Implement MSC4009 to widen the allowed Matrix ID grammar (#15536)

Behind a configuration flag this adds + to the list of allowed
characters in Matrix IDs. The main feature this enables is
using full E.164 phone numbers as Matrix IDs.

2 files changed, 18 insertions, 15 deletions

diff --git a/synapse/handlers/register.py b/synapse/handlers/register.py
index 61c4b833bd..c80946c2e9 100644
--- a/synapse/handlers/register.py
+++ b/synapse/handlers/register.py
@@ -46,7 +46,7 @@ from synapse.replication.http.register import (
     ReplicationRegisterServlet,
 )
 from synapse.spam_checker_api import RegistrationBehaviour
-from synapse.types import RoomAlias, UserID, create_requester
+from synapse.types import GUEST_USER_ID_PATTERN, RoomAlias, UserID, create_requester
 from synapse.types.state import StateFilter
 
 if TYPE_CHECKING:
@@ -143,10 +143,15 @@ class RegistrationHandler:
         assigned_user_id: Optional[str] = None,
         inhibit_user_in_use_error: bool = False,
     ) -> None:
-        if types.contains_invalid_mxid_characters(localpart):
+        if types.contains_invalid_mxid_characters(
+            localpart, self.hs.config.experimental.msc4009_e164_mxids
+        ):
+            extra_chars = (
+                "=_-./+" if self.hs.config.experimental.msc4009_e164_mxids else "=_-./"
+            )
             raise SynapseError(
                 400,
-                "User ID can only contain characters a-z, 0-9, or '=_-./'",
+                f"User ID can only contain characters a-z, 0-9, or '{extra_chars}'",
                 Codes.INVALID_USERNAME,
             )
 
@@ -195,16 +200,12 @@ class RegistrationHandler:
                         errcode=Codes.FORBIDDEN,
                     )
 
-        if guest_access_token is None:
-            try:
-                int(localpart)
-                raise SynapseError(
-                    400,
-                    "Numeric user IDs are reserved for guest users.",
-                    errcode=Codes.INVALID_USERNAME,
-                )
-            except ValueError:
-                pass
+        if guest_access_token is None and GUEST_USER_ID_PATTERN.fullmatch(localpart):
+            raise SynapseError(
+                400,
+                "Numeric user IDs are reserved for guest users.",
+                errcode=Codes.INVALID_USERNAME,
+            )
 
     async def register_user(
         self,
diff --git a/synapse/handlers/sso.py b/synapse/handlers/sso.py
index 92c3742625..25fd2eb3a1 100644
--- a/synapse/handlers/sso.py
+++ b/synapse/handlers/sso.py
@@ -225,6 +225,8 @@ class SsoHandler:
 
         self._consent_at_registration = hs.config.consent.user_consent_at_registration
 
+        self._e164_mxids = hs.config.experimental.msc4009_e164_mxids
+
     def register_identity_provider(self, p: SsoIdentityProvider) -> None:
         p_id = p.idp_id
         assert p_id not in self._identity_providers
@@ -711,7 +713,7 @@ class SsoHandler:
         # Since the localpart is provided via a potentially untrusted module,
         # ensure the MXID is valid before registering.
         if not attributes.localpart or contains_invalid_mxid_characters(
-            attributes.localpart
+            attributes.localpart, self._e164_mxids
         ):
             raise MappingException("localpart is invalid: %s" % (attributes.localpart,))
 
@@ -944,7 +946,7 @@ class SsoHandler:
             localpart,
         )
 
-        if contains_invalid_mxid_characters(localpart):
+        if contains_invalid_mxid_characters(localpart, self._e164_mxids):
             raise SynapseError(400, "localpart is invalid: %s" % (localpart,))
         user_id = UserID(localpart, self._server_name).to_string()
         user_infos = await self._store.get_users_by_id_case_insensitive(user_id)