def filter_gram(pv_word_gram):
scraper.pprint("--filter gram")
return_list = []
sorted_gram = sorted(pv_word_gram, key = lambda i:i["count"], reverse=True)
r_count = 0
for i in sorted_gram:
if r_count < 5:
if not i["word"] in config.stop_words:
return_list.append(i)
r_count+=1
if r_count ==config.term_count:
break
return return_list
def strip_punctuation(pv_list):
scraper.pprint("--strip punctuation")
clean_words = []
for item in pv_list:
item = item.strip(',')
item = item.strip('.')
item = item.strip('?')
item = item.strip('!')
item = item.strip('@')
item = item.strip('#')
item = item.replace('\n','')
item = item.replace('\r','')
item = item.rstrip()
clean_words.append(item)
return clean_words
def read_distros(pv_list):
scraper.pprint("--read distros")
scraper.pprint("== DISTRIBUTIONS == [TOP 5 WORDS]")
for i in pv_list:
tmp_string = ' SHOP {}, ID {} = ({}:{}, {}:{}, {}:{}, {}:{}, {}:{})'.format(i[0]['shop_name'],i[0]['shop_id'], i[1]['word'], i[1]['count'],i[2]['word'], i[2]['count'], i[3]['word'], i[3]['count'], i[4]['word'], i[4]['count'], i[5]['word'], i[5]['count'])
scraper.pprint(tmp_string)
scraper.pprint(' ')
return pv_list
def save(distributions, path):
scraper.pprint("--save")
s_t = '{}'.format(time.time())
time_split = s_t.split('.')
e = time_split[0]
lcl_path = path + '_{}.csv'.format(e)
scraper.pprint('make {}'.format(lcl_path))
with open(lcl_path, "w", newline="") as csvfile:
writer = csv.writer(csvfile)
writer.writerow(["id", "shop_id", "distribution", "e_time"])
lcl_id = 1
for d in distributions:
count_f = {}
count_f['count'] = lcl_id
d.append(str(time.time()))
d.insert(0, count_f)
writer.writerow(d)
lcl_id += 1
def get_recent_run():
#find the most timestamp of shops added to the archive
scraper.pprint('--get recent run')
files = os.listdir("./data")
dates = []
for f in files:
sub_f = f[:-15]
if sub_f == "shops":
lcl_t = f[6:]
lcl_t = lcl_t[:-4]
dates.append(lcl_t)
dates = sorted(dates, reverse=True)
try:
date = dates[0]
return date
except IndexError:
scraper.pprint('Data Folder is missing shop lists')
return 0
def get_shops(pv_timestamp, key):
#get the shop list from ./data/shops_'timestamp'
scraper.pprint("--get shops, pv_timestamp: {}".format(pv_timestamp))
lcl_path = "./data/shops_{}.csv".format(pv_timestamp)
shops = []
try:
with open(lcl_path, newline='') as csvfile:
reader = csv.reader(csvfile, delimiter=',', quotechar='|')
count = 0
for row in reader:
shop = {}
if count > 0:
shop['count'] = row[0]
shop['id'] = row[1]
shop['name'] = row[2]
shop['listings'] = row[3]
shop['key'] = key
shops.append(shop)
count += 1
except FileNotFoundError:
scraper.pprint(
"FILE NOT FOUND ERROR!! Did you provide an incorrect timestamp?")
scraper.pprint(" ")
shop = {}
shop['count'] = 0
shop['id'] = 0
shop['name'] = "Error, FILE NOT FOUND"
shop['listings'] = "0"
shop['key'] = "KEY NOT SAVED"
shops.append(shop)
return shops
def save(distributions, save_path, pv_timestamp):
scraper.pprint("--save")
s_t = '{}'.format(time.time())
time_split = s_t.split('.')
e = time_split[0]
lcl_path = save_path+'_{}${}.csv'.format(e,pv_timestamp)
scraper.pprint('make {}'.format(lcl_path))
with open(lcl_path, "w", newline="") as csvfile:
writer = csv.writer(csvfile)
writer.writerow(
[
"id"
,"distribution"
,"e_time"
]
)
lcl_id = 1
for d in distributions:
lcl_list = []
lcl_list.append(lcl_id)
lcl_list.append(d)
lcl_list.append(str(time.time()))
writer.writerow(lcl_list)
lcl_id+=1
def word_counter(shop):
print("shop id: {}, name: {}, listings: {}".format(shop['id'],shop['name'], shop['listings']))
shop_id = shop['id']
key = shop['key']
#returns a distribution chart of the 5 most common terms related to one shop
scraper.pprint("--word counter, shop_id: {}".format(shop_id))
#gather data
url = "https://openapi.etsy.com/v2/shops/{}/listings/active?limit=25&offset=0&api_key={}".format(shop_id, key)
headers = {'user-agent': 'my-app/0.0.1'}
r = requests.get(url, headers=headers)
r_status=r.status_code
scraper.pprint("API RESPONSE: {}".format(r_status))
return_list = []
if r_status == 200:
content = r.content
d_content = content.decode("utf-8")
content_json = json.loads(d_content)
result_string = ''
#create string
for result in content_json['results']:
result_string+=str(result['title'])
result_string+=str(result['description'])
result_list = result_string.lower().split(' ')
clean_words = strip_punctuation(result_list)
word_set = get_unique_words(clean_words)
return_frame = {}
return_frame["shop_id"] = shop_id
return_frame["shop_name"] = shop["name"]
return_list.append(return_frame)
for w in filter_gram(count_words(word_set, clean_words)):
return_list.append(w)
scraper.pprint(" Top Terms: {}".format(return_list))
else:
return_dict = {}
return_dict["word"] = 'Error Code Status {}'.format(r_status)
return_dict["count"] = 1
return_list.append(return_dict)
return return_list
def main():
scraper.pprint("--'main, analyzer.py'")
key = scraper.get_key()
distributions = []
save_path = './data/distribution'
arg_length = len(sys.argv)
if not arg_length == 2:
timestamp = get_recent_run()
if arg_length == 2:
timestamp = sys.argv[1]
scraper.pprint('timestamp: {}'.format(timestamp))
if not timestamp == 0:
shops = get_shops(timestamp, key)
lcl_distributions = threaded_counter(shops)
for lcl in lcl_distributions:
distributions.append(lcl)
save(read(distributions), save_path)
else:
scraper.pprint("Please run 'scraper.py' first.")
def word_counter(shop):
print("shop: {}".format(shop))
shop_id = shop['id']
key = shop['key']
#returns a distribution chart of the 5 most common terms related to one shop
scraper.pprint("--word counter, shop_id: {}".format(shop_id))
#gather data
url = "https://openapi.etsy.com/v2/shops/{}/listings/active?limit=25&offset=0&api_key={}".format(
shop_id, key)
headers = {'user-agent': 'my-app/0.0.1'}
r = requests.get(url, headers=headers)
r_status = r.status_code
scraper.pprint("API RESPONSE: {}".format(r_status))
if r_status == 200:
content = r.content
d_content = content.decode("utf-8")
content_json = json.loads(d_content)
result_string = ''
#create string
for result in content_json['results']:
result_string += str(result['title'])
result_string += str(result['description'])
result_list = result_string.lower().split(' ')
word_set = set()
clean_words = []
#process data
for item in result_list:
item = item.strip(',')
item = item.strip('.')
item = item.strip('\n')
word_set.add(item)
clean_words.append(item)
return_list = []
return_frame = {}
return_frame["shop_id"] = shop_id
return_list.append(return_frame)
word_gram = []
#use word set to build set of unique words
for w_set in word_set:
tmp_word = w_set
count = 0
#count the occurances of the clean words
for w_word in clean_words:
if w_set == w_word:
count += 1
word_frame = {}
word_frame['word'] = tmp_word
word_frame['count'] = count
word_gram.append(word_frame)
sorted_gram = sorted(word_gram, key=lambda i: i["count"], reverse=True)
r_count = 0
for i in sorted_gram:
if r_count < 5:
if not i["word"] == '':
if not i["word"] == '–':
return_list.append(i)
r_count += 1
if r_count == config.term_count:
break
scraper.pprint(" Top Terms: {}".format(return_list))
else:
return_dict = {}
return_dict["word"] = 'Error Code Status {}'.format(r_status)
return_dict["count"] = 1
return_list.append(return_dict)
return return_list
def threaded_counter(shops):
scraper.pprint('--threader')
pool = ThreadPool(4)
results = pool.map(word_counter, shops)
return results
def read(pv_list):
scraper.pprint("--read")
for item in pv_list:
scraper.pprint(item)
return pv_list
