class DataExtractor:
def __init__(self, response, url):
self.response = response
self.url = url
self.processors = Processors()
self.brands = Brands()
self.memory = Memory()
self.storages = Storages()
def parse(self):
data = {}
r = self.response
# produtos da shoptime
data['store'] = 'shoptime'
# nome do produto
try:
data['name'] = r.find('h1', {'id': 'main-product-name'})
data['name'] = self.validate_field(data, 'name')
data['name'] = re.sub(re.compile(u' \(Cód\. ([0-9])+\)'), '',
data['name'])
except (ValueError, TypeError, AttributeError):
data['name'] = ''
# url como variavel global da classe
data['url'] = self.url
# preço do produto
try:
data['price'] = r.find(
'span', {'data-partner-value': True})['data-partner-value']
data['price'] = self.normalize_price(data['price'])
except (ValueError, TypeError, AttributeError):
data['price'] = 0.0
# disponibilidade
try:
data['available'] = self.set_available(r, data['price'])
except (ValueError, TypeError, AttributeError):
data['available'] = False
try:
# processador
data['processor'] = r.find(
'th', text=re.compile(r'Processador')).parent.find('td').text
data['processor'] = self.normalize_processor(data['processor'])
except (ValueError, TypeError, AttributeError):
data['processor'] = ''
# marca
try:
data['brand'] = r.find(
'th', text=re.compile(r'Marca')).parent.find('td').text
data['brand'] = (self.normalize_brand(data['name'])).strip()
except (ValueError, TypeError, AttributeError):
data['brand'] = ''
# memória ram
try:
data['ram_memory'] = r.find(
'th', text=re.compile(u'Memória RAM')).parent.find('td').text
except (ValueError, TypeError, AttributeError):
data['ram_memory'] = ''
# sku para identificação
try:
data['sku'] = r.find('div', {'data-sku': True})['data-sku']
except (ValueError, TypeError, AttributeError):
data['sku'] = ''
# armazenamento (SSD/HD)
try:
try:
hd = r.find('th',
text=re.compile(r'HD')).parent.find('td').text
except (ValueError, TypeError, AttributeError):
hd = ''
try:
ssd = r.find('th',
text=re.compile(r'SSD')).parent.find('td').text
except (ValueError, TypeError, AttributeError):
ssd = ''
data['storage'] = self.normalize_storage(hd, ssd)
except (ValueError, TypeError, AttributeError):
data['storage'] = self.normalize_storage('', '')
# tamanho da tela
try:
data['display_size'] = self.normalize_display_size(
r.find('th', text=re.compile(
r'Polegadas da Tela')).parent.find('td').text)
except (ValueError, TypeError, AttributeError):
data['display_size'] = ''
try:
data['img_url'] = (r.findAll(
'img', {'class': 'p-image'})[0]['src']).strip()
except (ValueError, TypeError, AttributeError, IndexError):
data['img_url'] = ''
return data
def set_available(self, response, price):
meta = response.find('div', {'class': 'product-unavailable'})
if (meta == None) and (price != 0.0):
return True
else:
return False
def validate_field(self, data, field):
if data[field] != None:
return (data[field].get_text().strip() if
(len(data[field]) > 0) else '')
def normalize_display_size(self, text):
if text != None and len(text) > 0:
return (text.strip(" \"").replace(",", ".") + "\"").strip()
else:
return ''
def normalize_storage(self, hd, ssd):
result = ''
if hd != None and len(hd) > 0:
result = re.search('\d+.+[TG]B', hd)
if result != None:
return self.get_storage_capacity(result.group())
if ssd != None and len(ssd) > 0:
result = re.search('\d+.+[TG]B', ssd)
if result != None:
return self.get_storage_capacity(result.group())
def normalize_memory(self, raw_data):
if re.search('16', raw_data, re.IGNORECASE) != None:
return self.memory.get_16GB()
elif re.search('12', raw_data, re.IGNORECASE) != None:
return self.memory.get_12GB()
elif re.search('14', raw_data, re.IGNORECASE) != None:
return self.memory.get_14GB()
elif re.search('10', raw_data, re.IGNORECASE) != None:
return self.memory.get_10GB()
elif re.search('8', raw_data, re.IGNORECASE) != None:
return self.memory.get_8GB()
elif re.search('6', raw_data, re.IGNORECASE) != None:
return self.memory.get_6GB()
elif re.search('4', raw_data, re.IGNORECASE) != None:
return self.memory.get_4GB()
elif re.search('2', raw_data, re.IGNORECASE) != None:
return self.memory.get_2GB()
elif re.search('1', raw_data, re.IGNORECASE) != None:
return self.memory.get_1GB()
def normalize_price(self, raw_data):
try: # transforma 1.000, 00 em 1000.00
raw_data = (raw_data if len(raw_data) > 0 else '')
price = float(raw_data)
price = round(price, 2)
return price
except ValueError:
return 0.0
def normalize_brand(
self, raw_data
): # ["Samsung", "Asus", "Acer", "Dell", "Apple", "Positivo", "LG", "Lenovo"]
if re.search('dell', raw_data, re.IGNORECASE) != None:
return self.brands.get_dell()
elif re.search('asus', raw_data, re.IGNORECASE) != None:
return self.brands.get_asus()
elif re.search('apple', raw_data, re.IGNORECASE) != None:
return self.brands.get_apple()
elif re.search('acer', raw_data, re.IGNORECASE) != None:
return self.brands.get_acer()
elif re.search('samsung', raw_data, re.IGNORECASE) != None:
return self.brands.get_samsung()
elif re.search('positivo', raw_data, re.IGNORECASE) != None:
return self.brands.get_positivo()
elif re.search('lenovo', raw_data, re.IGNORECASE) != None:
return self.brands.get_lenovo()
elif re.search('lg', raw_data, re.IGNORECASE) != None:
return self.brands.get_lg()
elif re.search('compaq', raw_data, re.IGNORECASE) != None:
return self.brands.get_compaq()
elif re.search('seagate', raw_data, re.IGNORECASE) != None:
return self.brands.get_seagate()
elif re.search('gigabyte', raw_data, re.IGNORECASE) != None:
return self.brands.get_gigabyte()
elif (re.search('hp', raw_data, re.IGNORECASE) != None):
return self.brands.get_hp()
elif (re.search('sony', raw_data, re.IGNORECASE) != None):
return self.brands.get_sony()
def normalize_processor(self, raw_data):
raw_data = re.sub('\\\u\w\w\w\w', '', raw_data)
if re.search('i3', raw_data, re.IGNORECASE) != None:
return self.processors.get_i3()
elif re.search('i5', raw_data, re.IGNORECASE) != None:
return self.processors.get_i5()
elif re.search('i7', raw_data, re.IGNORECASE) != None:
return self.processors.get_i7()
elif re.search('Pentium', raw_data, re.IGNORECASE) != None:
return self.processors.get_pentium_quad()
elif re.search('byt|baytrail', raw_data, re.IGNORECASE) != None:
return self.processors.get_baytrail()
elif re.search('amd.+dual core', raw_data, re.IGNORECASE) \
!= None:
return self.processors.get_amd_dual()
elif re.search('atom', raw_data, re.IGNORECASE) != None:
return self.processors.get_atom()
elif re.search('Intel.+Core.+M', raw_data, re.IGNORECASE) \
!= None:
return self.processors.get_core_m()
elif re.search('Celeron', raw_data, re.IGNORECASE) != None:
return self.processors.get_celeron()
elif re.search('arm', raw_data, re.IGNORECASE) != None:
return self.processors.get_arm_a9()
elif re.search('samsung', raw_data, re.IGNORECASE) != None:
return self.processors.get_samsung()
# normalização de capacidade
def get_storage_capacity(self, raw_data):
if (re.search('2TB|2 TB', raw_data, re.IGNORECASE) != None):
return self.storages.get_2tb()
elif (re.search('1TB|1 TB', raw_data, re.IGNORECASE) != None):
return self.storages.get_1tb()
elif (re.search('750 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_750()
elif (re.search('640 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_640()
elif (re.search('500 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_500()
elif (re.search('320GB|320 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_320()
elif (re.search('256GB|256 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_256()
elif (re.search('160GB|160 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_160()
elif (re.search('128GB|128 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_128()
elif (re.search('80GB|80 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_80()
elif (re.search('64GB|64 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_64()
elif (re.search('32GB|32 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_32()
elif (re.search('16GB|16 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_16()
class DataExtractor():
def __init__(self, response, url):
self.response = response
self.url = url
self.processors = Processors()
self.brands = Brands()
self.memory = Memory()
self.storages = Storages()
#{ _id, available, brand, color, display_feature, display_size, graphics_processor_name, graphics_processor, name, operating_system, price, processor, ram_memory, sku, screen_resolution, storage, storage_type, url, img_url}
# TODO: display_feature, display_size, graphics_processor_name, graphics_processor, operating_system, screen_resolution, storage_type, img_url
def parse(self):
data = {}
# produtos das casas bahia
data['store'] = "casas_bahia"
# nome do produto
data['name'] = self.response.findAll("b", {"itemprop": "name"})
data['name'] = self.validate_field(data, 'name')
# url como variavel global da classe
data['url'] = self.url
# preco do produto
data['price'] = self.response.findAll("i", {"class": "sale price"})
data['price'] = self.normalize_price(data['price'])
# disponibilidade: nas casas bahia, se o produto possuir preco, o produto esta disponivel
data['available'] = data['price'] != None and data['price'] != 0.0
data['img_url'] = self.response.findAll('img', {'itemprop': 'image'})
data['img_url'] = self.normalize_img_url(data['img_url'])
# processador
data['processor'] = self.response.findAll("", {"class": "Processador"})
data['processor'] = self.normalize_processor(
self.validate_field(data, 'processor'))
# marca
data['brand'] = self.normalize_brand(data['name'])
# memoria ram
data['ram_memory'] = self.response.findAll("dl",
{"class": "Memoria-RAM"})
data['ram_memory'] = self.normalize_memory(
self.validate_field(data, 'ram_memory'))
# sku para identificacao
data['sku'] = self.url.split('?')[0].split('-')[-1].split('.')[0]
# armazenamento (SSD/HD)
hd = self.response.findAll(
"dl", {"class": ["Disco-rigido--HD-", "Memoria-Flash--SSD-"]})
data['storage'] = self.normalize_storage(hd)
# tamanho de tela
data['display_size'] = self.response.findAll(
"dl", {"class": "Tamanho-da-tela"})
data['display_size'] = data['display_size'][0].find(
'dd').get_text().strip() if (len(data["display_size"]) > 0) else ""
return data
def validate_field(self, data, field):
return (data[field][0].get_text().strip() if
(len(data[field]) > 0) else "")
def normalize_img_url(self, img_url):
return img_url[0]['src'] if (len(img_url) > 0) else None
def normalize_storage(self, hd):
if (len(hd) > 0):
hd = hd[0].find('dd').get_text()
result = ''
if hd != None and len(hd) > 0:
result = re.search('\d+.+[TG]B', hd)
if result != None:
result = self.get_storage_capacity(result.group())
return result
def normalize_memory(self, raw_data):
if (re.search('16GB|16 GB', raw_data, re.IGNORECASE) != None):
return self.memory.get_16GB()
elif (re.search('12GB|12 GB', raw_data, re.IGNORECASE) != None):
return self.memory.get_12GB()
elif (re.search('14GB|14 GB', raw_data, re.IGNORECASE) != None):
return self.memory.get_14GB()
elif (re.search('10GB|10 GB', raw_data, re.IGNORECASE) != None):
return self.memory.get_10GB()
elif (re.search('8GB|8 GB', raw_data, re.IGNORECASE) != None):
return self.memory.get_8GB()
elif (re.search('6GB|6 GB', raw_data, re.IGNORECASE) != None):
return self.memory.get_6GB()
elif (re.search('4GB|4 GB', raw_data, re.IGNORECASE) != None):
return self.memory.get_4GB()
elif (re.search('2GB|2 GB', raw_data, re.IGNORECASE) != None):
return self.memory.get_2GB()
elif (re.search('1GB|1 GB', raw_data, re.IGNORECASE) != None):
return self.memory.get_1GB()
# transforma 1.000,00 em 1000.00
def normalize_price(self, raw_data):
try:
raw_data = raw_data[0].get_text() if (len(raw_data) > 0) else ""
raw_data = raw_data.replace('.', '').replace(',', '.')
return float(raw_data)
except ValueError:
return 0.0
def normalize_brand(self, raw_data):
if (re.search('dell', raw_data, re.IGNORECASE) != None):
return self.brands.get_dell()
elif (re.search('asus', raw_data, re.IGNORECASE) != None):
return self.brands.get_asus()
elif (re.search('apple', raw_data, re.IGNORECASE) != None):
return self.brands.get_apple()
elif (re.search('acer', raw_data, re.IGNORECASE) != None):
return self.brands.get_acer()
elif (re.search('samsung', raw_data, re.IGNORECASE) != None):
return self.brands.get_samsung()
elif (re.search('positivo', raw_data, re.IGNORECASE) != None):
return self.brands.get_positivo()
elif (re.search('lenovo', raw_data, re.IGNORECASE) != None):
return self.brands.get_lenovo()
elif (re.search('lg', raw_data, re.IGNORECASE) != None):
return self.brands.get_lg()
elif (re.search('hp', raw_data, re.IGNORECASE) != None):
return self.brands.get_hp()
elif (re.search('sony', raw_data, re.IGNORECASE) != None):
return self.brands.get_sony()
# ['Intel Core i3', 'Intel Core i5', 'Intel Core i7', 'Intem Pentium Quad Core', 'Intel Baytrail', 'AMD Dual Core', 'Item Atom', 'Intel Core M', 'Intel Celeron']
def normalize_processor(self, raw_data):
# remove erros de enconding (ex: \u84d2)
raw_data = re.sub('\\\u\w\w\w\w', '', raw_data)
if (re.search("i3", raw_data, re.IGNORECASE) != None):
return self.processors.get_i3()
elif (re.search("i5", raw_data, re.IGNORECASE) != None):
return self.processors.get_i5()
elif (re.search("i7", raw_data, re.IGNORECASE) != None):
return self.processors.get_i7()
elif (re.search("Pentium", raw_data, re.IGNORECASE) != None):
return self.processors.get_pentium_quad()
elif (re.search("byt|baytrail", raw_data, re.IGNORECASE) != None):
return self.processors.get_baytrail()
elif (re.search('Intel.+[Dd]ual [Cc]ore', raw_data, re.IGNORECASE) !=
None):
return self.processors.get_intel_dual()
elif (re.search('Intel.+[Qq]uad [Cc]ore', raw_data, re.IGNORECASE) !=
None):
return self.processors.get_intel_quad()
elif (re.search("amd.+dual core", raw_data, re.IGNORECASE) != None):
return self.processors.get_amd_dual()
elif (re.search("atom", raw_data, re.IGNORECASE) != None):
return self.processors.get_atom()
elif (re.search("Intel.+Core.+M", raw_data, re.IGNORECASE) != None):
return self.processors.get_core_m()
elif (re.search("Celeron", raw_data, re.IGNORECASE) != None):
return self.processors.get_celeron()
elif (re.search("arm", raw_data, re.IGNORECASE) != None):
return self.processors.get_arm_a9()
elif (re.search("samsung", raw_data, re.IGNORECASE) != None):
return self.processors.get_samsung()
# normalização de capacidade
def get_storage_capacity(self, raw_data):
if (re.search('2TB|2 TB', raw_data, re.IGNORECASE) != None):
return self.storages.get_2tb()
elif (re.search('1TB|1 TB', raw_data, re.IGNORECASE) != None):
return self.storages.get_1tb()
elif (re.search('750 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_750()
elif (re.search('640 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_640()
elif (re.search('500 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_500()
elif (re.search('320GB|320 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_320()
elif (re.search('256GB|256 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_256()
elif (re.search('160GB|160 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_160()
elif (re.search('128GB|128 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_128()
elif (re.search('80GB|80 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_80()
elif (re.search('64GB|64 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_64()
elif (re.search('32GB|32 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_32()
elif (re.search('16GB|16 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_16()
class DataExtractor():
def __init__(self, response, url):
self.response = response
self.url = url
self.processors = Processors();
self.brands = Brands();
self.memory = Memory();
def parse(self):
namespaces = {'re': "http://exslt.org/regular-expressions"}
data = {}
data["store"] = "eletro_shopping"
data["name"] = self.response.xpath('//div[@itemprop="name"]/h1//text()')
data["name"] = self.validate_field(data, 'name')
data["url"] = self.url
data["price"] = self.response.xpath('//span[@itemprop="lowPrice"]//text()')
data["price"] = self.normalize_price(data["price"])
data["available"] = data["price"] != None and data["price"] != 0.0
data["processor"] = self.response.xpath('//td[text()="Processador - Modelo"]/following-sibling::td//text()')
data["processor"] = self.normalize_processor(self.validate_field(data, "processor"))
data["brand"] = self.normalize_brand(data["name"])
data["ram_memory"] = self.response.xpath('//td[contains(text(), "Mem") and contains(text(), "Capacidade")]/following-sibling::td//text()', namespaces = {'re' : 'http://exslt.org/regular-expressions'})
data["ram_memory"] = self.normalize_memory(self.validate_field(data, "ram_memory"))
data["sku"] = self.response.xpath('//div[@itemprop="name"]/span//text()')
if(len(data['sku']) > 0):
data["sku"] = data["sku"][0].split(':')[1].strip()
else:
data['sku'] = ''
data["disco"] = {}
disco = self.response.xpath('//td[text()="HD - Tipo"]/following-sibling::td//text()', namespaces={'re': "http://exslt.org/regular-expressions"})
hd = ''
ssd = ''
if(len(disco) > 0):
disco = disco[0]
if(re.search('HD', disco, re.IGNORECASE) != None and re.search('SSD', disco, re.IGNORECASE) != None):
aux = self.response.xpath('//td[text()="HD - Capacidade"]/following-sibling::td//text()', namespaces={'re': "http://exslt.org/regular-expressions"})
if(len(aux) > 0):
aux = aux[0];
if(re.search('\\+', aux) != None):
hd = aux.split('+')[0]
ssd = aux.split('+')[1]
else:
hd = disco.split('HD')[0].strip();
ssd = disco.split(' ')
ssd = ssd[ssd.index('SSD') - 1]
elif(re.search('HD', disco, re.IGNORECASE) != None):
hd = self.response.xpath('//td[text()="HD - Capacidade"]/following-sibling::td//text()', namespaces={'re': "http://exslt.org/regular-expressions"})
ssd = ''
elif(re.search('SSD', disco, re.IGNORECASE) != None):
hd = ''
ssd = self.response.xpath('//td[text()="HD - Capacidade"]/following-sibling::td//text()', namespaces={'re': "http://exslt.org/regular-expressions"})
data['disco']['hd'] = hd
data['disco']['ssd'] = ssd
data["display_size"] = self.response.xpath('//td[text()="Tela - Tamanho"]/following-sibling::td//text()', namespaces={'re': "http://exslt.org/regular-expressions"})
data["display_size"] = self.validate_field(data, "display_size")
return data
def validate_field(self, data, field):
return (data[field][0].strip() if (len(data[field]) > 0) else "")
# normalize storage
def normalize_memory(self, raw_data):
if (re.search('16', raw_data, re.IGNORECASE) != None):
return self.memory.get_16GB()
elif (re.search('14', raw_data, re.IGNORECASE) != None):
return self.memory.get_14GB()
elif (re.search('12', raw_data, re.IGNORECASE) != None):
return self.memory.get_12GB()
elif (re.search('10', raw_data, re.IGNORECASE) != None):
return self.memory.get_10GB()
elif (re.search('8', raw_data, re.IGNORECASE) != None):
return self.memory.get_8GB()
elif (re.search('6', raw_data, re.IGNORECASE) != None):
return self.memory.get_6GB()
elif (re.search('4', raw_data, re.IGNORECASE) != None):
return self.memory.get_4GB()
elif (re.search('2', raw_data, re.IGNORECASE) != None):
return self.memory.get_2GB()
elif (re.search('1', raw_data, re.IGNORECASE) != None):
return self.memory.get_1GB()
def normalize_price(self, raw_data):
try:
raw_data = raw_data[0].strip() if (len(raw_data) > 0) else ""
raw_data = raw_data.replace(".", "").replace(",", ".")
return float(raw_data)
except ValueError:
return 0.0
def normalize_brand(self, raw_data):
if (re.search("dell", raw_data, re.IGNORECASE) != None):
return self.brands.get_dell()
elif (re.search('asus', raw_data, re.IGNORECASE) != None):
return self.brands.get_asus()
elif (re.search('apple', raw_data, re.IGNORECASE) != None):
return self.brands.get_apple()
elif (re.search('acer', raw_data, re.IGNORECASE) != None):
return self.brands.get_acer()
elif (re.search('samsung', raw_data, re.IGNORECASE) != None):
return self.brands.get_samsung()
elif (re.search('positivo', raw_data, re.IGNORECASE) != None):
return self.brands.get_positivo()
elif (re.search('lenovo', raw_data, re.IGNORECASE) != None):
return self.brands.get_lenovo()
elif (re.search('lg', raw_data, re.IGNORECASE) != None):
return self.brands.get_lg()
def normalize_processor(self, raw_data):
# remove erros de enconding (ex: \u84d2)
raw_data = re.sub('\\\u\w\w\w\w', '', raw_data)
if (re.search("i3", raw_data, re.IGNORECASE) != None):
return self.processors.get_i3()
elif (re.search("i5", raw_data, re.IGNORECASE) != None):
return self.processors.get_i5()
elif (re.search("i7", raw_data, re.IGNORECASE) != None):
return self.processors.get_i7()
elif (re.search("Pentium", raw_data, re.IGNORECASE) != None):
return self.processors.get_pentium_quad()
elif (re.search("byt|baytrail", raw_data, re.IGNORECASE) != None):
return self.processors.get_baytrail()
elif (re.search("amd.+dual core", raw_data, re.IGNORECASE) != None):
return self.processors.get_amd_dual()
elif (re.search("atom", raw_data, re.IGNORECASE) != None):
return self.processors.get_atom()
elif (re.search("Intel.+Core.+M", raw_data, re.IGNORECASE) != None):
return self.processors.get_core_m()
elif (re.search("Celeron", raw_data, re.IGNORECASE) != None):
return self.processors.get_celeron()
elif (re.search("arm", raw_data, re.IGNORECASE) != None):
return self.processors.get_arm_a9()
elif (re.search("samsung", raw_data, re.IGNORECASE) != None):
return self.processors.get_samsung()
class DataExtractor():
def __init__(self, response, url):
self.response = response
self.url = url
self.processors = Processors()
self.brands = Brands()
self.memory = Memory()
#{ _id, available, brand, color, display_feature, display_size, graphics_processor_name, graphics_processor, name, operating_system, price, processor, ram_memory, sku, screen_resolution, storage, storage_type, url, img_url}
# TODO: display_feature, display_size, graphics_processor_name, graphics_processor, operating_system, screen_resolution, storage_type, img_url
def parse(self):
data = {}
r = self.response
# produtos das casas bahia
data['store'] = "ibyte_computadores"
# nome do produto
data['name'] = self.response.findAll("li", {"class": "product"})
data['name'] = self.validate_field(data, 'name')
# url como variavel global da classe
data['url'] = self.url
# nome do produto
data['price'] = self.response.find("div", {
"class": "preco-produto"
}).parent.findAll('span', {"class": "price"})
data['price'] = self.normalize_price(data['price'])
# disponibilidade: nas casas bahia, se o produto possuir preco, o produto esta disponivel
data['available'] = data['price'] != None and data['price'] != 0.0
try:
# processador
data['processor'] = r.find(
'td', text=re.compile(r'Processador:')).parent.find(
'td', {
'width': '570'
}).text
data['processor'] = self.normalize_processor(data['processor'])
except (ValueError, TypeError, AttributeError):
data['processor'] = ''
# marca
try:
data['brand'] = r.find('td',
text=re.compile(r'Marca:')).parent.find(
'td', {
'width': '570'
}).text.strip()
except (ValueError, TypeError, AttributeError):
data['brand'] = ''
# memória ram
try:
data['ram_memory'] = r.find(
'td', text=re.compile(u'Memória RAM:')).parent.find(
'td', {
'width': '570'
}).text.strip()
data['ram_memory'] = self.normalize_memory(data['ram_memory'])
except (ValueError, TypeError, AttributeError):
data['ram_memory'] = ''
# sku para identificação
try:
data['sku'] = r.find('div', {
'class': 'product-essential'
}).parent.find('h3').text.split(' ')[-1].split(')')[0]
except (ValueError, TypeError, AttributeError):
data['sku'] = ''
# armazenamento (SSD/HD)
try:
try:
hd = r.find('td', text=re.compile(r'HD:')).parent.find(
'td', {
'width': '570'
}).text
except (ValueError, TypeError, AttributeError):
hd = ''
try:
ssd = r.find('td', text=re.compile(r'SSD:')).parent.find(
'td', {
'width': '570'
}).text
except (ValueError, TypeError, AttributeError):
ssd = ''
data['storage'] = self.normalize_storage(hd, ssd)
except (ValueError, TypeError, AttributeError):
data['storage'] = {}
# tamanho da tela
try:
data['display_size'] = r.find(
'td', text=re.compile(r'Polegadas da Tela:')).parent.find(
'td', {
'width': '570'
}).text.strip()
except (ValueError, TypeError, AttributeError):
data['display_size'] = ''
#imagem do produto
try:
data['img_url'] = (r.findAll('img',
{'id': 'image'})[0]['src']).strip()
except (ValueError, TypeError, AttributeError, IndexError):
data['img_url'] = ''
return data
def validate_field(self, data, field):
return (data[field][0].get_text().strip() if
(len(data[field]) > 0) else "")
def normalize_storage(self, hd, ssd):
result = {}
if hd != None and len(hd) > 0:
result["HD"] = re.search('\d+TB', hd)
if result["HD"] != None:
result["HD"] = result["HD"].group()
if ssd != None and (len(ssd) > 0) and result == None:
result["SSD"] = re.search('\d+TB', ssd)
if result["SSD"] != None:
result["SSD"] = result["SSD"].group()
return result
def normalize_memory(self, raw_data):
if (re.search('16', raw_data, re.IGNORECASE) != None):
return self.memory.get_16GB()
elif (re.search('12', raw_data, re.IGNORECASE) != None):
return self.memory.get_12GB()
elif (re.search('14', raw_data, re.IGNORECASE) != None):
return self.memory.get_14GB()
elif (re.search('10', raw_data, re.IGNORECASE) != None):
return self.memory.get_10GB()
elif (re.search('8', raw_data, re.IGNORECASE) != None):
return self.memory.get_8GB()
elif (re.search('6', raw_data, re.IGNORECASE) != None):
return self.memory.get_6GB()
elif (re.search('4', raw_data, re.IGNORECASE) != None):
return self.memory.get_4GB()
elif (re.search('2', raw_data, re.IGNORECASE) != None):
return self.memory.get_2GB()
elif (re.search('1', raw_data, re.IGNORECASE) != None):
return self.memory.get_1GB()
def normalize_price(self, raw_data):
try:
# transforma 1.000,00 em 1000.00
raw_data = raw_data[0].get_text() if (len(raw_data) > 0) else ""
raw_data = raw_data.replace('.', '').replace(',', '.')
raw_data = raw_data.replace('R$', '').replace(',', '.')
return float(raw_data)
except ValueError:
return 0.0
def normalize_brand(self, raw_data):
# ["Samsung", "Asus", "Acer", "Dell", "Apple", "Positivo", "LG", "Lenovo"]
if (re.search('dell', raw_data, re.IGNORECASE) != None):
return self.brands.get_dell()
elif (re.search('asus', raw_data, re.IGNORECASE) != None):
return self.brands.get_asus()
elif (re.search('apple', raw_data, re.IGNORECASE) != None):
return self.brands.get_apple()
elif (re.search('acer', raw_data, re.IGNORECASE) != None):
return self.brands.get_acer()
elif (re.search('samsung', raw_data, re.IGNORECASE) != None):
return self.brands.get_samsung()
elif (re.search('positivo', raw_data, re.IGNORECASE) != None):
return self.brands.get_positivo()
elif (re.search('lenovo', raw_data, re.IGNORECASE) != None):
return self.brands.get_lenovo()
elif (re.search('lg', raw_data, re.IGNORECASE) != None):
return self.brands.get_lg()
def normalize_processor(self, raw_data):
# ['Intel Core i3', 'Intel Core i5', 'Intel Core i7', 'Intem Pentium Quad Core', 'Intel Baytrail', 'AMD Dual Core', 'Item Atom', 'Intel Core M', 'Intel Celeron']
# remove erros de enconding (ex: \u84d2)
raw_data = re.sub('\\\u\w\w\w\w', '', raw_data)
if (re.search("i3", raw_data, re.IGNORECASE) != None):
return self.processors.get_i3()
elif (re.search("i5", raw_data, re.IGNORECASE) != None):
return self.processors.get_i5()
elif (re.search("i7", raw_data, re.IGNORECASE) != None):
return self.processors.get_i7()
elif (re.search("Pentium", raw_data, re.IGNORECASE) != None):
return self.processors.get_pentium_quad()
elif (re.search("byt|baytrail", raw_data, re.IGNORECASE) != None):
return self.processors.get_baytrail()
elif (re.search("amd.+dual core", raw_data, re.IGNORECASE) != None):
return self.processors.get_amd_dual()
elif (re.search("atom", raw_data, re.IGNORECASE) != None):
return self.processors.get_atom()
elif (re.search("Intel.+Core.+M", raw_data, re.IGNORECASE) != None):
return self.processors.get_core_m()
elif (re.search("Celeron", raw_data, re.IGNORECASE) != None):
return self.processors.get_celeron()
elif (re.search("arm", raw_data, re.IGNORECASE) != None):
return self.processors.get_arm_a9()
elif (re.search("samsung", raw_data, re.IGNORECASE) != None):
return self.processors.get_samsung()
class DataExtractor():
def __init__(self, response, url):
self.response = response
self.url = url
self.processors = Processors()
self.brands = Brands()
self.memory = Memory()
self.storages = Storages()
#{ _id, available, brand, color, display_feature, display_size, graphics_processor_name, graphics_processor, name, operating_system, price, processor, ram_memory, sku, screen_resolution, storage, storage_type, url, img_url}
# TODO: display_feature, display_size, graphics_processor_name, graphics_processor, operating_system, screen_resolution, storage_type, img_url
def parse(self):
data = {}
r = self.response
# produtos das casas bahia
data['store'] = "mega_eletronicos"
# nome do produto
try:
data['name'] = r.find('div', {
'class': 'col-md-12 col-sm-12'
}).parent.find('h1').text
data['name'] = self.validate_field(data, 'name')
except (ValueError, TypeError, AttributeError):
data['name'] = ''
# url como variavel global da classe
data['url'] = self.url
# preço do produto
try:
data['price'] = r.find('h3', {
'class': 'real'
}).parent.find('span').text
data['price'] = self.normalize_price(data['price'])
except (ValueError, TypeError, AttributeError):
data['price'] = 0.0
# disponibilidade: nas casas bahia, se o produto possuir preco, o produto esta disponivel
data['available'] = data['price'] != None and data['price'] != 0.0
try:
# processador
data['processor'] = r.find(
'td',
text=re.compile(r'(Processor|Processador|Intel)')).parent.find(
'td', {
'width': '65%'
}).text
data['processor'] = self.normalize_processor(data['processor'])
except (ValueError, TypeError, AttributeError):
data['processor'] = ''
# marca
try:
data['brand'] = r.find('div', {
'class': 'col-md-12 col-sm-12'
}).parent.find('h1').text
data['brand'] = self.normalize_brand(data['brand'])
except (ValueError, TypeError, AttributeError):
data['brand'] = ''
# memória ram
try:
data['ram_memory'] = r.find(
'td', text=re.compile(r'(DDR|SDRAM|RAM)')).parent.find(
'td', {
'width': '65%'
}).text
data['ram_memory'] = self.normalize_memory(data['ram_memory'])
except (ValueError, TypeError, AttributeError):
data['ram_memory'] = ''
# sku para identificacao
data['sku'] = self.url.split('?')[0].split('-')[-1].split('.')[0]
# armazenamento (SSD/HD)
try:
data['storage'] = self.normalize_storage(
r.find('td', text=re.compile(r'(rpm|HDD|SSD)')).parent.find(
'td', {
'width': '65%'
}).text)
except (ValueError, TypeError, AttributeError):
data['storage'] = ''
# tamanho da tela
try:
#data['display_size'] = r.find('td', text=re.compile(r'(LED|LCD)')).parent.find('td', {'width': '65%'}).text
#data['display_size'] = re.sub(r'([a-z])', '', data['display_size'])
#data['display_size'] = re.sub(r'([A-Z])', '', data['display_size'])
# comentei tudo pois não consegui formalizar os tamanhos da tela
data['display_size'] = ''
except (ValueError, TypeError, AttributeError):
data['display_size'] = ''
return data
def validate_field(self, data, field):
return (data[field][0].get_text().strip() if
(len(data[field]) > 0) else "")
def normalize_storage(self, hd):
result = ''
if hd != None and len(hd) > 0:
result = re.search('\d+.+[TG]B', hd)
if result != None:
result = result.group()
return self.get_storage_capacity(result)
def normalize_memory(self, raw_data):
if (re.search('16', raw_data, re.IGNORECASE) != None):
return self.memory.get_16GB()
elif (re.search('12', raw_data, re.IGNORECASE) != None):
return self.memory.get_12GB()
elif (re.search('14', raw_data, re.IGNORECASE) != None):
return self.memory.get_14GB()
elif (re.search('10', raw_data, re.IGNORECASE) != None):
return self.memory.get_10GB()
elif (re.search('8', raw_data, re.IGNORECASE) != None):
return self.memory.get_8GB()
elif (re.search('6', raw_data, re.IGNORECASE) != None):
return self.memory.get_6GB()
elif (re.search('4', raw_data, re.IGNORECASE) != None):
return self.memory.get_4GB()
elif (re.search('2', raw_data, re.IGNORECASE) != None):
return self.memory.get_2GB()
elif (re.search('1', raw_data, re.IGNORECASE) != None):
return self.memory.get_1GB()
def normalize_price(self, raw_data):
try:
# transforma 1.000,00 em 1000.00
#raw_data = raw_data[0].get_text() if (len(raw_data) > 0) else ""
raw_data = raw_data.replace('.', '').replace(',', '.')
return float(raw_data)
except ValueError:
return 0.0
def normalize_brand(self, raw_data):
# ["Samsung", "Asus", "Acer", "Dell", "Apple", "Positivo", "LG", "Lenovo"]
if (re.search('dell', raw_data, re.IGNORECASE) != None):
return self.brands.get_dell()
elif (re.search('asus', raw_data, re.IGNORECASE) != None):
return self.brands.get_asus()
elif (re.search('apple', raw_data, re.IGNORECASE) != None):
return self.brands.get_apple()
elif (re.search('acer', raw_data, re.IGNORECASE) != None):
return self.brands.get_acer()
elif (re.search('samsung', raw_data, re.IGNORECASE) != None):
return self.brands.get_samsung()
elif (re.search('positivo', raw_data, re.IGNORECASE) != None):
return self.brands.get_positivo()
elif (re.search('lenovo', raw_data, re.IGNORECASE) != None):
return self.brands.get_lenovo()
elif (re.search('lg', raw_data, re.IGNORECASE) != None):
return self.brands.get_lg()
def normalize_processor(self, raw_data):
# ['Intel Core i3', 'Intel Core i5', 'Intel Core i7', 'Intem Pentium Quad Core', 'Intel Baytrail', 'AMD Dual Core', 'Item Atom', 'Intel Core M', 'Intel Celeron']
# remove erros de enconding (ex: \u84d2)
raw_data = re.sub('\\\u\w\w\w\w', '', raw_data)
if (re.search("i3", raw_data, re.IGNORECASE) != None):
return self.processors.get_i3()
elif (re.search("i5", raw_data, re.IGNORECASE) != None):
return self.processors.get_i5()
elif (re.search("i7", raw_data, re.IGNORECASE) != None):
return self.processors.get_i7()
elif (re.search("Pentium", raw_data, re.IGNORECASE) != None):
return self.processors.get_pentium_quad()
elif (re.search("byt|baytrail", raw_data, re.IGNORECASE) != None):
return self.processors.get_baytrail()
elif (re.search("amd.+dual core", raw_data, re.IGNORECASE) != None):
return self.processors.get_amd_dual()
elif (re.search("amd.+quad core", raw_data, re.IGNORECASE) != None):
return self.processors.get_amd_quad()
elif (re.search("atom", raw_data, re.IGNORECASE) != None):
return self.processors.get_atom()
elif (re.search("Intel.+Core.+M", raw_data, re.IGNORECASE) != None):
return self.processors.get_core_m()
elif (re.search("Celeron", raw_data, re.IGNORECASE) != None):
return self.processors.get_celeron()
elif (re.search("arm", raw_data, re.IGNORECASE) != None):
return self.processors.get_arm_a9()
elif (re.search("samsung", raw_data, re.IGNORECASE) != None):
return self.processors.get_samsung()
# normalização de capacidade
def get_storage_capacity(self, raw_data):
if (re.search('2TB|2 TB', raw_data, re.IGNORECASE) != None):
return self.storages.get_2tb()
elif (re.search('1TB|1 TB', raw_data, re.IGNORECASE) != None):
return self.storages.get_1tb()
elif (re.search('750 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_750()
elif (re.search('640 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_640()
elif (re.search('500 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_500()
elif (re.search('320GB|320 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_320()
elif (re.search('256GB|256 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_256()
elif (re.search('160GB|160 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_160()
elif (re.search('128GB|128 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_128()
elif (re.search('80GB|80 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_80()
elif (re.search('64GB|64 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_64()
elif (re.search('32GB|32 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_32()
elif (re.search('16GB|16 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_16()
class DataExtractor():
def __init__(self, response, url):
self.response = response
self.url = url
self.processors = Processors()
self.brands = Brands()
self.memory = Memory()
#{ _id, available, brand, color, display_feature, display_size, graphics_processor_name, graphics_processor, name, operating_system, price, processor, ram_memory, sku, screen_resolution, storage, storage_type, url, img_url}
# TODO: display_feature, display_size, graphics_processor_name, graphics_processor, operating_system, screen_resolution, storage_type, img_url
def parse(self):
data = {}
# produtos das casas bahia
data['store'] = "submarino"
# nome do produto
data['name'] = self.self.response.xpath('/html/body/div[5]/section/div/div/div[2]/div/div[1]/h1/span/text()')[0].strip()
data['name'] = self.validate_field(data, 'name')
# url como variavel global da classe
data['url'] = self.url
# # preco do produto
data['price'] = self.response.findAll("span", {"itemprop": "price/salesPrice"})
data['price'] = self.normalize_price(data['price'])
# disponibilidade: nas casas bahia, se o produto possuir preco, o produto esta disponivel
data['available'] = data['price'] != None and data['price'] != 0.0
# processador
data['processor'] = self.response.xpath('//*[@id="productdetails"]/div[3]/section/table/tbody/tr[6]/td/text()')
data['processor'] = self.normalize_processor(self.validate_field(data, 'processor'))
# marca
data['brand'] = self.normalize_brand(data['name'])
# memoria ram
data['ram_memory'] = self.response.xpath('//*[@id="productdetails"]/div[3]/section/table/tbody/tr[9]/td/text()')[0].strip()
data['ram_memory'] = self.normalize_memory(self.validate_field(data, 'ram_memory'))
# sku para identificacao
data['sku'] = self.url.split('/')[2].split('/')[-1]
# # armazenamento (SSD/HD) ------ observar busca por 'SSD'
hd = self.response.xpath('//*[@id="productdetails"]/div[3]/section/table/tbody/tr[11]/td/text()')[0].strip()
# ssd = self.reponse.findAll("SSD")
# if (ssd)
ssd = self.response.xpath('//*[@id="productdetails"]/div[3]/section/table/tbody/tr[10]/td/text()')[0].strip
# elif
# ssd = ""
data['storage'] = self.normalize_storage(hd, ssd)
# tamanho de tela
data['display_size'] = self.response.xpath('//*[@id="productdetails"]/div[3]/section/table/tbody/tr[4]/td/text()')[0].strip()
data['display_size'] = data['display_size'][0].find('dd').get_text().strip() if (len(data["display_size"]) > 0) else ""
return data
def validate_field(self, data, field):
return (data[field][0].get_text().strip() if (len(data[field]) > 0) else "")
def normalize_storage(self, hd, ssd):
if (len(hd) > 0):
hd = hd[0].find('dd').get_text()
if (len(ssd) > 0):
ssd = ssd[0].find('dd').get_text()
result = ''
if hd != None and len(hd) > 0:
result = re.search('\d+.+[TG]B', hd)
if result != None:
return self.get_storage_capacity(result.group())
if ssd != None and len(ssd) > 0:
result = re.search('\d+.+[TG]B', ssd)
if result != None:
return self.get_storage_capacity(result.group())
def normalize_memory(self, raw_data):
if (re.search('16', raw_data, re.IGNORECASE) != None):
return self.memory.get_16GB()
elif (re.search('12', raw_data, re.IGNORECASE) != None):
return self.memory.get_12GB()
elif (re.search('14', raw_data, re.IGNORECASE) != None):
return self.memory.get_14GB()
elif (re.search('10', raw_data, re.IGNORECASE) != None):
return self.memory.get_10GB()
elif (re.search('8', raw_data, re.IGNORECASE) != None):
return self.memory.get_8GB()
elif (re.search('6', raw_data, re.IGNORECASE) != None):
return self.memory.get_6GB()
elif (re.search('4', raw_data, re.IGNORECASE) != None):
return self.memory.get_4GB()
elif (re.search('2', raw_data, re.IGNORECASE) != None):
return self.memory.get_2GB()
elif (re.search('1', raw_data, re.IGNORECASE) != None):
return self.memory.get_1GB()
def normalize_price(self, raw_data):
try:
# transforma 1.000,00 em 1000.00
raw_data = raw_data[0].get_text() if (len(raw_data) > 0) else ""
raw_data = raw_data.replace('.', '').replace(',', '.')
return float(raw_data)
except ValueError:
return 0.0
def normalize_brand(self, raw_data):
# ["Samsung", "Asus", "Acer", "Dell", "Apple", "Positivo", "LG", "Lenovo"]
if (re.search('dell', raw_data, re.IGNORECASE) != None):
return self.brands.get_dell()
elif (re.search('asus', raw_data, re.IGNORECASE) != None):
return self.brands.get_asus()
elif (re.search('apple', raw_data, re.IGNORECASE) != None):
return self.brands.get_apple()
elif (re.search('acer', raw_data, re.IGNORECASE) != None):
return self.brands.get_acer()
elif (re.search('samsung', raw_data, re.IGNORECASE) != None):
return self.brands.get_samsung()
elif (re.search('positivo', raw_data, re.IGNORECASE) != None):
return self.brands.get_positivo()
elif (re.search('lenovo', raw_data, re.IGNORECASE) != None):
return self.brands.get_lenovo()
elif (re.search('lg', raw_data, re.IGNORECASE) != None):
return self.brands.get_lg()
def normalize_processor(self, raw_data):
# ['Intel Core i3', 'Intel Core i5', 'Intel Core i7', 'Intem Pentium Quad Core', 'Intel Baytrail', 'AMD Dual Core', 'Item Atom', 'Intel Core M', 'Intel Celeron']
# remove erros de enconding (ex: \u84d2)
raw_data = re.sub('\\\u\w\w\w\w', '', raw_data)
if (re.search("i3", raw_data, re.IGNORECASE) != None):
return self.processors.get_i3()
elif (re.search("i5", raw_data, re.IGNORECASE) != None):
return self.processors.get_i5()
elif (re.search("i7", raw_data, re.IGNORECASE) != None):
return self.processors.get_i7()
elif (re.search("Pentium", raw_data, re.IGNORECASE) != None):
return self.processors.get_pentium_quad()
elif (re.search("byt|baytrail", raw_data, re.IGNORECASE) != None):
return self.processors.get_baytrail()
elif (re.search("amd.+dual core", raw_data, re.IGNORECASE) != None):
return self.processors.get_amd_dual()
elif (re.search("atom", raw_data, re.IGNORECASE) != None):
return self.processors.get_atom()
elif (re.search("Intel.+Core.+M", raw_data, re.IGNORECASE) != None):
return self.processors.get_core_m()
elif (re.search("Celeron", raw_data, re.IGNORECASE) != None):
return self.processors.get_celeron()
elif (re.search("arm", raw_data, re.IGNORECASE) != None):
return self.processors.get_arm_a9()
elif (re.search("samsung", raw_data, re.IGNORECASE) != None):
return self.processors.get_samsung()
def get_storage_capacity(self, raw_data):
if (re.search('2TB|2 TB', raw_data, re.IGNORECASE) != None):
return self.storages.get_2tb()
elif (re.search('1TB|1 TB', raw_data, re.IGNORECASE) != None):
return self.storages.get_1tb()
elif (re.search('750 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_750()
elif (re.search('640 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_640()
elif (re.search('500 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_500()
elif (re.search('320GB|320 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_320()
elif (re.search('256GB|256 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_256()
elif (re.search('160GB|160 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_160()
elif (re.search('128GB|128 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_128()
elif (re.search('80GB|80 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_80()
elif (re.search('64GB|64 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_64()
elif (re.search('32GB|32 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_32()
elif (re.search('16GB|16 GB', raw_data, re.IGNORECASE) != None):
return self.storages.get_16()
class DataExtractor():
DOWNLOAD_DELAY = 5
def __init__(self, response, url):
self.response = response
self.url = url
self.processors = Processors()
self.brands = Brands()
self.memory = Memory()
def parse(self):
namespaces = {'re': "http://exslt.org/regular-expressions"}
data = {}
data["store"] = "novo_mundo"
data["name"] = self.response.xpath(
'//div[@class="productName"]//text()')
print self.response.xpath('//div[@class="productName"]')
data['name'] = self.validate_field(data, 'name')
data['url'] = self.url
data['price'] = self.response.xpath(
'//strong[@class="skuBestPrice"]//text()')[0].split(' ')[1]
data['price'] = self.normalize_price(data['price'])
data["available"] = data["price"] != None and data["price"] != 0.0
data['processor'] = self.response.xpath(
'//td[@class="value-field" and @class="Processador"]//text()')
data['processor'] = self.normalize_processor(
self.validate_field(data, 'processor'))
data['brand'] = self.normalize_brand(data['name'])
data['ram_memory'] = self.response.xpath(
'//td[@class="value-field" and @class="Memoria"]//text()')
data['ram_memory'] = self.normalize_memory(
self.validate_field(data, 'ram_memory'))
data['sku'] = self.response.xpath(
'//div[@class="skuReference"]//text()')[0]
data['sku'] = self.validate_field(data, 'sku')
hd = self.response.xpath(
'//td[@class="value-field" and @class="HD"]//text()')[0]
data['display_size'] = self.response.xpath(
'//td[@class="value-field" and @class="Tela"]//text()')
data['display_size'] = data['display_size'].split('\n')[0].split(
' ')[1]
data['display_size'] = self.validate_field(data, 'display_size')
return data
def validate_field(self, data, field):
return (data[field][0].strip() if (len(data[field]) > 0) else "")
# normalize storage
def normalize_memory(self, raw_data):
if (re.search('16', raw_data, re.IGNORECASE) != None):
return self.memory.get_16GB()
elif (re.search('14', raw_data, re.IGNORECASE) != None):
return self.memory.get_14GB()
elif (re.search('12', raw_data, re.IGNORECASE) != None):
return self.memory.get_12GB()
elif (re.search('10', raw_data, re.IGNORECASE) != None):
return self.memory.get_10GB()
elif (re.search('8', raw_data, re.IGNORECASE) != None):
return self.memory.get_8GB()
elif (re.search('6', raw_data, re.IGNORECASE) != None):
return self.memory.get_6GB()
elif (re.search('4', raw_data, re.IGNORECASE) != None):
return self.memory.get_4GB()
elif (re.search('2', raw_data, re.IGNORECASE) != None):
return self.memory.get_2GB()
elif (re.search('1', raw_data, re.IGNORECASE) != None):
return self.memory.get_1GB()
def normalize_price(self, raw_data):
try:
raw_data = raw_data[0].strip() if (len(raw_data) > 0) else ""
raw_data = raw_data.replace(".", "").replace(",", ".")
return float(raw_data)
except ValueError:
return 0.0
def normalize_brand(self, raw_data):
if (re.search("dell", raw_data, re.IGNORECASE) != None):
return self.brands.get_dell()
elif (re.search('asus', raw_data, re.IGNORECASE) != None):
return self.brands.get_asus()
elif (re.search('apple', raw_data, re.IGNORECASE) != None):
return self.brands.get_apple()
elif (re.search('acer', raw_data, re.IGNORECASE) != None):
return self.brands.get_acer()
elif (re.search('samsung', raw_data, re.IGNORECASE) != None):
return self.brands.get_samsung()
elif (re.search('positivo', raw_data, re.IGNORECASE) != None):
return self.brands.get_positivo()
elif (re.search('lenovo', raw_data, re.IGNORECASE) != None):
return self.brands.get_lenovo()
elif (re.search('lg', raw_data, re.IGNORECASE) != None):
return self.brands.get_lg()
def normalize_processor(self, raw_data):
# remove erros de enconding (ex: \u84d2)
raw_data = re.sub('\\\u\w\w\w\w', '', raw_data)
if (re.search("i3", raw_data, re.IGNORECASE) != None):
return self.processors.get_i3()
elif (re.search("i5", raw_data, re.IGNORECASE) != None):
return self.processors.get_i5()
elif (re.search("i7", raw_data, re.IGNORECASE) != None):
return self.processors.get_i7()
elif (re.search("Pentium", raw_data, re.IGNORECASE) != None):
return self.processors.get_pentium_quad()
elif (re.search("byt|baytrail", raw_data, re.IGNORECASE) != None):
return self.processors.get_baytrail()
elif (re.search("amd.+dual core", raw_data, re.IGNORECASE) != None):
return self.processors.get_amd_dual()
elif (re.search("atom", raw_data, re.IGNORECASE) != None):
return self.processors.get_atom()
elif (re.search("Intel.+Core.+M", raw_data, re.IGNORECASE) != None):
return self.processors.get_core_m()
elif (re.search("Celeron", raw_data, re.IGNORECASE) != None):
return self.processors.get_celeron()
elif (re.search("arm", raw_data, re.IGNORECASE) != None):
return self.processors.get_arm_a9()
elif (re.search("samsung", raw_data, re.IGNORECASE) != None):
return self.processors.get_samsung()
class DataExtractor():
def __init__(self, response, url):
self.response = response
self.url = url
self.processors = Processors()
self.brands = Brands()
self.memory = Memory()
def parse(self):
namespaces = {'re': "http://exslt.org/regular-expressions"}
data = {}
data["store"] = "eletrosom"
data['name'] = self.response.xpath(
'//div[@class="about"]/div[@class="meta"]/h1//text()')
data['name'] = self.validate_field(data, 'name')
data['url'] = self.url
data['price'] = self.response.xpath(
'//span[@class="regular-price"]/span/strong//text()')[0]
data['price'] = data['price'].split('$')[1]
data['price'] = self.normalize_price(data['price'])
if not self.response.xpath('//div[@class="indisponivel"]'):
data['available'] = True
else:
data['available'] = False
data['processor'] = self.response.xpath(
'//td[text()="Processador"]/following-sibling::td//text()')
data["processor"] = self.normalize_processor(
self.validate_field(data, "processor"))
data['brand'] = self.normalize_brand(data['name'])
data['ram_memory'] = self.response.xpath(
'//td[contains(text(), "Mem") and contains(text(), "ria") or contains(text(), "RAM")]/following-sibling::td//text()',
namespaces={'re': 'http://exslt.org/regular-expressions'})
data["ram_memory"] = self.normalize_memory(
self.validate_field(data, "ram_memory"))
data['sku'] = self.response.xpath('//p[@class="code"]//text()')
data['sku'] = data['sku'][0].split(':')[1].split('/')[0].strip()
data["disco"] = {}
disco = self.response.xpath(
'//td[contains(text(), "Disco") and contains(text(), "gido")]//text()'
)[0]
if disco:
if (re.search('ssd', disco, re.IGNORECASE) != None):
ssd = self.response.xpath(
'//td[contains(text(), "Disco") and contains(text(), "gido")]/following-sibling::td//text()'
)
hd = ''
else:
ssd = ''
hd = self.response.xpath(
'//td[contains(text(), "Disco") and contains(text(), "gido")]/following-sibling::td//text()'
)
data['disco']['hd'] = hd
data['disco']['ssd'] = ssd
if self.response.xpath('//td[text()="Tela"]'):
data["display_size"] = self.response.xpath(
'//td[text()="Tela"]/following-sibling::td//text()')
else:
data["display_size"] = self.response.xpath(
'//td[contains(text(), "Tamanho") and contains(text(), "Tela")]/following-sibling::td//text()'
)
data["display_size"] = self.validate_field(data, "display_size")
return data
def validate_field(self, data, field):
return (data[field][0].strip() if (len(data[field]) > 0) else "")
# normalize storage
def normalize_memory(self, raw_data):
if (re.search('1GB|1 GB', raw_data, re.IGNORECASE) != None):
return self.memory.get_1GB()
elif (re.search('2GB|2 GB', raw_data, re.IGNORECASE) != None):
return self.memory.get_2GB()
elif (re.search('4GB|4 GB', raw_data, re.IGNORECASE) != None):
return self.memory.get_4GB()
elif (re.search('6GB|6 GB', raw_data, re.IGNORECASE) != None):
return self.memory.get_6GB()
elif (re.search('8GB|8 GB', raw_data, re.IGNORECASE) != None):
return self.memory.get_8GB()
elif (re.search('10GB|10 GB', raw_data, re.IGNORECASE) != None):
return self.memory.get_10GB()
elif (re.search('12GB|12 GB', raw_data, re.IGNORECASE) != None):
return self.memory.get_12GB()
elif (re.search('14GB|14 GB', raw_data, re.IGNORECASE) != None):
return self.memory.get_14GB()
elif (re.search('16GB|16 GB', raw_data, re.IGNORECASE) != None):
return self.memory.get_16GB()
def normalize_price(self, raw_data):
try:
# raw_data = raw_data[0].strip() if (len(raw_data) > 0) else ""
raw_data = raw_data.replace(".", "").replace(",", ".")
return float(raw_data)
except ValueError:
return 0.0
def normalize_brand(self, raw_data):
if (re.search("dell", raw_data, re.IGNORECASE) != None):
return self.brands.get_dell()
elif (re.search('asus', raw_data, re.IGNORECASE) != None):
return self.brands.get_asus()
elif (re.search('apple', raw_data, re.IGNORECASE) != None):
return self.brands.get_apple()
elif (re.search('acer', raw_data, re.IGNORECASE) != None):
return self.brands.get_acer()
elif (re.search('samsung', raw_data, re.IGNORECASE) != None):
return self.brands.get_samsung()
elif (re.search('positivo', raw_data, re.IGNORECASE) != None):
return self.brands.get_positivo()
elif (re.search('lenovo', raw_data, re.IGNORECASE) != None):
return self.brands.get_lenovo()
elif (re.search('lg', raw_data, re.IGNORECASE) != None):
return self.brands.get_lg()
def normalize_processor(self, raw_data):
# remove erros de enconding (ex: \u84d2)
raw_data = re.sub('\\\u\w\w\w\w', '', raw_data)
if (re.search("i3", raw_data, re.IGNORECASE) != None):
return self.processors.get_i3()
elif (re.search("i5", raw_data, re.IGNORECASE) != None):
return self.processors.get_i5()
elif (re.search("i7", raw_data, re.IGNORECASE) != None):
return self.processors.get_i7()
elif (re.search("Pentium", raw_data, re.IGNORECASE) != None):
return self.processors.get_pentium_quad()
elif (re.search("byt|baytrail", raw_data, re.IGNORECASE) != None):
return self.processors.get_baytrail()
elif (re.search("amd.+dual core", raw_data, re.IGNORECASE) != None):
return self.processors.get_amd_dual()
elif (re.search("atom", raw_data, re.IGNORECASE) != None):
return self.processors.get_atom()
elif (re.search("Intel.+Core.+M", raw_data, re.IGNORECASE) != None):
return self.processors.get_core_m()
elif (re.search("Celeron", raw_data, re.IGNORECASE) != None):
return self.processors.get_celeron()
elif (re.search("arm", raw_data, re.IGNORECASE) != None):
return self.processors.get_arm_a9()
elif (re.search("samsung", raw_data, re.IGNORECASE) != None):
return self.processors.get_samsung()
class DataExtractor():
def __init__(self, response, url):
self.response = response
self.url = url
self.processors = Processors()
self.brands = Brands()
self.memory = Memory()
#{ _id, available, brand, color, display_feature, display_size, graphics_processor_name, graphics_processor, name, operating_system, price, processor, ram_memory, sku, screen_resolution, storage, storage_type, url, img_url}
# TODO: display_feature, display_size, graphics_processor_name, graphics_processor, operating_system, screen_resolution, storage_type, img_url
def parse(self):
data = {}
# produtos da havan
data['store'] = "havan"
# nome do produto
data['name'] = self.response.findAll("",
{"class": "product-qd-v1-name"})
data['name'] = self.validate_field(data, 'name')
# url como variavel global da classe
data['url'] = self.url
# preco do produto
data['price'] = self.response.findAll("", {"class": "skuBestPrice"})
data['price'] = self.normalize_price(data['price'])
# disponibilidade: nas casas bahia, se o produto possuir preco, o produto esta disponivel
data['available'] = data['price'] != None and data['price'] != 0.0
# processador
data['processor'] = self.response.findAll(
"", {"class": "value-field Modelo-do-Processador"})
data['processor'] = self.normalize_processor(
self.validate_field(data, 'processor'))
# marca
data['brand'] = self.normalize_brand(data['name'])
# memoria ram
data['ram_memory'] = self.response.findAll(
"td", {"class": "value-field Memoria-Fisica-Disponivel"})
data['ram_memory'] = self.normalize_memory(
self.validate_field(data, 'ram_memory'))
# sku para identificacao
data['sku'] = self.response.findAll("", {"class": "skuReference"})
data['sku'] = self.validate_field(data, 'sku')
# armazenamento (SSD/HD)
data['storage'] = self.response.findAll(
"", {"class": "value-field Capacidade-do-Disco-Rigido-HD-"})
data['storage'] = self.validate_field(data, 'storage')
#ssd = self.response.findAll("", {"class": "value-field Unidade-de-Estado-Solida-SSD-"})
# tamanho de tela
display = self.response.findAll(
"td", {"class": "value-field Tamanho-da-Tela"})
data['display_size'] = self.normalize_display(display) if (
len(display) > 0) else ""
return data
def validate_field(self, data, field):
return (data[field][0].get_text().strip() if
(len(data[field]) > 0) else "")
def normalize_memory(self, raw_data):
if (re.search('16', raw_data, re.IGNORECASE) != None):
return self.memory.get_16GB()
elif (re.search('12', raw_data, re.IGNORECASE) != None):
return self.memory.get_12GB()
elif (re.search('14', raw_data, re.IGNORECASE) != None):
return self.memory.get_14GB()
elif (re.search('10', raw_data, re.IGNORECASE) != None):
return self.memory.get_10GB()
elif (re.search('8', raw_data, re.IGNORECASE) != None):
return self.memory.get_8GB()
elif (re.search('6', raw_data, re.IGNORECASE) != None):
return self.memory.get_6GB()
elif (re.search('4', raw_data, re.IGNORECASE) != None):
return self.memory.get_4GB()
elif (re.search('2', raw_data, re.IGNORECASE) != None):
return self.memory.get_2GB()
elif (re.search('1', raw_data, re.IGNORECASE) != None):
return self.memory.get_1GB()
def normalize_price(self, raw_data):
try:
# transforma 1.000,00 em 1000.00
raw_data = raw_data[0].get_text() if (len(raw_data) > 0) else ""
raw_data = raw_data.replace('.',
'').replace(',',
'.').replace('R$', '')
return float(raw_data)
except ValueError:
return 0.0
def normalize_display(self, raw_data):
try:
raw_data = raw_data[0].get_text() if (len(raw_data) > 0) else ""
raw_data = raw_data.replace('"', '')
return raw_data
except ValueError:
return ""
def normalize_brand(self, raw_data):
# ["Samsung", "Asus", "Acer", "Dell", "Apple", "Positivo", "LG", "Lenovo"]
if (re.search('dell', raw_data, re.IGNORECASE) != None):
return self.brands.get_dell()
elif (re.search('asus', raw_data, re.IGNORECASE) != None):
return self.brands.get_asus()
elif (re.search('apple', raw_data, re.IGNORECASE) != None):
return self.brands.get_apple()
elif (re.search('acer', raw_data, re.IGNORECASE) != None):
return self.brands.get_acer()
elif (re.search('samsung', raw_data, re.IGNORECASE) != None):
return self.brands.get_samsung()
elif (re.search('positivo', raw_data, re.IGNORECASE) != None):
return self.brands.get_positivo()
elif (re.search('lenovo', raw_data, re.IGNORECASE) != None):
return self.brands.get_lenovo()
elif (re.search('lg', raw_data, re.IGNORECASE) != None):
return self.brands.get_lg()
def normalize_processor(self, raw_data):
# ['Intel Core i3', 'Intel Core i5', 'Intel Core i7', 'Intem Pentium Quad Core', 'Intel Baytrail', 'AMD Dual Core', 'Item Atom', 'Intel Core M', 'Intel Celeron']
# remove erros de enconding (ex: \u84d2)
raw_data = re.sub('\\\u\w\w\w\w', '', raw_data)
if (re.search("i3", raw_data, re.IGNORECASE) != None):
return self.processors.get_i3()
elif (re.search("i5", raw_data, re.IGNORECASE) != None):
return self.processors.get_i5()
elif (re.search("i7", raw_data, re.IGNORECASE) != None):
return self.processors.get_i7()
elif (re.search("Pentium", raw_data, re.IGNORECASE) != None):
return self.processors.get_pentium_quad()
elif (re.search("byt|baytrail", raw_data, re.IGNORECASE) != None):
return self.processors.get_baytrail()
elif (re.search("amd.+dual core", raw_data, re.IGNORECASE) != None):
return self.processors.get_amd_dual()
elif (re.search("atom", raw_data, re.IGNORECASE) != None):
return self.processors.get_atom()
elif (re.search("Intel.+Core.+M", raw_data, re.IGNORECASE) != None):
return self.processors.get_core_m()
elif (re.search("Celeron", raw_data, re.IGNORECASE) != None):
return self.processors.get_celeron()
elif (re.search("arm", raw_data, re.IGNORECASE) != None):
return self.processors.get_arm_a9()
elif (re.search("samsung", raw_data, re.IGNORECASE) != None):
return self.processors.get_samsung()
