Convert the CS flash cards by jwasham to CSV for Anki import.

jwasham-anki-builder.py

#!/usr/bin/env python
"""Convert the CS flash cards by jwasham to CSV for Anki import.

On import:

  - Choose tab as the field separator

  - Ignore lines where first field matches existing note

  - Allow HTML in fields

Some tags are added (c, c++, python, code, and vocab) to distinguish the type of cards
and their content. Try importing into a separate deck in order to avoid contaminating
existing ones.

For the source data, see: https://github.com/jwasham/computer-science-flash-cards
"""
import csv
import re
import sqlite3
import sys
from argparse import ArgumentParser
from contextlib import contextmanager
from enum import IntEnum


@contextmanager
def fopen(filename=None, mode="r"):
    if filename is None or filename == "-":
        f = sys.stdout
        try:
            yield f
        finally:
            f.close()
    else:
        with open(filename, mode) as f:
            yield f


class Type(IntEnum):
    VOCAB = 1
    CODE = 2


def convert(input, output):
    conn = sqlite3.connect(input)
    conn.text_factory = str
    cursor = conn.cursor()
    cursor.execute("SELECT type, front, back FROM cards")
    items = cursor.fetchall()
    write_to_csv(output, items)


def infer_coding_language(front, back, tags):
    """Heuristically infer the coding language used."""
    if re.search(r"\bC\+\+(?!\w)", front, re.IGNORECASE):
        tags.add("c++")
    if re.search(r"\bPython\b", front, re.IGNORECASE):
        tags.add("python")
    if re.search(r"\bC\b", front, re.IGNORECASE):
        tags.add("c")

    if re.search(r"\bdef [a-zA-Z]+", back):
        tags.add("python")

    if re.search(r"\b(void|int) [a-zA-Z]+", back):
        tags.add("c")


def write_to_csv(output, items, tags=None):
    common_tags = tags or set()
    with fopen(output, "w") as f:
        writer = csv.writer(f, delimiter="\t", lineterminator="\n")
        if tags:
            writer.writerow([f"tags: {' '.join(tags)}"])

        for item_type, front, back in items:
            tags = common_tags.copy() if common_tags else set()

            infer_coding_language(front, back, tags)

            if item_type == Type.CODE:
                tags.add("code")
                back = f"<pre>{back}</pre>"
            elif item_type == Type.VOCAB:
                tags.add("vocab")

            back = back.replace("\r\n", "<br>")

            writer.writerow([front, back, " ".join(tags)])


if __name__ == "__main__":
    parser = ArgumentParser()
    parser.add_argument("input", help="SQLite database file")
    parser.add_argument("--output", "-o", help="Output CSV file", default=None)
    args = parser.parse_args()
    convert(args.input, args.output)