修改时间为时间戳格式，同时运行三个spider写入一个文件会出现脏数据

news_spider/news_spider/commands/crawlall.py

@@ -35,5 +35,4 @@ class Command(ScrapyCommand):
 	for spidername in args or spider_loader.list():
 	    print "*********cralall spidername************" + spidername
 	    self.crawler_process.crawl(spidername, **opts.spargs)
-
         self.crawler_process.start()

news_spider/news_spider/items.py

@@ -13,3 +13,10 @@ class NewsSpiderItem(scrapy.Item):
 	title = scrapy.Field()
 	time = scrapy.Field()
 	content = scrapy.Field()
+	url = scrapy.Field()
+
+class TitleSpiderItem(scrapy.Item):
+	title = scrapy.Field()
+	time = scrapy.Field()
+	url = scrapy.Field()
+

news_spider/news_spider/pipelines.py

@@ -6,21 +6,34 @@
 # See: http://doc.scrapy.org/en/latest/topics/item-pipeline.html
 import codecs
 import json
+from items import TitleSpiderItem
+import fcntl
 
 
 class NewsSpiderPipeline(object):
 	
-	def process_item(self,item,spider):
-		return item
-
-
-class TouTiaoPipeline(object):
 	def __init__(self):
-#		self.file = codecs.open('toutiao.json','wb',encoding='utf-8')
+		self.file = open('news.json','wb')
-		self.file = open('toutiao.json','wb')
 
 	def process_item(self,item,spider):
+		fcntl.flock(self.file,fcntl.LOCK_EX)
 		line = json.dumps(dict(item))+'\n'
-#		self.file.write(line.decode("unicode_escape"))
 		self.file.write(line)
+		fcntl.flock(self.file,fcntl.LOCK_UN)
+		return item
+
+
+class TitlePipeline(object):
+	def __init__(self):
+		file_title = open('title.json','wb')
+
+	def process_item(self,item,spider):
+		fcntl.flock(file_title,fcntl.LOCK_EX)
+		title_item = TitleSpiderItem()
+		title_item['title'] = item['title']
+		title_item['time'] = item['time']
+		title_item['url'] = item['url']
+		line = json.dumps(dict(title_item))+'\n'
+		file_title.write(line)
+		fcntl.flock(file_title,fcntl.LOCK_UN)
 		return item

news_spider/news_spider/settings.py

@@ -63,9 +63,10 @@ USER_AGENT = 'Mozilla/5.0 (Windows NT 6.3; Win64; x64) AppleWebKit/537.36 (KHTML
 
 # Configure item pipelines
 # See http://scrapy.readthedocs.org/en/latest/topics/item-pipeline.html
-#ITEM_PIPELINES = {
+ITEM_PIPELINES = {
 #    'news_spider.pipelines.TouTiaoPipeline': 300,
-#}
+    'news_spider.pipelines.TitlePipeline': 500,
+}
 
 # Enable and configure the AutoThrottle extension (disabled by default)
 # See http://doc.scrapy.org/en/latest/topics/autothrottle.html

news_spider/news_spider/spiders/NetEase.py

@@ -12,14 +12,14 @@ class NetEaseSpider(scrapy.Spider):
 	allowed_domains=['news.163.com']
 
 	base_url = 'http://snapshot.news.163.com/wgethtml/http+!!news.163.com!'
-	year = ['2016','2015']
+#	year = ['2016','2015']
-	month = ['12','11','10','09','08','07','06','05','04','03','02','01']
+#	month = ['12','11','10','09','08','07','06','05','04','03','02','01']
-	day = ['31','30','29','28','27','26','25','24','23','22','21',
+#	day = ['31','30','29','28','27','26','25','24','23','22','21',
-		   '20','19','18','17','16','15','14','13','12','11','10',
+#		   '20','19','18','17','16','15','14','13','12','11','10',
-		   '09','08','07','06','05','04','03','02','01']
+#		   '09','08','07','06','05','04','03','02','01']
-#	day = ['31']
+	day = ['31']
-#	year = ['2016']
+	year = ['2016']
-#	month = ['03']
+	month = ['03']
 
 	def parse(self,response):
 		for y in self.year:
@@ -37,14 +37,16 @@ class NetEaseSpider(scrapy.Spider):
 	def parseNews(self,response):
 		data = response.xpath("//div[@class='post_content_main']")
 		item = NewsSpiderItem()
-		time = data.xpath("//div[@class='post_time_source']/text()").extract()
+		timee = data.xpath("//div[@class='post_time_source']/text()").extract()
 		title = data.xpath("//h1/text()").extract()
 		content = data.xpath("//div[@class='post_text']/p/text()").extract()
 
 		time_pattern = re.compile("[0-9]{4}-[0-9]{2}-[0-9]{2}\s[0-9]{2}:[0-9]{2}")
-		if(len(time)!=0 and len(title)!=0 and len(content)!=0):
+		if(len(timee)!=0 and len(title)!=0 and len(content)!=0):
-			item['time'] = time_pattern.findall(time[0])[0]
+			tm = time_pattern.findall(timee[0])[0]
+			item['time'] = int(time.mktime(time.strptime(tm,'%Y-%m-%d %H:%M')))
 			item['title'] = title[0]
+			item['url'] = response.url
 			cc=''
 			if(len(content)!=0):
 				for c in content:

news_spider/news_spider/spiders/Tencent.py

@@ -12,13 +12,17 @@ class TencentSpider(scrapy.Spider):
 	allowed_domains=['news.qq.com']
 
 	base_url = 'http://news.qq.com/b/history/index'
-	year = ['2016','2015','2014']
+#	year = ['2016','2015','2014']
-	month = ['12','11','10','09','08','07','06','05','04','03','02','01']
+#	month = ['12','11','10','09','08','07','06','05','04','03','02','01']
-	day = ['31','30','29','28','27','26','25','24','23','22','21',
+#	day = ['31','30','29','28','27','26','25','24','23','22','21',
-		   '20','19','18','17','16','15','14','13','12','11','10',
+#		   '20','19','18','17','16','15','14','13','12','11','10',
-		   '09','08','07','06','05','04','03','02','01']
+#		   '09','08','07','06','05','04','03','02','01']
 	tp = ['am','pm']
 
+	day = ['31']
+	year = ['2016']
+	month = ['03']
+
 	def parse(self,response):
 		for y in self.year:
 			for m in self.month:
@@ -37,14 +41,16 @@ class TencentSpider(scrapy.Spider):
 	def parseNews(self,response):
 		data = response.xpath("//div[@id='C-Main-Article-QQ']")
 		item = NewsSpiderItem()
-		time = data.xpath("//span[@class='article-time']/text()").extract()
+		timee = data.xpath("//span[@class='article-time']/text()").extract()
 		title = data.xpath("//div[@class='hd']//h1/text()").extract()
 		content = data.xpath("//p/text()").extract()
 
 		time_pattern = re.compile("[0-9]{4}-[0-9]{2}-[0-9]{2}\s[0-9]{2}:[0-9]{2}")
-		if(len(time)!=0 and len(title)!=0 and len(content)!=0):
+		if(len(timee)!=0 and len(title)!=0 and len(content)!=0):
-			item['time'] = time_pattern.findall(time[0])[0]
+			tm = time_pattern.findall(timee[0])[0]
+			item['time'] = int(time.mktime(time.strptime(tm,'%Y-%m-%d %H:%M')))
 			item['title'] = title[0]
+			item['url'] = response.url
 			cc=''
 			if(len(content)!=0):
 				for c in content:

news_spider/news_spider/spiders/TouTiaoSpider.py

@@ -12,7 +12,7 @@ class TouTiaoSpider(scrapy.Spider):
 	]
 	base_class_url = 'http://toutiao.com/articles_news_society'
 	base_url = 'http://toutiao.com'
-	maxpage = 501;#允许爬的最大的页数
+	maxpage = 10;#允许爬的最大的页数
 	category = ['articles_news_society','articles_news_entertainment',
 	'articles_movie','articles_news_tech','articles_digital',
 	'articels_news_sports','articles_news_finance','articles_news_military',
@@ -36,12 +36,15 @@ class TouTiaoSpider(scrapy.Spider):
 #解析具体新闻内容 
 	def parseNews(self,response):
 		articles = response.xpath("//div[@id='pagelet-article']")
-		for article in articles:
+		item = NewsSpiderItem()
-			item = NewsSpiderItem()
+		title = articles.xpath("//div[@class='article-header']/h1/text()").extract()[0]
-			item['title'] = article.xpath("//div[@class='article-header']/h1/text()").extract()[0]
+		tm = articles.xpath("//div[@id='pagelet-article']//span[@class='time']/text()").extract()[0]
-			item['time'] = article.xpath("//div[@id='pagelet-article']//span[@class='time']/text()").extract()[0]
+		content = articles.xpath("//div[@class='article-content']//p/text()").extract()
-			content = article.xpath("//div[@class='article-content']//p/text()").extract()
+
-#item['content'] = article.xpath("//div[@class='article-content']//p/text()").extract()
+		if(len(title)!=0 and len(tm)!=0 and len(content)!=0):
+			item['title'] = title
+			item['time'] = int(time.mktime(time.strptime(tm,'%Y-%m-%d %H:%M')))
+			item['url'] = response.url
 			cc=''
 			if(len(content) != 0):
 				for c in content:

news_spider/scrapy.cfg

@@ -7,5 +7,5 @@
 default = news_spider.settings
 
 [deploy]
-#url = http://localhost:6800/
+url = http://localhost:6800/
 project = news_spider

news_spider/show.py

@@ -5,13 +5,19 @@ reload(sys)
 sys.setdefaultencoding( "utf-8" )
 
 file = open(sys.argv[1])
-data = json.load(file)
+while 1:
+	line = file.readline()
+	if not line:
+		break
+	data = json.loads(line)
+	print data['time'],data['title'],data['url']
 
-c = 0
+#data = json.load(file)
-for article in data:
+#c = 0
-	c+=1
+#for article in data:
-	print article['time'],"--------",article['title']
+#	c+=1
 #	print "[----Time-----]\n",article['time'],article['title']
 #	print "[----Title----]\n",article['title']
+#	print article['time'],article['title']
 #	print "[----Article--]\n",article['content'],"\n\n"
-print c
+

news_spider/test.py

@@ -1,6 +1,7 @@
 import re
+import time
 
-time = " - -- 2015-06-15 15:34   "
+timee = " - -- 2015-06-15 15:34   "
 
 day = ['31','30','29','28','27','26','25','24','23','22','21',
 	   '20','19','18','17','16','15','14','13','12','11','10',
@@ -8,4 +9,7 @@ day = ['31','30','29','28','27','26','25','24','23','22','21',
 pattern = re.compile("[0-9]{4}-[0-9]{2}-[0-9]{2}\s[0-9]{2}:[0-9]{2}")
 #pattern = re.compile("[0-9]{4}-[0-9]{2}-[0-9]{2}")
 #pattern = re.compile("[0-9]")
-print pattern.findall(time)[0]
+tm = pattern.findall(timee)[0]
+
+a = time.mktime(time.strptime(tm,'%Y-%m-%d %H:%M'))
+print int(a)