Disable strict XML parsing

This change should improve parsing of broken XML feeds.

See https://golang.org/pkg/encoding/xml/#Decoder

2 files changed, 20 insertions, 0 deletions

diff --git a/reader/rdf/parser.go b/reader/rdf/parser.go
index f942317..861ce8c 100644
--- a/reader/rdf/parser.go
+++ b/reader/rdf/parser.go
@@ -18,6 +18,7 @@ func Parse(data io.Reader) (*model.Feed, *errors.LocalizedError) {
 	feed := new(rdfFeed)
 	decoder := xml.NewDecoder(data)
 	decoder.Entity = xml.HTMLEntity
+	decoder.Strict = false
 	decoder.CharsetReader = encoding.CharsetReader
 
 	err := decoder.Decode(feed)
diff --git a/reader/rdf/parser_test.go b/reader/rdf/parser_test.go
index 734b763..4f3d033 100644
--- a/reader/rdf/parser_test.go
+++ b/reader/rdf/parser_test.go
@@ -403,3 +403,22 @@ func TestParseFeedWithHTMLEntity(t *testing.T) {
 		t.Errorf(`Incorrect title, got: %q`, feed.Title)
 	}
 }
+
+func TestParseFeedWithInvalidCharacterEntity(t *testing.T) {
+	data := `<?xml version="1.0" encoding="utf-8"?>
+	<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns="http://purl.org/rss/1.0/">
+	  <channel>
+			<title>Example Feed</title>
+			<link>http://example.org/a&b</link>
+	  </channel>
+	</rdf:RDF>`
+
+	feed, err := Parse(bytes.NewBufferString(data))
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	if feed.SiteURL != "http://example.org/a&b" {
+		t.Errorf(`Incorrect URL, got: %q`, feed.SiteURL)
+	}
+}