Merge pull request #4 from TeamHG-Memex/domain-limits-fix

don't go out of domain for pagination URLs

undercrawler/spiders/base_spider.py

@@ -75,7 +75,7 @@ def _pagination_urls(self, response):
         return [
             url for url in
             unique(canonicalize_url(url) for url in autopager.urls(response))
-            if url.startswith('http')
+            if self.link_extractor.matches(url)
         ]