From 4b68bc8916ae0340a73c15028a882de0a084e093 Mon Sep 17 00:00:00 2001
From: l3akage <info@martinpoppen.de>
Date: Mon, 9 Jun 2025 12:10:45 +0200
Subject: [PATCH 1/3] Fix br tag conversion to newlines in goodreads book
 descriptions

- Replace <br /> tags with <br> before html2text processing
- html2text library doesn't handle <br /> (with space) format
- Add test to verify br tag replacement works correctly

Fixes #16
---
 goodreads/book.go      |  4 +++-
 goodreads/book_test.go | 37 +++++++++++++++++++++++++++++++++++++
 2 files changed, 40 insertions(+), 1 deletion(-)

diff --git a/goodreads/book.go b/goodreads/book.go
index 2e82830..455bdec 100644
--- a/goodreads/book.go
+++ b/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 is a supported one
+	e.Description = breakTagRegex.ReplaceAllString(e.Description, "<br>")
 	e.Description = html2text.HTML2TextWithOptions(e.Description, html2text.WithUnixLineBreaks())
 	e.Description = strings.TrimSpace(e.Description)
 
diff --git a/goodreads/book_test.go b/goodreads/book_test.go
index aed5d4d..006f351 100644
--- a/goodreads/book_test.go
+++ b/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[This is a test description.<br />This should be on a new line.<br/>Another line.<br>Final 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.\nThis should be on a new line.\nAnother line.\nFinal line.")
+
+	// Verify that no HTML br tags remain
+	require.NotContains(t, description, "<br")
+	require.NotContains(t, description, "br>")
+}

From 49b3624288cc2de664878e5d77b25619d6e4a7dd Mon Sep 17 00:00:00 2001
From: Martin <l3akage@users.noreply.github.com>
Date: Wed, 11 Jun 2025 21:18:10 +0200
Subject: [PATCH 2/3] Update goodreads/book.go

Co-authored-by: Arran Hobson Sayers <32173585+ahobsonsayers@users.noreply.github.com>
---
 goodreads/book.go | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/goodreads/book.go b/goodreads/book.go
index 455bdec..e52a35d 100644
--- a/goodreads/book.go
+++ b/goodreads/book.go
@@ -142,7 +142,7 @@ func (e *Edition) Sanitise() {
 	// not convert them to new lines properly
 	e.Description = descriptionAlternativeCoverRegex.ReplaceAllString(e.Description, "")
 	// HACK: html2text only handles <br> and <br/>, goodreads uses <br />
-	// replace unsupported br tag is a supported one
+	// 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)

From ee49bc731da0808894908c7333ca0da13949d032 Mon Sep 17 00:00:00 2001
From: l3akage <info@martinpoppen.de>
Date: Fri, 13 Jun 2025 07:07:19 +0200
Subject: [PATCH 3/3] Fix make linter happy and shortend a line

---
 goodreads/book_test.go | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/goodreads/book_test.go b/goodreads/book_test.go
index 006f351..a57a466 100644
--- a/goodreads/book_test.go
+++ b/goodreads/book_test.go
@@ -44,7 +44,7 @@ func TestBookUnmarshalBrTagReplacement(t *testing.T) {
 		<book>
 			<id>123</id>
 			<title>Test Book</title>
-			<description><![CDATA[This is a test description.<br />This should be on a new line.<br/>Another line.<br>Final line.]]></description>
+			<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>
@@ -68,7 +68,7 @@ func TestBookUnmarshalBrTagReplacement(t *testing.T) {
 	t.Logf("Description after processing: %q", description)
 
 	// Verify that <br> tags have been correctly converted to newlines
-	require.Contains(t, description, "test description.\nThis should be on a new line.\nAnother line.\nFinal line.")
+	require.Contains(t, description, "Test description\n2. line\n3. line\n4. line")
 
 	// Verify that no HTML br tags remain
 	require.NotContains(t, description, "<br")