bookmark-parser

Zero dependency bookmark parser exports into a shared bookmark node array.

Install

pnpm add @amoklab/bookmark-parser

API

The package exports:

• htmlBookmarkParser(text, options?)
• jsonBookmarkParser(text, options?)
• flattenBookmark(nodes, options?)
• BookmarkEntry
• BookmarkFolder
• BookmarkNode
• ParserOptions

Data model

All parsers return the same node shape:

• BookmarkFolder for folders
• BookmarkEntry for links
• BookmarkNode for either one

Folders keep their nested children. Both folders and bookmarks can keep a prevNode chain for tree navigation. The parsers return only the export's bookmark children, not a synthetic root node.

jsonBookmarkParser

Parses the Firefox JSON export format.

import { jsonBookmarkParser } from '@amoklab/bookmark-parser';

const bookmarks = jsonBookmarkParser(firefoxJsonText);
const flattened = jsonBookmarkParser(firefoxJsonText, {
  flatten: true,
});

The parser expects a Firefox bookmarks JSON export and returns the top-level bookmark nodes.

htmlBookmarkParser

Parses Netscape bookmark HTML exports produced by both Chrome and Firefox.

import { htmlBookmarkParser } from '@amoklab/bookmark-parser';

const bookmarks = htmlBookmarkParser(bookmarksHtmlText);
const flattened = htmlBookmarkParser(bookmarksHtmlText, {
  flatten: true,
});

Options

ParserOptions is accepted by the parsers:

export interface ParserOptions {
  flatten?: boolean;
  setPrevNode?: (node: BookmarkNode) => any;
}

• flatten: when true, parsers return a flattened bookmark array
• setPrevNode: optional hook that transforms the prevNode object used during parsing flattening

flattenBookmark(nodes, options?) accepts the setPrevNode hook only and still returns a flat preorder list.