def parse_hotel(
self, response: scrapy.http.response.html.HtmlResponse, name, text_eng, text_hr
):
if text_hr is None:
text_hr = "".join(
response.xpath("//dd[@class='hotel-description']//text()").extract()
)
new_link = response._get_url().replace("esky.hr/hoteli", "esky.com/hotels")
return scrapy.Request(
new_link,
callback=self.parse_hotel,
cb_kwargs={"name": name, "text_eng": None, "text_hr": text_hr},
)
else:
text_eng = "".join(
response.xpath("//dd[@class='hotel-description']//text()").extract()
)
if text_hr != text_eng:
return {
"name": name,
"text_eng": text_eng.strip(),
"text_hr": text_hr.strip(),
}
else:
return None
def parse_cinema(self,
response: scrapy.http.response.html.HtmlResponse) -> dict:
titles = self.get_titles(response)
movies_times = list()
for movie in response.xpath(
'//div[@class="table-responsive-wrapper"]'):
times = list()
for showtime in movie.xpath('.//tr'):
times += showtime.xpath('./td/text()').getall()
movies_times.append(times)
cinema = Cinema(
name=self.get_name(response),
description=self.get_description(response),
address=Address(street=self.get_street(response),
postal_code=self.get_postal_code(response),
district=self.get_district(response),
city='Berlin',
country='Germany'),
contact=Contact(telephone=self.get_telephone(response)),
prices=self.get_prices(response),
shows=self.create_shows(titles, movies_times))
self.logger.info(f'Scraped cinema: {cinema.name}')
yield cinema.to_dict()
def parse(self, response: scrapy.http.response.html.HtmlResponse) -> typing.Iterator[dict]:
# FIXME: when xpath has <1 match, .get() cheerfully returns None.
# FIXME: when xpath has >1 match, .get() cheerfully returns the first.
# How do I get exceptions for both cases?
for quote_etree in response.xpath('//*[@itemscope]'):
yield {
'author': quote_etree.xpath('.//*[@itemprop="author"]/text()').get(),
'text': quote_etree.xpath('.//*[@itemprop="text"]/text()').get(),
'tags': quote_etree.xpath('.//*[@class="tag"]/text()').getall()}
# Recursively descend the next page.
# Follow the "next page" link
for next_url in response.xpath('//li[@class="next"]/a/@href').getall():
yield scrapy.Request(
response.urljoin(next_url),
callback=self.parse)
def parse_eng(self, response: scrapy.http.response.html.HtmlResponse,
text_hr: str):
text_eng = "".join(
response.xpath(
"//dd[@class='hotel-description']//text()").extract())
if text_hr != text_eng:
with open("c_output_hr.txt", "a", encoding="utf-8") as f:
f.write(text_hr.replace("\t", "").replace("\n", ""))
with open("c_output_en.txt", "a", encoding="utf-8") as f:
f.write(text_eng.replace("\t", "").replace("\n", ""))
def parse(self, response: scrapy.http.response.html.HtmlResponse):
text_hr = "".join(
response.xpath(
"//dd[@class='hotel-description']//text()").extract())
if text_hr == "":
return None
new_link = response._get_url().replace("esky.hr/hoteli",
"esky.com/hotels")
return scrapy.Request(new_link,
callback=self.parse_eng,
cb_kwargs={"text_hr": text_hr})
def parse(self, response: scrapy.http.response.html.HtmlResponse):
selectors = response.xpath('//div[@class="controls"]/select/option')
# current movies:
base_kinodetail_url = 'https://www.berlin.de/kino/_bin/kinodetail.php/'
hrefs = [
base_kinodetail_url + sel.attrib['value'] for sel in selectors
if is_positiveinteger(sel.attrib['value'])
]
for href in hrefs:
self.logger.info(f'Scraping: {href}')
yield response.follow(href, self.parse_cinema)
def get_telephone(self,
response: scrapy.http.response.html.HtmlResponse) -> str:
return response.xpath(
'//span[contains(text(), "Telefon")]/following-sibling::span/text()'
).get()
def get_district(self,
response: scrapy.http.response.html.HtmlResponse) -> str:
return response.xpath('//span[@class="locality"]/text()').get()
def get_postal_code(
self, response: scrapy.http.response.html.HtmlResponse) -> str:
return response.xpath('//span[@class="postal-code"]/text()').get()
def get_street(self,
response: scrapy.http.response.html.HtmlResponse) -> str:
return response.xpath('//span[@class="street-address"]/text()').get()
def get_description(
self, response: scrapy.http.response.html.HtmlResponse) -> str:
return response.xpath('//div[@class="kinodetail echo"]/p/text()').get()
def get_prices(
self,
response: scrapy.http.response.html.HtmlResponse) -> List[str]:
return response.xpath(
'//section[@class="infoblock oeffnungszeiten"]/div/*/text()'
).getall()
def analyze_website(self, website_url: str,
response: scrapy.http.response.html.HtmlResponse):
gallery_page_matched_by_url = False
page_with_gallery_matched_by_url = False
can_be_page_with_gallery = True
number_of_imgs_matched_by_a_href = 0
number_of_images_by_img_src = 0
img_src_values = response.xpath("//img/@src").extract()
a_href_values = response.xpath("//a/@href").extract()
some_config_matched = False
for config in self.configs:
if not re.match(f'.*{config.domain}.*', website_url):
continue
some_config_matched = True
if config.pageWithGalleryUrlMatchesRegexp:
rexp_match = re.match(config.pageWithGalleryUrlMatchesRegexp,
website_url)
page_with_gallery_matched_by_url = page_with_gallery_matched_by_url or rexp_match is not None
if config.galleryUrlMatchesRegexp:
rexp_match = re.match(config.galleryUrlMatchesRegexp,
website_url)
gallery_page_matched_by_url = gallery_page_matched_by_url or rexp_match is not None
if config.pageWithGalleryUrlHasToMatchRegexp:
rexp_match = re.match(
config.pageWithGalleryUrlHasToMatchRegexp, website_url)
can_be_page_with_gallery = rexp_match is not None
if config.pageWithGalleryContainsImgSrcRegexp:
for img_src_val in img_src_values:
if re.match(config.pageWithGalleryContainsImgSrcRegexp,
img_src_val):
number_of_images_by_img_src += 1
if config.pageWithGalleryContainsAnchorHrefRegexp:
for a_href_val in a_href_values:
if re.match(config.pageWithGalleryContainsAnchorHrefRegexp,
a_href_val):
number_of_imgs_matched_by_a_href += 1
has_match_for_page_with_gallery_by_imgs = number_of_imgs_matched_by_a_href >= 1 \
or number_of_images_by_img_src >= 1
has_match_for_page_with_gallery_by_imgs = has_match_for_page_with_gallery_by_imgs \
and can_be_page_with_gallery
has_match = page_with_gallery_matched_by_url \
or gallery_page_matched_by_url \
or has_match_for_page_with_gallery_by_imgs
page_analysis_results = PageAnalysisResults()
page_analysis_results[
'number_of_images_by_a_href'] = number_of_imgs_matched_by_a_href
page_analysis_results[
'number_of_images_by_img_src'] = number_of_images_by_img_src
page_analysis_results['has_match'] = has_match
page_analysis_results[
'url_matched_for_gallery_page'] = gallery_page_matched_by_url
page_analysis_results[
'url_matched_for_page_with_gallery'] = page_with_gallery_matched_by_url
return page_analysis_results
