def get_seller_data(self, url):
# Caller :: get_artwork_data_slave and get_seller_id
visited.discard(url)
soup = TheMiner.fetch_page(url)
if soup is not None:
# Seller's Name
print(seller_name)
# Code will break if seller's name is not found
# Location
try:
print(location)
except AttributeError:
location = None
# Website
try:
print(website)
except AttributeError:
website = None
except TypeError:
website = None
bundle = [url, self.website.platform, seller_name, location, website]
print(bundle)
TheAuthour.write_seller(*bundle)
def get_seller_data(self, url):
# Caller :: get_artwork_data_slave
visited.discard(url)
soup = TheMiner.fetch_page(url)
# print("A")
if soup is not None:
# print("B")
A = soup.find('div', id='top-seller')
seller_name = A.h1.text.strip()
# print(seller_name)
# Code will break if seller's name is not found
try:
location = A.find('p',
class_="subtitle").text.strip().split(',')
location = location[-1].strip()
# print(location)
except AttributeError:
location = None
try:
website = str(soup.find('ul', id="websites").a['href']).strip()
# print(website)
except AttributeError:
website = None
except TypeError:
website = None
bundle = [
url, self.website.platform, seller_name, location, website
]
# print(bundle)
TheAuthour.write_seller(*bundle)
def get_seller_id(self, seller_url) -> int:
# Fetches seller_data, writes it in db, and returns seller_id.
# bundle = [seller_url, self.website.platform, 'KAZoART', None, url]
seller_id = None
if seller_url is not None:
if seller_url in SELLER_INFO.keys():
seller_id = SELLER_INFO.get(seller_url)
# print(seller_id)
else:
# If code reaches here then the entry for seller doesn't already exists. Let's call get_seller_data
# again with seller_url
self.get_seller_data(seller_url)
# wait for a second to make sure that transaction is smooth. Activate this line if errors are thrown.
# time.sleep(1)
# Try to fetch seller data again.
if seller_url in SELLER_INFO.keys():
seller_id = SELLER_INFO.get(seller_url)
else:
# Make a Kazoart style bundle, and write it to obtain a seller_id.
bundle = [seller_url, self.website.platform, 'BAREBONES', None, seller_url]
# Writing to db.
TheAuthour.write_seller(*bundle)
# This should generate the seller_id we so desperately desire.
# time.sleep(1)
if seller_url in SELLER_INFO.keys():
seller_id = SELLER_INFO.get(seller_url)
else:
print("FATAL ERROR :: Seller_id not found.")
else:
print("FATAL ERROR :: Seller_id not found.")
# Let's return seller_id, even if it's None. This will stop the get_artwork_Data_slave from gathering
# data beyond rule 3 check .
return seller_id
def get_seller_id(self, seller_url) -> int:
# Fetches seller_data, writes it in db, and returns seller_id.
# bundle = [seller_url, self.website.platform, 'KAZoART', None, url]
seller_id = None
if seller_url is not None:
if seller_url in SELLER_INFO.keys():
seller_id = SELLER_INFO.get(seller_url)
# print(seller_id)
else:
# Process and create the bundle here.
bundle = [
seller_url, self.website.platform, 'KAZoART', None,
seller_url
]
# Writing to db.
TheAuthour.write_seller(*bundle)
if seller_url in SELLER_INFO.keys():
seller_id = SELLER_INFO.get(seller_url)
else:
print("FATAL ERROR :: Seller_id not found.")
else:
print("FATAL ERROR :: Seller_id not found.")
# Let's return seller_id, even if it's None.
return seller_id
def get_seller_id(self, seller_url) -> int:
# Fetches seller_data, writes it in db, and returns seller_id.
# bundle = [seller_url, self.website.platform, 'KAZoART', None, url]
# print("GET SELLER ID")
seller_id = None
if seller_url is not None:
if seller_url in SELLER_INFO.keys():
seller_id = SELLER_INFO.get(seller_url)
return seller_id
# print(seller_id)
else:
# If code reaches here then the entry for seller doesn't already exists. Let's call get_seller_data
# with seller_url
self.get_seller_data(seller_url)
# Try to fetch seller data again.
if seller_url in SELLER_INFO.keys():
seller_id = SELLER_INFO.get(seller_url)
# If it is not a url, get_seller_data will fail to make an entry.In that case we move to the next part.
else:
# Make a Kazoart style bundle, and write it to obtain a seller_id.
# bundle = [seller_url, platform, Seller's name, location, website]
bundle = [
seller_url, self.website.platform, seller_url, None,
seller_url
]
# Writing to db.
TheAuthour.write_seller(*bundle)
# This should generate the seller_id we so desperately desire.
# time.sleep(1)
if seller_url in SELLER_INFO.keys():
# This will always run, unless the program is failing unexpectedly.
seller_id = SELLER_INFO.get(seller_url)
else:
print("FATAL ERROR :: Seller_id not found.")
else:
print("FATAL ERROR :: Seller_id not found.")
# Let's return seller_id, even if it's None. This will stop the get_artwork_Data_slave from gathering
# data beyond rule 3 check .
return seller_id
def get_artwork_data_slave(self, url):
soup = TheMiner.fetch_page(url)
if soup is not None:
# Field initiation :: Artwork_title, artist, price, seller_id ,
# medium, type, dimension, frame, authenticity, about :: year, support, signature
# artist_id, Image_loc = None
seller_id = None
artist = None
artwork = None
price = None
medium = None # (painting or sculpture)
technique = "" # Material and style
type_ = None
dimensions = None
frame = None
authenticity = None
about = None
artist_id = None
image_loc = None
year = None
support = None
signature = None
try:
# PRICE
A = soup.find('section', id='informations')
price = str(A.find('p',
class_='media-price price').text).strip()
number = ''
for p in price:
if p == '-':
break
if p.isdigit():
number += str(p)
if p == ".":
number += str(p)
price = float(number)
# print(price)
except AttributeError:
pass
except TypeError:
pass
# Rule : 5
if price is not None:
# Seller_id
try:
seller_url = soup.find(
'div', id='top-seller').find('a').get('href')
if 'galeries-d-art' in str(seller_url):
seller_url = re.sub('galeries-d-art', 'art-galleries',
seller_url)
# If seller_url is found.
seller_id = self.get_seller_id(seller_url)
except AttributeError:
# seller_id = None
# There are pages where the seller has no other page. Then we make the url ourselves.
seller_url = soup.find('div', id='top-seller').find(
'p', class_='highlight-title').text
seller_url = str(seller_url).strip()
if seller_url in SELLER_INFO:
seller_id = SELLER_INFO[seller_url]
else:
location = soup.find('div', id='top-seller').find(
'p', class_='subtitle').text.strip().split(',')
location = str(location[-1]).strip()
seller_name = seller_url
bundle = [
seller_url, self.website.platform, seller_name,
location, None
]
# We write the seller info directly and fetch the seller_id
TheAuthour.write_seller(*bundle)
seller_id = SELLER_INFO[seller_url]
# Artist_id
try:
artist_url = soup.find('section', id='informations').find(
'div', class_='relative').a.get('href')
if "oeuvres-d-art-contemporain" in artist_url:
re.sub("oeuvres-d-art-contemporain",
"contemporary-artworks", artist_url)
artist_id = self.get_artist_id(artist_url)
except AttributeError:
artist_id = None
print("\n\n\n\n\n")
print(url)
print("\n\n\n\n\n")
time.sleep(50)
# Medium must always have "Painting" or "Sculpture" (RULE :: 2)
la = str(url).split('/')
if 'painting' in la:
medium = "Painting " # (painting or sculpture)
elif 'sculpture' in la:
medium = "Sculpture"
else:
# So that url leaks don't break the code.
medium = None
# IF either the seller id or artist_id are missing, escape the rest. (Rule : 3)
# If medium is neither Paintings, not Sculptures. We don't fetch data. ( Rule : 2)
if seller_id is not None and artist_id is not None and medium is not None:
# ______________________________MAIN DATA FETCHING________________________
A = soup.find('section', id='informations')
B = A.find('div', class_='relative')
# ARTIST'S NAME
artist = B.find('span',
class_='primary-title').text.strip()
# print(artist)
# ARTWORK'S NAME
C = B.find('span', class_='secondary-title').text.strip()
artwork_ = C.split(',')
artwork_title = ""
for a in range(len(artwork_) - 1):
if a == 0:
artwork_title = artwork_[a]
continue
artwork_title = artwork_title + ", " + artwork_[
a].strip()
artwork = artwork_title
# print(artwork)
try:
# ARTWORK YEAR
year = C.split(',')[-1].strip()
# print(year)
except IndexError:
pass
# year = None
try:
# Image url
B = A.find('div', id='img-container')
image_loc = B.find('img',
id='img_original')['data-src']
# print(image_loc)
except AttributeError:
pass
# Contains:: image, dimensions, medium, type, Frame, Support, authenticity, signature
try:
D = soup.find('div', id='tabs-description').ul
E = D.find_all('li')
for e in E:
a = e.text
# Dimensions
if 'Dimensions' in a and 'About the artwork' not in a and 'Support' not in a:
dimensions = e.find(
'p',
class_='pull-right').strong.text.strip()
dim = True
# print(dimensions)
continue
# Medium (Sculpture/Painting) and Technique
if 'Medium' in a and 'About the artwork' not in a:
technique = e.find(
'p', class_='pull-right').text.split(" ")
# print(technique)
temp = ""
for t in technique:
if t != "":
temp += t.strip()
temp += " "
# medium = medium[0]
# technique = medium[1]
technique = temp
# print(technique)
continue
# Type
if 'Type' in a and 'About the artwork' not in a:
type_ = e.find('p',
class_='pull-right text-right'
).text.strip().split(' ')[0]
# print(type_)
continue
# Support (base)
if 'Support' in a and 'About the artwork' not in a:
try:
f = e.find('p',
class_='pull-right text-right'
).text.strip().split(' ')
support = f[0] + '. ' + f[1].strip('\n')
f = e.find(
'p', class_='pull-right text-right'
).strong.text.strip().strip('\n')
support += f
except IndexError:
support = e.find(
'p', class_='pull-right text-right'
).text.strip()
# print(support)
continue
# Framing
if 'Framing' in a and 'About the artwork' not in a:
frame = e.find(
'p', class_='pull-right').text.strip()
# print(frame)
continue
# Signature
if 'Signature' in a and 'About the artwork' not in a:
signature = e.find(
'p', class_='pull-right').text.strip()
# print(signature)
continue
# Authenticity
if 'Authenticity' in a and 'About the artwork' not in a:
authenticity = e.find(
'p',
class_='pull-right text-right').text.strip(
)
# print(authenticity)
continue
# Artwork Description
if 'About the artwork' in a:
about = e.find('p', class_="marg-bot-10")
if about is not None:
a = e.find(
'div',
class_=
"description-catalog see-more text-justify"
).text.strip()
about = about.text.strip()
about += a
else:
about = e.find('p', class_='').text.strip()
continue
# print(about)
except AttributeError:
pass
# self, artwork_title=None, artist_name=None, year=None, price=None, Dimensions=None, Medium=None,
# Type=None, Support=None, Frame=None, Signature=None, Authenticity=None, About=None,
# platform=None, image_addr=None, seller_id=None, artist_id=None)
artwork_bundle = {
"artwork_title": artwork,
"artist_name": artist,
"year": year,
"price": price,
"Medium": medium,
"Type": type_,
"Dimensions": dimensions,
"Support": support,
"Frame": frame,
"Signature": signature,
"Authenticity": authenticity,
"About": about,
"platform": self.website.platform,
"image_addr": image_loc,
"seller_id": seller_id,
"artist_id": artist_id,
"url": url,
"technique": technique
}
# print(artwork_bundle)
TheAuthour.write_artwork_price_image(**artwork_bundle)
else:
print(
f"SELLER ID :: {seller_id},\nARTIST ID :: {artist_id}")
else:
# If the price is not available, we skip the entire process.
print(f"PRICE NOT FOUND : {price} at {url}")
else:
print(f"\n\n\n\n\nURL DIDN'T RETURN : {url}\n\n\n\n\n")
def get_seller_data(self, url):
# Caller :: get_artwork_data_slave and get_seller_id
# We get to here only after we do not find the seller's info in SELLER_INFO
# print("GET SELLER DATA")
visited.discard(url)
soup = TheMiner.fetch_page(url)
if soup is not None:
# print("GET SELLER DATA: SOUP RETURNED")
seller_name = None
try:
# Seller's Name
seller_box = soup.find('div', id='jumpto--PartnerHeader')
seller_name = seller_box.h1.text.strip()
# print(seller_name)
# Code will break if seller's name is not found
except AttributeError:
pass
if seller_name is not None:
# print(f"SELLER NAME : {seller_name}")
# Location
try:
# Location is not available here.
location = ""
locatio = seller_box.h1.nextSibling()
# print(type(locatio))
try:
location = locatio.text
except AttributeError:
for l in locatio:
location += l.text
location += " "
# print(location)
except AttributeError:
location = None
except TypeError:
location = None
# Website
try:
website = soup.find_all('a')
for web in website:
if "http" in str(web.get('href')):
website = web.get('href')
print(web.get('href'))
break
# print(website)
except AttributeError:
website = None
except IndexError:
website = None
bundle = [
url, self.website.platform, seller_name, location, website
]
# print(bundle)
TheAuthour.write_seller(*bundle)
def get_artwork_data_slave(self, url, driver):
driver.get(url)
soup = BeautifulSoup(driver.page_source, url)
if soup is not None:
# Field initiation ::
artwork = None
price = None
type_ = None
dimensions = None
frame = None
authenticity = None
about = None
artist_id = None
image_loc = None
year = None
support = None
signature = None
# Material to be added to technique
technique = ""
seller_id = None
artist = None
medium = None
# Medium must always have "Painting" or "Sculpture" (RULE :: 2)
# if "/painting/" in str(url):
# medium = "Painting" # (painting or sculpture)
# elif "/sculpture/" in str(url):
# medium = "Sculpture"
# else:
# # So that url leaks don't break the code.
# medium = None
# Seller_id
try:
seller_url = soup.find('div', class_='WncCi').find('a')['href']
seller_id = self.get_seller_id(seller_url)
except AttributeError or TypeError:
# Seller doesn't have a page.
try:
seller_url = soup.find('div', class_='WncCi').text.strip()
if seller_url in SELLER_INFO.keys():
seller_id = SELLER_INFO.get(seller_url)
else:
# Make a Kazoart style bundle, and write it to obtain a seller_id.
# [seller_url, platform_id(from name), Seller's name, Location, website]
bundle = [
seller_url, self.website.platform,
'EMERGINGARTISTPLATFOM', None, None
]
# Writing to db.
TheAuthour.write_seller(*bundle)
# This should generate the seller_id we so desperately desire.
# time.sleep(1)
seller_id = SELLER_INFO.get(seller_url)
except AttributeError:
pass
# We'll let the seller name be seller_url if the url is not found.
# Artist_id
try:
artist_url = soup.find('div', class_='WncCi').a.get('href')
if str(artist_url).endswith(".com"):
artist_url = re.sub('.com', "", artist_url)
artist_url = re.sub('emergingartistplatform',
'emergingartistplatform.com',
artist_url)
artist_id = self.get_artist_id(artist_url)
except AttributeError:
try:
artist_url = soup.find('div', class_='WncCi').text.strip()
country = None
a = soup.find_all('pre')
for b in a:
if b.get('data-hook') == 'description':
p = b.find_all('p')
for j in p:
if 'Country' in j.text or 'country' in j.text or 'COUNTRY' in j.text:
title = j.text.split(":")
country = title[-1].strip()
artist_data_pack = [artist_url, None, country, None]
# artist_data_pack = [name, born, country, about]
# pack = [name, born, country, about]
# Updating KEY_INFO dictionary.
KEY_INFO[artist_url] = db.Artist.key_maker(
artist_data_pack)
key = KEY_INFO.get(artist_url)
# Updating the dB with artist listings.
TheAuthour.write_artist(*artist_data_pack)
artist_id = ARTIST_INFO[key]
except AttributeError:
artist_id = None
# Continue fetching data only if seller_id, artist_id and medium are found. (RULE :: 3, 4)
if seller_id is not None and artist_id is not None:
try:
a = soup.find_all('span')
t = ""
for b in a:
if b.get('data-hook') == "formatted-primary-price":
# print(b.text)
for p in b.text:
if str(p).isnumeric() or str(p) == ".":
t += p
price = float(t) * rate
# print(price)
# Price
# print(price)
except AttributeError:
price = None
except ValueError:
price = None
# RULE : 5
if price is not None:
# Find artist, artwork, year, type_(N/A), dimensions, support, frame, signature, authenticity,
# about, image_loc(actual url of the image), and technique
# Wish the code to break if either Artist's name or Artwork's name are not found.
# Artist
artist = soup.find('div', class_='WncCi').text.strip()
# print(artist)
# Artwork
a = soup.find_all('pre')
for b in a:
if b.get('data-hook') == 'description':
p = b.find_all('p')
for j in p:
if 'Title' in j.text or 'title' in j.text or 'TITLE' in j.text:
title = j.text.split(":")
artwork = title[-1].strip()
if len(artwork) >= 255:
artwork = artwork[0:255]
# print(artwork)
if 'Date' in j.text:
date = j.text.split(":")
year = date[-1].strip()
# print(year)
if 'Size' in j.text:
dimensions = j.text.split(":")
dimensions = dimensions[-1].strip()
# print(dimensions)
if 'Medium' in j.text:
technique = j.text.split(":")
technique = technique[-1].strip()
# print(technique)
if len(j.text.split(
":")) == 1 and about is None:
about = j.text[-1].strip()
# Medium (RULE : 3)
if "Sculptures" in self.website.start_url:
medium = "Sculpture"
else:
medium = "Painting"
# image_loc
image = soup.find('div',
class_='main-media-image-wrapper-hook')
image = image.find('div', id='get-image-item-id')
image_loc = image.get('href')
# print(image_loc)
artwork_bundle = {
"artwork_title": artwork,
"artist_name": artist,
"year": year,
"price": price,
"Medium": medium,
"Type": type_,
"Dimensions": dimensions,
"Support": support,
"Frame": frame,
"Signature": signature,
"Authenticity": authenticity,
"About": about,
"platform": self.website.platform,
"image_addr": image_loc,
"seller_id": seller_id,
"artist_id": artist_id,
"url": url,
"technique": technique
}
TheAuthour.write_artwork_price_image(**artwork_bundle)
else:
print(f"Skipping {url}\n PRICE : {price}")
else:
print(
f"Skipping : {url}\nSeller_id = {seller_id}, Artist_id = {artist_id}, medium = {medium}"
)
else:
print(f"Soup not returned for {url}")
