hedgedoc-migrator/hedgedoc-image.py
2022-09-15 19:31:00 +02:00

143 lines
4.3 KiB
Python

import requests
import json
import re
import argparse
import os
from urllib.parse import urlparse
from typing import List
from os.path import exists
PATH = "images/"
NEW_NETLOC = ""
seen_pads = set()
pads_json = []
os.makedirs(PATH, exist_ok=True)
os.makedirs(PATH + "uploads", exist_ok=True)
class Image:
def __init__(self, _old_url: str) -> None:
self.old_url = _old_url
old = urlparse(self.old_url)
self.new_url = "." + old.path;
def download(self):
r = requests.get(self.old_url)
p = PATH + urlparse(self.old_url).path
with open(p, 'wb') as f:
f.write(r.content)
print(f"Downloaded image {self.old_url} to {p}")
class PadPage:
def __init__(self, _old_url: str):
self.old_url: str = _old_url
#self.name: str = _name
self.images: List = []
self.linked_pads = []
self.content: str = ""
def gather_linked_pads(self):
regex = r'https://[\w\d.]+/(?!upload)[\w\-_]+'
matches = re.findall(regex, self.content)
full_url = urlparse(self.old_url)
for match in matches:
print(f"match: {match}")
url = urlparse(match)
if url.netloc == full_url.netloc:
self.linked_pads.append(PadPage(match))
else:
print("Dropped pad, wrong netloc")
def to_dict(self) -> dict:
old = urlparse(self.old_url)
new = old._replace(netloc="pad.hacknang.de")
return {"old_url": self.old_url, "new_url": new.geturl()}
def to_json(self) -> str:
old = urlparse(self.old_url)
new = old._replace(netloc=NEW_NETLOC)
return json.dumps({"old_url": self.old_url, "new_url": new.geturl()})
def download_and_process_content(self):
print(f"Downloading pad at {self.old_url}")
seen_pads.add(self.old_url)
r = requests.get(self.old_url + "/download")
if r.status_code == 200:
self.content = r.text
num = self._find_images()
print(f"Found {num} images")
else:
print(f"Error downloading Pad {self.old_url}, got HTTP status code {r.status_code}")
# returns number of images found
def _find_images(self) -> int:
regex = r'https://[\w\d.]+/uploads/[\w\d]+\.(?:png|jpg|jpeg|webp)'
matches = re.findall(regex, self.content)
full_url = urlparse(self.old_url)
for match in matches:
print(f"match: {match}")
url = urlparse(match)
if url.netloc == full_url.netloc:
i = Image(match)
self.images.append(i)
else:
print("Dropped pad, wrong netloc")
return len(matches)
def download_images(self):
for i in self.images:
i.download()
def recursive(self):
for pad in self.linked_pads:
if pad.old_url not in seen_pads:
print(f"New pad found: {pad.old_url}")
pad.download_and_process_content()
pad.download_images()
pad.gather_linked_pads()
pads_json.append(pad.to_dict())
pad.recursive()
if __name__ == "__main__":
parser = argparse.ArgumentParser(description="Recursivly downloads all images from a hedgedoc pad")
parser.add_argument("pad_url", metavar="N", type=str, nargs="+",
help="urls of the pad to start searching")
parser.add_argument("--replace", dest="new_pad_url", metavar="B", type=str, nargs=1,
help="url of the new pad for generating the json")
args = parser.parse_args()
old_file = []
if exists("pads.json"):
with open("pads.json", "r") as f:
old_file = json.load(f)
for entry in old_file:
seen_pads.add(entry["old_url"])
pads_json.extend(old_file)
print(f"Seen Pads: {seen_pads}")
NEW_NETLOC = args.new_pad_url[0]
for pad_url in args.pad_url:
pad = PadPage(pad_url)
pad.download_and_process_content()
pad.download_images()
pad.gather_linked_pads()
pads_json.append(pad.to_dict())
pad.recursive()
print(f"We have seen {len(seen_pads)} which were {seen_pads}")
with open("pads.json", "w") as f:
f.write(json.dumps(pads_json))