fix: false negative on shortcut type ref in no-missing-label-refs

[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 used *foo*, **foo**, ~~foo~~, ~~**foo**~~, ~~***foo***~~, and similar formats for the definition. Based on this, I expected patterns like the ones below to be reported as missing label references, but some were not detected.

What did you expect to happen?

According to the AST, patterns like *foo*, **foo**, ~~foo~~, ~~**foo**~~, and ~~***foo***~~ can be recognized as valid label references:

So, I expected the following patterns to be reported by this rule:

<!-- eslint markdown/no-missing-label-refs: "error" -->
- [*foo*]

- [**foo**]

- [~~foo~~]

- [~~**foo**~~]

- [~~***foo***~~]

Link to minimal reproducible Example

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

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

<!-- Not OK -->
- [*foo*]

- [**foo**]

- [~~foo~~]

- [~~**foo**~~]

- [~~***foo***~~]

What changes did you make? (Give an overview)

The current implementation only checks the Text node when detecting missing label reference syntax.

However, this causes an issue when Strong, Emphasis, or Delete markers (represented by *, _, or ~) are used in the syntax.

With these markers, a label like [**foo**] will have its Text node split into [, foo, and ] as shown in the AST below, making it impossible to detect the missing label reference pattern.

---

To resolve this issue, I examined the Paragraph, Heading, and TableCell nodes instead of simply examining the Text node.

This is a common pattern in Markdown rules when examining error-prone syntax:

markdown/src/rules/no-bare-urls.js

Lines 143 to 156 in 3067607

	"heading:exit"() {
	linkNodes.push(...tempLinkNodes);
	reset();
	},
	"paragraph:exit"() {
	linkNodes.push(...tempLinkNodes);
	reset();
	},
	"tableCell:exit"() {
	linkNodes.push(...tempLinkNodes);
	reset();
	},

markdown/src/rules/no-reversed-media-syntax.js

Lines 146 to 171 in 3067607

	"heading :matches(html, inlineCode)"(node) {
	skipRanges.push(sourceCode.getRange(node));
	},
	"heading:exit"(node) {
	findReversedMediaSyntax(node);
	skipRanges = [];
	},
	"paragraph :matches(html, inlineCode)"(node) {
	skipRanges.push(sourceCode.getRange(node));
	},
	"paragraph:exit"(node) {
	findReversedMediaSyntax(node);
	skipRanges = [];
	},
	"tableCell :matches(html, inlineCode)"(node) {
	skipRanges.push(sourceCode.getRange(node));
	},
	"tableCell:exit"(node) {
	findReversedMediaSyntax(node);
	skipRanges = [];
	},

markdown/src/rules/no-space-in-emphasis.js

Lines 258 to 286 in 3067607

	heading(node) {
	bufferManager.enter(node);
	},
	"heading > text"(node) {
	bufferManager.addText(node);
	},
	"heading:exit"(node) {
	bufferManager.exit(node);
	},
	paragraph(node) {
	bufferManager.enter(node);
	},
	"paragraph > text"(node) {
	bufferManager.addText(node);
	},
	"paragraph:exit"(node) {
	bufferManager.exit(node);
	},
	tableCell(node) {
	bufferManager.enter(node);
	},
	"tableCell > text"(node) {
	bufferManager.addText(node);
	},
	"tableCell:exit"(node) {
	bufferManager.exit(node);
	},

Related Issues

N/A

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

• Prerequisite: feat: fix false positives and negatives for consecutive backslashes #490

[lumirlumir]

I’ll take a look at this PR within the next two weeks, as I currently have some higher-priority PRs in progress.

[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.

[mdjermanovic]

Prerequisite: feat: fix false positives and negatives for consecutive backslashes #490

This has been merged now.