def all_listings_page(self, page_no):
"""Search result details for a single page"""
soup = self.soup if page_no == 1 else self._get_soup(page_no)
listings = soup.find_all(name='li', attrs={'class': 'srp clearfix '}) + \
soup.find_all(name='li', attrs={'class': 'srp clearfix premium-listing premium-listing--branded '})
df = pd.DataFrame()
df['listing_id'] = [i['data-listing-id'] for i in listings]
df['listing_price'] = [currency_to_num(
i.find(name='a',
attrs={'class': 'listing-results-price text-price'}).text.strip().split('\n')[0]) for i in listings]
df['price_modifier'] = [text_or_none(i.find(name='span', attrs={'class': 'price-modifier'})) for i in listings]
df['address'] = [i.find(name='a', attrs={'class': 'listing-results-address'}).text for i in listings]
df['summary'] = [i.find(name='a', attrs={'style': 'text-decoration:underline;'}).text for i in listings]
df['num_beds'] = [text_or_none(i.find(name='span', attrs={'class': 'num-icon num-beds'}), data_type=str)
for i in listings]
df['num_baths'] = [text_or_none(i.find(name='span', attrs={'class': 'num-icon num-baths'}), data_type=str)
for i in listings]
df['num_receptions'] = [text_or_none(
i.find(name='span', attrs={'class': 'num-icon num-reception'}), data_type=str) for i in listings]
df['description'] = [i.find(name='p').text.strip() for i in listings]
df['listing_date'] = [to_datetime(
i.find_all(name='small')[-1].text.split('Listed on \n')[1].split('\n')[0].strip()) for i in listings]
df['estate_agent'] = [i.find(name='p', attrs={'class': 'top-half listing-results-marketed'}).span.text
for i in listings]
df['station1'] = [get_station_name(i.find_all('li', attrs={'class': 'clearfix'}), 0) for i in listings]
df['distance1'] = [dist_to_num(i.find_all('li', attrs={'class': 'clearfix'})[0]) for i in listings]
df['station2'] = [get_station_name(i.find_all('li', attrs={'class': 'clearfix'}), 1) for i in listings]
df['distance2'] = [dist_to_num(i.find_all('li', attrs={'class': 'clearfix'})[1]) for i in listings]
df['date_generated'] = datetime.now()
return df
def sales_history(self, dataframe=False):
# historical listings available at https://www.zoopla.co.uk/property-history/{listing_id}
history = {
'date': [
i.text for i in self.soup.find_all(
name='span', attrs={'class': 'pdp-history__date'})
],
'status': [
i.text for i in self.soup.find_all(
name='span', attrs={'class': 'pdp-history__status'})
],
'price': [
currency_to_num(i.text.replace('View listing', ''))
for i in self.soup.find_all(
name='span', attrs={'class': 'pdp-history__price'})
],
'listing_id': []
}
for listing in self.soup.find_all(
name='span', attrs={'class': 'pdp-history__price'}):
try:
history['listing_id'].append(listing.a['href'].split('/')[-1])
except TypeError:
history['listing_id'].append(np.nan)
return pd.DataFrame(history) if dataframe else history
def _price_history(self):
dates = [i.text for i in self.soup.find_all(name='span', attrs={'class': 'dp-price-history__item-date'})]
prices = [currency_to_num(i.text) for i in
self.soup.find_all(name='span', attrs={'class': 'dp-price-history__item-price'})]
details = [i.text.strip() for i in
self.soup.find_all(name='span', attrs={'class': 'dp-price-history__item-detail'})]
return {'date': dates, 'price': prices, 'detail': details}
def property_value(self):
values = [
currency_to_num(i.text)
for i in self.soup.find_all(name='p',
attrs={'class': 'pdp-estimate__price'})
]
ranges = [
currency_to_num(i) for i in list(
chain(*[
i.text[7:].split(' - ') for i in self.soup.find_all(
name='p', attrs={'class': 'pdp-estimate__range'})
]))
]
try:
conf = float(
self.soup.find(name='span',
attrs={
'class': 'pdp-confidence-rating__copy'
}).text.strip().split('%')[0])
except AttributeError:
conf = np.nan
if len(values) == 1:
values.append(np.nan)
ranges.extend([np.nan, np.nan])
elif not values:
values.extend([np.nan, np.nan])
ranges.extend([np.nan, np.nan, np.nan, np.nan])
return {
'buy': {
'value': values[0],
'lower_bound': ranges[0],
'upper_bound': ranges[1]
},
'rent': {
'value': values[1],
'lower_bound': ranges[2],
'upper_bound': ranges[3]
},
'confidence': conf
}
def value_change(self):
period = [
i.text for i in self.soup.find_all(
name='span', attrs={'class': 'pdp-value-change__label'})
]
changes = [
currency_to_num(i.text) for i in self.soup.find_all(
name='span', attrs={'class': 'pdp-value-change__value'})
]
diffs = [
float(i.text.replace('%', '')) for i in self.soup.find_all(
name='span', attrs={'class': 'pdp-value-change__difference'})
]
df = pd.DataFrame(list(zip(period, changes, diffs)),
columns=['period', 'value_change', 'perc_change'])
df['value'] = self.property_value()['buy']['value'] / (
1 + df.perc_change / 100)
df['value'] = df.value.apply(myround)
df = df[['period', 'value', 'value_change', 'perc_change']]
return df