blog/twitfetch.py

from twitter import *

import pprint

import json
import pathlib
import urllib

import twauth

forcerefetch = False

base_dir = pathlib.Path('content') / 'twitter'
imgbase_dir = pathlib.Path('content') / 'media' / 'twitter'

# Info for getting alt text.
# Requires a project, which requires a phone number, which it rejects my home phone
# https://developer.twitter.com/en/docs/twitter-api/data-dictionary/object-model/media

def write_media(media):
	mediaurl = media['media_url_https']

	parts = urllib.parse.urlparse(mediaurl)
	fname = pathlib.PurePosixPath(parts.path).name
	imgpath = imgbase_dir / fname

	if not forcerefetch and imgpath.exists():
		print('media exists', fname)
		return

	resp = urllib.request.urlopen(mediaurl)
	try:
		if resp.getcode() == 200:
			with open(imgpath, 'wb') as fp:
				d = None
				while d != b'':
					d = resp.read1()
					fp.write(d)
	finally:
		resp.close()

def write_tweet(tweet):
	twpath = base_dir / (tweet['id_str'] + '.yaml')

	if not forcerefetch and twpath.exists():
		print('exists', tweet['id_str'])
		return

	#pprint.pprint(repr(tweet))

	try:
		# Has media to d/l
		for i in tweet['entities']['media']:
			write_media(i)
	except KeyError as e:
		pass

	with open(twpath, 'w') as fp:
		print('write:', tweet['id_str'])

		# wrap in --- to mark as metadata for hyde/jinja2
		print('---', file=fp)
		json.dump(tweet, fp)
		# dump doesn't terminate w/ nl
		print('', file=fp)
		print('---', file=fp)

def act_my_thread(tw, start_tw_id, stop_tw_id):
	tw_id = int(stop_tw_id)

	while True:
		# https://alexwlchan.net/2022/11/tweet-alt-text/
		base_tw = tw.statuses.show(_id=tw_id, tweet_mode='extended', include_ext_alt_text=True)

		write_tweet(base_tw)

		if start_tw_id == tw_id:
			break

		tw_id = base_tw['in_reply_to_status_id']

if __name__ == '__main__':
	args, kwargs = twauth.twit_argskw
	tw = Twitter(*args, **kwargs)

	# https://developer.twitter.com/en/docs/twitter-api/v1/tweets/post-and-engage/api-reference/get-statuses-show-id
	#print(repr(tw.statuses.show(_id=1556349580932833280)))

	with open('tweets.txt') as fp:
		for i in fp:
			action, tw_id, *extra = i.split()
			tw_id = int(tw_id)

			locals()['act_%s' % action](tw, tw_id, *extra)

	print('done')