def parse(self, response):
item_list = response.xpath("//div[@class='post_item']")
for item in item_list:
blogItem = BaiduItem()
# blogItem['title'] = item.xpath('./div[2]/h3[1]/a/text()')[0].extract().strip()
# blogItem['titleLink'] = item.xpath('./div[2]/h3[1]/a/@href')[0].extract()
# blogItem['suggestNum'] = item.xpath('./div[1]/div[1]/span/text()')[0].extract().strip()
# content = item.xpath('./div[2]/p[1]/text()')
# for c in content:
# if c.extract().strip() != '':
# blogItem['content'] = c.extract().strip()
# else:
# blogItem['content'] = ''
# blogItem['author'] = item.xpath('./div[2]/div[1]/a/text()')[0].extract().strip()
# blogItem['publishDate'] = item.xpath('./div[2]/div[1]/text()')[1].extract().strip()
yield blogItem
# handle next page
next = response.xpath(
"//div[@class='pager']/a[last()]/text()")[0].extract().strip()
print(next)
if 'Next' in next:
nextUrlStr = response.xpath(
"//div[@class='pager']/a[last()]/@href")[0].extract().strip()
nextNum = nextUrlStr[-1]
nextNum = 2
if int(nextNum) < 5:
nextUrl = 'https://www.cnblogs.com/#p/{0}'.format(nextNum)
print(nextUrl)
nextNum += 1
yield scrapy.Request(nextUrl, callback=self.parse)
def parse_rel(self, html):
"""解析相关搜索"""
item = BaiduItem()
for row in html.find_all('a'):
item['url'] = row['href']
item['title'] = row.get_text().strip()
yield item
def parse(self, response):
print(response.text)
# 模拟浏览器
self.driver.get("http://www.baidu.com")
# 根据id找到百度的输入框
input_element = self.driver.find_element_by_id('kw')
# 模拟输入搜索内容
input_element.send_keys("爬虫")
# 根据id找到百度的确认按钮
click_button = self.driver.find_element_by_id('su')
#模拟点击事件
click_button.click()
# 等5s
self.driver.implicitly_wait(5)
# 根据classname获取列表数据
result_list = self.driver.find_elements_by_class_name('result')
for result in result_list:
item = BaiduItem()
print("列表标题：" + result.text)
print(
"==========================================================================="
)
item["name"] = result.text
yield item
def parse_lyric(self, response):
hxs = HtmlXPathSelector(response)
item = BaiduItem()
content = hxs.select('//*[@id="lyricCont"]/text()')
##lyric
lyric = ''
for word in content:
lyric += word.extract()
item['lyric'] = lyric
## hot: define how popular the song
hot = hxs.select('//span[@class="num"]/text()').extract()[0]
item['hot'] = hot
## singer
singer = hxs.select(
'//span[@class="author_list"]/a/text()').extract()[0]
patt = '\s'
item['singer'] = re.sub(patt, '', singer)
## songName
songName = hxs.select(
'//ul[@class="path-list clearfix"]/li[4]/text()').extract()[0]
item['songName'] = songName
return item
def parse_item(self, response):
sel = Selector(response)
items = []
item = BaiduItem()
title = sel.xpath('//span[@class="ask-title "]/text()').extract()
for i in title:
items.append(i)
item['TitleName'] = items
return item
def parse(self, response):
filename = response.url.split(".")[-1] + '.txt'
for h2 in response.xpath('//article[@class="entry-common clearfix"]'):
item = BaiduItem()
item['title'] = str.strip(
h2.xpath('header/h2/a/text()').extract()[0])
item['url'] = h2.xpath('header/h2/a/@href').extract()[0]
# item['desc'] =str.strip(h2.xpath('header/div[2]/span/text()').extract()[0])
yield item
def parse(self, response):
jsonresp = json.loads(response.body_as_unicode())
for o in jsonresp['data']:
if 'thumbURL' in o:
l = ItemLoader(item=BaiduItem(), response=response)
#image_url
l.add_value('image_urls', o['thumbURL'])
l.add_value('image_type', tag)
yield l.load_item()
def get_tab2(self, response):
items = []
for rank in response.xpath('./div[@class="grayborder"]/table[@class="list-table"]/tbody/tr[7]/following-sibling::tr'):
item = BaiduItem()
item['rank_category'] = 'zongyi'
item['rank_name'] = rank.xpath('./td[@class="keyword"]/a[@class="list-title"]/text()').extract()
item['rank_index'] = rank.xpath('./td[@class="last"]/span/text()').extract()
items.append(item)
return items
def parse_week(self, browser):
# TODO: 定义变量，收集结果
lst=[]
for actived in ['搜索指数','资讯指数']:
# import ipdb; ipdb.set_trace()
# act=browser.find_element_by_xpath('//*[text()[contains(.,"搜索指数")]]').text
if actived == '搜索指数':
browser.find_element_by_xpath('//*[text()[contains(.,"搜索指数")]]').click()
else:
browser.find_element_by_xpath('//*[text()[contains(.,"资讯指数")]]').click()
try:
items = browser.find_elements_by_css_selector('.tab-content .list > .list-item')
# import ipdb; ipdb.set_trace()
data = browser.find_elements_by_css_selector('.date-text')[0].text
one = browser.find_elements_by_css_selector('.actived')[0].text
for item in items:
# 排名
rank = item.find_element_by_css_selector('.content .rank').text.strip()
# 姓名
name = item.find_element_by_css_selector('.content .name').text.strip()
# 行指数
line = item.find_element_by_css_selector('.content .line-light')
# 行指数-真实值
real_value = float(line.value_of_css_property('width').replace('px', ''))
# 行指数-最大值元素
line_max = line.find_element_by_xpath('..')
# 最大值
max_value = float(line_max.value_of_css_property('width').replace('px', ''))
# 指数值
index = round(100 * real_value / max_value, 2)
# 指数字符串
index_str = str(index).rstrip('0').rstrip('.') + '%'
cc=BaiduItem()
cc['one']=one
cc['data']=data
cc['actived']=actived
cc['rank']=rank
cc['name']=name
cc['index_str']=index_str
print(index_str)
print('-' * 30)
# yield cc
lst.append(cc)
# import ipdb; ipdb.set_trace()
except (NoSuchElementException, StaleElementReferenceException):
import ipdb; ipdb.set_trace()
pass
# TODO: 返回结果
return lst
def parse(self, response):
sel = Selector(response)
items = []
item = BaiduItem()
title = sel.xpath('//div[@class="question-title"]/a/text()').extract()
for i in title:
items.append(i)
item['TitleName'] = items
# print(item['TitleName'])
return item
def parse_week(self, browser, nav_name, someweek, exponent):
'''
搜素指数
'''
# 定义变量，收集结果
# import ipdb; ipdb.set_trace()
result = []
try:
# 获取每一页的每一行的数据
items = browser.find_elements_by_css_selector('.tab-content .list > .list-item')
# 分别便利获取每个娱乐圈人的数据
for item in items:
baidu = BaiduItem()
# 排名
baidu['rank'] = item.find_element_by_css_selector('.content .rank').text.strip()
# 姓名
baidu['name'] = item.find_element_by_css_selector('.content .name').text.strip()
if nav_name == '周上升榜':
# 趋势
# import ipdb; ipdb.set_trace()
trend = item.find_element_by_xpath('//div[@class="trend"]/span').text
baidu['trend'] = trend
else:
# 趋势
trend = item.find_element_by_xpath('//span[@class="trend-icon"]/i').get_attribute('class')
baidu['trend'] = re.sub('icon trend-', '', trend)
# 行指数
line = item.find_element_by_css_selector('.content .line-light')
# 行指数-真实值
real_value = float(line.value_of_css_property('width').replace('px', ''))
# 行指数-最大值元素
line_max = line.find_element_by_xpath('..')
# 最大值
max_value = float(line_max.value_of_css_property('width').replace('px', ''))
# 指数值
index = round(100 * real_value / max_value, 2)
# 指数字符串
index_str = str(index).rstrip('0').rstrip('.') + '%'
# 指数比例
baidu['index_str'] = index_str
# 棒单名
baidu['nav_name'] = nav_name
# 搜索指数 和资讯指数
baidu['exponent'] = exponent
# 日期
baidu['week'] = someweek
result.append(baidu)
print(nav_name, '--->', exponent, '--->', someweek, '--->', index_str)
print('-' * 30)
# import ipdb; ipdb.set_trace()
except (NoSuchElementException, StaleElementReferenceException):
import ipdb; ipdb.set_trace()
# 返回结果
return result
def parse(self, response):
item = BaiduItem()
soup = BeautifulSoup(response.text, 'html.parser')
title = soup.select('.lemmaWgt-lemmaTitle-title h1')[0].text
subhead = soup.select('.lemmaWgt-lemmaTitle-title h2')
if len(subhead) is not 0:
# print(subhead[0].text)
title = title + subhead[0].text
item['title'] = title
info_list = soup.select('.lemma-summary div')
info = ''
for temp in info_list:
# 截取其中文字信息
info += temp.text
# 如果有超链接，则继续爬取
a_list = temp.select('a')
if len(a_list) is not 0:
for a in a_list:
if a.has_attr('href'):
yield Request(self.base_url + a['href'], headers=self.headers)
item['info'] = info
properties_list = soup.select('.basicInfo-block dt')
properties = ''
for pro in properties_list:
properties += '###' + pro.text.strip().replace('\n', '')
# 如果有超链接，则继续爬取
a_list = pro.select('a')
if len(a_list) is not 0:
for a in a_list:
if a.has_attr('href'):
yield Request(self.base_url + a['href'], headers=self.headers)
item['properties'] = properties
values_list = soup.select('.basicInfo-block dd')
values = ''
for val in values_list:
values += '###' + val.text.strip().replace('\n', '')
# 如果有超链接，则继续爬取
a_list = val.select('a')
if len(a_list) is not 0:
for a in a_list:
if a.has_attr('href'):
yield Request(self.base_url + a['href'], headers=self.headers)
item['values'] = values
if len(soup.select('.summary-pic img')) is not 0:
item['img'] = soup.select('.summary-pic img')[0]['src']
print(item['title'])
yield item
def parse(self, response):
# 将响应内容生成Selector，用于数据清洗
sel = Selector(response)
items = []
# 定义BaiduItem对象
item = BaiduItem()
title = sel.xpath('//div[@class="question-title"]/a/text()').extract()
for i in title:
items.append(i)
item['TitleName'] = items
return item
def DownlordImg(self, response):
img = BaiduItem()
content = response.xpath("//*[@id='big-pic']/p/a/img")
if content:
img['image_urls'] = content.xpath(".//@src").extract()
yield img
links = response.xpath("/html/body/div[3]/div[2]/div[9]/ul/li")
for j in links:
if j.xpath(".//a/text()").extract_first() == '下一页':
next_link = 'https://www.aitaotu.com/' + j.xpath(
".//a/@href").extract_first()
yield scrapy.Request(next_link, callback=self.DownlordImg)
def parse(self, response):
num = 0
company = response.meta["company"]
list_content = response.css("#content_left").xpath("./div")
for i in list_content:
url = i.css(".t").xpath("./a/@href").get()
title = i.css(".t").xpath("./a").xpath("string(.)").get()
summary = i.css(".c-abstract").xpath("string(.)").get()
item = BaiduItem()
if url and title and summary:
num+=1
item["url"] = url
item["title"] = title
item["summary"] = summary
yield item
if num <10:
item1 = BaiduItem()
item1["error"] = company
return item1
def get_tieba(self, response):
'''
根据分类爬取此类型所有的贴吧信息
'''
tiebas = response.xpath('//*[@id="ba_list"]/div')
for tieba in tiebas:
item = BaiduItem()
# 初始化item并进行赋值保存
try:
try:
item['title'] = response.meta['title']
except KeyError:
item['title'] = 'title invalid!'
try:
item['tag'] = response.meta['tag']
except:
item['tag'] = 'tag invalid!'
item['name'] = tieba.xpath(
'a/div/p[@class="ba_name"]/text()').extract()[0]
item['img'] = tieba.xpath('a/img/@src').extract()[0]
item['user_num'] = tieba.xpath(
'a/div/p[@class="ba_num clearfix"]/span[@class="ba_m_num"]/text()'
).extract()[0]
item['topic_num'] = tieba.xpath(
'a/div/p[@class="ba_num clearfix"]/span[@class="ba_p_num"]/text()'
).extract()[0]
try:
item['description'] = tieba.xpath(
'a/div/p[@class="ba_desc"]/text()').extract()[0]
except KeyError:
item['description'] = 'not find description!'
except IndexError:
item['description'] = 'not find description!'
item['url'] = 'http://tieba.baidu.com' + tieba.xpath(
'a/@href').extract()[0]
except Exception as e:
print('------------------------')
print(e)
yield item
next_url = response.xpath('//*[@class="next"]/@href').extract()
if next_url:
# 按段是否存在下一页，存在则翻页抓取
url = 'http://tieba.baidu.com/' + next_url[0]
# print(url)
yield Request(url,
callback=self.get_tieba,
meta={
'title': response.meta['title'],
'tag': response.meta['tag']
})
def parse(self, response):
url = response.url
title = response.xpath(
"//*[@class='lemmaWgt-lemmaTitle-title']/h1/text()").extract()[0]
new_urls = response.xpath(
"//*[starts-with(@href,'/item/')]/@href").extract()
summary = response.xpath("//*[@class='lemma-summary']/div")
summary = summary.xpath('string(.)').extract()[0]
item = BaiduItem(url=url, title=title, summary=summary)
yield item
for url in new_urls:
self.x += 1
if self.x > 100:
break
self.url = url
yield Request(response.urljoin(url=url), callback=self.parse)
def parse(self, response):
div_list = response.xpath('//div[@class="t_con cleafix"]/div[2]/div')
for div in div_list:
item = BaiduItem()
item["author"] = div.xpath('./div[2]/span/@title').extract_first()
item["title"] = div.xpath("./div/a/text()").extract_first()
item["href"] = div.xpath("./div/a/@href").extract_first()
item["time"] = div.xpath("./div[2]/span[2]/text()").extract_first()
if item["author"] and item["title"] and item["href"]:
item["href"] = "http://tieba.baidu.com" + item['href']
yield scrapy.Request(item["href"],
callback=self.parse_detail,
meta=item)
next_url = response.xpath(
'//a[@class="next pagination-item "]/@href').extract_first()
if next_url is not None:
next_url = response.urljoin(next_url)
yield scrapy.Request(next_url, callback=self.parse)
def parse(self, response):
sel = Selector(response)
sel_0 = sel.xpath('/html/body/div[@id="body"]')
sel_1 = sel_0.xpath('.//ul/li')
base_url = get_base_url(response)
no = 0
for site in sel_1:
no += 1
title = site.xpath('a/text()').extract()
if (len(title) != 0 and len(''.join(title)) > 4):
link = site.xpath('a/@href').extract()
url_m = (str(link))[3:-2]
url_new = urljoin(base_url, url_m)
item = BaiduItem(title=title,
link=link,
manufacturer='baiduguonei')
yield scrapy.Request(url_new,
callback=self.parse_body,
meta={'item': item})
def parse_search(self, response):
import hashlib
params = response.meta.get('params')
keywords = response.meta.get('keywords')
timestamp = json.loads(response.body_as_unicode())['timestamp']
json_res = json.loads(response.body_as_unicode())['data']
for news in json_res.get('list'):
item = BaiduLoader(item=BaiduItem())
# 只爬取最近7天的新闻
if timestamp - news.get('publicTime') > 7*24*3600:
return None
item.add_value('post_time',
time.strftime("%Y-%m-%dT%H:%M:%S+08:00", time.localtime(float(news.get('publicTime')))))
item.add_value('search_word', params.get('word'))
item.add_value('author', news.get('author'))
item.add_value('title', news.get('title'))
item.add_value('link', news.get('url'))
item.add_value('hash_id', hashlib.md5(news.get('url').encode('utf8')).hexdigest())
item.add_value('cover_urls', news.get('imgUrl')) # cover image url
intro = item.load_item()
new_url = 'https://news.baidu.com/news?tn=bdapiinstantfulltext&src=' + news.get('url')
if keywords is None:
print('error')
print(news.get('url'))
yield Request(url=new_url,
meta={'intro': intro,
'enter_url': news.get('url'),
'keywords': response.meta.get('keywords').split(','),
},
callback=self.parse)
total = json_res.get('total')
is_continue = len(json_res.get('list')) == params['rn'] and params['pn'] < total
if is_continue:
params['pn'] = params['pn'] + params['rn']
url = "http://m.news.baidu.com/news?tn=bdapinewsearch&" + urllib.parse.urlencode(params)
yield scrapy.Request(url=url,
meta={'params': params, 'keywords': keywords},
callback=self.parse_search)
def parse(self, response):
try:
print("进入try..........................")
item = BaiduItem()
item['comment_num'] = re.findall(
r'<span class="red" style="margin-right:3px">(.+?)</span>',
response.text)
item['read_num'] = ''
item['up_num'] = ''
item['zhuanfa'] = ''
item['url'] = response.meta.get('url')
item["MonitorName"] = response.meta.get('MonitorName')
if item['comment_num'] == []:
item['comment_num'] = ''
print("---------------------------------")
else:
item['comment_num'] = item['comment_num'][0]
print(item["url"], item["MonitorName"], item['comment_num'],
"====================")
except:
print("进入except.......................")
yield item
def parse_phone(self, response):
item = BaiduItem()
sort = 0
soup = bs(response.body, 'lxml')
urlarr = self.get_url_query(response.url)
search = urlarr['wd'] if 'wd' in urlarr.keys() else urlarr['word']
if soup.find('div', class_='rw-list') is not None:
rel = self.parse_rel(soup.find('div', class_='rw-list'))
else:
rel = None
if soup.find('div', class_='hint-toprq-tips') is not None:
rec = self.parse_rel(soup.find('div', class_='hint-toprq-tips'))
else:
rec = None
if soup.find('div', id='page-controller').find(
'a', class_='new-nextpage') is not None:
next_page_url = soup.find('div', id='page-controller').find(
'a', class_='new-nextpage')['href']
else:
next_page_url = soup.find('div', id='page-controller').find(
'a', class_='new-nextpage-only')['href']
if not re.search('ms\=1', next_page_url):
next_page_url = next_page_url + '&ms=1'
connects = soup.find('div', id='results').find_all('div',
class_='c-result')
for index, connect in enumerate(connects):
if (self.page_num == 1):
url = 'https://m.baidu.com/su?pre=1&p=3&json=1&wd=%s&sugmode=2&_=1493098255100' % search
recommendList = self.get_recommend(url)
recLen = len(recommendList)
if recLen - 1 >= index:
item['recom_search'] = recommendList[index]
else:
item['recom_search'] = ''
else:
item['recom_search'] = ''
sort += 1
tag_a = connect.find('div', class_='c-container').find('a')
item['title'] = tag_a.get_text()
if rel is not None:
try:
data = next(rel)
item['rel_url'] = response.url + data['url']
item['rel_tit'] = data['title']
except StopIteration:
item['rel_url'] = ''
item['rel_tit'] = ''
else:
item['rel_url'] = ''
item['rel_tit'] = ''
if rec is not None:
try:
data = next(rec)
item['rec_url'] = self.parse_url(response.url +
data['url'])
item['rec_tit'] = data['title']
except StopIteration:
item['rec_url'] = ''
item['rec_tit'] = ''
else:
item['rec_url'] = ''
item['rec_tit'] = ''
if connect['data-log'] is not None:
data_log = json.loads(connect['data-log'].replace("'", '"'))
item['sort'] = data_log['order']
if data_log['mu'] != '':
item['url'] = data_log['mu']
else:
item['url'] = self.parse_url(tag_a['href'])
else:
item['url'] = self.parse_url(tag_a['href'])
if connect.find('span', class_='c-gray') is not None:
item['time'] = connect.find('span', class_='c-gray').get_text()
else:
item['time'] = ''
item['page'] = '第%s页' % self.page_num
# self.write_log(tag_a.get_text())
yield item
if next_page_url != '' and self.page_num < self.total_page:
self.page_num += 1
yield scrapy.Request(url=next_page_url, callback=self.parse_phone)
def parse_pc(self, response):
"""提取相关数据"""
sort = 0
item = BaiduItem()
soup = bs(response.body, 'lxml')
urlarr = self.get_url_query(response.url)
search = urlarr['wd'] if urlarr['wd'] else urlarr['word']
connects = soup.find_all(attrs={"class": "c-container"})
if soup.find('span', 'hint_toprq_tips_items') is not None:
rec = self.parse_rec(soup.find('span', 'hint_toprq_tips_items'))
else:
rec = None
if soup.find('div', id='rs') is not None and sort == 0:
rel = self.parse_rel(soup.find('div', id='rs'))
else:
rel = None
for index, connect in enumerate(connects):
if (self.page_num == 1):
url = 'https://sp0.baidu.com/5a1Fazu8AA54nxGko9WTAnF6hhy/su?wd=%s&cb=&json=1' % search
recommendList = self.get_recommend(url)
recLen = len(recommendList)
if recLen - 1 >= index:
item['recom_search'] = recommendList[index]
else:
item['recom_search'] = ''
else:
item['recom_search'] = ''
sort += 1
if connect.find('h3', 't') is None:
item['title'] = connect.find('a').get_text()
item['url'] = connect.find('a')['href']
else:
item['title'] = connect.find('h3').get_text().strip()
item['url'] = self.parse_url(connect.find('h3').a['href'])
item['page'] = '第%s页' % self.page_num
item['sort'] = sort
if connect.find('span', ' newTimeFactor_before_abs m') is not None:
item['time'] = connect.find(
'span', ' newTimeFactor_before_abs m').get_text().replace(
u'\xa0-\xa0', '')
else:
item['time'] = ''
if rec is not None:
try:
data = next(rec)
item['rec_url'] = self.parse_url(response.url +
data['url'])
item['rec_tit'] = data['title']
except StopIteration:
item['rec_url'] = ''
item['rec_tit'] = ''
else:
item['rec_url'] = ''
item['rec_tit'] = ''
if rel is not None:
try:
data = next(rel)
item['rel_url'] = response.url + data['url']
item['rel_tit'] = data['title']
except StopIteration:
item['rel_url'] = ''
item['rel_tit'] = ''
else:
item['rel_url'] = ''
item['rel_tit'] = ''
yield item
if self.page_num < self.total_page:
self.page_num += 1
next_url = response.urljoin(
soup.find('div',
id='page').find('strong').next_sibling['href'])
yield scrapy.Request(next_url, callback=self.parse_pc)
def parse_week(self, browser, ban, index_t):
# TODO: 定义变量，收集结果
data = []
try:
# 第几周
zhou = browser.find_element_by_css_selector(
'.date-text').text.strip()
items = browser.find_elements_by_css_selector(
'.tab-content .list > .list-item')
for item in items:
# 排名
rank = item.find_element_by_css_selector(
'.content .rank').text.strip()
# 姓名
name = item.find_element_by_css_selector(
'.content .name').text.strip()
# 判断是否是‘周上升榜’
if ban.text == '周上升榜':
# 上升位数
index_str = item.find_element_by_css_selector(
'.value.upvalue').text.strip()
else:
# 行指数
line = item.find_element_by_css_selector(
'.content .line-light')
# 行指数-真实值
real_value = float(
line.value_of_css_property('width').replace('px', ''))
# 行指数-最大值元素
line_max = line.find_element_by_xpath('..')
# 最大值
max_value = float(
line_max.value_of_css_property('width').replace(
'px', ''))
# 指数值
index = round(100 * real_value / max_value, 2)
# 指数字符串
index_str = str(index).rstrip('0').rstrip('.') + '%'
trend = None
if len(item.find_elements_by_css_selector(
'.icon.trend-fair')) > 0:
trend = 'fair'
elif len(item.find_elements_by_css_selector(
'.icon.trend-down')) > 0:
trend = 'down'
else:
trend = 'up'
print('排名:{0}, 姓名:{1}, {2}:{3}, {4}:{5}'.format(
rank, name, ban.text, zhou, index_t.text, index_str))
print('-' * 30)
db_item = BaiduItem()
db_item['rank'] = rank
db_item['name'] = name
db_item['index'] = index_str
db_item['trend'] = trend
db_item['tab_item'] = ban.text
db_item['index_type'] = index_t.text
db_item['zhou'] = zhou
data.append(db_item)
except (NoSuchElementException, StaleElementReferenceException):
import ipdb
ipdb.set_trace()
# TODO: 返回结果
return data
def parse(self, response, typeid, category):
print "---> begin execute parse method"
item = BaiduItem()
items = []
# print response.body
print "---> tags: " + typeid
data = response.body
# print data
print "---> category: " + category
s = json.loads(data)
# print s
MyData = s["data"]
# print "Data++++++++"
# print data
num = len(MyData)
for i in xrange(0, num - 1):
# print MyData[i]
print '\n--------------------------------------------------------------------------%d---------------------------------------------------\n' % (
i)
print '---> now is : ' + str(
time.strftime('%Y-%m-%d %H:%M %S', time.localtime(
time.time())))
# print MyData[i]["hoverURL"]
width = MyData[i]["width"]
height = MyData[i]["height"]
size = str(width) + "*" + str(height)
print '---> title: ' + MyData[i]["fromPageTitleEnc"]
#测试下载使用gif地址
# image = str(MyData[i]["middleURL"])
# print "image------->" + image
simid = MyData[i]["os"]
Hash = simid.split(",")[0]
# now = datetime.utcnow().replace(microsecond=0).isoformat(' ')
now = time.strftime('%Y%m%d%H%M%S', time.localtime(time.time()))
item['imgsize'] = size
item['imgid'] = Hash
item['category'] = category
item['tag'] = typeid
item['updateTime'] = now
item['scrawl_time'] = now
item['title'] = MyData[i]["fromPageTitleEnc"]
item['fromURLHost'] = MyData[i]["fromURLHost"]
item['author'] = MyData[i]["fromURLHost"]
ObjURL = MyData[i]["objURL"]
fromURL = MyData[i]["fromURL"]
e = {
"w": "a",
"k": "b",
"v": "c",
"1": "d",
"j": "e",
"u": "f",
"2": "g",
"i": "h",
"t": "i",
"3": "j",
"h": "k",
"s": "l",
"4": "m",
"g": "n",
"5": "o",
"r": "p",
"q": "q",
"6": "r",
"f": "s",
"p": "t",
"7": "u",
"e": "v",
"o": "w",
"8": "1",
"d": "2",
"n": "3",
"9": "4",
"c": "5",
"m": "6",
"0": "7",
"b": "8",
"l": "9",
"a": "0",
"_z2C$q": ":",
"_z&e3B": ".",
"AzdH3F": "/"
}
#破译obj编码 使用字符替换 没有使用正则(正则学的太烂) TODO:正则
print ObjURL
ekeys = e.keys()
# print ekeys
strLength = len(ekeys)
URL_O = ObjURL
URL_O = URL_O.replace("_z2C$q", ":")
URL_O = URL_O.replace("_z&e3B", ".")
URL_O = URL_O.replace("AzdH3F", "/")
URLLength = len(URL_O)
print URL_O
URL_F = fromURL
URL_F = URL_F.replace("_z2C$q", ":")
URL_F = URL_F.replace("_z&e3B", ".")
URL_F = URL_F.replace("AzdH3F", "/")
URL_FLength = len(URL_F)
s_f = ""
#解析fromURL TODO
for j in xrange(0, URL_FLength):
URLKey = URL_F[j]
url = ord(URLKey)
if (url >= ord('a')
and url <= ord('w')) or (url >= ord('0')
and url <= ord('9')):
str_url = e[str(URLKey)]
s_f = s_f + str_url
else:
s_f = s_f + URLKey
print s_f
s = ""
for j in xrange(0, URLLength):
URLKey = URL_O[j]
url = ord(URLKey)
if (url >= ord('a')
and url <= ord('w')) or (url >= ord('0')
and url <= ord('9')):
str_url = e[str(URLKey)]
s = s + str_url
else:
s = s + URLKey
hash_url = hashlib.md5(s).hexdigest()[8:-8]
item['linkmd5id'] = hash_url
print "---> hash_url: " + hash_url
print "---> url_orgin: " + s
#图片下载本地地址
# path = "/Users/chenxingwang/Desktop/"+ category +"/"+hash_url[:2]+"/"+hash_url
uploadUrl = "/" + category + "/" + hash_url[:2] + "/" + hash_url
path = "../output/gif" + uploadUrl
print "---> folder for saving image: " + path
isExists = os.path.exists(path)
if not isExists:
os.makedirs(path)
origin_filename = "origin.gif"
static_filename = "static.jpg"
thumb_filename = "thumb.gif"
detail_filename = "detail.gif"
url_orgin = s
url_thumb = MyData[i]["thumbURL"] + ".gif"
url_hover = MyData[i]["hoverURL"]
url_middle = MyData[i]["middleURL"] + ".gif"
item['fromURL'] = s_f
item['objURL'] = uploadUrl + "/" + origin_filename
item['hoverURL'] = uploadUrl + "/" + static_filename
item['thumbURL'] = uploadUrl + "/" + thumb_filename
item['middleURL'] = uploadUrl + "/" + detail_filename
item['filesize'] = 0
item['frame'] = 0
yield item
sleep(1)
# items.append(item)
# STEP 1. 下载缩略图
print "---> STEP 1. begin download thumb image: " + url_thumb
req = urllib2.Request(
url_thumb,
headers={
"Upgrade-Insecure-Requests": "1",
"X-DevTools-Emulate-Network-Conditions-Client-Id":
"7A55439C-E6CF-420D-B294-7635B17E648B",
"User-Agent":
"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_11_4) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/51.0.2704.103 Safari/537.36",
"Accept":
"text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8",
"Accept-Encoding": "gzip, deflate, sdch",
"Accept-Language": "zh-CN,zh;q=0.8"
})
try:
# print "thumb image to download ----------->" + url_thumb
img_path = path + "/" + thumb_filename
imgData = urllib2.urlopen(req).read()
f = file(img_path, "wb")
f.write(imgData)
f.close()
# generate static.jpg
Image.open(img_path).convert('RGB').save(path + "/" +
static_filename)
print('------> thumb image downloaded. saved to ' + img_path)
except IOError as err:
print("------> IO error:{0}".format(err))
except:
print '------> download thumb image occured some error'
print("------> Unexpected error:", sys.exc_info())
file_name = path + "/" + origin_filename
# STEP 2. 下载原图
print "---> STEP 2. begin download origin image: " + url_orgin
try:
# 2 / 0
urllib.urlretrieve(url_orgin, '%s' % (file_name))
print('------> origin image downloaded. saved to ' + file_name)
except IOError as err:
print("------> IO error:{0}".format(err))
except:
print("------> Unexpected error:", sys.exc_info())
print '------> download origin image occurred some error, skip this item. !!!!!!!!!!!!!!!!!!!'
item[
'customer_exceptions'] = 'download origin image occurred some error'
yield item
# STEP 3. 生成详情图：detail.gif
print "---> STEP 3. begin generate detail image"
try:
# extract fileinfo of origin.gif
im = Image.open(file_name)
# in KB
origin_size = os.stat(file_name).st_size / 1024
item['filesize'] = str(origin_size)
origin_frame_count = 1
try:
while 1:
im.seek(im.tell() + 1)
origin_frame_count = origin_frame_count + 1
except EOFError:
pass # end of sequence
item['frame'] = str(origin_frame_count)
print "------> origin image info : size-" + str(
origin_size) + "KB, frames-" + str(origin_frame_count)
# generate detail.gif
origin_size_threshold = 1.5
if origin_size > origin_size_threshold * 1024:
print "------> origin image is bigger than " + str(
origin_size_threshold) + "M"
im = Image.open(file_name)
tmp_path = path + "/temp/"
if not os.path.exists(tmp_path):
os.makedirs(tmp_path)
print '------> origin file info: ' + str(im.info)
if 'duration' in im.info.keys():
origin_duration = im.info['duration'] / 1000.00
else:
origin_duration = 0
temp_filenames = []
# index = 1
reader = imageio.get_reader(file_name)
for i, tmp_im in enumerate(reader):
imageio.imwrite("%sframe%d.png" % (tmp_path, i),
tmp_im)
temp_filenames.append("%sframe%d.png" % (tmp_path, i))
# for frame in ImageSequence.Iterator(im):
# frame.save("%sframe%d.png" % (tmp_path, index))
# temp_filenames.append("%sframe%d.png" % (tmp_path, index))
# index += 1
# print temp_filenames
with imageio.get_writer(
path + "/" + detail_filename,
mode='I',
duration=origin_duration) as writer:
for temp_filename in temp_filenames:
tmp_im = Image.open(temp_filename)
tmp_im.thumbnail((230, 230))
tmp_im.save(temp_filename)
image = imageio.imread(temp_filename)
writer.append_data(image)
shutil.rmtree(tmp_path)
print '------> end: generate detail.gif'
else:
print "------> copy origin.gif as detail.gif"
shutil.copyfile(file_name, path + "/" + detail_filename)
# a = 2 / 0
except IOError as err:
print("------> IO error:{0}".format(err))
except:
print("------> Unexpected error:", sys.exc_info())
print '------> generate detail.gif occured some error'
print "---> finished data parse : " + hash_url
yield item
sleep(1)
def parse(self, response, typeid, category):
print "---> begin execute parse method"
item = BaiduItem()
items = []
# print response.body
print "---> tags: " + typeid
data = response.body
# print data
print "---> category: " + category
s = json.loads(data)
# print s
MyData = s["data"]
# print "Data++++++++"
# print data
num = len(MyData)
for i in xrange(0, num - 1):
# print MyData[i]
print '\n--------------------------------------------------------------------------%d---------------------------------------------------\n' % (
i)
print '---> now is : ' + str(
time.strftime('%Y-%m-%d %H:%M %S', time.localtime(
time.time())))
# print MyData[i]["hoverURL"]
width = MyData[i]["width"]
height = MyData[i]["height"]
size = str(width) + "*" + str(height)
print '---> title: ' + MyData[i]["fromPageTitleEnc"]
#测试下载使用gif地址
# image = str(MyData[i]["middleURL"])
# print "image------->" + image
simid = MyData[i]["os"]
Hash = simid.split(",")[0]
# now = datetime.utcnow().replace(microsecond=0).isoformat(' ')
now = time.strftime('%Y%m%d%H%M%S', time.localtime(time.time()))
item['imgsize'] = size
item['imgid'] = Hash
item['category'] = category
item['tag'] = typeid
item['updateTime'] = now
item['scrawl_time'] = now
item['title'] = MyData[i]["fromPageTitleEnc"]
item['fromURLHost'] = MyData[i]["fromURLHost"]
item['author'] = MyData[i]["fromURLHost"]
ObjURL = MyData[i]["objURL"]
fromURL = MyData[i]["fromURL"]
e = {
"w": "a",
"k": "b",
"v": "c",
"1": "d",
"j": "e",
"u": "f",
"2": "g",
"i": "h",
"t": "i",
"3": "j",
"h": "k",
"s": "l",
"4": "m",
"g": "n",
"5": "o",
"r": "p",
"q": "q",
"6": "r",
"f": "s",
"p": "t",
"7": "u",
"e": "v",
"o": "w",
"8": "1",
"d": "2",
"n": "3",
"9": "4",
"c": "5",
"m": "6",
"0": "7",
"b": "8",
"l": "9",
"a": "0",
"_z2C$q": ":",
"_z&e3B": ".",
"AzdH3F": "/"
}
#破译obj编码 使用字符替换 没有使用正则(正则学的太烂) TODO:正则
print ObjURL
ekeys = e.keys()
# print ekeys
strLength = len(ekeys)
URL_O = ObjURL
URL_O = URL_O.replace("_z2C$q", ":")
URL_O = URL_O.replace("_z&e3B", ".")
URL_O = URL_O.replace("AzdH3F", "/")
URLLength = len(URL_O)
print URL_O
URL_F = fromURL
URL_F = URL_F.replace("_z2C$q", ":")
URL_F = URL_F.replace("_z&e3B", ".")
URL_F = URL_F.replace("AzdH3F", "/")
URL_FLength = len(URL_F)
s_f = ""
#解析fromURL TODO
for j in xrange(0, URL_FLength):
URLKey = URL_F[j]
url = ord(URLKey)
if (url >= ord('a')
and url <= ord('w')) or (url >= ord('0')
and url <= ord('9')):
str_url = e[str(URLKey)]
s_f = s_f + str_url
else:
s_f = s_f + URLKey
print s_f
s = ""
for j in xrange(0, URLLength):
URLKey = URL_O[j]
url = ord(URLKey)
if (url >= ord('a')
and url <= ord('w')) or (url >= ord('0')
and url <= ord('9')):
str_url = e[str(URLKey)]
s = s + str_url
else:
s = s + URLKey
hash_url = hashlib.md5(s).hexdigest()[8:-8]
item['linkmd5id'] = hash_url
print "---> hash_url: " + hash_url
print "---> url_orgin: " + s
#图片下载本地地址
# path = "/Users/chenxingwang/Desktop/"+ category +"/"+hash_url[:2]+"/"+hash_url
uploadUrl = "/" + category + "/" + hash_url[:2] + "/" + hash_url
path = "../output/gif" + uploadUrl
print "---> folder for saving image: " + path
# isExists=os.path.exists(path)
# if not isExists:
# os.makedirs(path)
origin_filename = "origin.gif"
static_filename = "static.jpg"
thumb_filename = "thumb.gif"
detail_filename = "detail.gif"
url_orgin = s
url_thumb = MyData[i]["thumbURL"] + ".gif"
url_hover = MyData[i]["hoverURL"]
url_middle = MyData[i]["middleURL"] + ".gif"
item['fromURL'] = s_f
item['objURL'] = uploadUrl + "/" + origin_filename
item['hoverURL'] = uploadUrl + "/" + static_filename
item['thumbURL'] = uploadUrl + "/" + thumb_filename
item['middleURL'] = uploadUrl + "/" + detail_filename
item['filesize'] = 0
item['frame'] = 0
item['source_thumb_url'] = url_thumb
item['source_original_url'] = s
sleep(1)
yield item
