Hexo to Hugo Article & Images migrator

_README.md

Hexo to Hugo Migrator

Usage:

1. cd to Hexo source/ directory. copy migrator.py here.
2. If you have image host CDN mapped to local directory, change LOCAL_CDN_DOMAIN and p_imgcdn accordingly
3. run ./migrator.py, pip package requests is the only dependency.
4. check output in _export/

migrator.py

import re
import shutil
from pathlib import Path
from tempfile import NamedTemporaryFile
from urllib.parse import urlparse

import requests

REGEX_IMAGE = re.compile(r'!\[(?P<comment>[^\]]*)\]\((?P<filename>.*?)(?=\"|\))(?P<optionalpart>\".*\")?\)')
WHITELIST = []
LOCAL_CDN_DOMAIN = "your-cdn.domain"

p_posts = Path("./_posts")
p_imgcdn = Path("../../img-cdn-local")
p_export = Path("./_export")


p_export.mkdir(parents=True, exist_ok=True)

for p in p_posts.rglob("*.md"):
    with p.open(encoding="utf-8") as f:
        content = f.read()
    images = list(REGEX_IMAGE.finditer(content))
    existed_image = set()
    dstdir = p_export / p.relative_to(p_posts).with_suffix("")
    dstdir.mkdir(parents=True, exist_ok=True)
    dstmd = dstdir / "index.zh.md"

    for i, img in enumerate(images):
        fpath = img.group("filename")
        if fpath in existed_image:
            print(p, img, "duplicated")
            continue
        elif not fpath:
            print(p, img, "empty")
            continue
        else:
            existed_image.add(fpath)
        if fpath.startswith("http"):
            url = urlparse(fpath)
            if url.netloc == LOCAL_CDN_DOMAIN:
                fpath = p_imgcdn / url.path[1:]
                assert fpath.exists(), f"{p}: CDN image {url.path} should exist on {fpath}"
                print(p, img, "found CDN image")
            elif url.netloc not in WHITELIST:
                print(p, img, "skipped not in whitelist.")
                continue
            else:
                resp = requests.get(fpath)
                suffix = Path(url.path).suffix
                with NamedTemporaryFile(suffix=suffix, delete=False) as tmpfile:
                    tmpfile.write(resp.content)
                    fpath = Path(tmpfile.name)
                    print(p, img, "downloaded to", fpath)
        else:
            fpath = Path(fpath)
            print(p, img, "is local image")
        if not fpath.exists():
            print(p, img, "not exists")
            continue

        dst = dstdir / f"{i:02d}{fpath.suffix}"
        shutil.copyfile(fpath, dst)

        cmt = img.group('comment')
        if (opt := img.group('optionalpart')):
            opt = f" {opt}"
        else:
            opt = ""
        content = content.replace(img.group(0), f"![{cmt}]({dst.name}{opt})")

    with dstmd.open("w", encoding="utf-8") as f:
        f.write(content)