(hi)[hi]
\\\(hi)[hi]

<!-- NOT OK -->
\\(hi)[hi]
\\\\(hi)[hi]

What changes did you make? (Give an overview)

In this PR, I've addressed false positives and negatives in no-missing-atx-heading-space, no-missing-label-refs, and no-reversed-media-syntax rules that were caused by consecutive backslash handling.

Previously, we used the following regex patterns to avoid matching backslash escapes:

• (?<!\\) and (?<!(?<!\\)\\)

However, these patterns are incorrect because they don't properly handle consecutive backslash escapes metioned above.

So, I've updated them to use the following regex correctly:

• (?<=(?<!\\)(?:\\{2})*)

(?<=(?<!\\)(?:\\{2})*)a

a           // even
\a          // odd
\\a         // even
\\\a        // odd
\\\\a       // even
\\\\\a      // odd
\\\\\\a     // even

Related Issues

Ref: #452 (comment)

Is there anything you'd like reviewers to focus on?

In JavaScript, \\\\\\\\ (8 backslashes) is interpreted as \\\\ (4 backslashes) in Markdown, and is ultimately recognized as \\ (2 backslashes). In this case, an even number of backslashes is not treated as an escape character and is recognized as a normal backslash.

[github-actions]

Hi everyone, it looks like we lost track of this pull request. Please review and see what the next steps are. This pull request will auto-close in 7 days without an update.