def show_submission(submission_id):
firebasedb.init_db_connection()
submission = firebasedb.get_submission(submission_id)
if submission:
timestamps = [
int((elem1 - submission.created_utc) / 60)
for elem1 in submission.timestamp
]
line = plt.plot(timestamps, submission.score)
plt.suptitle(f'{submission.author_name} {submission.title}')
plt.show()
submission = Submission_model.from_reddit_submission(reddit_submission)
# log.info(f"GameOverBot_fetcher fetched submission: {submission}")
try:
firebasedb.add_active_submission(submission)
except Exception as e:
log.error(
f"GameOverBot_fetcher error in add_active_submission: {e}")
if __name__ == '__main__':
# Parsing command line arguments
parse_args()
global log
log = logger_creator.init_logger("GameOverBot_fetcher_logger", is_debug)
log.info("[[[[[ GameOverBot_fetcher v0.9.1 ]]]]]")
# connect db
try:
firebasedb.init_db_connection()
except Exception as e:
log.error(e)
# reddit login
reddit = praw_auth.authenticate()
log.info(
f"[[[[[ GameOverBot_fetcher authenticated as {reddit.user.me()} ]]]]]"
)
# fetch submissions
fetch_submissions(reddit)
def show_submissions():
online = False
if online:
firebasedb.init_db_connection()
submissions = firebasedb.get_frozen_submissions()
# Dictionary for aggregating submissions author_submissions_data by author
# Key - author_name
# Value - list of [timestamp, score] tuples
author_data = defaultdict(list)
# Dictionary for counting author submissions
# Key - author_name
# Value - submission counter
author_submission_counter = defaultdict(int)
if online:
# Aggregate all submission author_submissions_data by author
for submission in submissions:
# Filter out submissions that don't have author_name (deleted submissions)
# Filter out submissions with less than 200 score
if submission.author_name and 100 <= submission.score[-1]:
author_submission_counter[submission.author_name] += 1
# Timestamp normalization
timestamps = [
int((timestamp - submission.created_utc) / 60)
for timestamp in submission.timestamp
]
for i in range(0, len(timestamps)):
# Create [timestamp, score] tuple and append it to the list by the author name key
author_data[submission.author_name].append(
(timestamps[i], submission.score[i]))
if online:
# Serialize object
json.dump(author_data, open("submissions_dump.json", 'w'))
json.dump(author_submission_counter,
open("author_submission_counter_dump.json", 'w'))
else:
# Restore serialized object
author_data = json.load(open("submissions_dump.json", 'r'))
author_submission_counter = json.load(
open("author_submission_counter_dump.json", 'r'))
line1 = None
fig, ax1 = plt.subplots()
ax1.set_xscale('log')
#ax1.set_yscale('log')
# Iterate through author_data keys
for author_name in author_data:
# Filter out authors with less than 10 submissions
if len(author_data[author_name]
) > 51 and author_submission_counter[author_name] >= 40:
author_submissions_data = author_data[author_name]
# Sort author_submissions_data by timestamp
author_submissions_data.sort(key=lambda tup: tup[0])
# Apply Savitzky-Golay filter
# window size 51, polynomial order 3
xhat = scipy.signal.savgol_filter(
list(zip(*author_submissions_data))[0], 51, 3)
#xhat = list(zip(*author_submissions_data))[0]
# window size 51, polynomial order 3
yhat = scipy.signal.savgol_filter(
list(zip(*author_submissions_data))[1], 51, 3)
plt.suptitle(
f'{author_name} - {author_submission_counter[author_name]}')
if author_name == "Eugen0242":
ax1.plot(xhat, yhat, color='tab:blue')
else:
line1 = ax1.plot(xhat, yhat, color='tab:green')
if line1[0]:
line1[0].set_alpha(0.1)
plt.show()
print('the end')
def show_submissions():
firebasedb.init_db_connection()
submissions = firebasedb.get_frozen_submissions()
line1 = None
line2 = None
#plt.xscale('log')
#plt.yscale('log')
fig, ax1 = plt.subplots()
ax1.set_xscale('log')
ax1.set_yscale('log')
#ax1.set_xlim([0, 360])
#ax1.set_ylim([0, 1000])
ax2 = ax1.twinx()
ax2.set_ylabel('ratio')
ax2.set_ylim([0.0, 1.1])
ax2.set_xscale('log')
# aggregate all submission data by a single author
total = []
for submission in submissions:
if submission.author_name and submission.author_name == "Lighthouse-scout":
timestamps = [
int((timestamp - submission.created_utc) / 60)
for timestamp in submission.timestamp
]
for i in range(0, len(timestamps)):
total.append(
(timestamps[i], submission.score[i], submission.ratio[i]))
print(total)
# filter data by Savitzky-Golay filter
x = []
y = []
for t in total:
x.append(t[0])
y.append(t[1])
yhat = scipy.signal.savgol_filter(total[0], 51,
3) # window size 51, polynomial order 3
d = []
for submission in submissions:
author_subs = []
if d[submission.author_name]:
author_subs = d[submission.author_name]
author_subs.append(submission)
for key in d:
submissions = d[key]
for submission in submissions:
if len(submission.score) > 60 and submission.score[-1] > 200 or True:
timestamps = [
int((elem1 - submission.created_utc) / 60)
for elem1 in submission.timestamp
]
if submission.author_name and submission.author_name == "Lighthouse-scout":
line1 = ax1.plot(timestamps,
submission.score,
color='tab:blue')
line2 = ax2.plot(timestamps,
submission.ratio,
color='tab:cyan')
#line1[0].set_alpha(0.1)
#line2[0].set_alpha(0.1)
else:
line1 = ax1.plot(timestamps,
submission.score,
color='tab:green')
line2 = ax2.plot(timestamps, submission.ratio, color='tab:red')
line1[0].set_alpha(0.1)
line2[0].set_alpha(0.1)
plt.suptitle(f'{submission.author_name} {submission.title}')
#print(submission)
plt.show()
print('the end')