Download Logseq repository images to local asset folder

LogseqImagesToAssetFolder.py

#!/usr/bin/env python
# coding: utf-8


# What is this?
# This script is for logseq webapp users that want to switch to the local version.
# It will download all images that are linked inside your graph and store them in the local "/assets/" folder.

# Remarks
# 1. Only supports Markdown files (for now)
# 2. Filnames are not preserved. The filenames will be a news random string. (Let me know if you need it otherwise.)
# 2. Make sure you close Logseq Desktop app and disconnect web app before starting!
# 3. Use at your own risk! Verify everything worked before permanently moving to the new local version!

# How to use
# 1. Download your logseq repository from Github to your local machine.
# 2. Run the script with "-f <Path to that repository>"
# 3. Reccomended if you know how: Check the uncommitted changes to the repository. This is an easy way to check if everything worked as intended.
# 3. Open that folder in the logseq client.

import glob
import re
import uuid
import urllib.request
from urllib.error import HTTPError
import shutil
from pathlib import Path
import argparse

regex_markdownFile = re.compile(r'!\[[^\]]*\]\(([^\)]*)\)') # markdown file syntax

def save_file_locally(logseq_dir, url):
    try:
        with urllib.request.urlopen(url) as r:
            file_ending = "." + r.info().get_content_subtype()
            filename = str(uuid.uuid4().hex) + str(uuid.uuid4().hex) + str(uuid.uuid4().hex) + file_ending

            with open(logseq_dir+"assets/"+filename, 'wb') as out_file:
                shutil.copyfileobj(r, out_file)
        return "../assets/"+ filename
    except HTTPError as err:
        print(F"Error! Could not load file from {url}!")
        print(err)
        return False
    except ValueError as err:
        print(F"Error! Could not load file for regex match: {url}!")
        print(err)
        return False



def run_dl(logseq_dir):
    for fPath in glob.glob(f"{logseq_dir}**/*.md", recursive=True):
        matches = []
        with open(fPath) as _file:
            filedata = _file.read()
            _file.seek(0)
            for i, line in enumerate(_file.readlines()):
                matches.extend(regex_markdownFile.findall(line))
        for match in matches:
            new_url = save_file_locally(logseq_dir=logseq_dir, url=match)
            if new_url:
                filedata = filedata.replace(match, new_url)

        with open(fPath, 'w') as _file:
            _file.write(filedata)
        print(fPath)
        
if __name__ == "__main__":
    # Program arguments.
    parser = argparse.ArgumentParser(
        description="Script to download the images of your git logseq graph to local asset folder."
    )
    parser.add_argument(
        "-f",
        type=str,
        default=None,
        metavar="<logseq graph>",
        help="Absolute path to your local repository copy of your logseq graph.",
    )


    args = parser.parse_args()
   
    if not args.f:
        print("Please give me the absolute path to your local repository copy of your logseq graph with:")
        print("-f <Path>")
    logseq_dir = args.f
    print(f"Starting download of files to: {logseq_dir}")
    run_dl(logseq_dir)

Thanks a lot. I've changed two things to make it work for me.
1- Replace line 30 to

regex_markdownFile = re.compile(r'!\[[^\]]*\]\((http.+)\)') # markdown file syntax

2- Add utf-8 to file open to overcome encoding errors. So run_dl becomes:

def run_dl(logseq_dir):
    for fPath in glob.glob(f"{logseq_dir}**/*.md", recursive=True):
        matches = []
        print(fPath)
        with open(fPath, encoding='utf-8') as _file:
            filedata = _file.read()
            _file.seek(0)
            for i, line in enumerate(_file.readlines()):
                matches.extend(regex_markdownFile.findall(line))
        for match in matches:
            new_url = save_file_locally(logseq_dir=logseq_dir, url=match)
            if new_url:
                filedata = filedata.replace(match, new_url)

        with open(fPath, 'w', encoding='utf-8') as _file:
            _file.write(filedata)
        print(fPath)