def get_thumbnails_from_s3(df):
def path_to_image_html(path):
return '<img src="' + path + '"width="200" >'
thumbnail = []
aws, bucket, s3 = s3_mongo_helper.initialize_s3()
temp_dir = tempfile.mkdtemp(dir=os.getcwd())
for link in df["s3_url"]:
if link is not None:
if link.split(".")[-1] == "mp4":
video_input_path = link
img_output_path = (temp_dir.split("/")[-1] + "/" +
link.split("/")[-1].split(".")[0] + ".jpg")
filename = link.split("/")[-1].split(".")[0] + ".jpg"
subprocess.call(
[
"ffmpeg",
"-i",
video_input_path,
"-ss",
"00:00:00.000",
"-vframes",
"1",
img_output_path,
],
stderr=subprocess.DEVNULL,
stdout=subprocess.DEVNULL,
)
s3_mongo_helper.upload_to_s3(
s3=s3,
file=img_output_path,
filename=filename,
bucket=bucket,
content_type="image/jpeg",
)
thumbnail.append(aws + bucket + "/" + filename)
elif link.split(".")[-1] == "txt":
thumbnail.append(None)
else: # if jpg/jpeg/png
thumbnail.append(link)
else: # if NaN
thumbnail.append(None)
df["thumbnail"] = np.array(thumbnail)
pd.set_option("display.max_colwidth", -1)
df_html = HTML(
df.to_html(
index=False,
escape=False,
formatters=dict(thumbnail=path_to_image_html),
render_links=True,
))
shutil.rmtree(temp_dir)
return df, df_html
def save_to_server(all_msgs: List[Msg], merged_msgs: List[Msg],
media_files: list, drive_id: str) -> None:
"""
Save msgs and media to the Tattle server.
This requires setting environment variables
"""
# 0. Initialize
all_coll, merged_coll = initialize_mongo()
aws, bucket, s3 = initialize_s3()
# 1. Insert all "raw" msgs
all_coll = initialize_mongo(var_prefix="whatsapp_all")
msgs_by_file = group_by_file(all_msgs)
to_insert = []
insert_dt = datetime.datetime.utcnow().isoformat()
for msgs in msgs_by_file.values():
to_insert.append({
'scrape_datetime': insert_dt,
'source': GOOGLE_DRIVE,
'source_loc': drive_id,
'msgs': [m.as_dict() for m in msgs]
})
all_coll.insert_many(to_insert)
# 2. Upsert merged msgs
merged_coll = initialize_mongo(var_prefix="whatsapp_merged")
msg_gids = [msg.group_id for msg in merged_msgs]
existing_msgs = merged_coll.find({"group_id": {"$in": msg_gids}})
if existing_msgs:
logging.warning("Not overwriting %d msgs already in server.",
len(existing_msgs))
merge_msgs_from_server(merged_msgs, existing_msgs)
merged_coll.insert_many([m.as_dict() for m in merged_msgs])
# 3. Upload media files to s3
for fl in media_files:
logging.info("Uploading %r", fl['hash'])
upload_to_s3(s3, fl['content'], fl['hash'], bucket, fl['media_mime_type'])
logging.info("Wrote %d files to S3. Done", len(media_files))
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
def trending_content_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()
# 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 as e:
print("Initialization failure")
print(logging.traceback.format_exc())
# Scrape data from Sharechat 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(
"get_data() returned empty dataframe. No posts were scraped."
)
else:
# Save data locally
sharechat_df.to_pickle("sharechat_df.pkl")
# Save data to S3 & Mongo DB
s3UploadSuccess = False
try:
print("S3 upload in progress ...")
sharechat_df = sharechat_helper.sharechat_s3_upload(
sharechat_df, aws, bucket,
s3) # the returned df includes s3 urls
s3UploadSuccess = True
print("Data uploaded to S3")
except Exception as e:
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_trending_data_preview.html",
"w") as f:
f.write(sharechat_df_html.data)
print("HTML preview file created")
except Exception as e:
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 as e:
print("MongoDB upload failed")
print(logging.traceback.format_exc())
pass
else:
pass
try:
print("CSV file creation in progress ... ")
sharechat_df.to_csv("sharechat_trending_data.csv")
print("CSV file created")
print("{} posts scraped".format(len(sharechat_df)))
except Exception as e:
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(
"get_data() 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_trending_data_preview.html", "w") as f:
f.write(sharechat_df_html.data)
print("HTML preview file created")
except Exception as e:
print("HTML preview file creation failed")
print(logging.traceback.format_exc())
pass
try:
print("CSV file creation in progress ... ")
sharechat_df.to_csv("sharechat_trending_data.csv")
print("CSV file created")
print("{} posts scraped".format(len(sharechat_df)))
except Exception as e:
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
