Improve email address validation

string_utils/string_utils.go

@@ -26,6 +26,45 @@ const (
 	regexIndexSubmatchEnd   = 3
 )
 
+var WhitespaceChars = []string{
+	// Standard whitespace characters
+	"\u0009", // Character tabulation
+	"\u000A", // Line feed
+	"\u000B", // Line tabulation
+	"\u000C", // Form feed
+	"\u000D", // Carriage return
+	"\u0020", // Space
+	"\u0085", // Next line
+	"\u00A0", // No-break space
+	"\u1680", // Ogham space mark
+	"\u2000", // En quad
+	"\u2001", // Em quad
+	"\u2002", // En space
+	"\u2003", // Em space
+	"\u2004", // Three-per-em space
+	"\u2005", // Four-per-em space
+	"\u2006", // Six-per-em space
+	"\u2007", // Figure space
+	"\u2008", // Punctuation space
+	"\u2009", // Thin space
+	"\u200A", // Hair space
+	"\u2028", // Line separator
+	"\u2029", // Paragraph separator
+	"\u202F", // Narrow no-break space
+	"\u205F", // Medium mathematical space
+	"\u3000", // Ideographic space
+}
+
+var NonOfficialWhitespaceChars = []string{
+	// Characters with property White_Space=no
+	"\u180E", // Mongolian vowel separator
+	"\u200B", // Zero width space
+	"\u200C", // Zero width non-joiner
+	"\u200D", // Zero width joiner
+	"\u2060", // Word joiner
+	"\uFEFF", // Zero width no-break space
+}
+
 var snakeCaseRegex = regexp.MustCompile("^" + snakeCasePattern + "$")
 
 func IsSnakeCase(name string) bool {
@@ -49,7 +88,11 @@ func ReplaceNonSpacingMarks(str string) string {
 }
 
 func RemoveAllWhiteSpaces(s string) string {
-	return strings.ReplaceAll(strings.ReplaceAll(s, " ", ""), "\t", "")
+	cleanedString := strings.ReplaceAll(strings.ReplaceAll(s, " ", ""), "\t", "")
+	for _, whitespaceChar := range WhitespaceChars {
+		cleanedString = strings.ReplaceAll(cleanedString, whitespaceChar, "")
+	}
+	return cleanedString
 }
 
 func ReplaceCaseInsensitive(string, toReplace, replaceWith string) string {

utils/utils.go

@@ -2,10 +2,10 @@ package utils
 
 import (
 	"bytes"
+	emailverifier "github.com/AfterShip/email-verifier"
 	"github.com/mohae/deepcopy"
 	"gitlab.bob.co.za/bob-public-utils/bobgroup-go-utils/errors"
 	"gitlab.bob.co.za/bob-public-utils/bobgroup-go-utils/string_utils"
-	"net/mail"
 	"net/url"
 	"os"
 	"reflect"
@@ -55,23 +55,68 @@ func PointerToValue[V any](value *V) V {
 }
 
 func ValidateEmailAddress(email string) (string, error) {
-	if email == "" {
+	// To lower
+	cleanedEmail := strings.ToLower(strings.TrimSpace(email))
+
+	// Remove all whitespaces
+	cleanedEmail = string_utils.RemoveAllWhiteSpaces(cleanedEmail)
+
+	// Also remove unofficial whitespaces
+	for _, char := range string_utils.NonOfficialWhitespaceChars {
+		cleanedEmail = strings.ReplaceAll(cleanedEmail, char, "")
+	}
+
+	// Strip invalid characters
+	cleanedEmail = stripInvalidCharacters(cleanedEmail)
+
+	// Make sure the email is not empty
+	if cleanedEmail == "" {
 		return "", errors.Error("email address is empty")
 	}
 
-	cleanEmail := strings.ToLower(strings.TrimSpace(email))
-	cleanEmail = string_utils.RemoveAllWhiteSpaces(cleanEmail)
+	// Parse and verify the email
+	verifier := emailverifier.NewVerifier()
+	result, err := verifier.Verify(cleanedEmail)
+	if err != nil || !result.Syntax.Valid {
+		return cleanedEmail, errors.Wrap(err, "could not parse email address")
+	}
+
+	return cleanedEmail, nil
+}
 
-	// Remove ZWSP ("\u200B") characters with an empty string to remove it
-	cleanEmail = strings.ReplaceAll(cleanEmail, "\u200B", "")
+func stripInvalidCharacters(email string) string {
+	cleanEmail := email
 
-	// We validate it but still return it since in some cases we don't want to break everything if the email is bad
-	_, err := mail.ParseAddress(cleanEmail)
-	if err != nil {
-		return cleanEmail, errors.Wrap(err, "could not parse email address")
+	// Replace quotes, asterisks, etc.
+	cleanEmail = strings.ReplaceAll(cleanEmail, "'", "")
+	cleanEmail = strings.ReplaceAll(cleanEmail, "*", "")
+	cleanEmail = strings.ReplaceAll(cleanEmail, "!", "")
+	cleanEmail = strings.ReplaceAll(cleanEmail, "+", "")
+
+	// Trim invalid characters, like underscore, so that it still fails if it's inside the email
+	cleanEmail = strings.Trim(cleanEmail, "_")
+
+	return cleanEmail
+}
+
+func SplitAndCleanEmailAddresses(emails string) []string {
+	var destinationEmails []string
+
+	splitEmails := string_utils.SplitString(emails, []rune{',', ';'})
+	if len(splitEmails) >= 1 {
+		// Success - return these emails
+		for _, email := range splitEmails {
+			cleanedEmail, err := ValidateEmailAddress(email)
+			if err == nil && cleanedEmail != "" {
+				destinationEmails = append(destinationEmails, cleanedEmail)
+			}
+		}
+		if len(destinationEmails) > 0 {
+			return destinationEmails
+		}
 	}
 
-	return cleanEmail, nil
+	return destinationEmails
 }
 
 func StripEmail(email string) (strippedEmail string, strippedDomain string) {