marcdantic

marcdantic is a Python package designed to parse, validate, and manipulate MARC (Machine-Readable Cataloging) records, a standard data format widely used by libraries and archives for bibliographic information.

Built on Pydantic for robust data validation, marcdantic supports parsing MARC records from both binary MARC (MRC) and MARC XML formats, producing clean, structured Python dictionaries for further processing or storage.

---

Features

• Binary MARC Parsing (from_mrc): Decode raw MARC21 records from their binary format with support for various encodings. Extracts leaders, fixed fields, variable fields, indicators, and subfields.
• MARC XML Parsing (from_xml): Parse MARC XML records, validate leaders and fields, and reconstruct MARC binary data with correct directory and field lengths.
• Pydantic Models for Queries: Define complex MARC search queries using Pydantic models with strong typing and operator support (exact match, regex, contains, etc.).
• Field Validation: Validates field tags, subfield codes, and indicator characters according to MARC standards.
• Flexible and Extensible: Easily customize field mappings, tag aliases, and add your own MARC processing logic.

Got it! Here’s the updated Installation section for your marcdantic README using the style and structure from your other lib’s README, but without the YAZ-related notes:

---

Installation

Installing from GitHub using version tag

You can install marcdantic directly from GitHub for a specific version tag:

pip install git+https://github.com/moravianlibrary/marcdantic.git@v1.2.3

Replace v1.2.3 with the desired version tag.

To always install the most recent version, use the latest tag:

pip install git+https://github.com/moravianlibrary/marcdantic.git@latest

Installing local dev environment

Install required dependencies using pip:

pip install -r requirements.txt

---

Basic Usage

Parsing a binary MARC record

from marcdantic.parsers import from_mrc

with open("record.mrc", "rb") as f:
    marc_bytes = f.read()

record = from_mrc(marc_bytes, encoding="utf-8")
print(record["leader"])
print(record["fixed_fields"])
print(record["variable_fields"])

Parsing a MARC XML record

from lxml import etree
from marcdantic.parsers import from_xml

xml_root = etree.parse("record.xml").getroot()
record = from_xml(xml_root)
print(record["leader"])
print(record["fixed_fields"])
print(record["variable_fields"])
print(record["marc"])  # reconstructed MARC binary

Constructing MARC search queries

from marcdantic.search import MarcCondition, MarcBoolQuery, MarcSearchRequest, SearchOperator

query = MarcBoolQuery(
    must=[
        MarcCondition(field="245", subfield="a", value="Python", operator=SearchOperator.Contains),
        MarcCondition(field="100", subfield="a", value="Guido van Rossum")
    ]
)

search_request = MarcSearchRequest(query=query, page=1, page_size=20)

---

Resources

• Ex Libris Aleph X-Services – Present Service
• Library of Congress MARC Bibliographic Standard
• A Proposal to Serialize MARC in JSON (Archived)

These references provide background on MARC structure, encoding, and usage in real-world library systems.