Slack exported conversation avatar replacement with locally downloaded avatars from original URL

fix_avatars.py

import os
import lxml.html as LH
import shutil
import requests

temp_dir = "/tmp/avatar_tmp"
if not os.path.isdir("html/avatars"):
    os.mkdir("html/avatars")

if os.path.isdir(temp_dir):
    shutil.rmtree(temp_dir, ignore_errors=True)

os.mkdir(temp_dir)

for html_file in os.listdir('html'):
    if os.path.isfile("html/" + html_file):
        with open('html/' + html_file, "rb") as html_string:
            root = LH.parse(html_string)
            for el in root.iter('img'):
                avatar_original_url = el.attrib['src']
                split_data = str(el.attrib['src']).split('/')

                if str(avatar_original_url).startswith('https://secure.gravatar.com/avatar/') or\
                        str(avatar_original_url).startswith('https://avatars.slack-edge.com/'):
                    avatar_folder_date = 'html/avatars/' + split_data[3]
                    if str(avatar_original_url).startswith('https://secure.gravatar.com/avatar/'):
                        avatar_file_name = split_data[4].split('?',1)[0]
                    else:
                        avatar_file_name = split_data[4]
                    avatar_folder_file = 'avatars/' + split_data[3] + '/' + avatar_file_name
                    print(avatar_folder_file)
                    if not os.path.isdir(avatar_folder_date):
                        os.mkdir(avatar_folder_date)

                    if not os.path.isfile(avatar_folder_date + '/' + avatar_file_name):
                        with open(avatar_folder_date + '/' + avatar_file_name, "wb") as avatar_destination:
                            avatar_destination.write(bytes(requests.get(avatar_original_url, stream=True).content))
                        print("New avatar found: " + avatar_folder_date + '/' + avatar_file_name)
                    # else:
                    #     print("Avatar " + avatar_folder_date + '/' + avatar_file_name + " already existing")
                    el.attrib['src'] = avatar_folder_file
            with open(temp_dir + '/' + html_file, "wb") as temp_file:
                temp_file.write(LH.tostring(root, pretty_print=True))
                os.remove('html/' + html_file)
                shutil.move(temp_dir + '/' + html_file, "html/")

This works as a second step on @levelsio gist to convert Slack raw message data to a more legible HTML.
At his gist I left a comment on how to use his code in a docker container.

Using this script in a docker container as in previous steps steps

Download gist code at output folder created with previous gist

cd $HOME/slack2html_output
wget -O fix_avatars.py \
https://gist.githubusercontent.com/droberin/74d8880610964518345d7285bc70f898/raw/17677ea13375697b3f25459df717ec1c65c6a523/fix_avatars.py

Execute docker and convert output

docker run -ti --rm -v $HOME/slack2html_output:/mnt/slack_data \
 -v $HOME/slack2html_output:/mnt/slack2html/ \
 python:3-slim bash -c 'cd /mnt/slack_data && pip3 install lxml requests && python3 fix_avatars.py'

Check renewed content

firefox $HOME/slack2html_output/html/

Remove python image if it ain't gonna be in use any longer.

It uses less than 200MiB of disk but... you know... clean it up...

docker rmi python:3-slim