slack_download.py

# Slackのアーカイブ中のトークンの付いたURLをダウンロードする。
# python3 slack_download.py 'xxx Slack export mmm dd yyyy - mmm dd yyyy/' out/ 'xoxe-...'

import sys
import os
import json
import urllib.parse
import urllib.request

indir = sys.argv[1]
outdir = sys.argv[2]
token = sys.argv[3]

def download(data):
    if type(data)==list:
        for d in data:
            download(d)
    if type(data)==dict:
        for d in data.values():
            download(d)
    if type(data)==str:
        if data.startswith("https://") and token in data:
            url = data
            print(url)
            path = [outdir]+urllib.parse.urlparse(data).path.split("/")
            os.makedirs(os.path.join(*path[:-1]), exist_ok=True)
            urllib.request.urlretrieve(url, os.path.join(*path))

for dirpath, dirnames, filenames in os.walk(indir):
    for filename in filenames:
        path = os.path.join(dirpath, filename)
        _, ext = os.path.splitext(path)
        if ext.lower()==".json":
            with open(path) as file:
                data = json.load(file)
                download(data)