Common EPUB2 data parser for Ridibooks services written in ES6

• Detailed parsing for EPUB2
• Supports package validation, decompression and style extraction with various parsing options
• Extract files within EPUB with various reading options

• Add encryption and decryption function
• Add readOptions.spine.truncate and readOption.spine.truncateMaxLength options
• Add readOptions.spine.minify and readOptions.css.minify options
• Support for EPUB3
• Support for CLI
• Support for other OCF spec (manifest.xml, metadata.xml, signatures.xml, encryption.xml, etc)

npm install @ridi/epub-parser

Basic:

import EpubParser from '@ridi/epub-parser';

const parser = new EpubParser('./foo/bar.epub');
parser.parse().then((book) => {
  const results = parser.read(book.spines);
  ...
});

Various inputs:

import fs from 'fs';
import EpubParser from '@ridi/epub-parser';

// Unzipped path of EPUB file.
new EpubParser('./foo/bar');

// EPUB file buffer.
const buffer = fs.readFileSync('./foo/bar.epub');
new EpubParser(buffer);

Book to Object, Object to Book:

import EpubParser from '@ridi/epub-parser';

const parser = new EpubParser('./foo/bar.epub');
parser.parse().then((book) => {
  const rawBook = book.toRaw();
  const newBook = new Book(rawBook);
  ...
});

Returns Promise<Book> with:

• Book: Instance with metadata, spine list, table of contents, etc.

Or throw exception.

Returns string or Object or string[] or Object[] with:

• string (readOptions.spine.extractBody is false)

• Object (readOptions.spine.extractAdapter is undefined):

  • body: Same reuslt as document.body.innerHTML
  • attrs: Attributes in body tag.

• Object (readOptions.spine.extractAdapter is defaultExtractAdapter):

  • content: extractBody output transformed by adapter.

Or throw exception.

• titles: string[]
• creators: Author[]
• subjects: string[]
• description: string?
• publisher: string?
• contributors: Author[]
• dates: DateTime[]
• type: string?
• format: string?
• identifiers: Identifier[]
• source: string?
• language: string?
• relation: string?
• rights: string?
• epubVersion: number?
• metas: Meta[]
• items: Item[]
• ncx: NcxItem?
• spines: SpintItem[]
• fonts: FontItem[]
• cover: ImageItem?
• images: ImageItem[]
• styles: CssItem[]
• guide: Guide[]
• deadItems: DeadItem[]

• name: string?
• role: string (Default: Author.Roles.UNDEFINED)

• value: strung?
• event: string (Default: DateTime.Events.UNDEFINED)

• value: string?
• scheme: string? (Default: Identifier.Schemes.UNDEFINED)

• name: string?
• content: string?

• title: string?
• type: string (Default: Guide.Types.UNDEFINED)
• href: string?
• item: Item?

• id: id?
• href: string?
• mediaType: string?
• size: number?
• isFileExists: boolean (size !== undefined)
• defaultEncoding: string?

• navPoints: NavPoint[]

• spineIndex: number (Default: -1)
• isLinear: boolean (Default: true)
• styles: CssItem[]?

• namespace: string?

• text: string?

• isCover: boolean (Default: false)

• raw: Object

• id: string?
• label: string?
• src: string?
• anchor: string?
• depth: number (Default: 0)
• children: NavPoint[]
• spine: SpineItem?

• validatePackage
• validateXml
• allowNcxFileMissing
• unzipPath
• createIntermediateDirectories
• removePreviousFile
• ignoreLinear
• useStyleNamespace
• styleNamespacePrefix

If true, validation package specifications in IDPF listed below.

• Zip header should not corrupt.
• mimetype file must be first file in archive.
• mimetype file should not compressed.
• mimetype file should only contain string application/epub+zip.
• Should not use extra field feature of ZIP format for mimetype file.

Default: false

If true, stop parsing when XML parsing errors occur.

Default: false

If false, stop parsing when NCX file not exists.

Default: true

If specified, uncompress to that path.

Only if input is buffer or file path of EPUB file.

Default: undefined

If true, creates intermediate directories for unzipPath.

Default: true

If true, removes a previous file from unzipPath.

Default: true

If true, ignore spineIndex difference caused by isLinear property of SpineItem.

// e.g. If left is false, right is true.
[{ spineIndex: 0, isLinear: true, ... },       [{ spineIndex: 0, isLinear: true, ... },
{ spineIndex: 1, isLinear: true, ... },        { spineIndex: 1, isLinear: true, ... },
{ spineIndex: -1, isLinear: false, ... },      { spineIndex: 2, isLinear: false, ... },
{ spineIndex: 2, isLinear: true, ... }]        { spineIndex: 3, isLinear: true, ... }]

Default: true

If true, One namespace is given per CSS file or inline style, and styles used for spine is described.

Otherwise it CssItem.namespace, SpineItem.styles is undefined.

In any list, InlineCssItem is always positioned after CssItem. (Book.styles, Book.items, SpineItem.styles, ...)

Default: false

Prepend given string to namespace for identification.

Default: 'ridi_style'

• encoding
• ignoreEntryNotFoundError
• basePath
• spine.extractBody
• spine.extractAdapter
• css.removeAtrules
• css.removeTags
• css.removeIds
• css.removeClasses

If specified then returns a string. Otherwise it returns a buffer.

If specify default, use Item.defaultEncoding.

Item.defaultEncoding // undefined (=buffer)
SpineItem.defaultEncoding // 'utf8'
CssItem.defaultEncoding // 'utf8'
InlineCssItem.defaultEncoding // 'utf8'
ImageItem.defaultEncoding // undefined (=buffer)

Default: 'default'

If false, throw Errors.ITEM_NOT_FOUND.

Default: true

If specified, change base path of paths used by spine and css.

HTML: SpineItem

...
  <!-- Before -->
  <div>
    <img src="../Images/cover.jpg">
  </div>
  <!-- After -->
  <div>
    <img src="{basePath}/OEBPS/Images/cover.jpg">
  </div>
...

CSS: CssItem, InlineCssItem

/* Before */
@font-face {
  font-family: NotoSansRegular;
  src: url("../Fonts/NotoSans-Regular.ttf");
}
/* After */
@font-face {
  font-family: NotoSansRegular;
  src: url("{basePath}/OEBPS/Fonts/NotoSans-Regular.ttf");
}

Default: undefined

If true, extract body. Otherwise it returns a full string.

true:

{
  body: '\n  <p>Extract style</p>\n  <img src=\"../Images/api-map.jpg\"/>\n',
  attrs: [
    {
      key: 'style',
      value: 'background-color: #000000;',
    },
    { // Only added if useStyleNamespace is true.
      key: 'class',
      value: '.ridi_style2, .ridi_style3, .ridi_style4, .ridi_style0, .ridi_style1',
    },
  ],
}

false:

'<!doctype><html>\n<head>\n</head>\n<body style="background-color: #000000;">\n  <p>Extract style</p>\n  <img src=\"../Images/api-map.jpg\"/>\n</body>\n</html>'

Default: false

If specified, transforms output of extractBody.

Define adapter:

const extractAdapter = (body, attrs) => {
  let string = '';
  attrs.forEach((attr) => {
    string += ` ${attr.key}=\"${attr.value}\"`;
  });
  return {
    content: `<article${string}>${body}</article>`,
  };
};

Result:

{
  content: '<article style=\"background-color: #000000;\" class=\".ridi_style2, .ridi_style3, .ridi_style4, .ridi_style0, .ridi_style1\">\n  <p>Extract style</p>\n  <img src=\"../Images/api-map.jpg\"/>\n</article>',
}

Default: defaultExtractAdapter

Remove at-rules.

Default: ['charset', 'import', 'keyframes', 'media', 'namespace', 'supports']

Remove selector that point to specified tags.

Default: []

Remove selector that point to specified ids.

Default: []

Remove selector that point to specified classes.

Default: []