search new markdown recently added, with text ANKI in it. Add to Anki Deck, as a raycast command

md_to_anki.py

# -*- coding: utf-8 -*-
# title: md_to_anki
# date: "2023-05-28"

# @raycast.title Markdown files 🪴 to ANKI
# @raycast.author HTLin the 🦎
# @raycast.description search new markdown recently added, with text ANKI in it. Add to Anki Deck
# @raycast.authorURL https://github.com/htlin222

# @raycast.icon 📚
# @raycast.mode silent
# @raycast.packageName System
# @raycast.schemaVersion 1

import glob
import mistune
import re
import os
import requests
import json
from datetime import datetime, timedelta

ANKI_DECK_NAME = "01_garden"
ANKI_NOTE_TYPE = 'Basic'
ANKI_FRONT_FIELD = 'Front'
ANKI_BACK_FIELD = 'Back'
ANKICONNECT_ENDPOINT = "http://localhost:8765"
KEY_WORD = 'ANKI'
FOLDER_TO_WATCH = '~/YOUR/PATH/WITH/MARKDOWN"


def search_markdown_files(directory):
    # Use glob to find Markdown files in the specified directory
    file_paths = glob.glob(directory + '/*.md')

    # Get the current time and calculate the time threshold
    current_time = datetime.now()
    threshold_time = current_time - timedelta(hours=24)

    for file_path in file_paths:
        # Get the modification timestamp of the file
        file_modified_time = datetime.fromtimestamp(os.path.getmtime(file_path))

        # Check if the file was modified or created within the last 24 hours
        if file_modified_time >= threshold_time:
            extract_note_content(file_path, KEY_WORD)


def extract_note_content(file_path, text):
    with open(file_path, 'r') as file:
        content = file.read()
        pattern = re.compile(text, re.IGNORECASE)
        matches = pattern.findall(content)
        pattern = re.compile(text, re.IGNORECASE)
        content = re.sub(pattern, '', content)
        if len(matches) > 0:
            print(f"found {file_path} with {KEY_WORD} in it")
            title_match = re.search(r'^#\s*(.*)', content, re.MULTILINE)
            if title_match:
                title = title_match.group(1).strip()
                back = content[title_match.end():]
                back = find_markdown_image_and_and_replace(back)
                back = remove_blank_lines(back)
                back = convert_markdown_to_html(back)
                print(f"after{back}")
            else:
                title = None
                back = None
            sent_to_anki(title, back)
    with open(file_path, 'w') as file:
        file.write(content)


def find_markdown_image_and_and_replace(md_content):
    pattern = r'\!\[(.*)\]\((http.*?)\)'

    # Find all remote image links in the Markdown content
    for match in re.finditer(pattern, md_content):
        image_alt_text = match.group(1)
        image_url = match.group(2)

        # Generate a local file path for the image based on its URL
        image_file_name = os.path.basename(image_url)
        payload = {
                    "action": "storeMediaFile",
                    "version": 6,
                    "params": {
                        "filename": image_file_name,
                        "url": image_url
                    }
                }
        response = requests.post(ANKICONNECT_ENDPOINT, data=json.dumps(payload))
        print(f"IMAGE: {response}")

        # Replace the remote image link with a local image path in the Markdown content
        md_content = md_content.replace(match.group(0), f'![{image_alt_text}]({image_file_name})')

    return md_content

def remove_blank_lines(text):
    lines = text.splitlines()
    non_blank_lines = [line for line in lines if line.strip()]
    return '\n'.join(non_blank_lines)


def convert_markdown_to_html(markdown_text):
    renderer = mistune.HTMLRenderer()
    markdown_parser = mistune.Markdown(renderer)
    html_text = markdown_parser(markdown_text)

    # Convert ![]() image syntax to HTML format
    html_text = re.sub(r'!\[([^\]]*)\]\((.*?)\)', r'<img alt="\1" src="\2">', html_text)

    # Add HTML line breaks after each line
    # html_text = html_text.replace('\n', '<br>')

    return html_text


def sent_to_anki(title, back):
    print(back)
    payload = {
        'action': 'addNote',
        'version': 6,
        'params': {
            'note': {
                'deckName': ANKI_DECK_NAME,
                'modelName': ANKI_NOTE_TYPE,
                'fields': {
                    ANKI_FRONT_FIELD: title,
                    ANKI_BACK_FIELD: back,
                },
                'options': {
                    'allowDuplicate': True,
                },
                'tags': ['from_garden'],
            },
        },
    }

    # Send the AnkiConnect request
    response = requests.post(ANKICONNECT_ENDPOINT, data=json.dumps(payload))

    print(f'{title} done')
    # Print the response
    print(response.content.decode())

if __name__ == '__main__':
    directory_path = os.path.expanduser(FOLDER_TO_WATCH)
    search_markdown_files(directory_path)
    print("😎 Done")