From 4f358aa0f3b1299378d677b02d36392e76848966 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Fr=C3=A9d=C3=A9ric=20Guillot?= <f@miniflux.net>
Date: Fri, 30 Oct 2020 23:20:44 -0700
Subject: [PATCH] Do not escape HTML for Atom 1.0 text content during parsing

Avoid encoding single quotes to HTML entities (&#39;).

Feed contents are sanitized after parsing.
---
 reader/atom/atom_10.go      |  5 +---
 reader/atom/atom_10_test.go | 59 ++++++++++++++++++++++++++++++++++++-
 2 files changed, 59 insertions(+), 5 deletions(-)

diff --git a/reader/atom/atom_10.go b/reader/atom/atom_10.go
index 4e49a8d4..25bdec37 100644
--- a/reader/atom/atom_10.go
+++ b/reader/atom/atom_10.go
@@ -6,7 +6,6 @@ package atom // import "miniflux.app/reader/atom"
 
 import (
 	"encoding/xml"
-	"html"
 	"strconv"
 	"strings"
 	"time"
@@ -221,10 +220,8 @@ func (a *atom10Text) String() string {
 	switch {
 	case a.Type == "xhtml":
 		content = a.XML
-	case a.Type == "html":
+	default:
 		content = a.Data
-	case a.Type == "text" || a.Type == "":
-		content = html.EscapeString(a.Data)
 	}
 
 	return strings.TrimSpace(content)
diff --git a/reader/atom/atom_10_test.go b/reader/atom/atom_10_test.go
index 1e763ca4..ad897440 100644
--- a/reader/atom/atom_10_test.go
+++ b/reader/atom/atom_10_test.go
@@ -359,7 +359,7 @@ func TestParseEntrySummaryWithPlainText(t *testing.T) {
 		t.Fatal(err)
 	}
 
-	if feed.Entries[0].Content != "&lt;Some text.&gt;" {
+	if feed.Entries[0].Content != "<Some text.>" {
 		t.Errorf("Incorrect entry content, got: %s", feed.Entries[0].Content)
 	}
 }
@@ -599,6 +599,63 @@ func TestParseInvalidXml(t *testing.T) {
 	}
 }
 
+func TestParseTitleWithSingleQuote(t *testing.T) {
+	data := `
+		<?xml version="1.0" encoding="utf-8"?>
+		<feed xmlns="http://www.w3.org/2005/Atom">
+			<title>' or ’</title>
+			<link href="http://example.org/"/>
+		</feed>
+	`
+
+	feed, err := Parse(bytes.NewBufferString(data))
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	if feed.Title != "' or ’" {
+		t.Errorf(`Incorrect title, got: %q`, feed.Title)
+	}
+}
+
+func TestParseTitleWithEncodedSingleQuote(t *testing.T) {
+	data := `
+		<?xml version="1.0" encoding="utf-8"?>
+		<feed xmlns="http://www.w3.org/2005/Atom">
+			<title type="html">Test&#39;s Blog</title>
+			<link href="http://example.org/"/>
+		</feed>
+	`
+
+	feed, err := Parse(bytes.NewBufferString(data))
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	if feed.Title != "Test's Blog" {
+		t.Errorf(`Incorrect title, got: %q`, feed.Title)
+	}
+}
+
+func TestParseTitleWithSingleQuoteAndHTMLType(t *testing.T) {
+	data := `
+		<?xml version="1.0" encoding="utf-8"?>
+		<feed xmlns="http://www.w3.org/2005/Atom">
+			<title type="html">O’Hara</title>
+			<link href="http://example.org/"/>
+		</feed>
+	`
+
+	feed, err := Parse(bytes.NewBufferString(data))
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	if feed.Title != "O’Hara" {
+		t.Errorf(`Incorrect title, got: %q`, feed.Title)
+	}
+}
+
 func TestParseWithHTMLEntity(t *testing.T) {
 	data := `
 		<?xml version="1.0" encoding="utf-8"?>