anki-sqlalchemy

Clean python interface for interacting with Anki's database


License
MIT
Install
pip install anki-sqlalchemy==0.3.0

Documentation

PyPi Version

Anki SQLAlchemy is an interface for interacting with the Anki sqlite database from python without having to either hack an Anki install or figure out the database structure and field serialization from scratch.

The goal of this project is not to support every version of Anki entirely. The current version supports at a signficant amount of the Anki 2.1.38.

Here is a small code snippet written first without anki_sqlalchemy to show how unintuivite the data format and columns names are without an wrapper.

# plain python without anki-sqlalchemy
import sqlite3

conn = sqlite3.connect('backup.db')
cursor = conn.execute("SELECT id, tags FROM notes WHERE mod >= ?", [1445394366])
note = cursor.fetchone()
note[0]  # 1428143940996
note[1]  # ' edit math probability wikipedia '

cursor = conn.execute("SELECT mod, type FROM cards WHERE nid = ?", [nid])
card = cursor.fetchone()
card[0]  # 1445394366
card[1]  # 2
# with anki-sqlalchemy
import datetime
from sqlalchemy import create_engine
from sqlalchemy.orm import sessionmaker
from anki_sqlalchemy import Card

engine = create_engine("sqlite:///backup.db", echo=True)
Session = sessionmaker(bind=engine)
session = Session()

note = session.query(Note).filter(Note.modification_time >= datetime.datetime(2017, 2, 5, 21, 29, 49)).first()
note.id  # 1428143940996
note.modification_time  # datetime.datetime(2017, 2, 5, 21, 29, 49)

card = note.cards[0]
card.modification_time  # datetime.datetime(2019, 11, 5, 22, 23, 3)
card.type  # <CardType.due: 2>

Anki SQLAlchemy also plays nicely with types too.

card: Card = session.query(Card).first()
reveal_type(card.modification_time)
# Revealed type is 'datetime.datetime*'

reveal_type(card.note.tags)
# Revealed type is 'builtins.list*[builtins.str]'

BEWARE!

This package can be used to make changes to your anki database. Before proceeding, please make a backup of your database file. You don't want to lose all your work with a bad query.

The Anki database typically lives in a collection.anki2 file.

Install

pip install anki_sqlalchemy