fix: make domain validation in list parser more lenient

go.mod

@@ -79,7 +79,6 @@ require (
 	github.com/KyleBanks/depth v1.2.1 // indirect
 	github.com/Masterminds/goutils v1.1.1 // indirect
 	github.com/alicebob/gopher-json v0.0.0-20200520072559-a9ecdc9d1d3a // indirect
-	github.com/asaskevich/govalidator v0.0.0-20210307081110-f21760c49a8d
 	github.com/beorn7/perks v1.0.1 // indirect
 	github.com/cespare/xxhash/v2 v2.1.2 // indirect
 	github.com/cpuguy83/go-md2man/v2 v2.0.2 // indirect

go.sum

@@ -62,8 +62,6 @@ github.com/alicebob/miniredis/v2 v2.30.1/go.mod h1:b25qWj4fCEsBeAAR2mlb0ufImGC6u
 github.com/antihax/optional v1.0.0/go.mod h1:uupD/76wgC+ih3iEmQUL+0Ugr19nfwCT1kdvxnR2qWY=
 github.com/asaskevich/EventBus v0.0.0-20200907212545-49d423059eef h1:2JGTg6JapxP9/R33ZaagQtAM4EkkSYnIAlOG5EI8gkM=
 github.com/asaskevich/EventBus v0.0.0-20200907212545-49d423059eef/go.mod h1:JS7hed4L1fj0hXcyEejnW57/7LCetXggd+vwrRnYeII=
-github.com/asaskevich/govalidator v0.0.0-20210307081110-f21760c49a8d h1:Byv0BzEl3/e6D5CLfI0j/7hiIEtvGVFPCZ7Ei2oq8iQ=
-github.com/asaskevich/govalidator v0.0.0-20210307081110-f21760c49a8d/go.mod h1:WaHUgvxTVq04UNunO+XhnAqY/wQc+bxr74GqbsZ/Jqw=
 github.com/avast/retry-go/v4 v4.3.3 h1:G56Bp6mU0b5HE1SkaoVjscZjlQb0oy4mezwY/cGH19w=
 github.com/avast/retry-go/v4 v4.3.3/go.mod h1:rg6XFaiuFYII0Xu3RDbZQkxCofFwruZKW8oEF1jpWiU=
 github.com/beorn7/perks v0.0.0-20180321164747-3a771d992973/go.mod h1:Dwedo/Wpr24TaqPxmxbtue+5NUziq4I4S80YR8gNf3Q=

lists/parsers/hosts.go

@@ -10,14 +10,21 @@ import (
 	"regexp"
 	"strings"
 
-	"github.com/asaskevich/govalidator"
 	"github.com/hashicorp/go-multierror"
 	"golang.org/x/net/idna"
 )
 
-const maxDomainNameLength = 255 // https://www.rfc-editor.org/rfc/rfc1034#section-3.1
+const (
+	maxDomainNameLength = 255 // https://www.rfc-editor.org/rfc/rfc1034#section-3.1
 
-var domainNameRegex = regexp.MustCompile(govalidator.DNSName)
+	dnsLabelPattern = `[a-zA-Z0-9_-]{1,63}`
+)
+
+// Validate a domain name, but with extra flexibility:
+// - no restriction on the start or end of labels
+//
+// https://www.rfc-editor.org/rfc/rfc1034#section-3.5
+var domainNameRegex = regexp.MustCompile(`^` + dnsLabelPattern + `(\.` + dnsLabelPattern + `)*[\._]?$`)
 
 // Hosts parses `r` as a series of `HostsIterator`.
 // It supports both the hosts file and host list formats.

lists/parsers/hosts_test.go

@@ -307,6 +307,10 @@ var _ = Describe("HostList", func() {
 
 				// Domain name w/ rune not supported by `idna.Lookup`
 				"domain_underscore.tld",
+
+				// invalid domain names we want to support
+				"-start-with-a-hyphen.com",
+				"end-with-a-hyphen-.com",
 			)
 		})
 
@@ -341,11 +345,21 @@ var _ = Describe("HostList", func() {
 			Expect(entry.String()).Should(Equal("domain_underscore.tld"))
 			Expect(sut.Position()).Should(Equal("line 8"))
 
+			entry, err = sut.Next(context.Background())
+			Expect(err).Should(Succeed())
+			Expect(entry.String()).Should(Equal("-start-with-a-hyphen.com"))
+			Expect(sut.Position()).Should(Equal("line 9"))
+
+			entry, err = sut.Next(context.Background())
+			Expect(err).Should(Succeed())
+			Expect(entry.String()).Should(Equal("end-with-a-hyphen-.com"))
+			Expect(sut.Position()).Should(Equal("line 10"))
+
 			_, err = sut.Next(context.Background())
 			Expect(err).ShouldNot(Succeed())
 			Expect(err).Should(MatchError(io.EOF))
 			Expect(IsNonResumableErr(err)).Should(BeTrue())
-			Expect(sut.Position()).Should(Equal("line 9"))
+			Expect(sut.Position()).Should(Equal("line 11"))
 		})
 	})