def parse_parts2(self, response):
log.msg("\tparse_parts time: %s" % int(time.time()), level=log.DEBUG)
ua = response.request.headers['User-Agent']
log.msg("\tua: %s" % ua, level=log.DEBUG)
for part in response.css('table.parts > tbody > tr'):
il = ItemLoader(item=CarPart(), selector=part)
il.add_xpath('shop_city', "td[@class='shop']/a/text()")
il.add_xpath('shop_name', "td[@class='shop']/a/strong/text()")
shop_url = il.get_xpath("td[@class='shop']/a/@href", TakeFirst())
photo_url = il.get_xpath("td[@class='photo']/a/@href", TakeFirst())
il.add_value('shop_url', urljoin(self.main_url, shop_url))
il.add_value('ext_link', urljoin(self.main_url, photo_url))
il.add_xpath('info', "td[@class='info']//text()")
il.add_xpath('price', "td[@class='price']//text()")
il.add_value('brand', response.meta.get('brand'))
il.add_value('model', response.meta.get('model'))
il.add_value('car_part', response.meta.get('car_part'))
il.add_value('category', response.meta.get('category'))
item = il.load_item()
if item.is_valid():
yield item
def Loader_content(response):
l = ItemLoader(item={}, response=response)
sub_title = ''
if len(l.get_xpath('//*[@class="pagenow"]/text()')) >= 1:
sub_title += '-' + l.get_xpath('//*[@class="pagenow"]/text()')[0]
l.add_value(
'title',
l.get_xpath('//*[@class="b_list-1a-1c"]/text()')[0] + sub_title)
l.add_value('src_url', response.url)
content_img = l.get_xpath('//*[@class="content-img"]/p/img/@src')
l.add_value('content', content_img)
l.add_value('image_urls', content_img)
print('正下载图片：', content_img)
time.sleep(len(content_img))
return l.load_item()
def parse(self, response):
item = PdfItem()
loader = ItemLoader(response=response)
pdf_path = '//*[contains(text(), "[PDF]")]'
pdf_url_path = '%s//following-sibling::*' % pdf_path
item['url'] = loader.get_xpath('%s' % pdf_url_path)
item['title'] = loader.get_xpath('%s/text()' % pdf_url_path,
TakeFirst())
summary_path = '%s//parent::*//parent::*/*[@class="s"]/*' % pdf_url_path
description_path = '%s/*[@class="st"]/*' % summary_path
item['description'] = loader.get_xpath(
'%s/text()|%s/*/text()' % (description_path, description_path))
similar_path = '%s/*[contains(@class, "f")]//a[contains(@href, "q=related:")]' % summary_path
# similar_url = loader.get_xpath('%s/@href' % similar_path, TakeFirst())
# yield Request(
# url=urlparse.urljoin(response.url, similar_url),
# callback=self.parse,
# meta=response.meta,
# dont_filter=True
# )
#
# next_path = '//*[@class="pn"]'
# next_url = loader.get_xpath('%s/@href' % next_path, TakeFirst())
# yield Request(
# url=urlparse.urljoin(response.url, next_url),
# callback=self.parse,
# meta=response.meta,
# dont_filter=True
# )
pdf_url = item['url']
print item
if pdf_url:
pdf_filename = os.path.basename(pdf_url)
pdf_filepath = '%s/%s/%s' % (DOWNLOAD_DIR, SEARCH_TERM,
pdf_filename)
if self.download_files:
self.download_file(pdf_url, pdf_filepath, response.url)
yield item
def Loader_content(response):
l = ItemLoader(item={}, response=response)
l.add_css('title', '.k_jianjie-3a-1-name::text')
l.add_value('date', l.get_xpath('//*[@class="k_jianjie-3a-2b"]/text()')[2])
#l.add_value('url',_response.url[len(self._scheme+"//"+self.allowed_domains[0]):])
l.add_css('down', '.k_jianjie-3a-5down::text', TrimAll())
conver_img = l.get_xpath('//*[@id="k_jianjie-2b"]/a/img/@src')
content_img = l.get_xpath('//*[@class="content"]/p/img/@src')
l.add_value('src_url', response.url)
l.add_value('preview', conver_img)
l.add_value('content', content_img)
l.add_value('image_urls', conver_img + content_img)
print('正下载图片：', conver_img + content_img)
#time.sleep(len(conver_img+content_img))
return l.load_item()
def parse(self, response):
item = PdfItem()
loader = ItemLoader(response=response)
pdf_path = '//*[contains(text(), "[PDF]")]'
pdf_url_path = '%s//following-sibling::*' % pdf_path
item['url'] = loader.get_xpath('%s' % pdf_url_path)
item['title'] = loader.get_xpath('%s/text()' % pdf_url_path, TakeFirst())
summary_path = '%s//parent::*//parent::*/*[@class="s"]/*' % pdf_url_path
description_path = '%s/*[@class="st"]/*' % summary_path
item['description'] = loader.get_xpath('%s/text()|%s/*/text()' % (description_path, description_path))
similar_path = '%s/*[contains(@class, "f")]//a[contains(@href, "q=related:")]' % summary_path
# similar_url = loader.get_xpath('%s/@href' % similar_path, TakeFirst())
# yield Request(
# url=urlparse.urljoin(response.url, similar_url),
# callback=self.parse,
# meta=response.meta,
# dont_filter=True
# )
#
# next_path = '//*[@class="pn"]'
# next_url = loader.get_xpath('%s/@href' % next_path, TakeFirst())
# yield Request(
# url=urlparse.urljoin(response.url, next_url),
# callback=self.parse,
# meta=response.meta,
# dont_filter=True
# )
pdf_url = item['url']
print item
if pdf_url:
pdf_filename = os.path.basename(pdf_url)
pdf_filepath = '%s/%s/%s' % (DOWNLOAD_DIR, SEARCH_TERM, pdf_filename)
if self.download_files:
self.download_file(pdf_url, pdf_filepath, response.url)
yield item
def parse(self, response):
l = ItemLoader(item=CoserItem(), response=response)
l.add_xpath('name', "//h1[@class='js-post-title']/text()")
l.add_xpath('info', "//div[@class='post__info']/div[@class='post__type post__info-group']/span/text()")
urls = l.get_xpath('//img[@class="detail_std detail_clickable"]/@src')
urls = [url.replace('/w650', '') for url in urls]
l.add_value('image_urls', urls)
l.add_value('url', response.url)
return l.load_item()
def parase_item(self,response):
l = ItemLoader(item = CoserspiderItem(),response=response)
l.add_xpath('name', "//h1[@class='js-post-title']/text()")
l.add_xpath('info',"//div[@class='post__info']/div[@class='post__type post__info-group']/span/text()")
urls = l.get_xpath('//img[@class="detail_std detail_clickable"]/@src')
urls = [url.replace('/w650','') for url in urls]
l.add_xpath('image_urls',urls)
l.add_xpath('url',response.url)
return l.load_item()
def Loader_index(self, item_selector):
l = ItemLoader(item={}, selector=item_selector)
conver_img = l.get_xpath('.//*[@class="lz_img"]/img/@src')
l.add_xpath('title', './/*[@class="k_list-lb-2"]/div[1]/a[1]/text()')
l.add_xpath('url', './/*[@class="k_list-lb-2"]/div[1]/a/@href')
l.add_value('preview', conver_img)
l.add_css('date', '#k_list-lb-2-f::text', re=r'(\d{4}-\d{2}-\d{2})')
l.add_value('image_urls', conver_img)
return l.load_item()
def parse(self, response):
try:
page = response.url.split("/")[-1].split(".")[0]
self.log('ID: %s' % page)
book_name = response.css('h2::text').extract_first()
#self.log('book_name: %s' % book_name)
book_author = response.css('h4::text').extract_first().replace(
'Tác giả: ', '')
#self.log('book_author: %s' % book_author)
book_category = response.css('h4 a::text').extract_first().replace(
'Thể loại: ', '')
#self.log('book_category: %s' % book_category)
book_cover = response.xpath(
'//img[@class="img-thumbnail"]//@src').extract_first()
# BookDownload = namedtuple('BookDownload', ['source', 'epub', 'mobi', 'pdf', 'azw3', 'prc'])
# book_downloads = []
# for book_download in response.css('div.book-download'):
# # print(book_download.css('a::text').extract_first())
# # bd = BookDownload._fields_defaults
# source = book_download.css('a::text').extract_first()
# epub = book_download.css('a')[1].extract()
# # book_downloads.append(bd)
# self.log('source: %s' % source)
# self.log('epub: %s' % epub)
# self.log('book_downloads: %s' % book_downloads)
loader = ItemLoader(response=response)
book_description = loader.get_xpath(
'//div[@class="book-description"]/node()', Join())
#self.log('book_description: %s' % book_description)
except:
self.log('ERROR in: %s', response.url)
yield {
'id': page,
'name': book_name,
'author': book_author,
'category': book_category,
'description': book_description,
'cover': book_cover
}
def parse(self, response):
if '%s/404' % DOMAIN_URL not in response.url:
item = BookItem()
loader = ItemLoader(response=response)
name_path = '//h1/text()|//h3/text()' # h1
item['name'] = " ".join(loader.get_xpath(name_path))
item['url'] = response.url
description_path = '//*[@itemprop="description"]' # span
item['description'] = "".join(loader.get_xpath('%s/text()|%s/*/text()' % (description_path, description_path)))
publisher_path = '//*[@itemprop="publisher"]/text()' # a
item['publisher'] = loader.get_xpath(publisher_path, TakeFirst())
by_path = '//*[@itemprop="author"]/text()' # b
item['author'] = loader.get_xpath(by_path, TakeFirst())
isbn_path = '//*[@itemprop="isbn"]/text()' # b
item['isbn13'] = loader.get_xpath(isbn_path, TakeFirst())
year_path = '//*[@itemprop="datePublished"]/text()' # b
item['year'] = loader.get_xpath(year_path, TakeFirst())
pages_path = '//*[@itemprop="numberOfPages"]/text()' # b
item['pages'] = loader.get_xpath(pages_path, TakeFirst())
language_path = '//*[@itemprop="inLanguage"]/text()' # b
item['language'] = loader.get_xpath(language_path, TakeFirst())
format_path = '//*[@itemprop="bookFormat"]/text()' # b
item['format'] = loader.get_xpath(format_path, TakeFirst())
url_path = '//a[contains(@href, "http://file")]/@href'
item['download_url'] = loader.get_xpath(url_path, TakeFirst())
buy_path = '//a[contains(@href, "http://isbn")]/@href'
item['buy'] = loader.get_xpath(buy_path, TakeFirst())
size_path = '//*[contains(text(), "size")]//following-sibling::*'
item['size'] = loader.get_xpath('%s/*/text()' % size_path, TakeFirst())
image_path = '//img[contains(@itemprop, "image")]'
item['image_url'] = loader.get_xpath('%s/@src' % image_path, TakeFirst())
related_path = '//td[contains(@width, "166")]/a'
item['related'] = {
'name': loader.get_xpath('%s/@title' % related_path, TakeFirst()),
'url': urlparse.urljoin(response.url, loader.get_xpath('%s/@href' % related_path, TakeFirst())),
}
pdf_url = item['download_url']
pdf_filename = './books/' + \
item['name'].replace(',', '_').replace('/', '\/').replace(' ', '_') + \
'-' + item['author'].replace(',', '').replace(' ', '_') + \
'-' + item['isbn13'].replace('-', '') + \
'.pdf'.replace('__', '_')
item['filename'] = pdf_filename
if self.download_files:
self.download_file(pdf_url, pdf_filename, response.url)
print response.url, item['image_url']
image_url = urlparse.urljoin(response.url, item['image_url'])
self.download_file(image_url, './%s' % item['image_url'].strip('/'), response.url)
yield item
def parse_post(self, response):
post = ItemLoader(item=ArabiaPostItem(), response=response)
post.default_output_processor = TakeFirst()
#post.add_xpath('id', '//*[@class="post_content replace_urls"]/@id', MapCompose(int), re=r'(\d+)')
post.add_xpath('id',
'//*[@class="short_url inputtext"]/@value',
MapCompose(int),
re=r'(\d+)')
post.add_xpath('title', '//*[@id="nav_title"]/a/text()')
post.add_xpath('up_votes',
'//*[@class="s_upvotes"]/text()',
MapCompose(int),
re=r'(\d+)')
post.add_xpath('down_votes',
'//*[@class="s_downvotes"]/text()',
MapCompose(int),
re=r'(\d+)')
post.add_xpath('points', '//*[@class="post_points ltr"]/text()',
MapCompose(int))
post.add_xpath('author_username',
'//*[@class="block username"]/text()')
post.add_xpath('author_fullname',
'//*[@class="block full_name"]/text()',
MapCompose(lambda value: value.replace(u'\xa0', u'')))
post.add_xpath('date', '//*[@class="icon-time"]/../text()')
post.add_xpath('community',
'//*[@class="icon-reorder"]/../a[1]/text()')
post.add_xpath('topics', '//*[@class="topic"]/text()',
MapCompose(string.strip))
post.add_xpath('url', '//*[@class="short_url inputtext"]/@value')
post.add_value(
'type',
'link' if post.get_xpath('//*[@id="nav_title"]/a/@rel',
TakeFirst()) == 'nofollow' else 'text')
if post.get_output_value('type') == 'link':
post.add_xpath('link', '//*[@id="nav_title"]/a/@href')
post.add_xpath('domain',
'//*[@class="post_domain"]/text()',
re=r'\((.+?)\)')
post.add_xpath('content', '//*[@class="post_content replace_urls"]/*',
Join('\n'))
post.add_value('item', 'post')
yield post.load_item()
comments = []
for row in response.selector.xpath(
'//*[contains(@class, "post_comment")]'):
comment = ItemLoader(item=ArabiaCommentItem(),
selector=row,
response=response)
comment.default_output_processor = TakeFirst()
comment.add_xpath('id', './@id', re=r'(\d+)')
comment.add_xpath('index',
'./@class',
MapCompose(int),
re=r'index(\d+)')
comment.add_value('post_id', post.get_output_value('id'))
#comment.add_value('parent_id', '')
comment.add_xpath('author_username',
'.//*[@class="comment_user"]/a/text()')
comment.add_xpath('date', './/*[@class="comment_date"]/text()')
comment.add_xpath('points',
'.//*[@class="comment_points ltr"]/text()')
comment.add_xpath(
'content',
'.//*[@class="post_content comment_content replace_urls"]/*',
Join('\n'))
#comment.add_xpath('url', './/*[@class="comment_short_url"]/a/@href')
comment.add_value(
'url', 'https://arabia.io/go/{0}/{1}'.format(
post.get_output_value('id'),
comment.get_output_value('id')))
comment.add_value('item', 'comment')
comments.append(comment)
for (index, comment) in enumerate(comments):
if comment.get_output_value('index') == 0:
comment.add_value('parent_id', 0)
continue
for comment_cursor in comments[:index][::-1]:
if comment_cursor.get_output_value(
'index') == comment.get_output_value('index') - 1:
comment.add_value('parent_id',
comment_cursor.get_output_value('id'))
break
for comment in comments:
yield comment.load_item()
def parse(self, response):
loader = ItemLoader(response=response)
xpaths = [
'//a[contains(text(), "Next")]',
'//a[contains(@href, "-ebook.htm")]/@href'
]
for xpath in xpaths:
for href in loader.get_xpath(xpath):
yield Request(url=urlparse.urljoin(response.url, href),
callback=self.parse,
meta=response.meta,
dont_filter=True)
if '-ebook.htm' in response.url:
item = BookItem()
name_path = '//*[@itemprop="name"]/text()' # h1
item['name'] = loader.get_xpath(name_path, TakeFirst())
item['url'] = response.url
description_path = '//div[contains(text(), ".")]/text()' # span
item['description'] = loader.get_xpath(description_path,
TakeFirst())
publisher_path = '//*[@itemprop="publisher"]/text()' # a
item['publisher'] = loader.get_xpath(publisher_path, TakeFirst())
by_path = '//*[@itemprop="author"]/text()' # b
item['author'] = loader.get_xpath(by_path, TakeFirst())
isbn_path = '//*[@itemprop="isbn"]/text()' # b
item['isbn10'] = loader.get_xpath(isbn_path, TakeFirst())
item['isbn13'] = loader.get_xpath(isbn_path).pop()
year_path = '//*[@itemprop="datePublished"]/text()' # b
item['year'] = loader.get_xpath(year_path, TakeFirst())
pages_path = '//*[@itemprop="numberOfPages"]/text()' # b
item['pages'] = loader.get_xpath(pages_path, TakeFirst())
language_path = '//*[@itemprop="inLanguage"]/text()' # b
item['language'] = loader.get_xpath(language_path, TakeFirst())
format_path = '//*[@itemprop="bookFormat"]/text()' # b
item['format'] = loader.get_xpath(format_path, TakeFirst())
edition_path = '//*[@itemprop="bookEdition"]/text()' # b
item['edition'] = loader.get_xpath(edition_path, TakeFirst())
series_path = '//*[contains(text(), "Series")]//following-sibling::*'
item['series'] = ", ".join(
loader.get_xpath('%s/*/text()' % series_path)).strip()
size_path = '//*[contains(text(), "Book size")]//following-sibling::*'
item['size'] = loader.get_xpath('%s/text()' % size_path,
TakeFirst())
related_url_path = '//h4/a/@href'
related_name_path = '//h4/a/text()'
item['related'] = {
'name':
loader.get_xpath(related_name_path, TakeFirst()),
'url':
urlparse.urljoin(
response.url,
loader.get_xpath(related_url_path, TakeFirst())),
}
url_path = '//div[contains(@id, "dl")]/script'
script = loader.get_xpath(url_path, TakeFirst())
download_url = script.split('href="')[1].split('" onclick')[0]
item['download_url'] = urlparse.urljoin(response.url, download_url)
pdf_url = item['download_url']
pdf_filename = './books_/' + item['name'] + ' - ' + item[
'author'] + ' - ' + item['isbn13'] + '.pdf'
item['filename'] = pdf_filename
if self.download_files:
# req = urllib2.Request(pdf_url)
# req.add_header('Referer', '%s' % response.url)
# r = urllib2.urlopen(req)
# r = r.read()
# with open(pdf_filename, 'wb') as out:
# print r
# out.write(r)
from socket import socket
host = DOMAIN
port = 80
path = response.url.replace(DOMAIN_URL, '')
xmlmessage = "<port>0</port>"
s = socket()
s.connect((host, port))
s.send("GET %s HTTP/1.1\r\n" % path)
s.send("Host: %s\r\n" % host)
s.send("Content-Type: text/html\r\n")
for line in s.makefile():
print line,
s.close()
yield item
for href in loader.get_xpath('//td/a/@href'):
yield Request(url=urlparse.urljoin(response.url, href),
callback=self.parse,
meta=response.meta,
dont_filter=True)
else:
for href in loader.get_xpath(
'//a[contains(@href, "-ebooks-")]/@href'):
yield Request(url=urlparse.urljoin(response.url, href),
callback=self.parse,
meta=response.meta,
dont_filter=True)
def parse(self, response):
loader = ItemLoader(response=response)
xpaths = [
'//a[contains(text(), "Next")]',
'//a[contains(@href, "-ebook.htm")]/@href'
]
for xpath in xpaths:
for href in loader.get_xpath(xpath):
yield Request(
url=urlparse.urljoin(response.url, href),
callback=self.parse,
meta=response.meta,
dont_filter=True
)
if '-ebook.htm' in response.url:
item = BookItem()
name_path = '//*[@itemprop="name"]/text()' # h1
item['name'] = loader.get_xpath(name_path, TakeFirst())
item['url'] = response.url
description_path = '//div[contains(text(), ".")]/text()' # span
item['description'] = loader.get_xpath(description_path, TakeFirst())
publisher_path = '//*[@itemprop="publisher"]/text()' # a
item['publisher'] = loader.get_xpath(publisher_path, TakeFirst())
by_path = '//*[@itemprop="author"]/text()' # b
item['author'] = loader.get_xpath(by_path, TakeFirst())
isbn_path = '//*[@itemprop="isbn"]/text()' # b
item['isbn10'] = loader.get_xpath(isbn_path, TakeFirst())
item['isbn13'] = loader.get_xpath(isbn_path).pop()
year_path = '//*[@itemprop="datePublished"]/text()' # b
item['year'] = loader.get_xpath(year_path, TakeFirst())
pages_path = '//*[@itemprop="numberOfPages"]/text()' # b
item['pages'] = loader.get_xpath(pages_path, TakeFirst())
language_path = '//*[@itemprop="inLanguage"]/text()' # b
item['language'] = loader.get_xpath(language_path, TakeFirst())
format_path = '//*[@itemprop="bookFormat"]/text()' # b
item['format'] = loader.get_xpath(format_path, TakeFirst())
edition_path = '//*[@itemprop="bookEdition"]/text()' # b
item['edition'] = loader.get_xpath(edition_path, TakeFirst())
series_path = '//*[contains(text(), "Series")]//following-sibling::*'
item['series'] = ", ".join(loader.get_xpath('%s/*/text()' % series_path)).strip()
size_path = '//*[contains(text(), "Book size")]//following-sibling::*'
item['size'] = loader.get_xpath('%s/text()' % size_path, TakeFirst())
related_url_path = '//h4/a/@href'
related_name_path = '//h4/a/text()'
item['related'] = {
'name': loader.get_xpath(related_name_path, TakeFirst()),
'url': urlparse.urljoin(response.url, loader.get_xpath(related_url_path, TakeFirst())),
}
url_path = '//div[contains(@id, "dl")]/script'
script = loader.get_xpath(url_path, TakeFirst())
download_url = script.split('href="')[1].split('" onclick')[0]
item['download_url'] = urlparse.urljoin(response.url, download_url)
pdf_url = item['download_url']
pdf_filename = './books_/' + item['name'] + ' - ' + item['author'] + ' - ' + item['isbn13'] + '.pdf'
item['filename'] = pdf_filename
if self.download_files:
# req = urllib2.Request(pdf_url)
# req.add_header('Referer', '%s' % response.url)
# r = urllib2.urlopen(req)
# r = r.read()
# with open(pdf_filename, 'wb') as out:
# print r
# out.write(r)
from socket import socket
host = DOMAIN
port = 80
path = response.url.replace(DOMAIN_URL, '')
xmlmessage = "<port>0</port>"
s = socket()
s.connect((host, port))
s.send("GET %s HTTP/1.1\r\n" % path)
s.send("Host: %s\r\n" % host)
s.send("Content-Type: text/html\r\n")
for line in s.makefile():
print line,
s.close()
yield item
for href in loader.get_xpath('//td/a/@href'):
yield Request(
url=urlparse.urljoin(response.url, href),
callback=self.parse,
meta=response.meta,
dont_filter=True
)
else:
for href in loader.get_xpath('//a[contains(@href, "-ebooks-")]/@href'):
yield Request(
url=urlparse.urljoin(response.url, href),
callback=self.parse,
meta=response.meta,
dont_filter=True
)
