Add ability to update url who are one week old and content modified

2 files changed, 51 insertions, 1 deletions

diff --git a/crawler/nevrax/spiders/scrape.py b/crawler/nevrax/spiders/scrape.py
index d27aecf..c9a8a53 100644
--- a/crawler/nevrax/spiders/scrape.py
+++ b/crawler/nevrax/spiders/scrape.py
@@ -8,7 +8,7 @@ import config
 import datetime
 
 class NevraxSpider(CrawlSpider):
-    name = "nevrax"
+    name = "nevrax_crawler"
     custom_settings = {
         'ITEM_PIPELINES': {
             'crawler.nevrax.pipelines.NevraxPipeline': 0
diff --git a/crawler/nevrax/spiders/update.py b/crawler/nevrax/spiders/update.py
new file mode 100644
index 0000000..b3f7aa1
--- /dev/null
+++ b/crawler/nevrax/spiders/update.py
@@ -0,0 +1,50 @@
+# -*- coding: utf-8 -*-
+import scrapy
+from scrapy.spiders import CrawlSpider, Rule
+from scrapy.linkextractors import LinkExtractor
+from scrapy import Selector
+
+import config
+import datetime
+
+from database.models import Nevrax
+
+from dateutil.relativedelta import *
+
+import logging
+
+class NevraxSpider(CrawlSpider):
+    name = "nevrax_updater"
+    custom_settings = {
+        'ITEM_PIPELINES': {
+            'crawler.nevrax.pipelines.NevraxPipeline': 0
+        }
+    }
+    allow_domains = [config.NEVRAX_URL]
+
+    datas = Nevrax.select(Nevrax.url).dicts()
+    datas_array = []
+    for value in datas:
+        datas_array.append(value['url'])
+    start_urls = datas_array
+
+    def start_requests(self):
+        for url in self.start_urls:
+            logging.info(url)
+            try:
+                page = Nevrax.get(Nevrax.url == url)
+                if page.date_updated < datetime.datetime.now()+relativedelta(weeks=-1):
+                    yield scrapy.Request(url, callback=self.parse_url, dont_filter=True)
+            except Nevrax.DoesNotExist:
+                yield scrapy.Request(url, callback=self.parse_url, dont_filter=True)
+                continue
+
+    def parse_url(self, response):
+        sel = Selector(response)
+        yield {
+                'url': response.url,
+                'title': response.css('title::text').extract_first(),
+                'content': ''.join(sel.select("//div[@class='bodya']//text()").extract()).strip(),
+                'content_length': len(response.body),
+                'date_updated': datetime.datetime.now()
+        }