Allow for identifier dicts in User Interactive Auth dicts

synapse/handlers/auth.py

@@ -45,12 +45,93 @@
 from synapse.module_api import ModuleApi
 from synapse.push.mailer import load_jinja2_templates
 from synapse.types import Requester, UserID
+from synapse.util.msisdn import phone_number_to_msisdn
 
 from ._base import BaseHandler
 
 logger = logging.getLogger(__name__)
 
 
+def client_dict_convert_legacy_fields_to_identifier(
+    submission: Dict[str, Union[str, Dict]]
+):
+    """Take a legacy-formatted login submission or User-Interactive Authentication dict and
+    updates it to feature an identifier dict instead.
+    Providing user-identifying information at the top-level of a login or UIA submission is
+    now deprecated and replaced with identifiers:
+    https://matrix.org/docs/spec/client_server/r0.6.1#identifier-types
+    Args:
+        submission: The client dict to convert. Passed by reference and modified
+    Raises:
+        SynapseError: if the dict contains a "medium" parameter that is anything other than
+            "email"
+    """
+    if "user" in submission:
+        submission["identifier"] = {"type": "m.id.user", "user": submission["user"]}
+        del submission["user"]
+
+    if "medium" in submission and "address" in submission:
+        submission["identifier"] = {
+            "type": "m.id.thirdparty",
+            "medium": submission["medium"],
+            "address": submission["address"],
+        }
+        del submission["medium"]
+        del submission["address"]
+
+    # We've converted valid, legacy login submissions to an identifier. If the
+    # dict still doesn't have an identifier, it's invalid
+    if "identifier" not in submission:
+        raise SynapseError(
+            400,
+            "Missing 'identifier' parameter in login submission",
+            errcode=Codes.MISSING_PARAM,
+        )
+
+    # Ensure the identifier has a type
+    if "type" not in submission["identifier"]:
+        raise SynapseError(
+            400, "'identifier' dict has no key 'type'", errcode=Codes.MISSING_PARAM,
+        )
+
+
+def login_id_phone_to_thirdparty(identifier: Dict[str, str]):
+    """Convert a phone login identifier type to a generic threepid identifier. Modifies
+    the identifier dict in place
+    Args:
+        identifier: Login identifier dict of type 'm.id.phone'
+    """
+    if "type" not in identifier:
+        raise SynapseError(
+            400, "Invalid phone-type identifier", errcode=Codes.MISSING_PARAM
+        )
+
+    if "country" not in identifier or (
+        # XXX: We used to require `number` instead of `phone`. The spec
+        # defines `phone`. So accept both
+        "phone" not in identifier
+        and "number" not in identifier
+    ):
+        raise SynapseError(
+            400, "Invalid phone-type identifier", errcode=Codes.INVALID_PARAM
+        )
+
+    # Accept both "phone" and "number" as valid keys in m.id.phone
+    phone_number = identifier.get("phone", identifier.get("number"))
+
+    # Convert user-provided phone number to a consistent representation
+    msisdn = phone_number_to_msisdn(identifier["country"], phone_number)
+
+    # Modify the passed dictionary by reference
+    del identifier["country"]
+    identifier.pop("number", None)
+    identifier.pop("phone", None)
+
+    identifier["type"] = "m.id.thirdparty"
+    identifier["medium"] = "msisdn"
+    identifier["address"] = msisdn
+
+
 class AuthHandler(BaseHandler):
     SESSION_EXPIRE_MS = 48 * 60 * 60 * 1000
 
@@ -522,6 +603,87 @@ async def _check_auth_dict(
         (canonical_id, callback) = await self.validate_login(user_id, authdict)
         return canonical_id
 
+    async def username_from_identifier(
+        self, identifier: Dict[str, str], password: Optional[str] = None
+    ) -> Optional[str]:
+        """Given a dictionary containing an identifier from a client, extract the
+        possibly unqualified username of the user that it identifies. Does *not*
+        guarantee that the user exists.
+        If this identifier dict contains a threepid, we attempt to ask password
+        auth providers about it or, failing that, look up an associated user in
+        the database.
+        Args:
+            identifier: The identifier dictionary provided by the client
+            password: The user provided password if one exists. Used for asking
+                password auth providers for usernames from 3pid+password combos.
+        Returns:
+            A username if one was found, or None otherwise
+        Raises:
+            SynapseError: If the identifier dict is invalid
+        """
+
+        # Convert phone type identifiers to generic threepid identifiers, which
+        # will be handled in the next step
+        if identifier["type"] == "m.id.phone":
+            login_id_phone_to_thirdparty(identifier)
+
+        # Convert a threepid identifier to an user identifier
+        if identifier["type"] == "m.id.thirdparty":
+            address = identifier.get("address")
+            medium = identifier.get("medium")
+
+            if not medium or not address:
+                # An error would've already been raised in
+                # `login_id_thirdparty_from_phone` if the original submission
+                # was a phone identifier
+                raise SynapseError(
+                    400, "Invalid thirdparty identifier", errcode=Codes.INVALID_PARAM,
+                )
+
+            if medium == "email":
+                # For emails, transform the address to lowercase.
+                # We store all email addresses as lowercase in the DB.
+                # (See add_threepid in synapse/handlers/auth.py)
+                address = address.lower()
+
+            # Check for auth providers that support 3pid login types
+            if password is not None:
+                canonical_user_id, _ = await self.check_password_provider_3pid(
+                    medium, address, password,
+                )
+                if canonical_user_id:
+                    # Authentication through password provider and 3pid succeeded
+                    return canonical_user_id
+
+            # Check local store
+            user_id = await self.hs.get_datastore().get_user_id_by_threepid(
+                medium, address
+            )
+            if not user_id:
+                # We were unable to find a user_id that belonged to the threepid returned
+                # by the password auth provider
+                return None
+
+            identifier = {"type": "m.id.user", "user": user_id}
+
+        # By this point, the identifier should be a `m.id.user`: if it's anything
+        # else, we haven't understood it.
+        if identifier["type"] != "m.id.user":
+            raise SynapseError(
+                400, "Unknown login identifier type", errcode=Codes.INVALID_PARAM,
+            )
+
+        # User identifiers have a "user" key
+        user = identifier.get("user")
+        if user is None:
+            raise SynapseError(
+                400,
+                "User identifier is missing 'user' key",
+                errcode=Codes.INVALID_PARAM,
+            )
+
+        return user
+
     def _get_params_recaptcha(self) -> dict:
         return {"public_key": self.hs.config.recaptcha_public_key}
 

synapse/rest/client/v1/login.py

@@ -17,6 +17,7 @@
 
 from synapse.api.errors import Codes, LoginError, SynapseError
 from synapse.api.ratelimiting import Ratelimiter
+from synapse.handlers.auth import client_dict_convert_legacy_fields_to_identifier
 from synapse.http.server import finish_request
 from synapse.http.servlet import (
     RestServlet,
@@ -27,47 +28,10 @@
 from synapse.rest.client.v2_alpha._base import client_patterns
 from synapse.rest.well_known import WellKnownBuilder
 from synapse.types import UserID
-from synapse.util.msisdn import phone_number_to_msisdn
 
 logger = logging.getLogger(__name__)
 
 
-def login_submission_legacy_convert(submission):
-    """
-    If the input login submission is an old style object
-    (ie. with top-level user / medium / address) convert it
-    to a typed object.
-    """
-    if "user" in submission:
-        submission["identifier"] = {"type": "m.id.user", "user": submission["user"]}
-        del submission["user"]
-
-    if "medium" in submission and "address" in submission:
-        submission["identifier"] = {
-            "type": "m.id.thirdparty",
-            "medium": submission["medium"],
-            "address": submission["address"],
-        }
-        del submission["medium"]
-        del submission["address"]
-
-
-def login_id_thirdparty_from_phone(identifier):
-    """
-    Convert a phone login identifier type to a generic threepid identifier
-    Args:
-        identifier(dict): Login identifier dict of type 'm.id.phone'
-
-    Returns: Login identifier dict of type 'm.id.threepid'
-    """
-    if "country" not in identifier or "number" not in identifier:
-        raise SynapseError(400, "Invalid phone-type identifier")
-
-    msisdn = phone_number_to_msisdn(identifier["country"], identifier["number"])
-
-    return {"type": "m.id.thirdparty", "medium": "msisdn", "address": msisdn}
-
-
 class LoginRestServlet(RestServlet):
     PATTERNS = client_patterns("/login$", v1=True)
     CAS_TYPE = "m.login.cas"
@@ -174,101 +138,52 @@ async def _do_other_login(self, login_submission):
             login_submission.get("address"),
             login_submission.get("user"),
         )
-        login_submission_legacy_convert(login_submission)
-
-        if "identifier" not in login_submission:
-            raise SynapseError(400, "Missing param: identifier")
-
-        identifier = login_submission["identifier"]
-        if "type" not in identifier:
-            raise SynapseError(400, "Login identifier has no type")
-
-        # convert phone type identifiers to generic threepids
-        if identifier["type"] == "m.id.phone":
-            identifier = login_id_thirdparty_from_phone(identifier)
-
-        # convert threepid identifiers to user IDs
-        if identifier["type"] == "m.id.thirdparty":
-            address = identifier.get("address")
-            medium = identifier.get("medium")
-
-            if medium is None or address is None:
-                raise SynapseError(400, "Invalid thirdparty identifier")
-
-            if medium == "email":
-                # For emails, transform the address to lowercase.
-                # We store all email addreses as lowercase in the DB.
-                # (See add_threepid in synapse/handlers/auth.py)
-                address = address.lower()
-
-            # We also apply account rate limiting using the 3PID as a key, as
-            # otherwise using 3PID bypasses the ratelimiting based on user ID.
-            self._failed_attempts_ratelimiter.ratelimit((medium, address), update=False)
-
-            # Check for login providers that support 3pid login types
-            (
-                canonical_user_id,
-                callback_3pid,
-            ) = await self.auth_handler.check_password_provider_3pid(
-                medium, address, login_submission["password"]
+        # Convert deprecated authdict formats to the current scheme
+        client_dict_convert_legacy_fields_to_identifier(login_submission)
+
+        # Check whether this attempt uses a threepid, if so, check if our failed attempt
+        # ratelimiter allows another attempt at this time
+        medium, address = (
+            login_submission.get("medium"),
+            login_submission.get("address"),
+        )
+        if medium and address:
+            self._failed_attempts_ratelimiter.ratelimit(
+                (medium.lower(), address.lower()), update=False
             )
-            if canonical_user_id:
-                # Authentication through password provider and 3pid succeeded
 
-                result = await self._complete_login(
-                    canonical_user_id, login_submission, callback_3pid
-                )
-                return result
+        # Extract a localpart or user ID from the values in the identifier
+        username = await self.auth_handler.username_from_identifier(
+            login_submission["identifier"], login_submission.get("password")
+        )
 
-            # No password providers were able to handle this 3pid
-            # Check local store
-            user_id = await self.hs.get_datastore().get_user_id_by_threepid(
-                medium, address
-            )
-            if not user_id:
-                logger.warning(
-                    "unknown 3pid identifier medium %s, address %r", medium, address
+        if not username:
+            if medium and address:
+                # The user attempted to login via threepid and failed
+                # Record this failed attempt
+                self._failed_attempts_ratelimiter.can_do_action(
+                    (medium.lower(), address.lower())
                 )
-                # We mark that we've failed to log in here, as
-                # `check_password_provider_3pid` might have returned `None` due
-                # to an incorrect password, rather than the account not
-                # existing.
-                #
-                # If it returned None but the 3PID was bound then we won't hit
-                # this code path, which is fine as then the per-user ratelimit
-                # will kick in below.
-                self._failed_attempts_ratelimiter.can_do_action((medium, address))
-                raise LoginError(403, "", errcode=Codes.FORBIDDEN)
-
-            identifier = {"type": "m.id.user", "user": user_id}
-
-        # by this point, the identifier should be an m.id.user: if it's anything
-        # else, we haven't understood it.
-        if identifier["type"] != "m.id.user":
-            raise SynapseError(400, "Unknown login identifier type")
-        if "user" not in identifier:
-            raise SynapseError(400, "User identifier is missing 'user' key")
-
-        if identifier["user"].startswith("@"):
-            qualified_user_id = identifier["user"]
-        else:
-            qualified_user_id = UserID(identifier["user"], self.hs.hostname).to_string()
-
-        # Check if we've hit the failed ratelimit (but don't update it)
-        self._failed_attempts_ratelimiter.ratelimit(
-            qualified_user_id.lower(), update=False
-        )
+
+                raise LoginError(403, "Unauthorized threepid", errcode=Codes.FORBIDDEN)
+
+            # The login failed for another reason
+            raise LoginError(403, "Invalid login", errcode=Codes.FORBIDDEN)
+
+        # We were able to extract a username successfully
+        # Check if we've hit the failed ratelimit for this user ID
+        self._failed_attempts_ratelimiter.ratelimit(username.lower(), update=False)
 
         try:
             canonical_user_id, callback = await self.auth_handler.validate_login(
-                identifier["user"], login_submission
+                username, login_submission
             )
         except LoginError:
             # The user has failed to log in, so we need to update the rate
             # limiter. Using `can_do_action` avoids us raising a ratelimit
             # exception and masking the LoginError. The actual ratelimiting
             # should have happened above.
-            self._failed_attempts_ratelimiter.can_do_action(qualified_user_id.lower())
+            self._failed_attempts_ratelimiter.can_do_action(username.lower())
             raise
 
         result = await self._complete_login(