From 0b4c723d2c310310efba41f58ee3a626510bcd3a Mon Sep 17 00:00:00 2001 From: Philipp Tanlak Date: Sat, 17 Feb 2024 21:47:43 +0100 Subject: Mark non 2XX status code as errors --- scrape.go | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'scrape.go') diff --git a/scrape.go b/scrape.go index 3e38e46..bf0e1d6 100644 --- a/scrape.go +++ b/scrape.go @@ -9,6 +9,7 @@ import ( "log" "net/http" "net/http/cookiejar" + "strconv" "sync" "github.com/cornelk/hashmap" @@ -187,6 +188,10 @@ func (s *Scraper) process(url string, depth int) { response.StatusCode = resp.StatusCode response.Headers = resp.Header + if response.StatusCode < 200 || response.StatusCode >= 300 { + response.Error = strconv.Itoa(response.StatusCode) + " " + http.StatusText(response.StatusCode) + } + response.Body, err = io.ReadAll(resp.Body) if err != nil { response.Error = err -- cgit v1.2.3