def ml_scraper(
USER_ID=None,
PASSCODE=None,
tag_hashes=None,
bucket_ids=None,
pages=None,
mode=None,
targeting=None,
):
if targeting == "bucket":
tag_hashes = sharechat_helper.get_tag_hashes(USER_ID, PASSCODE,
bucket_ids)
delay = uniform(10, 15)
elif targeting == "tag":
delay = uniform(30, 35)
if mode == "archive":
print("Scraping in archive mode")
start_time = time.time()
print("Initializing ...")
initializationSuccess = False
try:
coll = sharechat_helper.ml_initialize_mongo()
aws, bucket, s3 = sharechat_helper.ml_initialize_s3()
initializationSuccess = True
print("Initialized successfully")
except Exception:
print("Initialization failure")
print(logging.traceback.format_exc())
# Scrape data from tags
if initializationSuccess:
print("Scraping in progress ...")
sharechat_df = sharechat_helper.get_trending_data(
USER_ID, PASSCODE, tag_hashes, pages, delay)
if len(sharechat_df) < 1:
raise ValueError(
"Returned empty dataframe. No posts were scraped.")
else:
# Save data locally
sharechat_df.to_pickle("sharechat_df.pkl")
s3UploadSuccess = False
# Save data to S3 & Mongo DB
try:
print("S3 upload in progress ... ")
sharechat_df = sharechat_helper.ml_sharechat_s3_upload(
sharechat_df, aws, bucket, s3)
s3UploadSuccess = True
print("Data uploaded to S3")
except Exception:
print("S3 upload failed")
print(logging.traceback.format_exc())
pass
if s3UploadSuccess:
try:
print("HTML preview file creation in progress ...")
(
sharechat_df,
sharechat_df_html,
) = sharechat_helper.get_thumbnails_from_s3(
sharechat_df)
with open("sharechat_ml_data_preview.html", "w") as f:
f.write(sharechat_df_html.data)
print("HTML preview file created")
except Exception:
print("HTML preview file creation failed")
print(logging.traceback.format_exc())
pass
try:
print("MongoDB upload in progress ...")
sharechat_helper.sharechat_mongo_upload(
sharechat_df, coll)
print("Data uploaded to MongoDB")
except Exception:
print("MongoDB upload failed")
print(logging.traceback.format_exc())
pass
else:
pass
try:
print("CSV file creation in progress ... ")
sharechat_df.to_csv("sharechat_ml_data.csv")
print("CSV file created")
print("{} posts scraped".format(len(sharechat_df)))
except Exception:
print("CSV file creation failed")
print(logging.traceback.format_exc())
pass
print("Scraping complete")
print("Time taken: %s seconds" % (time.time() - start_time))
return sharechat_df
elif mode == "local":
print("Scraping in local mode")
start_time = time.time()
print("Scraping in progress ...")
sharechat_df = sharechat_helper.get_trending_data(
USER_ID, PASSCODE, tag_hashes, pages, delay)
if len(sharechat_df) < 1:
raise ValueError(
"Returned empty dataframe. No posts were scraped.")
else:
# Save data locally
sharechat_df.to_pickle("sharechat_df.pkl")
try:
print("HTML preview file creation in progress ...")
(
sharechat_df,
sharechat_df_html,
) = sharechat_helper.get_thumbnails_from_sharechat(sharechat_df)
with open("sharechat_ml_data_preview.html", "w") as f:
f.write(sharechat_df_html.data)
print("HTML preview file created")
except Exception:
print("HTML preview file creation failed")
print(logging.traceback.format_exc())
pass
try:
print("CSV file creation in progress ... ")
sharechat_df.to_csv("sharechat_ml_data.csv")
print("CSV file created")
print("{} posts scraped".format(len(sharechat_df)))
except Exception:
print("CSV file creation failed")
print(logging.traceback.format_exc())
pass
print("Scraping complete")
print("Time taken: %s seconds" % (time.time() - start_time))
return sharechat_df
def fresh_content_scraper(
USER_ID=None,
PASSCODE=None,
tag_hashes=None,
bucket_ids=None,
pages=None,
unix_timestamp=None,
mode=None,
targeting=None,
):
if targeting == "bucket":
tag_hashes = sharechat_helper.get_tag_hashes(USER_ID, PASSCODE,
bucket_ids)
delay = uniform(10, 15)
elif targeting == "tag":
delay = uniform(30, 35)
if mode == "archive":
print("Scraping in archive mode")
start_time = time.time()
# Initialize S3 and Mongo DB
print("Initializing ...")
initializationSuccess = False
try:
aws, bucket, s3 = s3_mongo_helper.initialize_s3()
coll = s3_mongo_helper.initialize_mongo()
initializationSuccess = True
print("Initialized successfully")
except Exception:
print("Initialization failure")
print(logging.traceback.format_exc())
# Scrape data from Sharechat tags
if initializationSuccess:
print("Scraping in progress ...")
sharechat_df = sharechat_helper.get_fresh_data(
USER_ID, PASSCODE, tag_hashes, pages, unix_timestamp, delay)
if len(sharechat_df) < 1:
raise ValueError(
"Returned empty dataframe. No posts were scraped.")
else:
# Save data to S3 & Mongo DB
s3UploadSuccess = False
try:
print("S3 upload in progress ...")
sharechat_df, tagwise_duplicates = sharechat_helper.sharechat_s3_upload(
sharechat_df, aws, bucket, s3,
coll) # the returned df includes s3 urls
s3UploadSuccess = True
print("Data uploaded to S3")
except Exception:
print("S3 upload failed")
print(logging.traceback.format_exc())
pass
if s3UploadSuccess:
aws, logbucket, s3 = sharechat_helper.initialize_s3_logbucket()
today = datetime.utcnow().strftime("%Y%m%d")
try:
print("HTML file creation in progress ...")
(
sharechat_df,
sharechat_df_html,
) = sharechat_helper.get_thumbnails_from_s3(sharechat_df)
with open("sharechat_fresh_data_preview.html", "w") as f:
f.write(sharechat_df_html.data)
print("HTML file created")
print("Uploading HTML file to S3 ...")
sharechat_helper.upload_logs(
s3=s3,
filename="sharechat_fresh_data_preview.html",
key="fresh_preview_" + today,
bucket=logbucket,
)
print("HTML file uploaded")
except Exception:
print("HTML file upload failed")
print(logging.traceback.format_exc())
pass
try:
print("Duplicates log creation in progress ...")
with open("tagwise_duplicates.json", "w") as fp:
json.dump(tagwise_duplicates, fp)
print("Duplicates log created")
print("Uploading duplicates log to S3 ...")
sharechat_helper.upload_logs(
s3=s3,
filename="tagwise_duplicates.json",
key="fresh_duplicates_" + today,
bucket=logbucket,
)
print("Duplicates log uploaded")
except Exception:
print("Duplicates log upload failed")
print(logging.traceback.format_exc())
pass
try:
print("CSV file creation in progress ... ")
sharechat_df.to_csv("sharechat_fresh_data.csv")
print("CSV file created")
print("Uploading CSV file to S3 ...")
sharechat_helper.upload_logs(
s3=s3,
filename="sharechat_fresh_data.csv",
key="fresh_posts_" + today,
bucket=logbucket,
)
print("CSV file uploaded")
except Exception:
print("CSV file upload failed")
print(logging.traceback.format_exc())
pass
try:
print("MongoDB upload in progress ...")
sharechat_helper.sharechat_mongo_upload(sharechat_df, coll)
print("Data uploaded to MongoDB")
print("{} posts saved".format(len(sharechat_df)))
except Exception:
print("MongoDB upload failed")
print(logging.traceback.format_exc())
pass
else:
pass
print("Scraping complete")
print("Time taken: %s seconds" % (time.time() - start_time))
return sharechat_df
elif mode == "local":
print("Scraping in local mode")
start_time = time.time()
print("Scraping in progress ...")
sharechat_df = sharechat_helper.get_fresh_data(USER_ID, PASSCODE,
tag_hashes, pages,
unix_timestamp, delay)
if len(sharechat_df) < 1:
raise ValueError(
"Returned empty dataframe. No posts were scraped.")
else:
# Save data locally
sharechat_df.to_pickle("sharechat_df.pkl")
try:
print("HTML preview file creation in progress ...")
(
sharechat_df,
sharechat_df_html,
) = sharechat_helper.get_thumbnails_from_sharechat(sharechat_df)
with open("sharechat_fresh_data_preview.html", "w") as f:
f.write(sharechat_df_html.data)
print("HTML preview file created")
except Exception:
print("HTML preview file creation failed")
print(logging.traceback.format_exc())
pass
try:
print("CSV file creation in progress ... ")
sharechat_df.to_csv("sharechat_fresh_data.csv")
print("CSV file created")
print("{} posts saved".format(len(sharechat_df)))
except Exception:
print("CSV file creation failed")
print(logging.traceback.format_exc())
pass
print("Scraping complete")
print("Time taken: %s seconds" % (time.time() - start_time))
return sharechat_df
