7.2 Fields per document type

Libiry extracts and manages book metadata from various sources and formats. All field names are configurable.

Field	EPUB	CBZ	Other
cover	Cover image in EPUB	First image	URL in Markdown
booktitle	OPF (dc:title)	ComicInfo.xml	Markdown
author	OPF (dc:creator)	ComicInfo.xml	Markdown
author_sort	OPF (opf:file-as)	Markdown	Markdown
isbn	OPF (dc:identifier)	Markdown	Markdown
rating	calibre:rating meta	ComicInfo.xml	Markdown
publisher	OPF (dc:publisher)	ComicInfo.xml	Markdown
publication date	OPF (dc:date)	year from ComicInfo.xml	Markdown
language	OPF (dc:language)	ComicInfo.xml (LanguageISO)	Markdown
pages	rendition:page-count or calibre:pages	ComicInfo.xml (PageCount)	Markdown
tags	OPF (dc:subject)	(Tags)	Markdown
series	calibre:series meta	ComicInfo.xml (Series)	Markdown
series index	calibre:series_index	ComicInfo.xml (Number)	Markdown
translator	OPF (dc:contributor role=trl)	ComicInfo.xml	Markdown
illustrator	OPF (dc:contributor role=ill)	ComicInfo.xml (Penciller)	Markdown
description	OPF (dc:description)	ComicInfo.xml (Summary)	Markdown
notes	calibre:user_notes	ComicInfo.xml (Notes)	Markdown

• rendition:page-count is an EPUB 3 standard meta element
• calibre:pages is a Calibre specific extension. It is only available if you create a custom column “pages” in Calibre

• MOBI, AZW and AZW3 are all Amazon Kindle formats. All metadata is stored in a sidecar for these formats, because they don’t support direct metadata editing
• ComicInfo.xml is a standard comic metadata format that is used in CBZ files
• CBR works with a protected RAR-format, so all metadata is stored in a sidecar
• When “Store metadata in sidecar” is True, the metadata for EPUB, PDF and CBZ are also maintained in sidecar files
• Please note that any HTML characters will be removed from field values before writing to sidecars, as are intermediate quotes
• Cover: a cover image URL or path should be placed here
• Language is a free format field, but is intended to have a language code in it
• ISBN: an ISBN-10 or ISBN-13 code is expected here
• Rating: Like Calibre, Libiry uses 0-10. It shows 0-5, though (automatic conversion)

Metadata extraction

EPUB Files

Metadata is read from the OPF file inside the EPUB:

<metadata>
  <dc:title>The Book Title</dc:title>
  <dc:creator>Author Name</dc:creator>
  <dc:subject>fiction</dc:subject>
  <dc:identifier>978-1234567890</dc:identifier>
</metadata>

PDF files

Metadata is read from the PDF’s document information dictionary:

• Title → booktitle
• Author → author
• Subject → description
• Keywords → tags (comma-separated)

MOBI/AZW Files

Basic metadata is extracted using the mobi library:

• Title → booktitle
• Author → author
• Tags → from OPF sidecar file

Comic files (CBR/CBZ)

Metadata is read from ComicInfo.xml:

<ComicInfo>
  <Title>Comic Title</Title>
  <Writer>Author Name</Writer>
  <Genre>Action, Adventure</Genre>
</ComicInfo>

Markdown files

---
cover: https://covers.openlibrary.org/b/id/11405920-M.jpg
booktitle: The Handmaid's Tale 
author: Margaret Atwood
isbn: 9780385490818 
rating: 5.0
language: en
tags:
  - distopian
  - scifi
path: Scifi/Atwood
size: 751.5 KB
type: epub
---
 
# The Handmaid's Tale 

• Uses YAML frontmatter (Obsidian-compatible)
• Numbers are parsed as integers
• Filename derived from book title
• Content section below frontmatter (optional)
• Compatible with Libiry for viewing

The Libiry BookSpineScanner can creates the following extra fields if you wish so:

Field	Description
size	Book file size
type	Book type
path	Original file name, including folder path. Backslashes in the path name are changed into forward slashes
book created	Creation date of the book - for tracking your reading habits
book modified	Modified date of the book - for tracking your reading habits

| scan_confidence: high (4%) | scan_source_api: openlibrary | scan_source: “20260531_102438 spaink.jpg” | scan_date: 2026-08-30T13:36:53.887Z

Libiry2Go creates the following extra fields:

Field	Description
size	Book file size
type	Book type
path	Original file name, including folder path. Backslashes in the path name are changed into forward slashes
book created	Creation date of the book - for tracking your reading habits
book modified	Modified date of the book - for tracking your reading habits

Tag management

Reading tags

Tags are read from (in priority order):

1. The sidecar file (if it exists)
2. The embedded metadata in the e-book

Writing tags

Tags are written to:

• EPUB: Embedded dc:subject elements
• CBZ: ComicInfo.xml
• Markdown: YAML frontmatter
• Other book formats: In sidecar file

Tag format

Tags are stored as:

• Comma-separated string in some formats
• Multiple dc:subject elements in EPUB/OPF
• Multi-line tags in sidecars. Arrays for tags are also supported ( tags: [fiction, fantasy]).

In markdown:

tags:
  - fiction
  - fantasy
  - read2026

Nested tags work too:

tags:
  - books/fantasy
  - status/read

Cover handling

Cover sources (priority order)

1. Embedded cover - Extracted from e-book file
2. Cover field - URL or path in metadata
3. Online lookup:
   • Open Library (by ISBN or title+author)
   • Google Books (by ISBN or title+author)
   • Europeana (European books)
   • Other book databases

Cover display

• Thumbnails are cached in an SQLite database
• Cache is cleared on startup for fresh covers
• Fallback display shows author + title text

Cover field formats

# URL
cover: https://example.com/cover.jpg
 
# Local path
cover: ./covers/book.jpg
 
# Obsidian syntax
cover: [[cover.jpg]]
cover: ![[cover.jpg]]

Duplicate detection

Is done using:

ISBN matching

Exact match on ISBN-10 or ISBN-13

Title + author matching

Normalization process:

1. Convert to lowercase
2. Remove articles: “the”, “a”, “an”, “de”, “het”, “een”
3. Remove author suffixes: “jr”, “sr”, “phd”
4. Sort author name parts alphabetically
5. Compare normalized strings

Example:

• “The Handmaid’s Tale” by “Margaret Atwood”
• “Handmaid’s Tale” by “Atwood, Margaret”
• Both normalize to the same value → detected as duplicate

Field name customization

Map Libiry fields to your existing metadata:

# For Obsidian
Field name booktitle: title
Field name tags: tags
 
# For custom setups
Field name booktitle: book_title
Field name author: writer

This ensures compatibility with:

• Goodreads exports
• Obsidian plugins (Book Search, Booksidian)
• Custom database schemas

Cover extraction priority

1. Embedded cover in e-book metadata
2. First image for comics (CBR/CBZ)
3. First page thumbnail for PDFs
4. Cover field in markdown files
5. Online lookup from Open Library, Google Books, Europeana

Performance notes

• For large PDFs it may take longer to extract covers
• Comics with many images are processed efficiently (only the first image)
• Cache files speed up processing