Convert boostnote cson format data to markdown

boostnote2md.py

#!/usr/bin/env python3
# coding: utf-8
"""
Author       : weaming
Created Time : 2018-05-26 21:32:59
Prerequisite:
    python3 -m pip install cson arrow
"""
import json
import os
import sys
import datetime
import cson
try:
    import arrow
    time_aware = True
except ImportError:
    print(
        'warning: datetime information will be discarded unless you install arrow'
    )
    time_aware = False


def read_file(fp):
    with open(fp) as f:
        return f.read()


def text_to_dict(text):
    """convert json or cson to dict"""
    try:
        return cson.loads(text)
    except:
        pass

    try:
        return json.loads(text)
    except:
        pass
    raise Exception("text should be in cson or json format")


def read_folder_names(fp):
    data = text_to_dict(read_file(fp))
    return {x['key']: x['name'] for x in data['folders']}


def write_boostnote_markdown(data, output, folder_map):
    """write boostnote dict to markdown"""
    target_dir = os.path.join(output, folder_map[data['folder']])
    if not os.path.exists(target_dir):
        os.makedirs(target_dir)

    target_file = os.path.join(target_dir, '{}.md'.format(data['title'].replace('/', '-')))
    with open(target_file, 'w') as f:
        f.write(data.get('content', ''))
        print(target_file)

    if time_aware:
        update_at = arrow.get(data['updatedAt'])
        update_at_epoch = int(update_at.timestamp())
        os.utime(target_file, (update_at_epoch, update_at_epoch))
        stat = os.stat(target_file)


def process_file(source, output, folder_map):
    data = text_to_dict(read_file(source))
    write_boostnote_markdown(data, output, folder_map)


def main(boostnote_dir, output):
    """
    :input: input folder path
    :output: output folder path
    """
    folder_map = read_folder_names(os.path.join(boostnote_dir, 'boostnote.json'))
    notes_dir = os.path.join(boostnote_dir, 'notes')
    for name in os.listdir(notes_dir):
        if not name.endswith('.cson'):
            continue

        source = os.path.join(notes_dir, name)
        process_file(source, output, folder_map)


if __name__ == '__main__':
    import argparse
    parser = argparse.ArgumentParser(
        description="convert boostnote cson format data to markdown")

    parser.add_argument(
        '-s',
        '--source',
        type=str,
        help="directory store the cson files",
        default=".")
    parser.add_argument(
        '-o', '--output', type=str, help="output directory", default="output")

    args = parser.parse_args()

    main(args.source, args.output)

This is awesome.
For those who find this, here's the usage:

usage: boostnote2md.py [-h] [-s SOURCE] [-o OUTPUT]

I had to pip install cson. In addition to the change noklam made, I also changed it to support unicode:

   with io.open(target_file, 'w', encoding='utf-8') as f:
        f.write(data.get('content', ''))
        print(target_file)

(need to import io for this to work).

Line 56 f.write(data['content'])
would be better to change to f.write(data.get('content', ''))` to avoid error when there are empty notes.

@noklam May "content" not exist?

I created new one empty note, and the content is

createdAt: "2018-12-25T12:26:26.948Z"
updatedAt: "2018-12-25T12:26:26.948Z"
type: "MARKDOWN_NOTE"
folder: "7487fac9e197583f5a35"
title: ""
tags: []
content: ""
isStarred: false
isTrashed: false

@weaming that's the problem, if content does not exist it will cause error. By using get it will get empty string instead of error.

@weaming THANK YOU! That saved me a lot of time :-D

I noticed the tags are not preserved. Therefore I added a line above line 56:
f.write(f"Tags: {', '.join(data.get('tags', ''))}\n")
Inserting the tags comma separated at the top of each note/file.

One more thing: When I discovered that all my snippet notes were empty after converting, I added code to not only export markdown notes but also snippet notes, putting the snippets sequentially under each other.

with open(target_file, 'w', encoding='utf-8') as f:
        if data["type"] == 'MARKDOWN_NOTE':
            f.write(f"Tags: {', '.join(data.get('tags', ''))}\n")
            f.write(data.get('content', ''))
        elif data["type"] == 'SNIPPET_NOTE':
            f.write(f"Tags: {', '.join(data.get('tags', ''))}\n")
            snippets = data.get('snippets', '')
            if snippets != '':
                for snippet in snippets:
                    if not snippet['name']:
                        f.write(data['title'])
                    else:
                        f.write(f"File/Description: {snippet['name']}\n")
                        try:
                            f.write(f'Language: {snippet["mode"]}\n')
                        except KeyError:
                            pass
                        f.write('Code:\n```')
                        try:
                            f.write(f'{str(snippet["mode"]).lower()}\n')
                        except KeyError:
                            f.write('\n')
                        f.write(snippet['content'])
                        f.write('\n```\n')
                        if len(snippets) > 1:
                            f.write('\n***\n\n')
            else:
                f.write('')
        else:
            sys.exit(1, 'Unknown note type - do not know how to handle, exiting!')
        print(target_file)

@mbrausen Thank you for sharing! But I will not update this gist, because I had stopped using boostnote and have no testing for your code.

Thank you for making this!, this was an excellent resource to convert and organize boostnote notes to Obsidian.

The time_aware logic from L59-L63 doesn't work correctly for me. Unless your local time is Zulu time (GMT+0), all timestamps of exported files will be off.

To export files with correct lastModified stat, I changed the logic as follows:

   if time_aware:
        update_at = arrow.get(data['updatedAt'])
        timestamp = update_at.timestamp()
        os.utime(target_file, (timestamp, timestamp))
        stat = os.stat(target_file)

@boonyasukd Updated as your suggestion. Thks!