From 43f96c6988c115779bec480325005a12a39118da Mon Sep 17 00:00:00 2001
From: Richard Patel <terorie@alphakevin.club>
Date: Tue, 18 Dec 2018 15:39:41 +0100
Subject: [PATCH] Benchmark: Reference parser

---
 crawl_test.go | 38 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 38 insertions(+)

diff --git a/crawl_test.go b/crawl_test.go
index 2ca3f84..406447f 100644
--- a/crawl_test.go
+++ b/crawl_test.go
@@ -1,7 +1,11 @@
 package main
 
 import (
+	"bytes"
+	"github.com/PuerkitoBio/goquery"
 	"github.com/terorie/od-database-crawler/fasturl"
+	"net/url"
+	"strings"
 	"testing"
 )
 
@@ -19,3 +23,37 @@ func BenchmarkParseDir(b *testing.B) {
 		}
 	}
 }
+
+func BenchmarkParseDirReference(b *testing.B) {
+	for n := 0; n < b.N; n++ {
+		u, err := url.Parse("http://archive.ubuntu.com/ubuntu/indices/")
+		if err != nil {
+			b.Fatal("Failed to parse URL", err)
+		}
+
+		_, err = referenceParseDir([]byte(apache2Listing), u)
+		if err != nil {
+			b.Fatal("Failed to extract links", err)
+		}
+	}
+}
+
+func referenceParseDir(body []byte, baseUri *url.URL) (links []*url.URL, err error) {
+	doc, err := goquery.NewDocumentFromReader(bytes.NewReader(body))
+	if err != nil { return nil, err }
+
+	doc.Find("a[href]").Each(func(i int, s *goquery.Selection) {
+		href, _ := s.Attr("href")
+
+		sub, err := baseUri.Parse(href)
+		if err != nil { return } // continue
+
+		if !strings.HasPrefix(sub.String(), baseUri.String()) {
+			return // continue
+		}
+
+		links = append(links, sub)
+	})
+
+	return
+}