Add FileExfiltrationWriter

synapse/app/admin_cmd.py

@@ -15,7 +15,11 @@
 # limitations under the License.
 import argparse
 import logging
+import os
 import sys
+import tempfile
+
+from canonicaljson import json
 
 from twisted.internet import defer, task
 
@@ -24,7 +28,7 @@ from synapse.app import _base
 from synapse.config._base import ConfigError
 from synapse.config.homeserver import HomeServerConfig
 from synapse.config.logger import setup_logging
-from synapse.handlers.admin import FileExfiltrationWriter
+from synapse.handlers.admin import ExfiltrationWriter
 from synapse.replication.slave.storage._base import BaseSlavedStore
 from synapse.replication.slave.storage.account_data import SlavedAccountDataStore
 from synapse.replication.slave.storage.appservice import SlavedApplicationServiceStore
@@ -107,6 +111,70 @@ def export_data_command(hs, args):
     print(res)
 
 
+class FileExfiltrationWriter(ExfiltrationWriter):
+    """An ExfiltrationWriter that writes the users data to a directory.
+    Returns the directory location on completion.
+
+    Note: This writes to disk on the main reactor thread.
+
+    Args:
+        user_id (str): The user whose data is being exfiltrated.
+        directory (str|None): The directory to write the data to, if None then
+            will write to a temporary directory.
+    """
+
+    def __init__(self, user_id, directory=None):
+        self.user_id = user_id
+
+        if directory:
+            self.base_directory = directory
+        else:
+            self.base_directory = tempfile.mkdtemp(
+                prefix="synapse-exfiltrate__%s__" % (user_id,)
+            )
+
+        os.makedirs(self.base_directory, exist_ok=True)
+        if list(os.listdir(self.base_directory)):
+            raise Exception("Directory must be empty")
+
+    def write_events(self, room_id, events):
+        room_directory = os.path.join(self.base_directory, "rooms", room_id)
+        os.makedirs(room_directory, exist_ok=True)
+        events_file = os.path.join(room_directory, "events")
+
+        with open(events_file, "a") as f:
+            for event in events:
+                print(json.dumps(event.get_pdu_json()), file=f)
+
+    def write_state(self, room_id, event_id, state):
+        room_directory = os.path.join(self.base_directory, "rooms", room_id)
+        state_directory = os.path.join(room_directory, "state")
+        os.makedirs(state_directory, exist_ok=True)
+
+        event_file = os.path.join(state_directory, event_id)
+
+        with open(event_file, "a") as f:
+            for event in state.values():
+                print(json.dumps(event.get_pdu_json()), file=f)
+
+    def write_invite(self, room_id, event, state):
+        self.write_events(room_id, [event])
+
+        # We write the invite state somewhere else as they aren't full events
+        # and are only a subset of the state at the event.
+        room_directory = os.path.join(self.base_directory, "rooms", room_id)
+        os.makedirs(room_directory, exist_ok=True)
+
+        invite_state = os.path.join(room_directory, "invite_state")
+
+        with open(invite_state, "a") as f:
+            for event in state.values():
+                print(json.dumps(event), file=f)
+
+    def finished(self):
+        return self.base_directory
+
+
 def start(config_options):
     parser = argparse.ArgumentParser(description="Synapse Admin Command")
     HomeServerConfig.add_arguments_to_parser(parser)