-
-
Save glensc/3ef3590f8e3c4bed0550b99e40220c32 to your computer and use it in GitHub Desktop.
Pretty-print a facebook message.json, fixing up broken encoding
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/env python3 | |
import sys | |
import json | |
import warnings | |
from datetime import datetime | |
def fixup_str(text): | |
return text.encode('latin1').decode('utf8') | |
def fixup_list(l): | |
return [fixup(e) for e in l] | |
def fixup_dict(dct): | |
return {fixup_str(k): fixup(v) for k, v in dct.items()} | |
def fixup(e): | |
if isinstance(e, dict): | |
return fixup_dict(e) | |
if isinstance(e, list): | |
return fixup_list(e) | |
if isinstance(e, str): | |
return fixup_str(e) | |
return e | |
def filter_messages(messages): | |
for message in messages: | |
if message['type'] == 'Generic': | |
yield message | |
def format_message(message): | |
sender = message['sender_name'] | |
timestamp = datetime.fromtimestamp(message['timestamp_ms']/1000) | |
if message['is_unsent']: | |
content = "<Unsent Message>" | |
elif 'content' in message: | |
content = message['content'] | |
elif 'photos' in message: | |
content = 'Photo: <{0}>'.format(message['photos'][0]['uri']) | |
elif 'gifs' in message: | |
content = "Gif: <{0}>".format(message['gifs'][0]['uri']) | |
elif 'audio_files' in message: | |
content = "Audio: <{0}>".format(message['audio_files'][0]['uri']) | |
elif 'videos' in message: | |
content = "Video: <{0}>".format(message['videos'][0]['uri']) | |
elif 'files' in message: | |
content = "Video: <{0}>".format(message['files'][0]['uri']) | |
elif 'sticker' in message: | |
content = "Sticker: <{0}>".format(message['sticker']['uri']) | |
else: | |
return None | |
return f'{timestamp}\t{sender}\t{content}' | |
def main(args): | |
if not args: | |
args = ['message_1.json'] | |
messages = [] | |
from pprint import pprint | |
for filename in reversed(args): | |
with open(filename, 'r') as f: | |
chunk = fixup(json.loads(f.read())['messages']) | |
chunk.reverse() | |
messages.extend(chunk) | |
for message in filter_messages(messages): | |
formatted = format_message(message) | |
if not formatted: | |
warnings.warn(f"Skipped {message}") | |
continue | |
print(formatted) | |
if __name__ == '__main__': | |
main(sys.argv[1:]) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment