In the case of HTML Paragraph, erroneous detection occurs.

[kemsakurai]

I applied the no-start-duplicated-conjunction rule to the HTML file.
The following Lint error occurred, but I think that this error is a false detection.

  916:1011  error    Don't repeat "</p>" in 2 phrases    no-start-duplicated-conjunction

In the case of HTML, the data of the node of Paragraph is as follows.

{ type: 'Paragraph',
  tagName: 'p',
  properties: {},
  children: 
   [ { type: 'Strong',
       tagName: 'strong',
       properties: {},
       children: [Array],
       loc: [Object],
       range: [Array],
       raw: '<strong>HTML と、Page クラスに Footer 部とする Container を追加する</strong>' },
     type: 'UNKNOWN' ],
  loc: 
   { start: { line: 1144, column: 36 },
     end: { line: 1144, column: 104 } },
  range: [ 46454, 46522 ],
  raw: '<p><strong>HTML と、Page クラスに Footer 部とする Container を追加する</strong></p>' }

In the case of Markdown, the data of the node is as follows.

{ type: 'Paragraph',
  children: 
   [ { type: 'Code',
       value: 'o',
       loc: [Object],
       range: [Array],
       raw: '`o`' },
     { type: 'Str',
       value: ' をつけたルールを一律 グローバルインストールします。    ',
       loc: [Object],
       range: [Array],
       raw: ' をつけたルールを一律 グローバルインストールします。    ' } ],
  loc: 
   { start: { line: 117, column: 0 },
     end: { line: 117, column: 34 } },
  range: [ 5761, 5795 ],
  raw: '`o` をつけたルールを一律 グローバルインストールします。    ' }

I thought false detection occurred because HTML tags included in the raw property of the Paragraph node generated from HTML.
I attempted to fix it. but I could not decide whether to remove HTML tags on the rule side or whether HTML parser should remove the HTML tag of raw property.

[azu]

HTML tags included in the raw property of the Paragraph node generated from HTML.

Paragraph node's raw includes raw HTML String is valid.
String node's raw includes <strong> HML string is valid.

For example, **text** in markdown, the node's raw include ** string.

https://textlint.github.io/astexplorer/

Is HTML text that is tested <p><strong>HTML と、Page クラスに Footer 部とする Container を追加する</strong></p> ?

---

Notes: maybe, this rule can use textlint-util-to-string.

const StringSource = require("textlint-util-to-string");
const source = new StringSource(node);
const text = source.toString();
// This `text` should be plain text.

insteadof

const text = getSource(node);

Or this rule collect Str nodes and concat it.

[kemsakurai]

Thank you for answering.
I understood the possible value of raw.

Is HTML text that is tested

Add HTML and Container to be the Footer part to the Page class </ strong> </ p>?

This value was as an example.
It is not a Node in which the problem actually occurred.
sorry.

I will try using textlint-util-to-string to see if it works well.