Allow SAML username provider plugins (#6411)

1 files changed, 126 insertions, 60 deletions

diff --git a/synapse/config/saml2_config.py b/synapse/config/saml2_config.py
index c5ea2d43a1..b91414aa35 100644
--- a/synapse/config/saml2_config.py
+++ b/synapse/config/saml2_config.py
@@ -14,17 +14,19 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-import re
+import logging
 
 from synapse.python_dependencies import DependencyException, check_requirements
-from synapse.types import (
-    map_username_to_mxid_localpart,
-    mxid_localpart_allowed_characters,
-)
-from synapse.util.module_loader import load_python_module
+from synapse.util.module_loader import load_module, load_python_module
 
 from ._base import Config, ConfigError
 
+logger = logging.getLogger(__name__)
+
+DEFAULT_USER_MAPPING_PROVIDER = (
+    "synapse.handlers.saml_handler.DefaultSamlMappingProvider"
+)
+
 
 def _dict_merge(merge_dict, into_dict):
     """Do a deep merge of two dicts
@@ -75,15 +77,69 @@ class SAML2Config(Config):
 
         self.saml2_enabled = True
 
-        self.saml2_mxid_source_attribute = saml2_config.get(
-            "mxid_source_attribute", "uid"
-        )
-
         self.saml2_grandfathered_mxid_source_attribute = saml2_config.get(
             "grandfathered_mxid_source_attribute", "uid"
         )
 
-        saml2_config_dict = self._default_saml_config_dict()
+        # user_mapping_provider may be None if the key is present but has no value
+        ump_dict = saml2_config.get("user_mapping_provider") or {}
+
+        # Use the default user mapping provider if not set
+        ump_dict.setdefault("module", DEFAULT_USER_MAPPING_PROVIDER)
+
+        # Ensure a config is present
+        ump_dict["config"] = ump_dict.get("config") or {}
+
+        if ump_dict["module"] == DEFAULT_USER_MAPPING_PROVIDER:
+            # Load deprecated options for use by the default module
+            old_mxid_source_attribute = saml2_config.get("mxid_source_attribute")
+            if old_mxid_source_attribute:
+                logger.warning(
+                    "The config option saml2_config.mxid_source_attribute is deprecated. "
+                    "Please use saml2_config.user_mapping_provider.config"
+                    ".mxid_source_attribute instead."
+                )
+                ump_dict["config"]["mxid_source_attribute"] = old_mxid_source_attribute
+
+            old_mxid_mapping = saml2_config.get("mxid_mapping")
+            if old_mxid_mapping:
+                logger.warning(
+                    "The config option saml2_config.mxid_mapping is deprecated. Please "
+                    "use saml2_config.user_mapping_provider.config.mxid_mapping instead."
+                )
+                ump_dict["config"]["mxid_mapping"] = old_mxid_mapping
+
+        # Retrieve an instance of the module's class
+        # Pass the config dictionary to the module for processing
+        (
+            self.saml2_user_mapping_provider_class,
+            self.saml2_user_mapping_provider_config,
+        ) = load_module(ump_dict)
+
+        # Ensure loaded user mapping module has defined all necessary methods
+        # Note parse_config() is already checked during the call to load_module
+        required_methods = [
+            "get_saml_attributes",
+            "saml_response_to_user_attributes",
+        ]
+        missing_methods = [
+            method
+            for method in required_methods
+            if not hasattr(self.saml2_user_mapping_provider_class, method)
+        ]
+        if missing_methods:
+            raise ConfigError(
+                "Class specified by saml2_config."
+                "user_mapping_provider.module is missing required "
+                "methods: %s" % (", ".join(missing_methods),)
+            )
+
+        # Get the desired saml auth response attributes from the module
+        saml2_config_dict = self._default_saml_config_dict(
+            *self.saml2_user_mapping_provider_class.get_saml_attributes(
+                self.saml2_user_mapping_provider_config
+            )
+        )
         _dict_merge(
             merge_dict=saml2_config.get("sp_config", {}), into_dict=saml2_config_dict
         )
@@ -103,22 +159,27 @@ class SAML2Config(Config):
             saml2_config.get("saml_session_lifetime", "5m")
         )
 
-        mapping = saml2_config.get("mxid_mapping", "hexencode")
-        try:
-            self.saml2_mxid_mapper = MXID_MAPPER_MAP[mapping]
-        except KeyError:
-            raise ConfigError("%s is not a known mxid_mapping" % (mapping,))
-
-    def _default_saml_config_dict(self):
+    def _default_saml_config_dict(
+        self, required_attributes: set, optional_attributes: set
+    ):
+        """Generate a configuration dictionary with required and optional attributes that
+        will be needed to process new user registration
+
+        Args:
+            required_attributes: SAML auth response attributes that are
+                necessary to function
+            optional_attributes: SAML auth response attributes that can be used to add
+                additional information to Synapse user accounts, but are not required
+
+        Returns:
+            dict: A SAML configuration dictionary
+        """
         import saml2
 
         public_baseurl = self.public_baseurl
         if public_baseurl is None:
             raise ConfigError("saml2_config requires a public_baseurl to be set")
 
-        required_attributes = {"uid", self.saml2_mxid_source_attribute}
-
-        optional_attributes = {"displayName"}
         if self.saml2_grandfathered_mxid_source_attribute:
             optional_attributes.add(self.saml2_grandfathered_mxid_source_attribute)
         optional_attributes -= required_attributes
@@ -207,33 +268,58 @@ class SAML2Config(Config):
           #
           #config_path: "%(config_dir_path)s/sp_conf.py"
 
-          # the lifetime of a SAML session. This defines how long a user has to
+          # The lifetime of a SAML session. This defines how long a user has to
           # complete the authentication process, if allow_unsolicited is unset.
           # The default is 5 minutes.
           #
           #saml_session_lifetime: 5m
 
-          # The SAML attribute (after mapping via the attribute maps) to use to derive
-          # the Matrix ID from. 'uid' by default.
+          # An external module can be provided here as a custom solution to
+          # mapping attributes returned from a saml provider onto a matrix user.
           #
-          #mxid_source_attribute: displayName
-
-          # The mapping system to use for mapping the saml attribute onto a matrix ID.
-          # Options include:
-          #  * 'hexencode' (which maps unpermitted characters to '=xx')
-          #  * 'dotreplace' (which replaces unpermitted characters with '.').
-          # The default is 'hexencode'.
-          #
-          #mxid_mapping: dotreplace
-
-          # In previous versions of synapse, the mapping from SAML attribute to MXID was
-          # always calculated dynamically rather than stored in a table. For backwards-
-          # compatibility, we will look for user_ids matching such a pattern before
-          # creating a new account.
+          user_mapping_provider:
+            # The custom module's class. Uncomment to use a custom module.
+            #
+            #module: mapping_provider.SamlMappingProvider
+
+            # Custom configuration values for the module. Below options are
+            # intended for the built-in provider, they should be changed if
+            # using a custom module. This section will be passed as a Python
+            # dictionary to the module's `parse_config` method.
+            #
+            config:
+              # The SAML attribute (after mapping via the attribute maps) to use
+              # to derive the Matrix ID from. 'uid' by default.
+              #
+              # Note: This used to be configured by the
+              # saml2_config.mxid_source_attribute option. If that is still
+              # defined, its value will be used instead.
+              #
+              #mxid_source_attribute: displayName
+
+              # The mapping system to use for mapping the saml attribute onto a
+              # matrix ID.
+              #
+              # Options include:
+              #  * 'hexencode' (which maps unpermitted characters to '=xx')
+              #  * 'dotreplace' (which replaces unpermitted characters with
+              #     '.').
+              # The default is 'hexencode'.
+              #
+              # Note: This used to be configured by the
+              # saml2_config.mxid_mapping option. If that is still defined, its
+              # value will be used instead.
+              #
+              #mxid_mapping: dotreplace
+
+          # In previous versions of synapse, the mapping from SAML attribute to
+          # MXID was always calculated dynamically rather than stored in a
+          # table. For backwards- compatibility, we will look for user_ids
+          # matching such a pattern before creating a new account.
           #
           # This setting controls the SAML attribute which will be used for this
-          # backwards-compatibility lookup. Typically it should be 'uid', but if the
-          # attribute maps are changed, it may be necessary to change it.
+          # backwards-compatibility lookup. Typically it should be 'uid', but if
+          # the attribute maps are changed, it may be necessary to change it.
           #
           # The default is 'uid'.
           #
@@ -241,23 +327,3 @@ class SAML2Config(Config):
         """ % {
             "config_dir_path": config_dir_path
         }
-
-
-DOT_REPLACE_PATTERN = re.compile(
-    ("[^%s]" % (re.escape("".join(mxid_localpart_allowed_characters)),))
-)
-
-
-def dot_replace_for_mxid(username: str) -> str:
-    username = username.lower()
-    username = DOT_REPLACE_PATTERN.sub(".", username)
-
-    # regular mxids aren't allowed to start with an underscore either
-    username = re.sub("^_", "", username)
-    return username
-
-
-MXID_MAPPER_MAP = {
-    "hexencode": map_username_to_mxid_localpart,
-    "dotreplace": dot_replace_for_mxid,
-}