class SogoAccount():
def __init__(self):
self.settings = Settings()
self.getSettings()
self.file = FileIOMiddleware()
self.requests = RequestsMiddleware()
self.doraemon = Doraemon()
self.doraemon.createFilePath(self.work_path_prd2)
self.doraemon.createFilePath(self.settings.LOG_PATH)
def getSettings(self):
settings_name = self.settings.CreateSettings('sogo')
self.work_path_prd2 = settings_name['WORK_PATH_PRD2']
self.mongo = settings_name['MONGO_URLS']
self.name = settings_name['NAME']
self.max_pool_size = settings_name['MAX_POOL_SIZE']
self.log_path = self.settings.LOG_PATH_PRD2
self.urls = settings_name['URLS']
self.restart_path = settings_name['RESTART_PATH']
self.restart_interval = settings_name['RESTART_INTERVAL']
self.valid_proxy_pool_sogo_account = self.settings.VALID_PROXY_POOL_SOGO_ACCOUNT
self.invalid_proxy_pool_sogo_account= self.settings.INVALID_PROXY_POOL_SOGO_ACCOUNT
self.finished_sogo_account = self.settings.FINISHED_SOGO_ACCOUNT
self.regx = re.compile("[0-9]{1,}.[0-9]{1,}.[0-9]{1,}.[0-9]{1,}:[0-9]{1,}")
def getProxy(self):
url = "http://ip.16yun.cn:817/myip/pl/c167cc62-6ad5-4876-bfd8-0cc423dab398/?s=wygafjcqjv&u=hellobee&count=2"
# url = "http://129.28.124.247:43059/get_ip.php?key=908299fbaefcacef4eb2c9e6ea18c5f2"
response = self.requests.requests_request(url, headers=None, host="ip.16yun.cn", referer="ip.16yun.cn")
proxy_list = response.text.strip().split('\n')
for proxy in proxy_list:
ip = proxy.strip()
isValidIp = self.regx.match(ip)
if self.doraemon.isEmpty(ip) is False and isValidIp is not None:
self.file.logger(self.log_path, "Proxy: {0} is available.".format(ip))
print "Proxy: {0} is available.".format(ip)
try:
self.doraemon.hashSet(self.valid_proxy_pool_sogo_account, ip, ip)
except Exception as e:
print "Exception to set redis for available sogo account of ip: {0}: {1}.".format(ip, e.message)
self.file.logger(self.log_path, "Exception to set redis for available sogo account of ip: {0}: {1}.".format(ip, e.message))
else:
self.file.logger(self.log_path, 'Fail to get proxy for sogo account.')
print "Fail to get proxy for sogo account."
def parse(self, response):
current_url = response['response'].current_url.encode('gbk')
print 'Start to parse: {0}'.format(current_url)
html = etree.HTML(response['response'].page_source)
key = response['request_title']
href_item = html.xpath(".//*[contains(@uigs, 'account_name_0')]/@href")
if len(href_item) == 0:
print 'Blocked and change for another proxy.'
self.doraemon.hashSet(self.invalid_proxy_pool_sogo_account, self.proxy, self.proxy)
self.doraemon.delHashSet(self.valid_proxy_pool_sogo_account, self.proxy)
all_valid_proxy = list(self.doraemon.getAllHasSet(self.valid_proxy_pool_sogo_account))
if len(all_valid_proxy) == 0:
print 'The proxy pool is empty and get proxy again.'
self.file.logger(self.log_path, 'The proxy pool is empty and get proxy again.')
self.getProxy()
all_valid_proxy = list(self.doraemon.getAllHasSet(self.valid_proxy_pool_sogo_account))
self.proxy = all_valid_proxy.pop()
return
href = href_item[0]
url = urlparse.urljoin(current_url, href)
self.doraemon.hashSet(self.name, key, url)
self.doraemon.hashSet(self.finished_sogo_account, key, key)
print 'Finished for {0}'.format(key)
self.current_url.pop()
if len(self.new_urls) > 0:
new_url = self.new_urls.pop()
print 'Start next: {0}'.format(new_url[0])
self.current_url.append(new_url)
print 'End to parse {0}, url: {1}'.format(key, href_item[0])
def start_requests(self):
if self.doraemon.isExceedRestartInterval(self.restart_path, self.restart_interval) is False:
return
self.file.logger(self.log_path, 'Start {0} requests'.format(self.name))
print 'Start {0} requests'.format(self.name)
try:
self.getProxy()
except Exception as e:
self.file.logger(self.settings.LOG_PATH, 'Exception to get proxy: {0}'.format(str(e.message)))
all_finished_id = list(self.doraemon.getAllHasSet(self.finished_sogo_account))
all_valid_proxy = list(self.doraemon.getAllHasSet(self.valid_proxy_pool_sogo_account))
if self.doraemon.isEmpty(all_valid_proxy):
self.file.logger(self.log_path, 'No available proxy for sogo account and return.')
print "No available proxy for sogo account and return."
return
self.new_urls = []
self.current_url = []
keys = []
content = self.file.readFromTxt(self.urls)
keys_list = content.split('\n')
for key in keys_list:
if self.doraemon.isEmpty(key) is False:
keys.append(key)
self.proxy = all_valid_proxy.pop()
for key in keys:
if key not in all_finished_id:
timestamp = '00'.join(str(time.time()).split('.'))
tmp_url = "https://weixin.sogou.com/weixin?type=1&s_from=input&query={0}&ie=utf8&_sug_=n&_sug_type_=".format(key)
self.new_urls.append([tmp_url, key])
request = BrowserRequest()
if len(self.new_urls) > 0:
self.current_url.append(self.new_urls.pop())
else:
print 'No more urls.'
while len(self.current_url) > 0:
print "Proxy :{0}".format(self.proxy)
if len(self.new_urls) > 0:
self.current_url.append(self.new_urls.pop())
else:
print 'No more urls.'
request.start_chrome(self.current_url, self.max_pool_size, self.log_path, self.proxy, callback=self.parse)
self.file.logger(self.log_path, 'End for requests of {0}.'.format(self.name))
class Huxiu():
def __init__(self):
self.settings = Settings()
self.getSettings()
self.file = FileIOMiddleware()
self.doraemon = Doraemon()
self.doraemon.createFilePath(self.work_path_prd1)
self.doraemon.createFilePath(self.settings.LOG_PATH)
def getSettings(self):
settings_name = self.settings.CreateSettings('huxiu')
self.source = settings_name['SOURCE_NAME']
self.work_path_prd1 = settings_name['WORK_PATH_PRD1']
self.finished_txt_path = '/home/dev/Data/rsyncData/huxiu_nlp/text/'
self.url_path = '/home/dev/Data/rsyncData/huxiu_nlp/huxiu_nlp.csv'
self.mongo = 'huxiu_nlp'
self.name = settings_name['NAME']
self.max_pool_size = 4
self.log_path = self.settings.LOG_PATH
self.today = self.settings.TODAY
self.is_open_cache = settings_name['IS_OPEN_CACHE']
def parse(self, response):
current_url = response['response'].current_url.encode('gbk')
print 'Start to parse: {0}'.format(current_url)
html = etree.HTML(response['response'].page_source)
data = {}
comment_number = ""
title = ""
url = ""
id = ""
share_number = ""
image_url = ""
content = ""
time = ""
author_url = ""
author_name = ""
valid = False
url = current_url
id = str(filter(str.isdigit, current_url.encode('gbk')))
title1 = html.xpath(".//*[contains(@class,'t-h1')]/text()")
comment_number1 = html.xpath(
".//*[contains(@class, 'article-pl pull-left')]/text()")
share_number1 = html.xpath(
".//*[contains(@class, 'article-share pull-left')]/text()")
image_url1 = html.xpath(
".//*[contains(@class, 'article-img-box')]/img/@src")
content1 = html.xpath(
".//div[contains(@class, 'article-content-wrap')]//text()")
time1 = html.xpath(".//*[contains(@class, 'article-time')]/text()")
author_url1 = html.xpath(
".//*[contains(@class, 'author-name')]/a/@href")
author_name1 = html.xpath(
".//*[contains(@class, 'author-name')]/a/text()")
if self.doraemon.isEmpty(title1) is False:
title = title1[0].strip()
if self.doraemon.isEmpty(comment_number1) is False:
comment_number = str(
filter(str.isdigit, comment_number1[0].encode('gbk'))).strip()
if self.doraemon.isEmpty(share_number1) is False:
share_number = str(
filter(str.isdigit, share_number1[0].encode('gbk'))).strip()
if self.doraemon.isEmpty(image_url1) is False:
image_url = image_url1[0].strip()
if self.doraemon.isEmpty(content1) is False:
content = ''.join(content1).strip()
valid = True
if self.doraemon.isEmpty(time1) is False:
time = ''.join(time1).strip()
time = self.doraemon.getDateFromString(time)
if self.doraemon.isEmpty(author_url1) is False:
author_url = urlparse.urljoin(current_url, author_url1[0].strip())
if self.doraemon.isEmpty(author_name1) is False:
author_name = ''.join(author_name1[0]).strip()
data = {
'title': title,
'comment_number': comment_number,
'share_number': share_number,
'image_url': image_url,
'url': url,
'public_time': time,
'author_url': author_url,
'author_name': author_name,
'id': id,
'download_time': self.today,
'is_open_cache': self.is_open_cache,
'source': self.source
}
print 'End to parse: {0}'.format(current_url)
if valid == True and self.doraemon.isEmpty(title) is False:
self.file.logger(self.log_path,
'Start to store mongo {0}'.format(data['url']))
print 'Start to store mongo {0}'.format(data['url'])
self.doraemon.storeMongodb(self.mongo, data)
self.file.logger(self.log_path,
'End to store mongo {0}'.format(data['url']))
print 'End to store mongo {0}'.format(data['url'])
self.doraemon.storeTxt(id, content, self.finished_txt_path,
self.name)
self.doraemon.storeFinished(self.doraemon.bf_huxiu_nlp,
response['request_title'])
else:
self.doraemon.storeFinished(self.doraemon.bf_huxiu_nlp,
response['request_title'])
del current_url, html, title, comment_number, share_number, image_url, url, content, time, author_url, author_name, id, data
gc.collect()
def start_requests(self):
self.file.logger(self.log_path, 'Start request: {0}'.format(self.name))
print 'Start ' + self.name + ' requests'
new_url_titles = self.doraemon.readNewUrls(self.doraemon.bf_huxiu_nlp,
self.url_path)
# new_url_titles = [['https://www.huxiu.com/article/36.html', '【WHAT】十年内10大互联网IPO']]
if len(new_url_titles) == 0:
self.file.logger(self.log_path,
'No new url for: {0}'.format(self.name))
print 'No new url for: {0}'.format(self.name)
return
request = BrowserRequest()
content = request.start_chrome(new_url_titles,
self.max_pool_size,
self.log_path,
None,
callback=self.parse)
self.file.logger(self.log_path,
'End requests: {0}'.format(str(len(content))))
print 'End requests: {0}'.format(str(len(content)))
del new_url_titles, request, content
gc.collect()
class Topbaidu():
def __init__(self):
self.getSettings()
self.file = FileIOMiddleware()
self.doraemon = Doraemon()
self.doraemon.createFilePath(self.work_path_prd2)
self.doraemon.createFilePath(Settings.LOG_PATH)
def getSettings(self):
self.work_path_prd2 = settings_name['WORK_PATH_PRD2']
self.mongo = settings_name['MONGO_URLS']
self.name = settings_name['NAME']
self.max_pool_size = settings_name['MAX_POOL_SIZE']
self.log_path = Settings.LOG_PATH_PRD2
self.urls = settings_name['URLS']
self.restart_path = settings_name['RESTART_PATH']
self.restart_interval = settings_name['RESTART_INTERVAL']
self.today = Settings.TODAY
def parse(self, response):
current_url = response['response'].current_url.encode('gbk')
print 'Start to parse: {0}'.format(current_url)
html = etree.HTML(response['response'].page_source)
href_items = html.xpath(".//*[contains(@class, 'article-item-title')]")
for item in href_items:
href = item.xpath("@href")
valid = True
if len(href) == 0:
continue
href_url = href[0]
hasId = str(filter(str.isdigit, href_url))
if len(hasId) == 0:
print 'Invalid url for no id: {0}'.format(href_url)
continue
for good in self.goodkeys:
if valid == True:
continue
if good in href_url:
valid = True
for bad in self.badkeys:
if valid == False:
continue
if bad in href_url:
valid = False
if valid:
short_url_parts = re.split(r'[., /, _]', href_url)
id = short_url_parts[len(short_url_parts) - 1]
url = urlparse.urljoin(current_url, href_url)
title = ""
title_list1 = item.xpath(".//text()")
if len(title_list1) > 0:
title = title_list1[0]
print title
is_title_empty = self.doraemon.isEmpty(title)
if (is_title_empty is False) and (
self.doraemon.isDuplicated(title) is False):
data = {
'title': title.strip(),
'url': url.strip(),
'id': id.strip(),
'download_time': self.today
}
self.file.logger(
self.log_path,
'Start to store mongo {0}'.format(data['url']))
print 'Start to store mongo {0}'.format(data['url'])
self.doraemon.storeMongodb(self.mongo, data)
self.file.logger(
self.log_path,
'End to store mongo {0}'.format(data['url']))
print 'End to store mongo {0}'.format(data['url'])
self.file.logger(self.log_path, 'Done for {0}'.format(url))
else:
if is_title_empty is True:
self.file.logger(self.log_path,
'Empty title for {0}'.format(url))
print 'Empty title for {0}'.format(url)
print 'Finished or Empty title for {0}'.format(url)
else:
self.file.logger(self.log_path, 'Invalid {0}'.format(href_url))
print 'Invalid {0}'.format(href_url)
print 'End to parse {0}'.format(href_url)
def start_requests(self):
if self.doraemon.isExceedRestartInterval(
self.restart_path, self.restart_interval) is False:
return
self.file.logger(self.log_path, 'Start {0} requests'.format(self.name))
print 'Start {0} requests'.format(self.name)
self.badkeys = []
self.goodkeys = []
new_urls = []
content = self.file.readFromTxt(self.urls)
url_list = content.split('\n')
for url in url_list:
if self.doraemon.isEmpty(url) is False:
new_urls.append([url, ''])
if len(new_urls) == 0:
print 'No url.'
return
request = BrowserRequest()
content = request.start_chrome(new_urls,
self.max_pool_size,
self.log_path,
None,
callback=self.parse)
self.file.logger(
self.log_path,
'End for {0} requests of {1}.'.format(str(len(content)),
self.name))
print 'End for {0} requests of {1}.'.format(str(len(content)),
self.name)
class WeixinSalticidae():
def __init__(self):
self.settings = Settings()
self.getSettings()
self.file = FileIOMiddleware()
self.doraemon = Doraemon()
self.doraemon.createFilePath(self.work_path_prd1)
self.doraemon.createFilePath(self.settings.LOG_PATH)
self.doraemon.createFilePath(self.finished_img_path)
def getSettings(self):
settings_name = self.settings.CreateSettings('weixin')
self.source = settings_name['SOURCE_NAME']
self.work_path_prd1 = settings_name['WORK_PATH_PRD1']
self.finished_img_path = settings_name['FINISHED_IMG_PATH']
self.finished_origin_html_path = settings_name[
'FINISHED_ORIGIN_HTML_PATH']
self.finished_processed_html_path = settings_name[
'FINISHED_PROCESSED_HTML_PATH']
self.finished_content_path = settings_name['FINISHED_CONTENT_PATH']
self.mongo = settings_name['MONGO']
self.name = settings_name['NAME']
self.max_pool_size = settings_name['MAX_POOL_SIZE']
self.url_deepinews_10002_article = self.settings.URL_DEEPINEWS_10002_ARTICLE
self.url_deepinews_10002_image = self.settings.URL_DEEPINEWS_10002_IMAGE
self.log_path = self.settings.LOG_PATH
self.today = self.settings.TODAY
self.restart_path = settings_name['RESTART_PATH']
self.restart_interval = settings_name['RESTART_INTERVAL']
self.regx_img = re.compile('<img(.*?)/>')
self.regx_date = re.compile(
'<em id="publish_time" class="rich_media_meta rich_media_meta_text">(.*?)</em>'
)
self.regx_img_type = re.compile('data-type="(.*?)"')
self.regx_img_data_src = re.compile('data-src="(.*?)"')
self.regx_img_src = re.compile('src="(.*?)"')
self.regx_img_class = re.compile('class="(.*?)"')
def getPostFixOfImage(self, image_type):
if image_type == 'jpeg':
return 'jpg'
if image_type == 'png':
return 'png'
if image_type == 'gif':
return 'gif'
else:
print 'Other type: {0}'.format(image_type)
def start_requests(self):
self.file.logger(self.log_path,
'Start dowload images for: {0} '.format(self.name))
print 'Start dowload images for: {0} '.format(self.name)
new_ids = self.doraemon.readNewImageIds(
self.doraemon.bf_finished_image_id, self.finished_content_path)
if len(new_ids) == 0:
self.file.logger(self.log_path,
'No new image id for {0}'.format(self.name))
print 'No new image id for {0}'.format(self.name)
return
self.doraemon.createFilePath(self.finished_processed_html_path)
self.doraemon.createFilePath(self.finished_img_path)
for id in new_ids:
print 'Start to remove pictures in: {0}'.format(id)
html_file = self.file.readFromHtml("{0}/{1}.html".format(
self.finished_origin_html_path, id))
img_list = re.findall(self.regx_img, html_file)
date_list = re.findall(self.regx_date, html_file)
new_html = ''
number = 0
for old_time in date_list:
new_date = self.doraemon.getDateFromString(old_time)
old_time_content = '<em id="publish_time" class="rich_media_meta rich_media_meta_text">{0}</em>'.format(
old_time)
new_time_content = '<em id="publish_time" class="rich_media_meta rich_media_meta_text">{0}</em>'.format(
new_date)
new_html = html_file.replace(old_time_content,
new_time_content)
html_file = new_html
for img in img_list:
old_img = img
image_id = "{0}_{1}".format(id, number)
image_data_src = ''.join(
re.findall(self.regx_img_data_src, img)).strip()
image_src = re.findall(self.regx_img_src, img)
image_type = ''.join(re.findall(self.regx_img_type,
img)).strip()
image_post_fix = self.getPostFixOfImage(image_type)
if (self.doraemon.isEmpty(image_data_src) is True) or \
(self.doraemon.isEmpty(image_src) is True) or \
(self.doraemon.isEmpty(image_type) is True):
continue
origin_image_path = "{0}/{1}.{2}".format(
self.finished_img_path, image_id, image_post_fix)
print 'Start to download image: {0}'.format(image_data_src)
self.doraemon.downloadImage(image_data_src, origin_image_path)
image_size = self.doraemon.getFileSize(origin_image_path)
if image_size > 60:
print 'Start to compress image: {0}'.format(image_data_src)
self.doraemon.compressImage(origin_image_path,
origin_image_path, 2)
print 'Finished to compress image: {0}'.format(
image_data_src)
print 'Finished to download image: {0}'.format(image_data_src)
print 'Start to replace image url: {0}'.format(image_id)
new_imgurl = "{0}{1}.{2}".format(
self.url_deepinews_10002_image, image_id, image_post_fix)
# new_imgurl = '/home/dev/Data/rsyncData/prd4/weixin/img/{0}.{1}'.format(image_id, image_post_fix)
src_list = re.findall(self.regx_img_src, img)
img_class_list = re.findall(self.regx_img_class, img)
for img_class in img_class_list:
new_img = img.replace(img_class, 'rich_pages')
img = new_img
for src in src_list:
new_img = img.replace(src, new_imgurl)
img = new_img
new_html = html_file.replace(old_img, img)
html_file = new_html
print 'Finished to replace image url: {0}'.format(image_id)
number += 1
self.doraemon.storeHtml(id, new_html,
self.finished_processed_html_path)
self.doraemon.storeFinished(self.doraemon.bf_finished_image_id, id)