Convert markdown string to plain text using marked.

marked_convert_to_plain.ts

import { marked, Marked } from "marked";

function listFactory(token, parser) {
  let depth = 0; // Track nesting level
  const { items, ordered, start } = token; // Destructure the token object

  const indent = " ".repeat(depth * 2);
  const startNum = ordered ? start || 1 : null;

  const listItems = items.map((item, index) => {
    const prefix = ordered ? `${startNum + index}. ` : "- ";
    depth++; // Increase depth for nested content
    const itemContent = parser.parse(item.tokens); // Parse nested tokens
    depth--; // Decrease depth after parsing

    // Split content into lines and filter out empty ones
    const contentLines = itemContent
      .split("\n")
      .filter((line) => line.trim() !== "");
    if (contentLines.length === 0) return indent + prefix;

    // Prepend prefix to the first line
    contentLines[0] = indent + prefix + contentLines[0];
    // Indent subsequent lines for nested content
    for (let i = 1; i < contentLines.length; i++) {
      contentLines[i] = " ".repeat((depth + 1) * 2) + contentLines[i];
    }
    return contentLines.join("\n");
  });

  // Join items with newlines and add an extra newline
  return listItems.join("\n") + "\n";
}

function removeNewlinesOutsidePre(htmlString: string) {
  const parts = htmlString.split(/(<pre(?:\s[^>]*)?>[\s\S]*?<\/pre>)/g);
  const processedParts = parts.map((part) => {
    if (!part.match(/^<pre(?:\s[^>]*)?>/)) {
      return part.replace(/\n/g, "");
    }
    return part;
  });
  return processedParts.join("");
}

function handleMarkdown(markdownText?: string, asPlain?: boolean): string {
  if (!markdownText) return markdownText;

  const baseRenderer = {
    paragraph({ tokens }) {
      const text = this.parser.parseInline(tokens);
      return text + "\n";
    },
  };

  const plainRenderer = {
    heading(token) {
      return token.raw;
    },
    hr(token) {
      return token.raw + '\n';
    },
    list(token) {
      return listFactory(token, this.parser);
    },
    link(token) {
      return token.href;
    },
    image(token) {
      return `[image: ${token.href}]`;
    },
    strong(token) {
      const text = this.parser.parseInline(token.tokens);
      return text;
    },
    em(token) {
      const text = this.parser.parseInline(token.tokens);
      return text;
    },
    table(token) {
      return token.raw
    },
    code(token) {
      // console.log(`text: `, token.text);
      return token.raw + '\n';
    },
    codespan(token) {
      return token.raw;
    },
    blockquote(token) {
      return token.raw + '\n';
    }
  };

  const advancedRenderer = {
    heading({ tokens }) {
      const text = this.parser.parseInline(tokens);
      return `<b>${text}</b><br>`;
    },
    list(token) {
      return listFactory(token, this.parser);
    },
  };

  const renderer = asPlain
    ? {
        ...baseRenderer,
        ...plainRenderer,
      }
    : {
        ...baseRenderer,
        ...advancedRenderer,
      };

  const renderedText = new Marked().use({ renderer }).parse(markdownText);
  return asPlain ? renderedText : removeNewlinesOutsidePre(renderedText);
}

const text = `
## thu nghiem

Xem the nao \`code\` xem the nao.

> blockquote nay ne
> hang thu hai

---

1. Item _thu nghiem_ \`code\` 1
2. Item 2
- List item 1
  - Nested item 1.1
  - Nestted item 1.2
- List item 2 https://abc.com

- [] asdasdas

Link: [text link](http://xxx.com)

\`\`\`
\\begin{align}
  x + y = 1
\\end{align}  
\`\`\`

![asdasd](https://xxx.com/ui.jpg)

| Name  | Age | City       |
|-------|-----|------------|
| Alice | 25  | New York   |
| Bob   | 30  | London     |
| Carol | 28  | Paris      |
`;

// handleMarkdown(text, true)
handleMarkdown(text, false);
console.log("\n 😄😄😄😄😄 \n");
handleMarkdown(text, true);