def main_loop():
# try to learn every hour
if time.time() - main_loop.last_learned > 60*60:
logger.info("Learning from today's top posts")
grow_from_top(timeperiod="day", limit=50)
main_loop.last_learned = time.time()
# Start writing comments
logger.debug("Finding comments to reply to")
comments_to_insert = []
for c in r.get_all_comments(limit=200):
# Keep from processing the same thing twice
if db.last_seen(c) is not None:
continue
# learn it
comments_to_insert.append(c)
# Don't reply to yourself
if c.author.name == config.username:
continue
response = scorer.get_best_response(c)
if response is None:
continue
logger.info("Responding to a user")
try:
c.reply(rewriter.prepare_for_post(response, c))
except HTTPError as err:
if err.response.status_code == 403:
logger.warning("HTTP 403. This probably means you were banned "
"from a subreddit.")
else:
logger.exception("Error responding to user")
db.insert_comments(*comments_to_insert, fast=True)
def main_loop():
# try to learn every hour
if time.time() - main_loop.last_learned > 60*60:
print("Learning from today's top posts")
grow_from_top(timeperiod="day", limit=50)
main_loop.last_learned = time.time()
# Start writing comments
print("Fetching more comments")
comments_to_insert = []
for c in r.get_all_comments(limit=200):
# Keep from processing the same thing twice
if db.last_seen(c) is not None:
continue
# learn it
comments_to_insert.append(c)
# Don't reply to yourself
if c.author.name == config.username:
continue
response = get_best_response(c)
if response is None:
continue
print("Someone said:\n %s\nSo I should say:\n %s" %
(c.body, response.body))
try:
c.reply(rewriter.prepare_for_post(response, c))
except:
traceback.print_exc()
db.insert_comments(*comments_to_insert, fast=True)