Merge pull request #4 from ryotosaito/master

Decode text/plain and text/html body with encoding
marcospgmelo · Oct 20, 2020 · da7f34e · da7f34e
2 parents 9e2f5b1 + 7ce522d
commit da7f34e
Show file tree

Hide file tree

Showing 2 changed files with 101 additions and 4 deletions.
diff --git a/parsemail.go b/parsemail.go
@@ -46,9 +46,31 @@ func Parse(r io.Reader) (email Email, err error) {
 		email.TextBody, email.HTMLBody, email.EmbeddedFiles, err = parseMultipartRelated(msg.Body, params["boundary"])
 	case contentTypeTextPlain:
 		message, _ := ioutil.ReadAll(msg.Body)
+		var reader io.Reader
+		reader, err = decodeContent(strings.NewReader(string(message[:])), msg.Header.Get("Content-Transfer-Encoding"))
+		if err != nil {
+			return
+		}
+
+		message, err = ioutil.ReadAll(reader)
+		if err != nil {
+			return
+		}
+
 		email.TextBody = strings.TrimSuffix(string(message[:]), "\n")
 	case contentTypeTextHtml:
 		message, _ := ioutil.ReadAll(msg.Body)
+		var reader io.Reader
+		reader, err = decodeContent(strings.NewReader(string(message[:])), msg.Header.Get("Content-Transfer-Encoding"))
+		if err != nil {
+			return
+		}
+
+		message, err = ioutil.ReadAll(reader)
+		if err != nil {
+			return
+		}
+
 		email.HTMLBody = strings.TrimSuffix(string(message[:]), "\n")
 	default:
 		email.Content, err = decodeContent(msg.Body, msg.Header.Get("Content-Transfer-Encoding"))
@@ -121,14 +143,26 @@ func parseMultipartRelated(msg io.Reader, boundary string) (textBody, htmlBody s
 
 		switch contentType {
 		case contentTypeTextPlain:
-			ppContent, err := ioutil.ReadAll(part)
+			message, _ := ioutil.ReadAll(part)
+			reader, err := decodeContent(strings.NewReader(string(message[:])), part.Header.Get("Content-Transfer-Encoding"))
+			if err != nil {
+				return textBody, htmlBody, embeddedFiles, err
+			}
+
+			ppContent, err := ioutil.ReadAll(reader)
 			if err != nil {
 				return textBody, htmlBody, embeddedFiles, err
 			}
 
 			textBody += strings.TrimSuffix(string(ppContent[:]), "\n")
 		case contentTypeTextHtml:
-			ppContent, err := ioutil.ReadAll(part)
+			message, _ := ioutil.ReadAll(part)
+			reader, err := decodeContent(strings.NewReader(string(message[:])), part.Header.Get("Content-Transfer-Encoding"))
+			if err != nil {
+				return textBody, htmlBody, embeddedFiles, err
+			}
+
+			ppContent, err := ioutil.ReadAll(reader)
 			if err != nil {
 				return textBody, htmlBody, embeddedFiles, err
 			}
@@ -178,14 +212,26 @@ func parseMultipartAlternative(msg io.Reader, boundary string) (textBody, htmlBo
 
 		switch contentType {
 		case contentTypeTextPlain:
-			ppContent, err := ioutil.ReadAll(part)
+			message, _ := ioutil.ReadAll(part)
+			reader, err := decodeContent(strings.NewReader(string(message[:])), part.Header.Get("Content-Transfer-Encoding"))
+			if err != nil {
+				return textBody, htmlBody, embeddedFiles, err
+			}
+
+			ppContent, err := ioutil.ReadAll(reader)
 			if err != nil {
 				return textBody, htmlBody, embeddedFiles, err
 			}
 
 			textBody += strings.TrimSuffix(string(ppContent[:]), "\n")
 		case contentTypeTextHtml:
-			ppContent, err := ioutil.ReadAll(part)
+			message, _ := ioutil.ReadAll(part)
+			reader, err := decodeContent(strings.NewReader(string(message[:])), part.Header.Get("Content-Transfer-Encoding"))
+			if err != nil {
+				return textBody, htmlBody, embeddedFiles, err
+			}
+
+			ppContent, err := ioutil.ReadAll(reader)
 			if err != nil {
 				return textBody, htmlBody, embeddedFiles, err
 			}
@@ -362,6 +408,8 @@ func decodeContent(content io.Reader, encoding string) (io.Reader, error) {
 		}
 
 		return bytes.NewReader(dd), nil
+	case "8bit":
+		return content, nil
 	case "":
 		return content, nil
 	default:

diff --git a/parsemail_test.go b/parsemail_test.go
@@ -403,6 +403,31 @@ So, "Hello".`,
 			htmlBody:  "<div dir=\"ltr\"><div>Time for the egg.</div><div><br></div><div><br><br></div></div>",
 			textBody: "Time for the egg.",
 		},
+		14: {
+			contentType: "multipart/alternative; boundary=\"000000000000ab2e1f05a26de586\"",
+			mailData:    base64Content,
+			subject:     "Saying Hello",
+			from: []mail.Address{
+				{
+					Name:    "John Doe",
+					Address: "[email protected]",
+				},
+			},
+			sender: mail.Address{
+				Name:    "Michael Jones",
+				Address: "[email protected]",
+			},
+			to: []mail.Address{
+				{
+					Name:    "Mary Smith",
+					Address: "[email protected]",
+				},
+			},
+			messageID: "[email protected]",
+			date:      parseDate("Fri, 21 Nov 1997 09:55:06 -0600"),
+			htmlBody:  "<div dir=\"ltr\">👍</div>",
+			textBody:  "👍",
+		},
 	}
 
 	for index, td := range testData {
@@ -946,3 +971,27 @@ Content-Disposition: attachment;
 
 --f403045f1dcc043a44054c8e6bbf--
 `
+
+var base64Content = `MIME-Version: 1.0
+From: John Doe <[email protected]>
+Sender: Michael Jones <[email protected]>
+To: Mary Smith <[email protected]>
+Subject: Saying Hello
+Date: Fri, 21 Nov 1997 09:55:06 -0600
+Message-ID: <[email protected]>
+Content-Type: multipart/alternative; boundary="000000000000ab2e1f05a26de586"
+
+--000000000000ab2e1f05a26de586
+Content-Type: text/plain; charset="UTF-8"
+Content-Transfer-Encoding: base64
+
+8J+RjQo=
+
+--000000000000ab2e1f05a26de586
+Content-Type: text/html; charset="UTF-8"
+Content-Transfer-Encoding: base64
+
+PGRpdiBkaXI9Imx0ciI+8J+RjTwvZGl2Pgo=
+
+--000000000000ab2e1f05a26de586--
+`