Converts dendron note structure to obsidian note structure

dendron-to-obsidian.py

# converts dendron note structure to obsidian note structure
# - removes frontmatter
# - adds tags
# - converts internal links
# - ignores empty (frontmatter-only) notes
#
# use at your own risk

import glob
import re
import os


PATH_TO_DENDRON_VAULT = "TODO"
PATH_OBSIDIAN_VAULT = "TODO"
TAGS = {
    "study": "#study",
    "study.k8s": "#k8s",
    "study.k8s.ckad": "#ckad",
    # ...
    "tools": "#tools",
    "tools.vscode": "#vscode",
}


def to_wikilink(dendron_link):
    segments = dendron_link.split("#")
    link_components = segments[0].split(".")
    if len(segments) == 1:
        link = dendron_link
        appendix = ""
        note_name = link_components[-1]
    else:
        link = segments[0]
        appendix = f"#{'#'.join(segments[1:])}"
        note_name = segments[-1]
    obsidian_link = link.replace(".", "/")
    return f"{obsidian_link}.md{appendix}|{note_name}"


def _cleanse_markdown(dendron_path):
    with open(dendron_path, "r") as f:
        raw_content = f.read()

    # remove frontmatter
    blocks = raw_content.split("---\n")
    md_content = "---\n".join(blocks[2:])

    # add tags
    file_name = os.path.basename(dendron_path)

    if not any([tag in file_name for tag in ["daily", "monthly"]]) and not _is_empty(
        md_content
    ):
        tags = [v for k, v in TAGS.items() if file_name.startswith(k)]
        md_content = f"{' '.join(tags)}\n{md_content}"

    # convert internal links
    for match in re.findall(r"\[\[(.*?)\]\]", md_content):
        md_content = md_content.replace(f"[[{match}]]", f"[[{to_wikilink(match)}]]")

    return md_content


def _is_empty(md_content):
    return len(md_content.strip()) == 0


def _convert_dendron_note(dendron_path):
    dendron_filename = os.path.basename(
        dendron_path
    )  # tech.powershell.commands.md | cmd.journal.daily.2023.03.09.md | cmd.journal.monthly.2022.10.md
    if "daily" in dendron_filename:
        obs_dirs = "journal/daily"
        date_part = dendron_filename.split(".")[-4:-1]
        obs_name = "-".join(date_part) + ".md"
    elif "monthly" in dendron_filename:
        year = dendron_filename.split(".")[-3]
        obs_dirs = f"journal/monthly/{year}"
        obs_name = dendron_filename.split(".")[-2] + ".md"
        print(obs_dirs, obs_name)
    else:
        obs_dirs = "/".join(dendron_filename.split(".")[:-2])
        obs_name = f"{dendron_filename.split('.')[-2]}.md"
    return obs_dirs, obs_name


def _create_note(obs_dirs, obs_note, md_content):
    if not os.path.exists(f"{PATH_OBSIDIAN_VAULT}/{obs_dirs}"):
        os.makedirs(f"{PATH_OBSIDIAN_VAULT}/{obs_dirs}")
    with open(f"{PATH_OBSIDIAN_VAULT}/{obs_dirs}/{obs_note}", "w") as f:
        f.write(md_content)


if __name__ == "__main__":
    for filepath in glob.iglob(f"{PATH_TO_DENDRON_VAULT}/*.md"):
        md_content = _cleanse_markdown(filepath)
        if not _is_empty(md_content):
            obs_dirs, obs_name = _convert_dendron_note(filepath)
            _create_note(obs_dirs, obs_name, md_content)
            print(filepath, "converted to", f"{obs_dirs}/{obs_name}")
        else:
            print("Skipping empty file", filepath)

Thanks 🙏