fixes, example queue from beginning of time
This commit is contained in:
File diff suppressed because it is too large
Load Diff
+35
-27
@@ -15,6 +15,7 @@ from twapi import TwAPI
|
||||
import talenttweet as tt
|
||||
|
||||
PROGRAM_ARGS = None
|
||||
safe_to_post_tweets = True
|
||||
|
||||
def write_user_timestamp(user_id, file, timestamp = None, error = False):
|
||||
if timestamp is None:
|
||||
@@ -42,15 +43,16 @@ def get_user_tweets(id, since_timestamp=None, limit=None):
|
||||
c.Hide_output = True
|
||||
c.Since = '' if since_timestamp == None else util.timestamp_to_tdate(since_timestamp)
|
||||
|
||||
user_str = f'{util.get_username_local(id)}'
|
||||
user_str = f'@{util.get_username_local(id)}'
|
||||
print(f'Scraping tweets from {user_str} since {"forever ago" if c.Since == "" else c.Since}...')
|
||||
try:
|
||||
twint.run.Search(c)
|
||||
except:
|
||||
print(f'Had trouble getting tweets from {user_str}')
|
||||
traceback.print_exc()
|
||||
|
||||
for twt in tweets:
|
||||
if twt.quote_url != '':
|
||||
if type(twt.quote_url) is str and twt.quote_url != '':
|
||||
qrt_count += 1
|
||||
|
||||
print(f'Scraped {len(tweets)} tweets, {qrt_count} of which are quote tweets.')
|
||||
@@ -89,13 +91,14 @@ async def get_cross_talent_tweets(queue_path):
|
||||
finished_user_timestamps = dict()
|
||||
ttweets_dict = dict()
|
||||
posted_ttweets = set() # TODO: don't add TTweet to ttweets_dict if its id exists in posted_ttweets
|
||||
global safe_to_post_tweets
|
||||
|
||||
# Populate structures with existing data from queue.txt
|
||||
try:
|
||||
with open(queue_path, 'r') as f:
|
||||
finished_user_timestamps = get_finished_user_timestamps(f)
|
||||
print(finished_user_timestamps)
|
||||
|
||||
with open(queue_path, 'r') as f: # reset seek head
|
||||
# Get existing queued TalentTweets
|
||||
for line in f:
|
||||
tokens = line.split()
|
||||
@@ -107,30 +110,27 @@ async def get_cross_talent_tweets(queue_path):
|
||||
except FileNotFoundError:
|
||||
print('queue.txt not found.')
|
||||
|
||||
# Pull tweets from twint
|
||||
# Begin getting tweets from online
|
||||
with open(queue_path, 'w') as f:
|
||||
print('Pulling tweets from online!')
|
||||
try:
|
||||
for talent_id in talent_lists.talents:
|
||||
if talent_id not in finished_user_timestamps or \
|
||||
finished_user_timestamps[talent_id] < datetime.datetime.now().timestamp():
|
||||
try:
|
||||
# tweets = get_user_tweets(talent_id, since_timestamp=1663698621) # shorten test runs
|
||||
tweets = get_user_tweets(talent_id, since_timestamp=finished_user_timestamps.get(talent_id, None))
|
||||
for tweet in tweets:
|
||||
if tweet.id not in ttweets_dict:
|
||||
ttweet = await tt.TalentTweet.create_from_twint_tweet(tweet)
|
||||
if ttweet.is_cross_company():
|
||||
ttweets_dict[ttweet.tweet_id] = ttweet
|
||||
except:
|
||||
print('Error occurred processing tweet data.')
|
||||
print(traceback.format_exc())
|
||||
write_user_timestamp(user_id=talent_id, file=f, error=True)
|
||||
else:
|
||||
write_user_timestamp(user_id=talent_id, file=f)
|
||||
for i, (talent_id, talent_username) in enumerate(talent_lists.talents.items()):
|
||||
print(f'[{i+1}/{len(talent_lists.talents)}]{util.get_username(talent_id)}----------------------------')
|
||||
try:
|
||||
# tweets = get_user_tweets(talent_id, since_timestamp=1663698621) # shorten test runs
|
||||
tweets = get_user_tweets(talent_id, since_timestamp=finished_user_timestamps.get(talent_id, None))
|
||||
for tweet in tweets:
|
||||
if tweet.id not in ttweets_dict:
|
||||
ttweet = await tt.TalentTweet.create_from_twint_tweet(tweet)
|
||||
if ttweet.is_cross_company():
|
||||
ttweets_dict[ttweet.tweet_id] = ttweet
|
||||
except:
|
||||
print('Error occurred processing tweet data.')
|
||||
safe_to_post_tweets = False
|
||||
print(traceback.format_exc())
|
||||
write_user_timestamp(user_id=talent_id, file=f, error=True)
|
||||
else:
|
||||
print(f'Skipping already completed {util.get_username_local(talent_id)}')
|
||||
write_user_timestamp(user_id=talent_id, file=f, timestamp=finished_user_timestamps[talent_id])
|
||||
write_user_timestamp(user_id=talent_id, file=f)
|
||||
f.write('\n')
|
||||
ttweets_dict = dict(sorted(ttweets_dict.items()))
|
||||
for ttweet in ttweets_dict.values():
|
||||
@@ -139,10 +139,11 @@ async def get_cross_talent_tweets(queue_path):
|
||||
print('Unhandled error occurred while pulling tweets.')
|
||||
traceback.print_exc()
|
||||
print('Saving queue.txt and exiting.')
|
||||
exit(1)
|
||||
safe_to_post_tweets = False
|
||||
|
||||
return ttweets_dict
|
||||
|
||||
# Return number of TalentTweets successfully posted
|
||||
async def process_queue(ttweets_dict: dict) -> int:
|
||||
global PROGRAM_ARGS
|
||||
ttweets_posted = 0
|
||||
@@ -177,13 +178,20 @@ async def process_queue(ttweets_dict: dict) -> int:
|
||||
|
||||
return ttweets_posted
|
||||
|
||||
# return True = no problems
|
||||
# return False = issue occurred where we couldn't post all past tweets properly
|
||||
async def run(program_args):
|
||||
global PROGRAM_ARGS
|
||||
global safe_to_post_tweets
|
||||
PROGRAM_ARGS = program_args
|
||||
queue_path = get_queue_path()
|
||||
while True:
|
||||
ttweets_dict = await get_cross_talent_tweets(queue_path)
|
||||
print(f'found {len(ttweets_dict)} cross-company tweets')
|
||||
if await process_queue(ttweets_dict) == 0:
|
||||
print('Posted no new tweets; we\'re caught up!')
|
||||
break
|
||||
if safe_to_post_tweets:
|
||||
if await process_queue(ttweets_dict) == 0:
|
||||
print('Posted no new tweets; we\'re caught up!')
|
||||
return True
|
||||
else:
|
||||
print('Tweets were not retrieved cleanly.')
|
||||
return False
|
||||
+3
-2
@@ -50,8 +50,9 @@ async def async_main():
|
||||
await listen.run()
|
||||
case 'c' | 'catchup':
|
||||
print('RUNNING IN CATCH-UP MODE\n')
|
||||
await catchup.run(PROGRAM_ARGS)
|
||||
await listen.run()
|
||||
if await catchup.run(PROGRAM_ARGS):
|
||||
print('CATCH-UP MODE DONE, GOING INTO LISTEN MODE')
|
||||
await listen.run()
|
||||
case 'd' | 'delete-all':
|
||||
print('WARNING: SELF-DESTRUCT MODE')
|
||||
await self_destruct()
|
||||
|
||||
@@ -16,7 +16,6 @@ def __create_dict(file, _dict):
|
||||
words = line.split()
|
||||
if len(words) == 2 and line[0] != '#':
|
||||
name, id = line.split()
|
||||
talents[int(id)] = name
|
||||
name = util.get_username_online(id, default=name) # attempt to get updated name
|
||||
talents[int(id)] = name
|
||||
_dict[int(id)] = name
|
||||
|
||||
@@ -0,0 +1 @@
|
||||
# TODO: move queue structures and file handling here
|
||||
Reference in New Issue
Block a user