From 81e16ffb3885a311e4a460880c2a7ec55eba4295 Mon Sep 17 00:00:00 2001 From: ale Date: Fri, 31 Aug 2018 11:09:12 +0100 Subject: Do not drop /index.html at the end of URLs --- crawler.go | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/crawler.go b/crawler.go index 9776cfc..d5c2e88 100644 --- a/crawler.go +++ b/crawler.go @@ -158,7 +158,7 @@ func (c *Crawler) Enqueue(link Outlink, depth int) error { } // Normalize the URL. - urlStr := purell.NormalizeURL(link.URL, purell.FlagsSafe|purell.FlagRemoveDotSegments|purell.FlagRemoveDuplicateSlashes|purell.FlagRemoveFragment|purell.FlagRemoveDirectoryIndex|purell.FlagSortQuery) + urlStr := purell.NormalizeURL(link.URL, purell.FlagsSafe|purell.FlagRemoveDotSegments|purell.FlagRemoveDuplicateSlashes|purell.FlagRemoveFragment|purell.FlagSortQuery) // Protect the read-modify-update below with a mutex. c.enqueueMx.Lock() -- cgit v1.2.3-54-g00ecf