Fix br tag conversion to newlines in goodreads book descriptions

goodreads/book.go

@@ -141,7 +141,9 @@ func (e *Edition) Sanitise() {
 	// Break tags need to be specially handled to add new lines as html2text does
 	// not convert them to new lines properly
 	e.Description = descriptionAlternativeCoverRegex.ReplaceAllString(e.Description, "")
-	e.Description = breakTagRegex.ReplaceAllString(e.Description, "\n")
+	// HACK: html2text only handles <br> and <br/>, goodreads uses <br />
+	// replace unsupported br tag with a supported one
+	e.Description = breakTagRegex.ReplaceAllString(e.Description, "<br>")
 	e.Description = html2text.HTML2TextWithOptions(e.Description, html2text.WithUnixLineBreaks())
 	e.Description = strings.TrimSpace(e.Description)
 

goodreads/book_test.go

@@ -37,3 +37,40 @@ func TestUnmarshalGenres(t *testing.T) {
 	expectedGenres := goodreads.Genres{"Fantasy", "Classic", "Fiction"}
 	require.Equal(t, expectedGenres, genres)
 }
+
+func TestBookUnmarshalBrTagReplacement(t *testing.T) {
+	testXML := `
+	<GoodreadsResponse>
+		<book>
+			<id>123</id>
+			<title>Test Book</title>
+			<description><![CDATA[Test description<br />2. line<br/>3. line<br>4. line]]></description>
+			<work>
+				<original_title>Test Book</original_title>
+				<ratings_sum>100</ratings_sum>
+				<ratings_count>10</ratings_count>
+			</work>
+			<popular_shelves>
+				<shelf name="fiction" count="100"/>
+			</popular_shelves>
+		</book>
+	</GoodreadsResponse>
+	`
+
+	var response struct {
+		Book goodreads.Book `xml:"book"`
+	}
+
+	err := xml.Unmarshal([]byte(testXML), &response)
+	require.NoError(t, err)
+
+	description := response.Book.BestEdition.Description
+	t.Logf("Description after processing: %q", description)
+
+	// Verify that <br> tags have been correctly converted to newlines
+	require.Contains(t, description, "Test description\n2. line\n3. line\n4. line")
+
+	// Verify that no HTML br tags remain
+	require.NotContains(t, description, "<br")
+	require.NotContains(t, description, "br>")
+}