From 41245d0fcf36ea318e297384c2764c1d2779d5d7 Mon Sep 17 00:00:00 2001 From: Yomguithereal Date: Wed, 15 Nov 2023 10:50:20 +0100 Subject: [PATCH] Fixing focus spider --- minet/crawl/focus.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/minet/crawl/focus.py b/minet/crawl/focus.py index 3119e01b6a..8c7bf01a2a 100644 --- a/minet/crawl/focus.py +++ b/minet/crawl/focus.py @@ -144,7 +144,7 @@ def process(self, job: CrawlJob, response: Response): if not should_follow_href(url): continue - unique_urls.add(response.resolve(url)) + unique_urls.add(response.urljoin(url)) has_relevant_content, relevant_size = self.content_relevance(content)