message/contact: ensure valid email address in cache

Filter out the (rare but existent) invalid email addresses from the
cache; use the new method Contact::has_valid_email for that.

lib/message/mu-contact.cc

@@ -20,6 +20,7 @@
 #include "mu-contact.hh"
 #include "mu-message.hh"
 #include "utils/mu-utils.hh"
+#include "utils/mu-regex.hh"
 #include "mu-mime-object.hh"
 
 #include <gmime/gmime.h>
@@ -49,6 +50,35 @@ Contact::display_name(bool quote) const
 		return address_rfc2047(*this);
 }
 
+
+static Regex email_rx;
+
+bool
+Contact::has_valid_email() const
+{
+	/* regexp as per:
+	 *   https://html.spec.whatwg.org/multipage/input.html#valid-e-mail-address
+	 *
+	 * "This requirement is a willful violation of RFC 5322, which defines a
+	 * syntax for email addresses that is simultaneously too strict (before
+	 * the "@" character), too vague (after the "@" character), and too lax
+	 * (allowing comments, whitespace characters, and quoted strings in
+	 * manners unfamiliar to most users) to be of practical use here."
+	 */
+	constexpr auto email_rx_str =
+		R"(^[a-zA-Z0-9.!#$%&'*+\/=?^_`{|}~-]+@[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?(?:\.[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?)*$)";
+	/* avoid std::regex here, since speed matters. PCRE is faster */
+	if (!email_rx) {
+		if (auto&& res = Regex::make(email_rx_str, G_REGEX_OPTIMIZE); !res) {
+			g_error("BUG: error in regex: %s", res.error().what()); /* terminates process */
+		} else
+			email_rx = res.value();
+	}
+
+	return email_rx.matches(email);
+}
+
+
 std::string
 Mu::to_string(const Mu::Contacts& contacts)
 {
@@ -94,6 +124,7 @@ test_ctor_foo()
 
 	assert_equal(c.email, "foo@example.com");
 	assert_equal(c.name, "Foo Bar");
+	g_assert_true(c.has_valid_email());
 	g_assert_true(*c.field_id() == Field::Id::Bcc);
 	g_assert_cmpuint(c.message_date,==,1645214647);
 
@@ -115,6 +146,7 @@ test_ctor_blinky()
 
 	assert_equal(c.email, "bar@example.com");
 	assert_equal(c.name, "Blinky");
+	g_assert_true(c.has_valid_email());
 	g_assert_true(c.personal);
 	g_assert_cmpuint(c.frequency,==,13);
 	g_assert_cmpuint(c.tstamp,==,12345);
@@ -138,6 +170,7 @@ test_ctor_cleanup()
 	assert_equal(c.email, "bar@example.com");
 	assert_equal(c.name, "Bli ky");
 	g_assert_true(c.personal);
+	g_assert_true(c.has_valid_email());
 	g_assert_cmpuint(c.frequency,==,13);
 	g_assert_cmpuint(c.tstamp,==,12345);
 	g_assert_cmpuint(c.message_date,==,1645215014);
@@ -159,6 +192,7 @@ test_encode()
 
 	assert_equal(c.email, "cassius@example.com");
 	assert_equal(c.name, "Ali, Muhammad \"The Greatest\"");
+	g_assert_true(c.has_valid_email());
 	g_assert_false(c.personal);
 	g_assert_cmpuint(c.frequency,==,333);
 	g_assert_cmpuint(c.tstamp,==,768);
@@ -177,6 +211,7 @@ test_sender()
 
 	assert_equal(c.email, "aa@example.com");
 	assert_equal(c.name, "Anders Ångström");
+	g_assert_true(c.has_valid_email());
 	g_assert_false(c.personal);
 	g_assert_cmpuint(c.frequency,==,1);
 	g_assert_cmpuint(c.message_date,==,54321);
@@ -191,6 +226,14 @@ test_misc()
 	g_assert_false(!!contact_type_from_field_id(Field::Id::Subject));
 }
 
+static void
+test_broken_email()
+{
+	Contact c{"a***@@booa@example..com", "abcdef",
+		Contact::Type::Sender, 54321};
+
+	g_assert_false(c.has_valid_email());
+}
 
 int
 main(int argc, char* argv[])
@@ -205,6 +248,7 @@ main(int argc, char* argv[])
 
 	g_test_add_func("/message/contact/sender", test_sender);
 	g_test_add_func("/message/contact/misc", test_misc);
+	g_test_add_func("/message/contact/broken-email", test_broken_email);
 
 	return g_test_run();
 }

lib/message/mu-contact.hh

@@ -96,6 +96,15 @@ struct Contact {
 	std::string display_name(bool quote_if_needed=false) const;
 
 
+	/**
+	 * Does the contact contain a valid email address as per
+	 *   https://html.spec.whatwg.org/multipage/input.html#valid-e-mail-address
+	 * ?
+	 *
+	 * @return true or false
+	 */
+	bool has_valid_email() const;
+
 	/**
 	 * Operator==; based on the hash values (ie. lowercase e-mail address)
 	 *
@@ -151,13 +160,13 @@ struct Contact {
 	 * data members
 	 */
 
-	std::string		email;		/**< Email address for this contact.Not empty */
-	std::string		name;		/**< Name for this contact; can be empty. */
-	Type			type;		/**< Type of contact */
-	int64_t			message_date;	/**< date of the contact's message */
-	bool			personal;	/**<  A personal message? */
-	size_t			frequency;	/**< Frequency of this contact */
-	int64_t			tstamp;		/**< Timestamp for this contact (internal use) */
+	std::string	email;		/**< Email address for this contact.Not empty */
+	std::string	name;		/**< Name for this contact; can be empty. */
+	Type		type;		/**< Type of contact */
+	int64_t		message_date;	/**< Date of the contact's message */
+	bool		personal;	/**<  A personal message? */
+	size_t		frequency;	/**< Frequency of this contact */
+	int64_t		tstamp;		/**< Timestamp for this contact (internal use) */
 
 private:
 	void cleanup_name() { // replace control characters by spaces.