Emit log messages as 'message' not 'log'.

Remove tests that got blacklisted while torturing was enabled, and add one that
fails.

.buildkite/merge_base_branch.sh

@@ -1,6 +1,6 @@
 #!/usr/bin/env bash
 
-set -ex
+set -e
 
 if [[ "$BUILDKITE_BRANCH" =~ ^(develop|master|dinsic|shhs|release-.*)$ ]]; then
     echo "Not merging forward, as this is a release branch"
@@ -18,6 +18,8 @@ else
     GITBASE=$BUILDKITE_PULL_REQUEST_BASE_BRANCH
 fi
 
+echo "--- merge_base_branch $GITBASE"
+
 # Show what we are before
 git --no-pager show -s
 

.buildkite/postgres-config.yaml

@@ -0,0 +1,21 @@
+# Configuration file used for testing the 'synapse_port_db' script.
+# Tells the script to connect to the postgresql database that will be available in the
+# CI's Docker setup at the point where this file is considered.
+server_name: "localhost:8800"
+
+signing_key_path: "/src/.buildkite/test.signing.key"
+
+report_stats: false
+
+database:
+  name: "psycopg2"
+  args:
+    user: postgres
+    host: postgres
+    password: postgres
+    database: synapse
+
+# Suppress the key server warning.
+trusted_key_servers:
+  - server_name: "matrix.org"
+suppress_key_server_warning: true

.buildkite/scripts/create_postgres_db.py

@@ -0,0 +1,36 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+# Copyright 2019 The Matrix.org Foundation C.I.C.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import logging
+from synapse.storage.engines import create_engine
+
+logger = logging.getLogger("create_postgres_db")
+
+if __name__ == "__main__":
+    # Create a PostgresEngine.
+    db_engine = create_engine({"name": "psycopg2", "args": {}})
+
+    # Connect to postgres to create the base database.
+    # We use "postgres" as a database because it's bound to exist and the "synapse" one
+    # doesn't exist yet.
+    db_conn = db_engine.module.connect(
+        user="postgres", host="postgres", password="postgres", dbname="postgres"
+    )
+    db_conn.autocommit = True
+    cur = db_conn.cursor()
+    cur.execute("CREATE DATABASE synapse;")
+    cur.close()
+    db_conn.close()

.buildkite/scripts/test_synapse_port_db.sh

@@ -0,0 +1,36 @@
+#!/bin/bash
+#
+# Test script for 'synapse_port_db', which creates a virtualenv, installs Synapse along
+# with additional dependencies needed for the test (such as coverage or the PostgreSQL
+# driver), update the schema of the test SQLite database and run background updates on it,
+# create an empty test database in PostgreSQL, then run the 'synapse_port_db' script to
+# test porting the SQLite database to the PostgreSQL database (with coverage).
+
+set -xe
+cd `dirname $0`/../..
+
+echo "--- Install dependencies"
+
+# Install dependencies for this test.
+pip install psycopg2 coverage coverage-enable-subprocess
+
+# Install Synapse itself. This won't update any libraries.
+pip install -e .
+
+echo "--- Generate the signing key"
+
+# Generate the server's signing key.
+python -m synapse.app.homeserver --generate-keys -c .buildkite/sqlite-config.yaml
+
+echo "--- Prepare the databases"
+
+# Make sure the SQLite3 database is using the latest schema and has no pending background update.
+scripts-dev/update_database --database-config .buildkite/sqlite-config.yaml
+
+# Create the PostgreSQL database.
+./.buildkite/scripts/create_postgres_db.py
+
+echo "+++ Run synapse_port_db"
+
+# Run the script
+coverage run scripts/synapse_port_db --sqlite-database .buildkite/test_db.db --postgres-config .buildkite/postgres-config.yaml

.buildkite/sqlite-config.yaml

@@ -0,0 +1,18 @@
+# Configuration file used for testing the 'synapse_port_db' script.
+# Tells the 'update_database' script to connect to the test SQLite database to upgrade its
+# schema and run background updates on it.
+server_name: "localhost:8800"
+
+signing_key_path: "/src/.buildkite/test.signing.key"
+
+report_stats: false
+
+database:
+  name: "sqlite3"
+  args:
+    database: ".buildkite/test_db.db"
+
+# Suppress the key server warning.
+trusted_key_servers:
+  - server_name: "matrix.org"
+suppress_key_server_warning: true

.buildkite/worker-blacklist

@@ -28,3 +28,14 @@ User sees updates to presence from other users in the incremental sync.
 Gapped incremental syncs include all state changes
 
 Old members are included in gappy incr LL sync if they start speaking
+
+# new failures as of https://github.com/matrix-org/sytest/pull/732
+Device list doesn't change if remote server is down
+Remote servers cannot set power levels in rooms without existing powerlevels
+Remote servers should reject attempts by non-creators to set the power levels
+
+# https://buildkite.com/matrix-dot-org/synapse/builds/6134#6f67bf47-e234-474d-80e8-c6e1868b15c5
+Server correctly handles incoming m.device_list_update
+
+# this fails reliably with a torture level of 100 due to https://github.com/matrix-org/synapse/issues/6536
+Outbound federation requests missing prev_events and then asks for /state_ids and resolves the state

.github/PULL_REQUEST_TEMPLATE.md

@@ -1,7 +1,8 @@
 ### Pull Request Checklist
 
-<!-- Please read CONTRIBUTING.rst before submitting your pull request -->
+<!-- Please read CONTRIBUTING.md before submitting your pull request -->
 
 * [ ] Pull request is based on the develop branch
-* [ ] Pull request includes a [changelog file](https://github.com/matrix-org/synapse/blob/master/CONTRIBUTING.rst#changelog)
-* [ ] Pull request includes a [sign off](https://github.com/matrix-org/synapse/blob/master/CONTRIBUTING.rst#sign-off)
+* [ ] Pull request includes a [changelog file](https://github.com/matrix-org/synapse/blob/master/CONTRIBUTING.md#changelog)
+* [ ] Pull request includes a [sign off](https://github.com/matrix-org/synapse/blob/master/CONTRIBUTING.md#sign-off)
+* [ ] Code style is correct (run the [linters](https://github.com/matrix-org/synapse/blob/master/CONTRIBUTING.md#code-style))

CHANGES.md

@@ -1,3 +1,193 @@
+Synapse 1.7.0rc2 (2019-12-11)
+=============================
+
+Bugfixes
+--------
+
+- Fix incorrect error message for invalid requests when setting user's avatar URL. ([\#6497](https://github.com/matrix-org/synapse/issues/6497))
+- Fix support for SQLite 3.7. ([\#6499](https://github.com/matrix-org/synapse/issues/6499))
+- Fix regression where sending email push would not work when using a pusher worker. ([\#6507](https://github.com/matrix-org/synapse/issues/6507), [\#6509](https://github.com/matrix-org/synapse/issues/6509))
+
+
+Synapse 1.7.0rc1 (2019-12-09)
+=============================
+
+Features
+--------
+
+- Implement per-room message retention policies. ([\#5815](https://github.com/matrix-org/synapse/issues/5815), [\#6436](https://github.com/matrix-org/synapse/issues/6436))
+- Add etag and count fields to key backup endpoints to help clients guess if there are new keys. ([\#5858](https://github.com/matrix-org/synapse/issues/5858))
+- Add `/admin/v2/users` endpoint with pagination. Contributed by Awesome Technologies Innovationslabor GmbH. ([\#5925](https://github.com/matrix-org/synapse/issues/5925))
+- Require User-Interactive Authentication for `/account/3pid/add`, meaning the user's password will be required to add a third-party ID to their account. ([\#6119](https://github.com/matrix-org/synapse/issues/6119))
+- Implement the `/_matrix/federation/unstable/net.atleastfornow/state/<context>` API as drafted in MSC2314. ([\#6176](https://github.com/matrix-org/synapse/issues/6176))
+- Configure privacy-preserving settings by default for the room directory. ([\#6355](https://github.com/matrix-org/synapse/issues/6355))
+- Add ephemeral messages support by partially implementing [MSC2228](https://github.com/matrix-org/matrix-doc/pull/2228). ([\#6409](https://github.com/matrix-org/synapse/issues/6409))
+- Add support for [MSC 2367](https://github.com/matrix-org/matrix-doc/pull/2367), which allows specifying a reason on all membership events. ([\#6434](https://github.com/matrix-org/synapse/issues/6434))
+
+
+Bugfixes
+--------
+
+- Transfer non-standard power levels on room upgrade. ([\#6237](https://github.com/matrix-org/synapse/issues/6237))
+- Fix error from the Pillow library when uploading RGBA images. ([\#6241](https://github.com/matrix-org/synapse/issues/6241))
+- Correctly apply the event filter to the `state`, `events_before` and `events_after` fields in the response to `/context` requests. ([\#6329](https://github.com/matrix-org/synapse/issues/6329))
+- Fix caching devices for remote users when using workers, so that we don't attempt to refetch (and potentially fail) each time a user requests devices. ([\#6332](https://github.com/matrix-org/synapse/issues/6332))
+- Prevent account data syncs getting lost across TCP replication. ([\#6333](https://github.com/matrix-org/synapse/issues/6333))
+- Fix bug: TypeError in `register_user()` while using LDAP auth module. ([\#6406](https://github.com/matrix-org/synapse/issues/6406))
+- Fix an intermittent exception when handling read-receipts. ([\#6408](https://github.com/matrix-org/synapse/issues/6408))
+- Fix broken guest registration when there are existing blocks of numeric user IDs. ([\#6420](https://github.com/matrix-org/synapse/issues/6420))
+- Fix startup error when http proxy is defined. ([\#6421](https://github.com/matrix-org/synapse/issues/6421))
+- Fix error when using synapse_port_db on a vanilla synapse db. ([\#6449](https://github.com/matrix-org/synapse/issues/6449))
+- Fix uploading multiple cross signing signatures for the same user. ([\#6451](https://github.com/matrix-org/synapse/issues/6451))
+- Fix bug which lead to exceptions being thrown in a loop when a cross-signed device is deleted. ([\#6462](https://github.com/matrix-org/synapse/issues/6462))
+- Fix `synapse_port_db` not exiting with a 0 code if something went wrong during the port process. ([\#6470](https://github.com/matrix-org/synapse/issues/6470))
+- Improve sanity-checking when receiving events over federation. ([\#6472](https://github.com/matrix-org/synapse/issues/6472))
+- Fix inaccurate per-block Prometheus metrics. ([\#6491](https://github.com/matrix-org/synapse/issues/6491))
+- Fix small performance regression for sending invites. ([\#6493](https://github.com/matrix-org/synapse/issues/6493))
+- Back out cross-signing code added in Synapse 1.5.0, which caused a performance regression. ([\#6494](https://github.com/matrix-org/synapse/issues/6494))
+
+
+Improved Documentation
+----------------------
+
+- Update documentation and variables in user contributed systemd reference file. ([\#6369](https://github.com/matrix-org/synapse/issues/6369), [\#6490](https://github.com/matrix-org/synapse/issues/6490))
+- Fix link in the user directory documentation. ([\#6388](https://github.com/matrix-org/synapse/issues/6388))
+- Add build instructions to the docker readme. ([\#6390](https://github.com/matrix-org/synapse/issues/6390))
+- Switch Ubuntu package install recommendation to use python3 packages in INSTALL.md. ([\#6443](https://github.com/matrix-org/synapse/issues/6443))
+- Write some docs for the quarantine_media api. ([\#6458](https://github.com/matrix-org/synapse/issues/6458))
+- Convert CONTRIBUTING.rst to markdown (among other small fixes). ([\#6461](https://github.com/matrix-org/synapse/issues/6461))
+
+
+Deprecations and Removals
+-------------------------
+
+- Remove admin/v1/users_paginate endpoint. Contributed by Awesome Technologies Innovationslabor GmbH. ([\#5925](https://github.com/matrix-org/synapse/issues/5925))
+- Remove fallback for federation with old servers which lack the /federation/v1/state_ids API. ([\#6488](https://github.com/matrix-org/synapse/issues/6488))
+
+
+Internal Changes
+----------------
+
+- Add benchmarks for structured logging and improve output performance. ([\#6266](https://github.com/matrix-org/synapse/issues/6266))
+- Improve the performance of outputting structured logging. ([\#6322](https://github.com/matrix-org/synapse/issues/6322))
+- Refactor some code in the event authentication path for clarity. ([\#6343](https://github.com/matrix-org/synapse/issues/6343), [\#6468](https://github.com/matrix-org/synapse/issues/6468), [\#6480](https://github.com/matrix-org/synapse/issues/6480))
+- Clean up some unnecessary quotation marks around the codebase. ([\#6362](https://github.com/matrix-org/synapse/issues/6362))
+- Complain on startup instead of 500'ing during runtime when `public_baseurl` isn't set when necessary. ([\#6379](https://github.com/matrix-org/synapse/issues/6379))
+- Add a test scenario to make sure room history purges don't break `/messages` in the future. ([\#6392](https://github.com/matrix-org/synapse/issues/6392))
+- Clarifications for the email configuration settings. ([\#6423](https://github.com/matrix-org/synapse/issues/6423))
+- Add more tests to the blacklist when running in worker mode. ([\#6429](https://github.com/matrix-org/synapse/issues/6429))
+- Refactor data store layer to support multiple databases in the future. ([\#6454](https://github.com/matrix-org/synapse/issues/6454), [\#6464](https://github.com/matrix-org/synapse/issues/6464), [\#6469](https://github.com/matrix-org/synapse/issues/6469), [\#6487](https://github.com/matrix-org/synapse/issues/6487)) 
+- Port synapse.rest.client.v1 to async/await. ([\#6482](https://github.com/matrix-org/synapse/issues/6482))
+- Port synapse.rest.client.v2_alpha to async/await. ([\#6483](https://github.com/matrix-org/synapse/issues/6483))
+- Port SyncHandler to async/await. ([\#6484](https://github.com/matrix-org/synapse/issues/6484))
+
+Synapse 1.6.1 (2019-11-28)
+==========================
+
+Security updates
+----------------
+
+This release includes a security fix ([\#6426](https://github.com/matrix-org/synapse/issues/6426), below). Administrators are encouraged to upgrade as soon as possible.
+
+Bugfixes
+--------
+
+- Clean up local threepids from user on account deactivation. ([\#6426](https://github.com/matrix-org/synapse/issues/6426))
+- Fix startup error when http proxy is defined. ([\#6421](https://github.com/matrix-org/synapse/issues/6421))
+
+
+Synapse 1.6.0 (2019-11-26)
+==========================
+
+Bugfixes
+--------
+
+- Fix phone home stats reporting. ([\#6418](https://github.com/matrix-org/synapse/issues/6418))
+
+
+Synapse 1.6.0rc2 (2019-11-25)
+=============================
+
+Bugfixes
+--------
+
+- Fix a bug which could cause the background database update hander for event labels to get stuck in a loop raising exceptions. ([\#6407](https://github.com/matrix-org/synapse/issues/6407))
+
+
+Synapse 1.6.0rc1 (2019-11-20)
+=============================
+
+Features
+--------
+
+- Add federation support for cross-signing. ([\#5727](https://github.com/matrix-org/synapse/issues/5727))
+- Increase default room version from 4 to 5, thereby enforcing server key validity period checks. ([\#6220](https://github.com/matrix-org/synapse/issues/6220))
+- Add support for outbound http proxying via http_proxy/HTTPS_PROXY env vars. ([\#6238](https://github.com/matrix-org/synapse/issues/6238))
+- Implement label-based filtering on `/sync` and `/messages` ([MSC2326](https://github.com/matrix-org/matrix-doc/pull/2326)). ([\#6301](https://github.com/matrix-org/synapse/issues/6301), [\#6310](https://github.com/matrix-org/synapse/issues/6310), [\#6340](https://github.com/matrix-org/synapse/issues/6340))
+
+
+Bugfixes
+--------
+
+- Fix LruCache callback deduplication for Python 3.8. Contributed by @V02460. ([\#6213](https://github.com/matrix-org/synapse/issues/6213))
+- Remove a room from a server's public rooms list on room upgrade. ([\#6232](https://github.com/matrix-org/synapse/issues/6232), [\#6235](https://github.com/matrix-org/synapse/issues/6235))
+- Delete keys from key backup when deleting backup versions. ([\#6253](https://github.com/matrix-org/synapse/issues/6253))
+- Make notification of cross-signing signatures work with workers. ([\#6254](https://github.com/matrix-org/synapse/issues/6254))
+- Fix exception when remote servers attempt to join a room that they're not allowed to join. ([\#6278](https://github.com/matrix-org/synapse/issues/6278))
+- Prevent errors from appearing on Synapse startup if `git` is not installed. ([\#6284](https://github.com/matrix-org/synapse/issues/6284))
+- Appservice requests will no longer contain a double slash prefix when the appservice url provided ends in a slash. ([\#6306](https://github.com/matrix-org/synapse/issues/6306))
+- Fix `/purge_room` admin API. ([\#6307](https://github.com/matrix-org/synapse/issues/6307))
+- Fix the `hidden` field in the `devices` table for SQLite versions prior to 3.23.0. ([\#6313](https://github.com/matrix-org/synapse/issues/6313))
+- Fix bug which casued rejected events to be persisted with the wrong room state. ([\#6320](https://github.com/matrix-org/synapse/issues/6320))
+- Fix bug where `rc_login` ratelimiting would prematurely kick in. ([\#6335](https://github.com/matrix-org/synapse/issues/6335))
+- Prevent the server taking a long time to start up when guest registration is enabled. ([\#6338](https://github.com/matrix-org/synapse/issues/6338))
+- Fix bug where upgrading a guest account to a full user would fail when account validity is enabled. ([\#6359](https://github.com/matrix-org/synapse/issues/6359))
+- Fix `to_device` stream ID getting reset every time Synapse restarts, which had the potential to cause unable to decrypt errors. ([\#6363](https://github.com/matrix-org/synapse/issues/6363))
+- Fix permission denied error when trying to generate a config file with the docker image. ([\#6389](https://github.com/matrix-org/synapse/issues/6389))
+
+
+Improved Documentation
+----------------------
+
+- Contributor documentation now mentions script to run linters. ([\#6164](https://github.com/matrix-org/synapse/issues/6164))
+- Modify CAPTCHA_SETUP.md to update the terms `private key` and `public key` to `secret key` and `site key` respectively. Contributed by Yash Jipkate. ([\#6257](https://github.com/matrix-org/synapse/issues/6257))
+- Update `INSTALL.md` Email section to talk about `account_threepid_delegates`. ([\#6272](https://github.com/matrix-org/synapse/issues/6272))
+- Fix a small typo in `account_threepid_delegates` configuration option. ([\#6273](https://github.com/matrix-org/synapse/issues/6273))
+
+
+Internal Changes
+----------------
+
+- Add a CI job to test the `synapse_port_db` script. ([\#6140](https://github.com/matrix-org/synapse/issues/6140), [\#6276](https://github.com/matrix-org/synapse/issues/6276))
+- Convert EventContext to an attrs. ([\#6218](https://github.com/matrix-org/synapse/issues/6218))
+- Move `persist_events` out from main data store. ([\#6240](https://github.com/matrix-org/synapse/issues/6240), [\#6300](https://github.com/matrix-org/synapse/issues/6300))
+- Reduce verbosity of user/room stats. ([\#6250](https://github.com/matrix-org/synapse/issues/6250))
+- Reduce impact of debug logging. ([\#6251](https://github.com/matrix-org/synapse/issues/6251))
+- Expose some homeserver functionality to spam checkers. ([\#6259](https://github.com/matrix-org/synapse/issues/6259))
+- Change cache descriptors to always return deferreds. ([\#6263](https://github.com/matrix-org/synapse/issues/6263), [\#6291](https://github.com/matrix-org/synapse/issues/6291))
+- Fix incorrect comment regarding the functionality of an `if` statement. ([\#6269](https://github.com/matrix-org/synapse/issues/6269))
+- Update CI to run `isort` over the `scripts` and `scripts-dev` directories. ([\#6270](https://github.com/matrix-org/synapse/issues/6270))
+- Replace every instance of `logger.warn` method with `logger.warning` as the former is deprecated. ([\#6271](https://github.com/matrix-org/synapse/issues/6271), [\#6314](https://github.com/matrix-org/synapse/issues/6314))
+- Port replication http server endpoints to async/await. ([\#6274](https://github.com/matrix-org/synapse/issues/6274))
+- Port room rest handlers to async/await. ([\#6275](https://github.com/matrix-org/synapse/issues/6275))
+- Remove redundant CLI parameters on CI's `flake8` step. ([\#6277](https://github.com/matrix-org/synapse/issues/6277))
+- Port `federation_server.py` to async/await. ([\#6279](https://github.com/matrix-org/synapse/issues/6279))
+- Port receipt and read markers to async/wait. ([\#6280](https://github.com/matrix-org/synapse/issues/6280))
+- Split out state storage into separate data store. ([\#6294](https://github.com/matrix-org/synapse/issues/6294), [\#6295](https://github.com/matrix-org/synapse/issues/6295))
+- Refactor EventContext for clarity. ([\#6298](https://github.com/matrix-org/synapse/issues/6298))
+- Update the version of black used to 19.10b0. ([\#6304](https://github.com/matrix-org/synapse/issues/6304))
+- Add some documentation about worker replication. ([\#6305](https://github.com/matrix-org/synapse/issues/6305))
+- Move admin endpoints into separate files. Contributed by Awesome Technologies Innovationslabor GmbH. ([\#6308](https://github.com/matrix-org/synapse/issues/6308))
+- Document the use of `lint.sh` for code style enforcement & extend it to run on specified paths only. ([\#6312](https://github.com/matrix-org/synapse/issues/6312))
+- Add optional python dependencies and dependant binary libraries to snapcraft packaging. ([\#6317](https://github.com/matrix-org/synapse/issues/6317))
+- Remove the dependency on psutil and replace functionality with the stdlib `resource` module. ([\#6318](https://github.com/matrix-org/synapse/issues/6318), [\#6336](https://github.com/matrix-org/synapse/issues/6336))
+- Improve documentation for EventContext fields. ([\#6319](https://github.com/matrix-org/synapse/issues/6319))
+- Add some checks that we aren't using state from rejected events. ([\#6330](https://github.com/matrix-org/synapse/issues/6330))
+- Add continuous integration for python 3.8. ([\#6341](https://github.com/matrix-org/synapse/issues/6341))
+- Correct spacing/case of various instances of the word "homeserver". ([\#6357](https://github.com/matrix-org/synapse/issues/6357))
+- Temporarily blacklist the failing unit test PurgeRoomTestCase.test_purge_room. ([\#6361](https://github.com/matrix-org/synapse/issues/6361))
+
+
 Synapse 1.5.1 (2019-11-06)
 ==========================
 

CONTRIBUTING.md

@@ -1,5 +1,4 @@
-Contributing code to Matrix
-===========================
+# Contributing code to Matrix
 
 Everyone is welcome to contribute code to Matrix
 (https://github.com/matrix-org), provided that they are willing to license
@@ -7,18 +6,17 @@ their contributions under the same license as the project itself. We follow a
 simple 'inbound=outbound' model for contributions: the act of submitting an
 'inbound' contribution means that the contributor agrees to license the code
 under the same terms as the project's overall 'outbound' license - in our
-case, this is almost always Apache Software License v2 (see LICENSE).
+case, this is almost always Apache Software License v2 (see [LICENSE](LICENSE)).
 
-How to contribute
-~~~~~~~~~~~~~~~~~
+## How to contribute
 
 The preferred and easiest way to contribute changes to Matrix is to fork the
-relevant project on github, and then create a pull request to ask us to pull
-your changes into our repo
-(https://help.github.com/articles/using-pull-requests/)
+relevant project on github, and then [create a pull request](
+https://help.github.com/articles/using-pull-requests/) to ask us to pull
+your changes into our repo.
 
 **The single biggest thing you need to know is: please base your changes on
-the develop branch - /not/ master.**
+the develop branch - *not* master.**
 
 We use the master branch to track the most recent release, so that folks who
 blindly clone the repo and automatically check out master get something that
@@ -30,9 +28,8 @@ use github's pull request workflow to review the contribution, and either ask
 you to make any refinements needed or merge it and make them ourselves. The
 changes will then land on master when we next do a release.
 
-We use `Buildkite <https://buildkite.com/matrix-dot-org/synapse>`_ for
-continuous integration.  Buildkite builds need to be authorised by a
-maintainer. If your change breaks the build, this will be shown in GitHub, so
+We use [Buildkite](https://buildkite.com/matrix-dot-org/synapse) for continuous
+integration. If your change breaks the build, this will be shown in GitHub, so
 please keep an eye on the pull request for feedback.
 
 To run unit tests in a local development environment, you can use:
@@ -47,41 +44,63 @@ To run unit tests in a local development environment, you can use:
   set up PostgreSQL yourself.
 
 Docker images are available for running the integration tests (SyTest) locally,
-see the `documentation in the SyTest repo
-<https://github.com/matrix-org/sytest/blob/develop/docker/README.md>`_ for more
+see the [documentation in the SyTest repo](
+https://github.com/matrix-org/sytest/blob/develop/docker/README.md) for more
 information.
 
-Code style
-~~~~~~~~~~
+## Code style
 
 All Matrix projects have a well-defined code-style - and sometimes we've even
 got as far as documenting it... For instance, synapse's code style doc lives
-at https://github.com/matrix-org/synapse/tree/master/docs/code_style.md.
+[here](docs/code_style.md).
+
+To facilitate meeting these criteria you can run `scripts-dev/lint.sh`
+locally. Since this runs the tools listed in the above document, you'll need
+python 3.6 and to install each tool:
+
+```
+# Install the dependencies
+pip install -U black flake8 isort
+
+# Run the linter script
+./scripts-dev/lint.sh
+```
+
+**Note that the script does not just test/check, but also reformats code, so you
+may wish to ensure any new code is committed first**. By default this script
+checks all files and can take some time; if you alter only certain files, you
+might wish to specify paths as arguments to reduce the run-time:
+
+```
+./scripts-dev/lint.sh path/to/file1.py path/to/file2.py path/to/folder
+```
+
+Before pushing new changes, ensure they don't produce linting errors. Commit any
+files that were corrected.
 
 Please ensure your changes match the cosmetic style of the existing project,
 and **never** mix cosmetic and functional changes in the same commit, as it
 makes it horribly hard to review otherwise.
 
-Changelog
-~~~~~~~~~
+
+## Changelog
 
 All changes, even minor ones, need a corresponding changelog / newsfragment
-entry. These are managed by Towncrier
-(https://github.com/hawkowl/towncrier).
+entry. These are managed by [Towncrier](https://github.com/hawkowl/towncrier).
 
-To create a changelog entry, make a new file in the ``changelog.d`` file named
-in the format of ``PRnumber.type``. The type can be one of the following:
+To create a changelog entry, make a new file in the `changelog.d` directory named
+in the format of `PRnumber.type`. The type can be one of the following:
 
-* ``feature``.
-* ``bugfix``.
-* ``docker`` (for updates to the Docker image).
-* ``doc`` (for updates to the documentation).
-* ``removal`` (also used for deprecations).
-* ``misc`` (for internal-only changes).
+* `feature`
+* `bugfix`
+* `docker` (for updates to the Docker image)
+* `doc` (for updates to the documentation)
+* `removal` (also used for deprecations)
+* `misc` (for internal-only changes)
 
 The content of the file is your changelog entry, which should be a short
-description of your change in the same style as the rest of our `changelog
-<https://github.com/matrix-org/synapse/blob/master/CHANGES.md>`_. The file can
+description of your change in the same style as the rest of our [changelog](
+https://github.com/matrix-org/synapse/blob/master/CHANGES.md). The file can
 contain Markdown formatting, and should end with a full stop ('.') for
 consistency.
 
@@ -89,20 +108,21 @@ Adding credits to the changelog is encouraged, we value your
 contributions and would like to have you shouted out in the release notes!
 
 For example, a fix in PR #1234 would have its changelog entry in
-``changelog.d/1234.bugfix``, and contain content like "The security levels of
-Florbs are now validated when recieved over federation. Contributed by Jane
+`changelog.d/1234.bugfix`, and contain content like "The security levels of
+Florbs are now validated when received over federation. Contributed by Jane
 Matrix.".
 
-Debian changelog
-----------------
+## Debian changelog
 
-Changes which affect the debian packaging files (in ``debian``) are an
+Changes which affect the debian packaging files (in `debian`) are an
 exception.
 
 In this case, you will need to add an entry to the debian changelog for the
-next release. For this, run the following command::
+next release. For this, run the following command:
 
-  dch
+```
+dch
+```
 
 This will make up a new version number (if there isn't already an unreleased
 version in flight), and open an editor where you can add a new changelog entry.
@@ -114,74 +134,77 @@ directory, you will need both a regular newsfragment *and* an entry in the
 debian changelog. (Though typically such changes should be submitted as two
 separate pull requests.)
 
-Sign off
-~~~~~~~~
+## Sign off
 
 In order to have a concrete record that your contribution is intentional
 and you agree to license it under the same terms as the project's license, we've adopted the
 same lightweight approach that the Linux Kernel
-`submitting patches process <https://www.kernel.org/doc/html/latest/process/submitting-patches.html#sign-your-work-the-developer-s-certificate-of-origin>`_, Docker
-(https://github.com/docker/docker/blob/master/CONTRIBUTING.md), and many other
+[submitting patches process](
+https://www.kernel.org/doc/html/latest/process/submitting-patches.html#sign-your-work-the-developer-s-certificate-of-origin>),
+[Docker](https://github.com/docker/docker/blob/master/CONTRIBUTING.md), and many other
 projects use: the DCO (Developer Certificate of Origin:
 http://developercertificate.org/). This is a simple declaration that you wrote
-the contribution or otherwise have the right to contribute it to Matrix::
+the contribution or otherwise have the right to contribute it to Matrix:
 
-    Developer Certificate of Origin
-    Version 1.1
+```
+Developer Certificate of Origin
+Version 1.1
 
-    Copyright (C) 2004, 2006 The Linux Foundation and its contributors.
-    660 York Street, Suite 102,
-    San Francisco, CA 94110 USA
+Copyright (C) 2004, 2006 The Linux Foundation and its contributors.
+660 York Street, Suite 102,
+San Francisco, CA 94110 USA
 
-    Everyone is permitted to copy and distribute verbatim copies of this
-    license document, but changing it is not allowed.
+Everyone is permitted to copy and distribute verbatim copies of this
+license document, but changing it is not allowed.
 
-    Developer's Certificate of Origin 1.1
+Developer's Certificate of Origin 1.1
 
-    By making a contribution to this project, I certify that:
+By making a contribution to this project, I certify that:
 
-    (a) The contribution was created in whole or in part by me and I
-        have the right to submit it under the open source license
-        indicated in the file; or
+(a) The contribution was created in whole or in part by me and I
+    have the right to submit it under the open source license
+    indicated in the file; or
 
-    (b) The contribution is based upon previous work that, to the best
-        of my knowledge, is covered under an appropriate open source
-        license and I have the right under that license to submit that
-        work with modifications, whether created in whole or in part
-        by me, under the same open source license (unless I am
-        permitted to submit under a different license), as indicated
-        in the file; or
+(b) The contribution is based upon previous work that, to the best
+    of my knowledge, is covered under an appropriate open source
+    license and I have the right under that license to submit that
+    work with modifications, whether created in whole or in part
+    by me, under the same open source license (unless I am
+    permitted to submit under a different license), as indicated
+    in the file; or
 
-    (c) The contribution was provided directly to me by some other
-        person who certified (a), (b) or (c) and I have not modified
-        it.
+(c) The contribution was provided directly to me by some other
+    person who certified (a), (b) or (c) and I have not modified
+    it.
 
-    (d) I understand and agree that this project and the contribution
-        are public and that a record of the contribution (including all
-        personal information I submit with it, including my sign-off) is
-        maintained indefinitely and may be redistributed consistent with
-        this project or the open source license(s) involved.
+(d) I understand and agree that this project and the contribution
+    are public and that a record of the contribution (including all
+    personal information I submit with it, including my sign-off) is
+    maintained indefinitely and may be redistributed consistent with
+    this project or the open source license(s) involved.
+```
 
 If you agree to this for your contribution, then all that's needed is to
-include the line in your commit or pull request comment::
+include the line in your commit or pull request comment:
 
-    Signed-off-by: Your Name <your@email.example.org>
+```
+Signed-off-by: Your Name <your@email.example.org>
+```
 
 We accept contributions under a legally identifiable name, such as
 your name on government documentation or common-law names (names
 claimed by legitimate usage or repute). Unfortunately, we cannot
 accept anonymous contributions at this time.
 
-Git allows you to add this signoff automatically when using the ``-s``
-flag to ``git commit``, which uses the name and email set in your
-``user.name`` and ``user.email`` git configs.
+Git allows you to add this signoff automatically when using the `-s`
+flag to `git commit`, which uses the name and email set in your
+`user.name` and `user.email` git configs.
 
-Conclusion
-~~~~~~~~~~
+## Conclusion
 
-That's it!  Matrix is a very open and collaborative project as you might expect
-given our obsession with open communication.  If we're going to successfully
+That's it! Matrix is a very open and collaborative project as you might expect
+given our obsession with open communication. If we're going to successfully
 matrix together all the fragmented communication technologies out there we are
-reliant on contributions and collaboration from the community to do so.  So
+reliant on contributions and collaboration from the community to do so. So
 please get involved - and we hope you have as much fun hacking on Matrix as we
 do!

INSTALL.md

@@ -36,7 +36,7 @@ that your email address is probably `user@example.com` rather than
 System requirements:
 
 - POSIX-compliant system (tested on Linux & OS X)
-- Python 3.5, 3.6, or 3.7
+- Python 3.5, 3.6, 3.7 or 3.8.
 - At least 1GB of free RAM if you want to join large public rooms like #matrix:matrix.org
 
 Synapse is written in Python but some of the libraries it uses are written in
@@ -109,8 +109,8 @@ Installing prerequisites on Ubuntu or Debian:
 
 ```
 sudo apt-get install build-essential python3-dev libffi-dev \
-                     python-pip python-setuptools sqlite3 \
-                     libssl-dev python-virtualenv libjpeg-dev libxslt1-dev
+                     python3-pip python3-setuptools sqlite3 \
+                     libssl-dev python3-virtualenv libjpeg-dev libxslt1-dev
 ```
 
 #### ArchLinux
@@ -133,9 +133,9 @@ sudo yum install libtiff-devel libjpeg-devel libzip-devel freetype-devel \
 sudo yum groupinstall "Development Tools"
 ```
 
-#### Mac OS X
+#### macOS
 
-Installing prerequisites on Mac OS X:
+Installing prerequisites on macOS:
 
 ```
 xcode-select --install
@@ -144,6 +144,14 @@ sudo pip install virtualenv
 brew install pkg-config libffi
 ```
 
+On macOS Catalina (10.15) you may need to explicitly install OpenSSL
+via brew and inform `pip` about it so that `psycopg2` builds:
+
+```
+brew install openssl@1.1
+export LDFLAGS=-L/usr/local/Cellar/openssl\@1.1/1.1.1d/lib/
+```
+
 #### OpenSUSE
 
 Installing prerequisites on openSUSE:
@@ -413,16 +421,18 @@ For a more detailed guide to configuring your server for federation, see
 
 ## Email
 
-It is desirable for Synapse to have the capability to send email. For example,
-this is required to support the 'password reset' feature.
+It is desirable for Synapse to have the capability to send email. This allows
+Synapse to send password reset emails, send verifications when an email address
+is added to a user's account, and send email notifications to users when they
+receive new messages.
 
 To configure an SMTP server for Synapse, modify the configuration section
-headed ``email``, and be sure to have at least the ``smtp_host``, ``smtp_port``
-and ``notif_from`` fields filled out. You may also need to set ``smtp_user``,
-``smtp_pass``, and ``require_transport_security``.
+headed `email`, and be sure to have at least the `smtp_host`, `smtp_port`
+and `notif_from` fields filled out.  You may also need to set `smtp_user`,
+`smtp_pass`, and `require_transport_security`.
 
-If Synapse is not configured with an SMTP server, password reset via email will
- be disabled by default.
+If email is not configured, password reset, registration and notifications via
+email will be disabled.
 
 ## Registering a user
 

UPGRADE.rst

@@ -75,6 +75,23 @@ for example:
      wget https://packages.matrix.org/debian/pool/main/m/matrix-synapse-py3/matrix-synapse-py3_1.3.0+stretch1_amd64.deb
      dpkg -i matrix-synapse-py3_1.3.0+stretch1_amd64.deb
 
+Upgrading to v1.7.0
+===================
+
+In an attempt to configure Synapse in a privacy preserving way, the default
+behaviours of ``allow_public_rooms_without_auth`` and
+``allow_public_rooms_over_federation`` have been inverted. This means that by
+default, only authenticated users querying the Client/Server API will be able
+to query the room directory, and relatedly that the server will not share
+room directory information with other servers over federation.
+
+If your installation does not explicitly set these settings one way or the other
+and you want either setting to be ``true`` then it will necessary to update
+your homeserver configuration file accordingly.
+
+For more details on the surrounding context see our `explainer
+<https://matrix.org/blog/2019/11/09/avoiding-unwelcome-visitors-on-private-matrix-servers>`_.
+
 
 Upgrading to v1.5.0
 ===================

changelog.d/6349.feature

@@ -0,0 +1 @@
+Implement v2 APIs for the `send_join` and `send_leave` federation endpoints (as described in [MSC1802](https://github.com/matrix-org/matrix-doc/pull/1802)).

changelog.d/6377.bugfix

@@ -0,0 +1 @@
+Prevent redacted events from being returned during message search.

changelog.d/6385.bugfix

@@ -0,0 +1 @@
+Prevent error on trying to search a upgraded room when the server is not in the predecessor room.

changelog.d/6394.feature

@@ -0,0 +1 @@
+Add a develop script to generate full SQL schemas.

changelog.d/6411.feature

@@ -0,0 +1 @@
+Allow custom SAML username mapping functinality through an external provider plugin.

changelog.d/6486.bugfix

@@ -0,0 +1 @@
+Improve performance of looking up cross-signing keys.

changelog.d/6496.misc

@@ -0,0 +1 @@
+Port synapse.handlers.initial_sync to async/await.

changelog.d/6501.misc

@@ -0,0 +1 @@
+Refactor get_events_from_store_or_dest to return a dict.

changelog.d/6502.removal

@@ -0,0 +1 @@
+Remove redundant code from event authorisation implementation.

changelog.d/6503.misc

@@ -0,0 +1 @@
+Move get_state methods into FederationHandler.

changelog.d/6504.misc

@@ -0,0 +1 @@
+Port handlers.account_data and handlers.account_validity to async/await.

changelog.d/6505.misc

@@ -0,0 +1 @@
+Make `make_deferred_yieldable` to work with async/await.

changelog.d/6506.misc

@@ -0,0 +1 @@
+Remove `SnapshotCache` in favour of `ResponseCache`.

changelog.d/6510.misc

@@ -0,0 +1 @@
+Change phone home stats to not assume there is a single database and report information about the database used by the main data store.

changelog.d/6511.misc

@@ -0,0 +1 @@
+Move database config from apps into HomeServer object.

changelog.d/6512.misc

@@ -0,0 +1 @@
+Silence mypy errors for files outside those specified.

changelog.d/6514.bugfix

@@ -0,0 +1 @@
+Fix race which occasionally caused deleted devices to reappear.

changelog.d/6515.misc

@@ -0,0 +1 @@
+Clean up some logging when handling incoming events over federation.

changelog.d/6517.misc

@@ -0,0 +1 @@
+Port some of FederationHandler to async/await.

changelog.d/6521.misc

@@ -0,0 +1 @@
+Refactor some code in the event authentication path for clarity.

changelog.d/6522.bugfix

@@ -0,0 +1 @@
+Prevent redacted events from being returned during message search.

changelog.d/6524.misc

@@ -0,0 +1,2 @@
+Improve sanity-checking when receiving events over federation.
+

changelog.d/6534.misc

@@ -0,0 +1 @@
+Test more folders against mypy.

changelog.d/6538.misc

@@ -0,0 +1 @@
+Adjust the sytest blacklist for worker mode.

contrib/experiments/test_messaging.py

@@ -78,7 +78,7 @@ class InputOutput(object):
             m = re.match("^join (\S+)$", line)
             if m:
                 # The `sender` wants to join a room.
-                room_name, = m.groups()
+                (room_name,) = m.groups()
                 self.print_line("%s joining %s" % (self.user, room_name))
                 self.server.join_room(room_name, self.user, self.user)
                 # self.print_line("OK.")
@@ -105,7 +105,7 @@ class InputOutput(object):
             m = re.match("^backfill (\S+)$", line)
             if m:
                 # we want to backfill a room
-                room_name, = m.groups()
+                (room_name,) = m.groups()
                 self.print_line("backfill %s" % room_name)
                 self.server.backfill(room_name)
                 return

contrib/systemd/README.md

@@ -0,0 +1,17 @@
+# Setup Synapse with Systemd
+This is a setup for managing synapse with a user contributed systemd unit 
+file. It provides a `matrix-synapse` systemd unit file that should be tailored 
+to accommodate your installation in accordance with the installation 
+instructions provided in [installation instructions](../../INSTALL.md).
+
+## Setup
+1. Under the service section, ensure the `User` variable matches which user
+you installed synapse under and wish to run it as. 
+2. Under the service section, ensure the `WorkingDirectory` variable matches
+where you have installed synapse.
+3. Under the service section, ensure the `ExecStart` variable matches the
+appropriate locations of your installation.
+4. Copy the `matrix-synapse.service` to `/etc/systemd/system/`
+5. Start Synapse: `sudo systemctl start matrix-synapse`
+6. Verify Synapse is running: `sudo systemctl status matrix-synapse`
+7. *optional* Enable Synapse to start at system boot: `sudo systemctl enable matrix-synapse`

contrib/systemd/matrix-synapse.service

@@ -4,8 +4,11 @@
 #    systemctl enable matrix-synapse
 #    systemctl start matrix-synapse
 #
+# This assumes that Synapse has been installed by a user named
+# synapse.
+#
 # This assumes that Synapse has been installed in a virtualenv in
-# /opt/synapse/env.
+# the user's home directory: `/home/synapse/synapse/env`.
 #
 # **NOTE:** This is an example service file that may change in the future. If you
 # wish to use this please copy rather than symlink it.
@@ -22,8 +25,8 @@ Restart=on-abort
 User=synapse
 Group=nogroup
 
-WorkingDirectory=/opt/synapse
-ExecStart=/opt/synapse/env/bin/python -m synapse.app.homeserver --config-path=/opt/synapse/homeserver.yaml
+WorkingDirectory=/home/synapse/synapse
+ExecStart=/home/synapse/synapse/env/bin/python -m synapse.app.homeserver --config-path=/home/synapse/synapse/homeserver.yaml
 SyslogIdentifier=matrix-synapse
 
 # adjust the cache factor if necessary

debian/changelog

@@ -1,3 +1,15 @@
+matrix-synapse-py3 (1.6.1) stable; urgency=medium
+
+  * New synapse release 1.6.1.
+
+ -- Synapse Packaging team <packages@matrix.org>  Thu, 28 Nov 2019 11:10:40 +0000
+
+matrix-synapse-py3 (1.6.0) stable; urgency=medium
+
+  * New synapse release 1.6.0.
+
+ -- Synapse Packaging team <packages@matrix.org>  Tue, 26 Nov 2019 12:15:40 +0000
+
 matrix-synapse-py3 (1.5.1) stable; urgency=medium
 
   * New synapse release 1.5.1.

docker/README.md

@@ -101,7 +101,7 @@ is suitable for local testing, but for any practical use, you will either need
 to use a reverse proxy, or configure Synapse to expose an HTTPS port.
 
 For documentation on using a reverse proxy, see
-https://github.com/matrix-org/synapse/blob/master/docs/reverse_proxy.rst.
+https://github.com/matrix-org/synapse/blob/master/docs/reverse_proxy.md.
 
 For more information on enabling TLS support in synapse itself, see
 https://github.com/matrix-org/synapse/blob/master/INSTALL.md#tls-certificates. Of
@@ -130,3 +130,15 @@ docker run -it --rm \
 This will generate the same configuration file as the legacy mode used, but
 will store it in `/data/homeserver.yaml` instead of a temporary location. You
 can then use it as shown above at [Running synapse](#running-synapse).
+
+## Building the image
+
+If you need to build the image from a Synapse checkout, use the following `docker
+ build` command from the repo's root:
+ 
+```
+docker build -t matrixdotorg/synapse -f docker/Dockerfile .
+```
+
+You can choose to build a different docker image by changing the value of the `-f` flag to
+point to another Dockerfile.

docker/start.py

@@ -169,11 +169,11 @@ def run_generate_config(environ, ownership):
     # log("running %s" % (args, ))
 
     if ownership is not None:
-        args = ["su-exec", ownership] + args
-        os.execv("/sbin/su-exec", args)
-
         # make sure that synapse has perms to write to the data dir.
         subprocess.check_output(["chown", ownership, data_dir])
+
+        args = ["su-exec", ownership] + args
+        os.execv("/sbin/su-exec", args)
     else:
         os.execv("/usr/local/bin/python", args)
 
@@ -217,8 +217,9 @@ def main(args, environ):
         # backwards-compatibility generate-a-config-on-the-fly mode
         if "SYNAPSE_CONFIG_PATH" in environ:
             error(
-                "SYNAPSE_SERVER_NAME and SYNAPSE_CONFIG_PATH are mutually exclusive "
-                "except in `generate` or `migrate_config` mode."
+                "SYNAPSE_SERVER_NAME can only be combined with SYNAPSE_CONFIG_PATH "
+                "in `generate` or `migrate_config` mode. To start synapse using a "
+                "config file, unset the SYNAPSE_SERVER_NAME environment variable."
             )
 
         config_path = "/compiled/homeserver.yaml"

docs/CAPTCHA_SETUP.md

@@ -4,7 +4,7 @@ The captcha mechanism used is Google's ReCaptcha. This requires API keys from Go
 
 ## Getting keys
 
-Requires a public/private key pair from:
+Requires a site/secret key pair from:
 
 <https://developers.google.com/recaptcha/>
 
@@ -15,8 +15,8 @@ Must be a reCAPTCHA v2 key using the "I'm not a robot" Checkbox option
 The keys are a config option on the home server config. If they are not
 visible, you can generate them via `--generate-config`. Set the following value:
 
-    recaptcha_public_key: YOUR_PUBLIC_KEY
-    recaptcha_private_key: YOUR_PRIVATE_KEY
+    recaptcha_public_key: YOUR_SITE_KEY
+    recaptcha_private_key: YOUR_SECRET_KEY
 
 In addition, you MUST enable captchas via:
 

docs/admin_api/media_admin_api.md

@@ -21,3 +21,20 @@ It returns a JSON body like the following:
     ]
 }
 ```
+
+# Quarantine media in a room
+
+This API 'quarantines' all the media in a room.
+
+The API is:
+
+```
+POST /_synapse/admin/v1/quarantine_media/<room_id>
+
+{}
+```
+
+Quarantining media means that it is marked as inaccessible by users. It applies
+to any local media, and any locally-cached copies of remote media.
+
+The media file itself (and any thumbnails) is not deleted from the server.

docs/admin_api/user_admin_api.rst

@@ -1,3 +1,48 @@
+List Accounts
+=============
+
+This API returns all local user accounts.
+
+The api is::
+
+    GET /_synapse/admin/v2/users?from=0&limit=10&guests=false
+
+including an ``access_token`` of a server admin.
+The parameters ``from`` and ``limit`` are required only for pagination.
+By default, a ``limit`` of 100 is used.
+The parameter ``user_id`` can be used to select only users with user ids that
+contain this value.
+The parameter ``guests=false`` can be used to exclude guest users,
+default is to include guest users.
+The parameter ``deactivated=true`` can be used to include deactivated users,
+default is to exclude deactivated users.
+If the endpoint does not return a ``next_token`` then there are no more users left.
+It returns a JSON body like the following:
+
+.. code:: json
+
+    {
+        "users": [
+            {
+                "name": "<user_id1>",
+                "password_hash": "<password_hash1>",
+                "is_guest": 0,
+                "admin": 0,
+                "user_type": null,
+                "deactivated": 0
+            }, {
+                "name": "<user_id2>",
+                "password_hash": "<password_hash2>",
+                "is_guest": 0,
+                "admin": 1,
+                "user_type": null,
+                "deactivated": 0
+            }
+        ],
+        "next_token": "100"
+    }
+
+
 Query Account
 =============
 

docs/saml_mapping_providers.md

@@ -0,0 +1,77 @@
+# SAML Mapping Providers
+
+A SAML mapping provider is a Python class (loaded via a Python module) that
+works out how to map attributes of a SAML response object to Matrix-specific
+user attributes. Details such as user ID localpart, displayname, and even avatar
+URLs are all things that can be mapped from talking to a SSO service.
+
+As an example, a SSO service may return the email address
+"john.smith@example.com" for a user, whereas Synapse will need to figure out how
+to turn that into a displayname when creating a Matrix user for this individual.
+It may choose `John Smith`, or `Smith, John [Example.com]` or any number of
+variations. As each Synapse configuration may want something different, this is
+where SAML mapping providers come into play.
+
+## Enabling Providers
+
+External mapping providers are provided to Synapse in the form of an external
+Python module. Retrieve this module from [PyPi](https://pypi.org) or elsewhere,
+then tell Synapse where to look for the handler class by editing the
+`saml2_config.user_mapping_provider.module` config option.
+
+`saml2_config.user_mapping_provider.config` allows you to provide custom
+configuration options to the module. Check with the module's documentation for
+what options it provides (if any). The options listed by default are for the
+user mapping provider built in to Synapse. If using a custom module, you should
+comment these options out and use those specified by the module instead.
+
+## Building a Custom Mapping Provider
+
+A custom mapping provider must specify the following methods:
+
+* `__init__(self, parsed_config)`
+   - Arguments:
+     - `parsed_config` - A configuration object that is the return value of the
+       `parse_config` method. You should set any configuration options needed by
+       the module here.
+* `saml_response_to_user_attributes(self, saml_response, failures)`
+    - Arguments:
+      - `saml_response` - A `saml2.response.AuthnResponse` object to extract user
+                          information from.
+      - `failures` - An `int` that represents the amount of times the returned
+                     mxid localpart mapping has failed.  This should be used
+                     to create a deduplicated mxid localpart which should be
+                     returned instead. For example, if this method returns
+                     `john.doe` as the value of `mxid_localpart` in the returned
+                     dict, and that is already taken on the homeserver, this
+                     method will be called again with the same parameters but
+                     with failures=1. The method should then return a different
+                     `mxid_localpart` value, such as `john.doe1`.
+    - This method must return a dictionary, which will then be used by Synapse
+      to build a new user. The following keys are allowed:
+       * `mxid_localpart` - Required. The mxid localpart of the new user.
+       * `displayname` - The displayname of the new user. If not provided, will default to
+                         the value of `mxid_localpart`.
+* `parse_config(config)`
+    - This method should have the `@staticmethod` decoration.
+    - Arguments:
+        - `config` - A `dict` representing the parsed content of the
+          `saml2_config.user_mapping_provider.config` homeserver config option.
+           Runs on homeserver startup. Providers should extract any option values
+           they need here.
+    - Whatever is returned will be passed back to the user mapping provider module's
+      `__init__` method during construction.
+* `get_saml_attributes(config)`
+    - This method should have the `@staticmethod` decoration.
+    - Arguments:
+        - `config` - A object resulting from a call to `parse_config`.
+    - Returns a tuple of two sets. The first set equates to the saml auth
+      response attributes that are required for the module to function, whereas
+      the second set consists of those attributes which can be used if available,
+      but are not necessary.
+
+## Synapse's Default Provider
+
+Synapse has a built-in SAML mapping provider if a custom provider isn't
+specified in the config. It is located at
+[`synapse.handlers.saml_handler.DefaultSamlMappingProvider`](../synapse/handlers/saml_handler.py).

docs/sample_config.yaml

@@ -54,15 +54,16 @@ pid_file: DATADIR/homeserver.pid
 #
 #require_auth_for_profile_requests: true
 
-# If set to 'false', requires authentication to access the server's public rooms
-# directory through the client API. Defaults to 'true'.
+# If set to 'true', removes the need for authentication to access the server's
+# public rooms directory through the client API, meaning that anyone can
+# query the room directory. Defaults to 'false'.
 #
-#allow_public_rooms_without_auth: false
+#allow_public_rooms_without_auth: true
 
-# If set to 'false', forbids any other homeserver to fetch the server's public
-# rooms directory via federation. Defaults to 'true'.
+# If set to 'true', allows any other homeserver to fetch the server's public
+# rooms directory via federation. Defaults to 'false'.
 #
-#allow_public_rooms_over_federation: false
+#allow_public_rooms_over_federation: true
 
 # The default room version for newly created rooms.
 #
@@ -72,7 +73,7 @@ pid_file: DATADIR/homeserver.pid
 # For example, for room version 1, default_room_version should be set
 # to "1".
 #
-#default_room_version: "4"
+#default_room_version: "5"
 
 # The GC threshold parameters to pass to `gc.set_threshold`, if defined
 #
@@ -287,7 +288,7 @@ listeners:
 # Used by phonehome stats to group together related servers.
 #server_context: context
 
-# Resource-constrained Homeserver Settings
+# Resource-constrained homeserver Settings
 #
 # If limit_remote_rooms.enabled is True, the room complexity will be
 # checked before a user joins a new remote room. If it is above
@@ -328,6 +329,69 @@ listeners:
 #
 #user_ips_max_age: 14d
 
+# Message retention policy at the server level.
+#
+# Room admins and mods can define a retention period for their rooms using the
+# 'm.room.retention' state event, and server admins can cap this period by setting
+# the 'allowed_lifetime_min' and 'allowed_lifetime_max' config options.
+#
+# If this feature is enabled, Synapse will regularly look for and purge events
+# which are older than the room's maximum retention period. Synapse will also
+# filter events received over federation so that events that should have been
+# purged are ignored and not stored again.
+#
+retention:
+  # The message retention policies feature is disabled by default. Uncomment the
+  # following line to enable it.
+  #
+  #enabled: true
+
+  # Default retention policy. If set, Synapse will apply it to rooms that lack the
+  # 'm.room.retention' state event. Currently, the value of 'min_lifetime' doesn't
+  # matter much because Synapse doesn't take it into account yet.
+  #
+  #default_policy:
+  #  min_lifetime: 1d
+  #  max_lifetime: 1y
+
+  # Retention policy limits. If set, a user won't be able to send a
+  # 'm.room.retention' event which features a 'min_lifetime' or a 'max_lifetime'
+  # that's not within this range. This is especially useful in closed federations,
+  # in which server admins can make sure every federating server applies the same
+  # rules.
+  #
+  #allowed_lifetime_min: 1d
+  #allowed_lifetime_max: 1y
+
+  # Server admins can define the settings of the background jobs purging the
+  # events which lifetime has expired under the 'purge_jobs' section.
+  #
+  # If no configuration is provided, a single job will be set up to delete expired
+  # events in every room daily.
+  #
+  # Each job's configuration defines which range of message lifetimes the job
+  # takes care of. For example, if 'shortest_max_lifetime' is '2d' and
+  # 'longest_max_lifetime' is '3d', the job will handle purging expired events in
+  # rooms whose state defines a 'max_lifetime' that's both higher than 2 days, and
+  # lower than or equal to 3 days. Both the minimum and the maximum value of a
+  # range are optional, e.g. a job with no 'shortest_max_lifetime' and a
+  # 'longest_max_lifetime' of '3d' will handle every room with a retention policy
+  # which 'max_lifetime' is lower than or equal to three days.
+  #
+  # The rationale for this per-job configuration is that some rooms might have a
+  # retention policy with a low 'max_lifetime', where history needs to be purged
+  # of outdated messages on a very frequent basis (e.g. every 5min), but not want
+  # that purge to be performed by a job that's iterating over every room it knows,
+  # which would be quite heavy on the server.
+  #
+  #purge_jobs:
+  #  - shortest_max_lifetime: 1d
+  #    longest_max_lifetime: 3d
+  #    interval: 5m:
+  #  - shortest_max_lifetime: 3d
+  #    longest_max_lifetime: 1y
+  #    interval: 24h
+
 
 ## TLS ##
 
@@ -743,11 +807,11 @@ uploads_path: "DATADIR/uploads"
 ## Captcha ##
 # See docs/CAPTCHA_SETUP for full details of configuring this.
 
-# This Home Server's ReCAPTCHA public key.
+# This homeserver's ReCAPTCHA public key.
 #
 #recaptcha_public_key: "YOUR_PUBLIC_KEY"
 
-# This Home Server's ReCAPTCHA private key.
+# This homeserver's ReCAPTCHA private key.
 #
 #recaptcha_private_key: "YOUR_PRIVATE_KEY"
 
@@ -955,7 +1019,7 @@ uploads_path: "DATADIR/uploads"
 # If a delegate is specified, the config option public_baseurl must also be filled out.
 #
 account_threepid_delegates:
-    #email: https://example.com     # Delegate email sending to example.org
+    #email: https://example.com     # Delegate email sending to example.com
     #msisdn: http://localhost:8090  # Delegate SMS sending to this local process
 
 # Users who register on this homeserver will automatically be joined
@@ -1186,33 +1250,58 @@ saml2_config:
   #
   #config_path: "CONFDIR/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
+  user_mapping_provider:
+    # The custom module's class. Uncomment to use a custom module.
+    #
+    #module: mapping_provider.SamlMappingProvider
 
-  # 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
+    # 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
 
-  # 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.
+      # 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'.
   #
@@ -1270,8 +1359,23 @@ password_config:
 #   smtp_user: "exampleusername"
 #   smtp_pass: "examplepassword"
 #   require_transport_security: false
-#   notif_from: "Your Friendly %(app)s Home Server <noreply@example.com>"
-#   app_name: Matrix
+#
+#   # notif_from defines the "From" address to use when sending emails.
+#   # It must be set if email sending is enabled.
+#   #
+#   # The placeholder '%(app)s' will be replaced by the application name,
+#   # which is normally 'app_name' (below), but may be overridden by the
+#   # Matrix client application.
+#   #
+#   # Note that the placeholder must be written '%(app)s', including the
+#   # trailing 's'.
+#   #
+#   notif_from: "Your Friendly %(app)s homeserver <noreply@example.com>"
+#
+#   # app_name defines the default value for '%(app)s' in notif_from. It
+#   # defaults to 'Matrix'.
+#   #
+#   #app_name: my_branded_matrix_server
 #
 #   # Enable email notifications by default
 #   #

docs/tcp_replication.md

@@ -199,7 +199,20 @@ client (C):
 
 #### REPLICATE (C)
 
-   Asks the server to replicate a given stream
+Asks the server to replicate a given stream. The syntax is:
+
+```
+    REPLICATE <stream_name> <token>
+```
+
+Where `<token>` may be either:
+ * a numeric stream_id to stream updates since (exclusive)
+ * `NOW` to stream all subsequent updates.
+
+The `<stream_name>` is the name of a replication stream to subscribe 
+to (see [here](../synapse/replication/tcp/streams/_base.py) for a list
+of streams). It can also be `ALL` to subscribe to all known streams,
+in which case the `<token>` must be set to `NOW`.
 
 #### USER_SYNC (C)
 

docs/user_directory.md

@@ -7,7 +7,6 @@ who are present in a publicly viewable room present on the server.
 
 The directory info is stored in various tables, which can (typically after
 DB corruption) get stale or out of sync.  If this happens, for now the
-solution to fix it is to execute the SQL here
-https://github.com/matrix-org/synapse/blob/master/synapse/storage/schema/delta/53/user_dir_populate.sql
+solution to fix it is to execute the SQL [here](../synapse/storage/data_stores/main/schema/delta/53/user_dir_populate.sql)
 and then restart synapse. This should then start a background task to
 flush the current tables and regenerate the directory.

mypy.ini

@@ -1,8 +1,11 @@
 [mypy]
-namespace_packages=True
-plugins=mypy_zope:plugin
-follow_imports=skip
-mypy_path=stubs
+namespace_packages = True
+plugins = mypy_zope:plugin
+follow_imports = silent
+check_untyped_defs = True
+show_error_codes = True
+show_traceback = True
+mypy_path = stubs
 
 [mypy-zope]
 ignore_missing_imports = True

scripts-dev/hash_history.py

@@ -27,7 +27,7 @@ class Store(object):
         "_store_pdu_reference_hash_txn"
     ]
     _store_prev_pdu_hash_txn = SignatureStore.__dict__["_store_prev_pdu_hash_txn"]
-    _simple_insert_txn = SQLBaseStore.__dict__["_simple_insert_txn"]
+    simple_insert_txn = SQLBaseStore.__dict__["simple_insert_txn"]
 
 
 store = Store()

scripts-dev/lint.sh

@@ -7,7 +7,15 @@
 
 set -e
 
-isort -y -rc synapse tests scripts-dev scripts
-flake8 synapse tests
-python3 -m black synapse tests scripts-dev scripts
+if [ $# -ge 1 ]
+then
+  files=$*
+else
+  files="synapse tests scripts-dev scripts"
+fi
+
+echo "Linting these locations: $files"
+isort -y -rc $files
+flake8 $files
+python3 -m black $files
 ./scripts-dev/config-lint.sh

scripts-dev/make_full_schema.sh

@@ -0,0 +1,184 @@
+#!/bin/bash
+#
+# This script generates SQL files for creating a brand new Synapse DB with the latest
+# schema, on both SQLite3 and Postgres.
+#
+# It does so by having Synapse generate an up-to-date SQLite DB, then running
+# synapse_port_db to convert it to Postgres. It then dumps the contents of both.
+
+POSTGRES_HOST="localhost"
+POSTGRES_DB_NAME="synapse_full_schema.$$"
+
+SQLITE_FULL_SCHEMA_OUTPUT_FILE="full.sql.sqlite"
+POSTGRES_FULL_SCHEMA_OUTPUT_FILE="full.sql.postgres"
+
+REQUIRED_DEPS=("matrix-synapse" "psycopg2")
+
+usage() {
+  echo
+  echo "Usage: $0 -p <postgres_username> -o <path> [-c] [-n] [-h]"
+  echo
+  echo "-p <postgres_username>"
+  echo "  Username to connect to local postgres instance. The password will be requested"
+  echo "  during script execution."
+  echo "-c"
+  echo "  CI mode. Enables coverage tracking and prints every command that the script runs."
+  echo "-o <path>"
+  echo "  Directory to output full schema files to."
+  echo "-h"
+  echo "  Display this help text."
+}
+
+while getopts "p:co:h" opt; do
+  case $opt in
+    p)
+      POSTGRES_USERNAME=$OPTARG
+      ;;
+    c)
+      # Print all commands that are being executed
+      set -x
+
+      # Modify required dependencies for coverage
+      REQUIRED_DEPS+=("coverage" "coverage-enable-subprocess")
+
+      COVERAGE=1
+      ;;
+    o)
+      command -v realpath > /dev/null || (echo "The -o flag requires the 'realpath' binary to be installed" && exit 1)
+      OUTPUT_DIR="$(realpath "$OPTARG")"
+      ;;
+    h)
+      usage
+      exit
+      ;;
+    \?)
+      echo "ERROR: Invalid option: -$OPTARG" >&2
+      usage
+      exit
+      ;;
+  esac
+done
+
+# Check that required dependencies are installed
+unsatisfied_requirements=()
+for dep in "${REQUIRED_DEPS[@]}"; do
+  pip show "$dep" --quiet || unsatisfied_requirements+=("$dep")
+done
+if [ ${#unsatisfied_requirements} -ne 0 ]; then
+  echo "Please install the following python packages: ${unsatisfied_requirements[*]}"
+  exit 1
+fi
+
+if [ -z "$POSTGRES_USERNAME" ]; then
+  echo "No postgres username supplied"
+  usage
+  exit 1
+fi
+
+if [ -z "$OUTPUT_DIR" ]; then
+  echo "No output directory supplied"
+  usage
+  exit 1
+fi
+
+# Create the output directory if it doesn't exist
+mkdir -p "$OUTPUT_DIR"
+
+read -rsp "Postgres password for '$POSTGRES_USERNAME': " POSTGRES_PASSWORD
+echo ""
+
+# Exit immediately if a command fails
+set -e
+
+# cd to root of the synapse directory
+cd "$(dirname "$0")/.."
+
+# Create temporary SQLite and Postgres homeserver db configs and key file
+TMPDIR=$(mktemp -d)
+KEY_FILE=$TMPDIR/test.signing.key # default Synapse signing key path
+SQLITE_CONFIG=$TMPDIR/sqlite.conf
+SQLITE_DB=$TMPDIR/homeserver.db
+POSTGRES_CONFIG=$TMPDIR/postgres.conf
+
+# Ensure these files are delete on script exit
+trap 'rm -rf $TMPDIR' EXIT
+
+cat > "$SQLITE_CONFIG" <<EOF
+server_name: "test"
+
+signing_key_path: "$KEY_FILE"
+macaroon_secret_key: "abcde"
+
+report_stats: false
+
+database:
+  name: "sqlite3"
+  args:
+    database: "$SQLITE_DB"
+
+# Suppress the key server warning.
+trusted_key_servers: []
+EOF
+
+cat > "$POSTGRES_CONFIG" <<EOF
+server_name: "test"
+
+signing_key_path: "$KEY_FILE"
+macaroon_secret_key: "abcde"
+
+report_stats: false
+
+database:
+  name: "psycopg2"
+  args:
+    user: "$POSTGRES_USERNAME"
+    host: "$POSTGRES_HOST"
+    password: "$POSTGRES_PASSWORD"
+    database: "$POSTGRES_DB_NAME"
+
+# Suppress the key server warning.
+trusted_key_servers: []
+EOF
+
+# Generate the server's signing key.
+echo "Generating SQLite3 db schema..."
+python -m synapse.app.homeserver --generate-keys -c "$SQLITE_CONFIG"
+
+# Make sure the SQLite3 database is using the latest schema and has no pending background update.
+echo "Running db background jobs..."
+scripts-dev/update_database --database-config "$SQLITE_CONFIG"
+
+# Create the PostgreSQL database.
+echo "Creating postgres database..."
+createdb $POSTGRES_DB_NAME
+
+echo "Copying data from SQLite3 to Postgres with synapse_port_db..."
+if [ -z "$COVERAGE" ]; then
+  # No coverage needed
+  scripts/synapse_port_db --sqlite-database "$SQLITE_DB" --postgres-config "$POSTGRES_CONFIG"
+else
+  # Coverage desired
+  coverage run scripts/synapse_port_db --sqlite-database "$SQLITE_DB" --postgres-config "$POSTGRES_CONFIG"
+fi
+
+# Delete schema_version, applied_schema_deltas and applied_module_schemas tables
+# This needs to be done after synapse_port_db is run
+echo "Dropping unwanted db tables..."
+SQL="
+DROP TABLE schema_version;
+DROP TABLE applied_schema_deltas;
+DROP TABLE applied_module_schemas;
+"
+sqlite3 "$SQLITE_DB" <<< "$SQL"
+psql $POSTGRES_DB_NAME -U "$POSTGRES_USERNAME" -w <<< "$SQL"
+
+echo "Dumping SQLite3 schema to '$OUTPUT_DIR/$SQLITE_FULL_SCHEMA_OUTPUT_FILE'..."
+sqlite3 "$SQLITE_DB" ".dump" > "$OUTPUT_DIR/$SQLITE_FULL_SCHEMA_OUTPUT_FILE"
+
+echo "Dumping Postgres schema to '$OUTPUT_DIR/$POSTGRES_FULL_SCHEMA_OUTPUT_FILE'..."
+pg_dump --format=plain --no-tablespaces --no-acl --no-owner $POSTGRES_DB_NAME | sed -e '/^--/d' -e 's/public\.//g' -e '/^SET /d' -e '/^SELECT /d' > "$OUTPUT_DIR/$POSTGRES_FULL_SCHEMA_OUTPUT_FILE"
+
+echo "Cleaning up temporary Postgres database..."
+dropdb $POSTGRES_DB_NAME
+
+echo "Done! Files dumped to: $OUTPUT_DIR"

scripts-dev/update_database

@@ -0,0 +1,100 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+# Copyright 2019 The Matrix.org Foundation C.I.C.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import argparse
+import logging
+import sys
+
+import yaml
+
+from twisted.internet import defer, reactor
+
+from synapse.config.homeserver import HomeServerConfig
+from synapse.metrics.background_process_metrics import run_as_background_process
+from synapse.server import HomeServer
+from synapse.storage import DataStore
+from synapse.storage.prepare_database import prepare_database
+
+logger = logging.getLogger("update_database")
+
+
+class MockHomeserver(HomeServer):
+    DATASTORE_CLASS = DataStore
+
+    def __init__(self, config, **kwargs):
+        super(MockHomeserver, self).__init__(
+            config.server_name, reactor=reactor, config=config, **kwargs
+        )
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(
+        description=(
+            "Updates a synapse database to the latest schema and runs background updates"
+            " on it."
+        )
+    )
+    parser.add_argument("-v", action="store_true")
+    parser.add_argument(
+        "--database-config",
+        type=argparse.FileType("r"),
+        required=True,
+        help="A database config file for either a SQLite3 database or a PostgreSQL one.",
+    )
+
+    args = parser.parse_args()
+
+    logging_config = {
+        "level": logging.DEBUG if args.v else logging.INFO,
+        "format": "%(asctime)s - %(name)s - %(lineno)d - %(levelname)s - %(message)s",
+    }
+
+    logging.basicConfig(**logging_config)
+
+    # Load, process and sanity-check the config.
+    hs_config = yaml.safe_load(args.database_config)
+
+    if "database" not in hs_config:
+        sys.stderr.write("The configuration file must have a 'database' section.\n")
+        sys.exit(4)
+
+    config = HomeServerConfig()
+    config.parse_config_dict(hs_config, "", "")
+
+    # Instantiate and initialise the homeserver object.
+    hs = MockHomeserver(config)
+
+    db_conn = hs.get_db_conn()
+    # Update the database to the latest schema.
+    prepare_database(db_conn, hs.database_engine, config=config)
+    db_conn.commit()
+
+    # setup instantiates the store within the homeserver object.
+    hs.setup()
+    store = hs.get_datastore()
+
+    @defer.inlineCallbacks
+    def run_background_updates():
+        yield store.db.updates.run_background_updates(sleep=False)
+        # Stop the reactor to exit the script once every background update is run.
+        reactor.stop()
+
+    # Apply all background updates on the database.
+    reactor.callWhenRunning(
+        lambda: run_as_background_process("background_updates", run_background_updates)
+    )
+
+    reactor.run()

scripts/move_remote_media_to_new_store.py

@@ -72,7 +72,7 @@ def move_media(origin_server, file_id, src_paths, dest_paths):
     # check that the original exists
     original_file = src_paths.remote_media_filepath(origin_server, file_id)
     if not os.path.exists(original_file):
-        logger.warn(
+        logger.warning(
             "Original for %s/%s (%s) does not exist",
             origin_server,
             file_id,

scripts/synapse_port_db

@@ -47,6 +47,7 @@ from synapse.storage.data_stores.main.media_repository import (
 from synapse.storage.data_stores.main.registration import (
     RegistrationBackgroundUpdateStore,
 )
+from synapse.storage.data_stores.main.room import RoomBackgroundUpdateStore
 from synapse.storage.data_stores.main.roommember import RoomMemberBackgroundUpdateStore
 from synapse.storage.data_stores.main.search import SearchBackgroundUpdateStore
 from synapse.storage.data_stores.main.state import StateBackgroundUpdateStore
@@ -54,6 +55,7 @@ from synapse.storage.data_stores.main.stats import StatsStore
 from synapse.storage.data_stores.main.user_directory import (
     UserDirectoryBackgroundUpdateStore,
 )
+from synapse.storage.database import Database
 from synapse.storage.engines import create_engine
 from synapse.storage.prepare_database import prepare_database
 from synapse.util import Clock
@@ -131,54 +133,22 @@ class Store(
     EventsBackgroundUpdatesStore,
     MediaRepositoryBackgroundUpdateStore,
     RegistrationBackgroundUpdateStore,
+    RoomBackgroundUpdateStore,
     RoomMemberBackgroundUpdateStore,
     SearchBackgroundUpdateStore,
     StateBackgroundUpdateStore,
     UserDirectoryBackgroundUpdateStore,
     StatsStore,
 ):
-    def __init__(self, db_conn, hs):
-        super().__init__(db_conn, hs)
-        self.db_pool = hs.get_db_pool()
-
-    @defer.inlineCallbacks
-    def runInteraction(self, desc, func, *args, **kwargs):
-        def r(conn):
-            try:
-                i = 0
-                N = 5
-                while True:
-                    try:
-                        txn = conn.cursor()
-                        return func(
-                            LoggingTransaction(txn, desc, self.database_engine, [], []),
-                            *args,
-                            **kwargs
-                        )
-                    except self.database_engine.module.DatabaseError as e:
-                        if self.database_engine.is_deadlock(e):
-                            logger.warn("[TXN DEADLOCK] {%s} %d/%d", desc, i, N)
-                            if i < N:
-                                i += 1
-                                conn.rollback()
-                                continue
-                        raise
-            except Exception as e:
-                logger.debug("[TXN FAIL] {%s} %s", desc, e)
-                raise
-
-        with PreserveLoggingContext():
-            return (yield self.db_pool.runWithConnection(r))
-
     def execute(self, f, *args, **kwargs):
-        return self.runInteraction(f.__name__, f, *args, **kwargs)
+        return self.db.runInteraction(f.__name__, f, *args, **kwargs)
 
     def execute_sql(self, sql, *args):
         def r(txn):
             txn.execute(sql, args)
             return txn.fetchall()
 
-        return self.runInteraction("execute_sql", r)
+        return self.db.runInteraction("execute_sql", r)
 
     def insert_many_txn(self, txn, table, headers, rows):
         sql = "INSERT INTO %s (%s) VALUES (%s)" % (
@@ -221,7 +191,7 @@ class Porter(object):
     def setup_table(self, table):
         if table in APPEND_ONLY_TABLES:
             # It's safe to just carry on inserting.
-            row = yield self.postgres_store._simple_select_one(
+            row = yield self.postgres_store.db.simple_select_one(
                 table="port_from_sqlite3",
                 keyvalues={"table_name": table},
                 retcols=("forward_rowid", "backward_rowid"),
@@ -231,12 +201,14 @@ class Porter(object):
             total_to_port = None
             if row is None:
                 if table == "sent_transactions":
-                    forward_chunk, already_ported, total_to_port = (
-                        yield self._setup_sent_transactions()
-                    )
+                    (
+                        forward_chunk,
+                        already_ported,
+                        total_to_port,
+                    ) = yield self._setup_sent_transactions()
                     backward_chunk = 0
                 else:
-                    yield self.postgres_store._simple_insert(
+                    yield self.postgres_store.db.simple_insert(
                         table="port_from_sqlite3",
                         values={
                             "table_name": table,
@@ -266,7 +238,7 @@ class Porter(object):
 
             yield self.postgres_store.execute(delete_all)
 
-            yield self.postgres_store._simple_insert(
+            yield self.postgres_store.db.simple_insert(
                 table="port_from_sqlite3",
                 values={"table_name": table, "forward_rowid": 1, "backward_rowid": 0},
             )
@@ -320,7 +292,7 @@ class Porter(object):
         if table == "user_directory_stream_pos":
             # We need to make sure there is a single row, `(X, null), as that is
             # what synapse expects to be there.
-            yield self.postgres_store._simple_insert(
+            yield self.postgres_store.db.simple_insert(
                 table=table, values={"stream_id": None}
             )
             self.progress.update(table, table_size)  # Mark table as done
@@ -361,7 +333,9 @@ class Porter(object):
 
                 return headers, forward_rows, backward_rows
 
-            headers, frows, brows = yield self.sqlite_store.runInteraction("select", r)
+            headers, frows, brows = yield self.sqlite_store.db.runInteraction(
+                "select", r
+            )
 
             if frows or brows:
                 if frows:
@@ -375,7 +349,7 @@ class Porter(object):
                 def insert(txn):
                     self.postgres_store.insert_many_txn(txn, table, headers[1:], rows)
 
-                    self.postgres_store._simple_update_one_txn(
+                    self.postgres_store.db.simple_update_one_txn(
                         txn,
                         table="port_from_sqlite3",
                         keyvalues={"table_name": table},
@@ -414,7 +388,7 @@ class Porter(object):
 
                 return headers, rows
 
-            headers, rows = yield self.sqlite_store.runInteraction("select", r)
+            headers, rows = yield self.sqlite_store.db.runInteraction("select", r)
 
             if rows:
                 forward_chunk = rows[-1][0] + 1
@@ -431,8 +405,8 @@ class Porter(object):
                     rows_dict = []
                     for row in rows:
                         d = dict(zip(headers, row))
-                        if "\0" in d['value']:
-                            logger.warn('dropping search row %s', d)
+                        if "\0" in d["value"]:
+                            logger.warning("dropping search row %s", d)
                         else:
                             rows_dict.append(d)
 
@@ -452,7 +426,7 @@ class Porter(object):
                         ],
                     )
 
-                    self.postgres_store._simple_update_one_txn(
+                    self.postgres_store.db.simple_update_one_txn(
                         txn,
                         table="port_from_sqlite3",
                         keyvalues={"table_name": "event_search"},
@@ -502,17 +476,14 @@ class Porter(object):
         self.progress.set_state("Preparing %s" % config["name"])
         conn = self.setup_db(config, engine)
 
-        db_pool = adbapi.ConnectionPool(
-            config["name"], **config["args"]
-        )
+        db_pool = adbapi.ConnectionPool(config["name"], **config["args"])
 
         hs = MockHomeserver(self.hs_config, engine, conn, db_pool)
 
-        store = Store(conn, hs)
+        store = Store(Database(hs), conn, hs)
 
-        yield store.runInteraction(
-            "%s_engine.check_database" % config["name"],
-            engine.check_database,
+        yield store.db.runInteraction(
+            "%s_engine.check_database" % config["name"], engine.check_database,
         )
 
         return store
@@ -520,7 +491,9 @@ class Porter(object):
     @defer.inlineCallbacks
     def run_background_updates_on_postgres(self):
         # Manually apply all background updates on the PostgreSQL database.
-        postgres_ready = yield self.postgres_store.has_completed_background_updates()
+        postgres_ready = (
+            yield self.postgres_store.db.updates.has_completed_background_updates()
+        )
 
         if not postgres_ready:
             # Only say that we're running background updates when there are background
@@ -528,9 +501,9 @@ class Porter(object):
             self.progress.set_state("Running background updates on PostgreSQL")
 
         while not postgres_ready:
-            yield self.postgres_store.do_next_background_update(100)
+            yield self.postgres_store.db.updates.do_next_background_update(100)
             postgres_ready = yield (
-                self.postgres_store.has_completed_background_updates()
+                self.postgres_store.db.updates.has_completed_background_updates()
             )
 
     @defer.inlineCallbacks
@@ -539,7 +512,9 @@ class Porter(object):
             self.sqlite_store = yield self.build_db_store(self.sqlite_config)
 
             # Check if all background updates are done, abort if not.
-            updates_complete = yield self.sqlite_store.has_completed_background_updates()
+            updates_complete = (
+                yield self.sqlite_store.db.updates.has_completed_background_updates()
+            )
             if not updates_complete:
                 sys.stderr.write(
                     "Pending background updates exist in the SQLite3 database."
@@ -580,22 +555,22 @@ class Porter(object):
                 )
 
             try:
-                yield self.postgres_store.runInteraction("alter_table", alter_table)
+                yield self.postgres_store.db.runInteraction("alter_table", alter_table)
             except Exception:
                 # On Error Resume Next
                 pass
 
-            yield self.postgres_store.runInteraction(
+            yield self.postgres_store.db.runInteraction(
                 "create_port_table", create_port_table
             )
 
             # Step 2. Get tables.
             self.progress.set_state("Fetching tables")
-            sqlite_tables = yield self.sqlite_store._simple_select_onecol(
+            sqlite_tables = yield self.sqlite_store.db.simple_select_onecol(
                 table="sqlite_master", keyvalues={"type": "table"}, retcol="name"
             )
 
-            postgres_tables = yield self.postgres_store._simple_select_onecol(
+            postgres_tables = yield self.postgres_store.db.simple_select_onecol(
                 table="information_schema.tables",
                 keyvalues={},
                 retcol="distinct table_name",
@@ -647,7 +622,7 @@ class Porter(object):
             if isinstance(col, bytes):
                 return bytearray(col)
             elif isinstance(col, string_types) and "\0" in col:
-                logger.warn(
+                logger.warning(
                     "DROPPING ROW: NUL value in table %s col %s: %r",
                     table,
                     headers[j],
@@ -685,11 +660,11 @@ class Porter(object):
             rows = txn.fetchall()
             headers = [column[0] for column in txn.description]
 
-            ts_ind = headers.index('ts')
+            ts_ind = headers.index("ts")
 
             return headers, [r for r in rows if r[ts_ind] < yesterday]
 
-        headers, rows = yield self.sqlite_store.runInteraction("select", r)
+        headers, rows = yield self.sqlite_store.db.runInteraction("select", r)
 
         rows = self._convert_rows("sent_transactions", headers, rows)
 
@@ -722,7 +697,7 @@ class Porter(object):
         next_chunk = yield self.sqlite_store.execute(get_start_id)
         next_chunk = max(max_inserted_rowid + 1, next_chunk)
 
-        yield self.postgres_store._simple_insert(
+        yield self.postgres_store.db.simple_insert(
             table="port_from_sqlite3",
             values={
                 "table_name": "sent_transactions",
@@ -735,7 +710,7 @@ class Porter(object):
             txn.execute(
                 "SELECT count(*) FROM sent_transactions" " WHERE ts >= ?", (yesterday,)
             )
-            size, = txn.fetchone()
+            (size,) = txn.fetchone()
             return int(size)
 
         remaining_count = yield self.sqlite_store.execute(get_sent_table_size)
@@ -782,10 +757,13 @@ class Porter(object):
     def _setup_state_group_id_seq(self):
         def r(txn):
             txn.execute("SELECT MAX(id) FROM state_groups")
-            next_id = txn.fetchone()[0] + 1
+            curr_id = txn.fetchone()[0]
+            if not curr_id:
+                return
+            next_id = curr_id + 1
             txn.execute("ALTER SEQUENCE state_group_id_seq RESTART WITH %s", (next_id,))
 
-        return self.postgres_store.runInteraction("setup_state_group_id_seq", r)
+        return self.postgres_store.db.runInteraction("setup_state_group_id_seq", r)
 
 
 ##############################################
@@ -866,7 +844,7 @@ class CursesProgress(Progress):
         duration = int(now) - int(self.start_time)
 
         minutes, seconds = divmod(duration, 60)
-        duration_str = '%02dm %02ds' % (minutes, seconds)
+        duration_str = "%02dm %02ds" % (minutes, seconds)
 
         if self.finished:
             status = "Time spent: %s (Done!)" % (duration_str,)
@@ -876,7 +854,7 @@ class CursesProgress(Progress):
                 left = float(self.total_remaining) / self.total_processed
 
                 est_remaining = (int(now) - self.start_time) * left
-                est_remaining_str = '%02dm %02ds remaining' % divmod(est_remaining, 60)
+                est_remaining_str = "%02dm %02ds remaining" % divmod(est_remaining, 60)
             else:
                 est_remaining_str = "Unknown"
             status = "Time spent: %s (est. remaining: %s)" % (
@@ -962,7 +940,7 @@ if __name__ == "__main__":
         description="A script to port an existing synapse SQLite database to"
         " a new PostgreSQL database."
     )
-    parser.add_argument("-v", action='store_true')
+    parser.add_argument("-v", action="store_true")
     parser.add_argument(
         "--sqlite-database",
         required=True,
@@ -971,12 +949,12 @@ if __name__ == "__main__":
     )
     parser.add_argument(
         "--postgres-config",
-        type=argparse.FileType('r'),
+        type=argparse.FileType("r"),
         required=True,
         help="The database config file for the PostgreSQL database",
     )
     parser.add_argument(
-        "--curses", action='store_true', help="display a curses based progress UI"
+        "--curses", action="store_true", help="display a curses based progress UI"
     )
 
     parser.add_argument(
@@ -1052,3 +1030,4 @@ if __name__ == "__main__":
     if end_error_exec_info:
         exc_type, exc_value, exc_traceback = end_error_exec_info
         traceback.print_exception(exc_type, exc_value, exc_traceback)
+        sys.exit(5)

snap/snapcraft.yaml

@@ -20,3 +20,23 @@ parts:
     source: .
     plugin: python
     python-version: python3
+    python-packages:
+      - '.[all]'
+    build-packages:
+      - libffi-dev
+      - libturbojpeg0-dev
+      - libssl-dev
+      - libxslt1-dev
+      - libpq-dev
+      - zlib1g-dev
+    stage-packages:
+      - libasn1-8-heimdal
+      - libgssapi3-heimdal
+      - libhcrypto4-heimdal
+      - libheimbase1-heimdal
+      - libheimntlm0-heimdal
+      - libhx509-5-heimdal
+      - libkrb5-26-heimdal
+      - libldap-2.4-2
+      - libpq5
+      - libsasl2-2

synapse/__init__.py

@@ -14,7 +14,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-""" This is a reference implementation of a Matrix home server.
+""" This is a reference implementation of a Matrix homeserver.
 """
 
 import os
@@ -36,7 +36,7 @@ try:
 except ImportError:
     pass
 
-__version__ = "1.5.1"
+__version__ = "1.7.0rc2"
 
 if bool(os.environ.get("SYNAPSE_TEST_PATCH_LOG_CONTEXTS", False)):
     # We import here so that we don't have to install a bunch of deps when

synapse/_scripts/register_new_matrix_user.py

@@ -144,8 +144,8 @@ def main():
     logging.captureWarnings(True)
 
     parser = argparse.ArgumentParser(
-        description="Used to register new users with a given home server when"
-        " registration has been disabled. The home server must be"
+        description="Used to register new users with a given homeserver when"
+        " registration has been disabled. The homeserver must be"
         " configured with the 'registration_shared_secret' option"
         " set."
     )
@@ -202,7 +202,7 @@ def main():
         "server_url",
         default="https://localhost:8448",
         nargs="?",
-        help="URL to use to talk to the home server. Defaults to "
+        help="URL to use to talk to the homeserver. Defaults to "
         " 'https://localhost:8448'.",
     )
 

synapse/api/auth.py

@@ -497,7 +497,7 @@ class Auth(object):
         token = self.get_access_token_from_request(request)
         service = self.store.get_app_service_by_token(token)
         if not service:
-            logger.warn("Unrecognised appservice access token.")
+            logger.warning("Unrecognised appservice access token.")
             raise InvalidClientTokenError()
         request.authenticated_entity = service.sender
         return defer.succeed(service)

synapse/api/constants.py

@@ -1,7 +1,8 @@
 # -*- coding: utf-8 -*-
 # Copyright 2014-2016 OpenMarket Ltd
 # Copyright 2017 Vector Creations Ltd
-# Copyright 2018 New Vector Ltd
+# Copyright 2018-2019 New Vector Ltd
+# Copyright 2019 The Matrix.org Foundation C.I.C.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -94,6 +95,8 @@ class EventTypes(object):
     ServerACL = "m.room.server_acl"
     Pinned = "m.room.pinned_events"
 
+    Retention = "m.room.retention"
+
 
 class RejectedReason(object):
     AUTH_ERROR = "auth_error"
@@ -138,3 +141,14 @@ class LimitBlockingTypes(object):
 
     MONTHLY_ACTIVE_USER = "monthly_active_user"
     HS_DISABLED = "hs_disabled"
+
+
+class EventContentFields(object):
+    """Fields found in events' content, regardless of type."""
+
+    # Labels for the event, cf https://github.com/matrix-org/matrix-doc/pull/2326
+    LABELS = "org.matrix.labels"
+
+    # Timestamp to delete the event after
+    # cf https://github.com/matrix-org/matrix-doc/pull/2228
+    SELF_DESTRUCT_AFTER = "org.matrix.self_destruct_after"

synapse/api/errors.py

@@ -457,7 +457,7 @@ def cs_error(msg, code=Codes.UNKNOWN, **kwargs):
 
 
 class FederationError(RuntimeError):
-    """  This class is used to inform remote home servers about erroneous
+    """  This class is used to inform remote homeservers about erroneous
     PDUs they sent us.
 
     FATAL: The remote server could not interpret the source event.

synapse/api/filtering.py

@@ -1,5 +1,8 @@
 # -*- coding: utf-8 -*-
 # Copyright 2015, 2016 OpenMarket Ltd
+# Copyright 2017 Vector Creations Ltd
+# Copyright 2018-2019 New Vector Ltd
+# Copyright 2019 The Matrix.org Foundation C.I.C.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -20,6 +23,7 @@ from jsonschema import FormatChecker
 
 from twisted.internet import defer
 
+from synapse.api.constants import EventContentFields
 from synapse.api.errors import SynapseError
 from synapse.storage.presence import UserPresenceState
 from synapse.types import RoomID, UserID
@@ -66,6 +70,10 @@ ROOM_EVENT_FILTER_SCHEMA = {
         "contains_url": {"type": "boolean"},
         "lazy_load_members": {"type": "boolean"},
         "include_redundant_members": {"type": "boolean"},
+        # Include or exclude events with the provided labels.
+        # cf https://github.com/matrix-org/matrix-doc/pull/2326
+        "org.matrix.labels": {"type": "array", "items": {"type": "string"}},
+        "org.matrix.not_labels": {"type": "array", "items": {"type": "string"}},
     },
 }
 
@@ -259,6 +267,9 @@ class Filter(object):
 
         self.contains_url = self.filter_json.get("contains_url", None)
 
+        self.labels = self.filter_json.get("org.matrix.labels", None)
+        self.not_labels = self.filter_json.get("org.matrix.not_labels", [])
+
     def filters_all_types(self):
         return "*" in self.not_types
 
@@ -282,6 +293,7 @@ class Filter(object):
             room_id = None
             ev_type = "m.presence"
             contains_url = False
+            labels = []
         else:
             sender = event.get("sender", None)
             if not sender:
@@ -300,10 +312,11 @@ class Filter(object):
             content = event.get("content", {})
             # check if there is a string url field in the content for filtering purposes
             contains_url = isinstance(content.get("url"), text_type)
+            labels = content.get(EventContentFields.LABELS, [])
 
-        return self.check_fields(room_id, sender, ev_type, contains_url)
+        return self.check_fields(room_id, sender, ev_type, labels, contains_url)
 
-    def check_fields(self, room_id, sender, event_type, contains_url):
+    def check_fields(self, room_id, sender, event_type, labels, contains_url):
         """Checks whether the filter matches the given event fields.
 
         Returns:
@@ -313,6 +326,7 @@ class Filter(object):
             "rooms": lambda v: room_id == v,
             "senders": lambda v: sender == v,
             "types": lambda v: _matches_wildcard(event_type, v),
+            "labels": lambda v: v in labels,
         }
 
         for name, match_func in literal_keys.items():

synapse/app/__init__.py

@@ -44,6 +44,8 @@ def check_bind_error(e, address, bind_addresses):
         bind_addresses (list): Addresses on which the service listens.
     """
     if address == "0.0.0.0" and "::" in bind_addresses:
-        logger.warn("Failed to listen on 0.0.0.0, continuing because listening on [::]")
+        logger.warning(
+            "Failed to listen on 0.0.0.0, continuing because listening on [::]"
+        )
     else:
         raise e

synapse/app/_base.py

@@ -269,7 +269,7 @@ def start(hs, listeners=None):
 
         # It is now safe to start your Synapse.
         hs.start_listening(listeners)
-        hs.get_datastore().start_profiling()
+        hs.get_datastore().db.start_profiling()
 
         setup_sentry(hs)
         setup_sdnotify(hs)

synapse/app/admin_cmd.py

@@ -45,7 +45,6 @@ from synapse.replication.slave.storage.registration import SlavedRegistrationSto
 from synapse.replication.slave.storage.room import RoomStore
 from synapse.replication.tcp.client import ReplicationClientHandler
 from synapse.server import HomeServer
-from synapse.storage.engines import create_engine
 from synapse.util.logcontext import LoggingContext
 from synapse.util.versionstring import get_version_string
 
@@ -229,14 +228,10 @@ def start(config_options):
 
     synapse.events.USE_FROZEN_DICTS = config.use_frozen_dicts
 
-    database_engine = create_engine(config.database_config)
-
     ss = AdminCmdServer(
         config.server_name,
-        db_config=config.database_config,
         config=config,
         version_string="Synapse/" + get_version_string(synapse),
-        database_engine=database_engine,
     )
 
     setup_logging(ss, config, use_worker_options=True)

synapse/app/appservice.py

@@ -34,7 +34,6 @@ from synapse.replication.slave.storage.events import SlavedEventStore
 from synapse.replication.slave.storage.registration import SlavedRegistrationStore
 from synapse.replication.tcp.client import ReplicationClientHandler
 from synapse.server import HomeServer
-from synapse.storage.engines import create_engine
 from synapse.util.httpresourcetree import create_resource_tree
 from synapse.util.manhole import manhole
 from synapse.util.versionstring import get_version_string
@@ -94,7 +93,7 @@ class AppserviceServer(HomeServer):
                 )
             elif listener["type"] == "metrics":
                 if not self.get_config().enable_metrics:
-                    logger.warn(
+                    logger.warning(
                         (
                             "Metrics listener configured, but "
                             "enable_metrics is not True!"
@@ -103,7 +102,7 @@ class AppserviceServer(HomeServer):
                 else:
                     _base.listen_metrics(listener["bind_addresses"], listener["port"])
             else:
-                logger.warn("Unrecognized listener type: %s", listener["type"])
+                logger.warning("Unrecognized listener type: %s", listener["type"])
 
         self.get_tcp_replication().start_replication(self)
 
@@ -143,8 +142,6 @@ def start(config_options):
 
     events.USE_FROZEN_DICTS = config.use_frozen_dicts
 
-    database_engine = create_engine(config.database_config)
-
     if config.notify_appservices:
         sys.stderr.write(
             "\nThe appservices must be disabled in the main synapse process"
@@ -159,10 +156,8 @@ def start(config_options):
 
     ps = AppserviceServer(
         config.server_name,
-        db_config=config.database_config,
         config=config,
         version_string="Synapse/" + get_version_string(synapse),
-        database_engine=database_engine,
     )
 
     setup_logging(ps, config, use_worker_options=True)

synapse/app/client_reader.py

@@ -62,7 +62,6 @@ from synapse.rest.client.v2_alpha.keys import KeyChangesServlet, KeyQueryServlet
 from synapse.rest.client.v2_alpha.register import RegisterRestServlet
 from synapse.rest.client.versions import VersionsRestServlet
 from synapse.server import HomeServer
-from synapse.storage.engines import create_engine
 from synapse.util.httpresourcetree import create_resource_tree
 from synapse.util.manhole import manhole
 from synapse.util.versionstring import get_version_string
@@ -153,7 +152,7 @@ class ClientReaderServer(HomeServer):
                 )
             elif listener["type"] == "metrics":
                 if not self.get_config().enable_metrics:
-                    logger.warn(
+                    logger.warning(
                         (
                             "Metrics listener configured, but "
                             "enable_metrics is not True!"
@@ -162,7 +161,7 @@ class ClientReaderServer(HomeServer):
                 else:
                     _base.listen_metrics(listener["bind_addresses"], listener["port"])
             else:
-                logger.warn("Unrecognized listener type: %s", listener["type"])
+                logger.warning("Unrecognized listener type: %s", listener["type"])
 
         self.get_tcp_replication().start_replication(self)
 
@@ -181,14 +180,10 @@ def start(config_options):
 
     events.USE_FROZEN_DICTS = config.use_frozen_dicts
 
-    database_engine = create_engine(config.database_config)
-
     ss = ClientReaderServer(
         config.server_name,
-        db_config=config.database_config,
         config=config,
         version_string="Synapse/" + get_version_string(synapse),
-        database_engine=database_engine,
     )
 
     setup_logging(ss, config, use_worker_options=True)

synapse/app/event_creator.py

@@ -57,7 +57,6 @@ from synapse.rest.client.v1.room import (
 )
 from synapse.server import HomeServer
 from synapse.storage.data_stores.main.user_directory import UserDirectoryStore
-from synapse.storage.engines import create_engine
 from synapse.util.httpresourcetree import create_resource_tree
 from synapse.util.manhole import manhole
 from synapse.util.versionstring import get_version_string
@@ -147,7 +146,7 @@ class EventCreatorServer(HomeServer):
                 )
             elif listener["type"] == "metrics":
                 if not self.get_config().enable_metrics:
-                    logger.warn(
+                    logger.warning(
                         (
                             "Metrics listener configured, but "
                             "enable_metrics is not True!"
@@ -156,7 +155,7 @@ class EventCreatorServer(HomeServer):
                 else:
                     _base.listen_metrics(listener["bind_addresses"], listener["port"])
             else:
-                logger.warn("Unrecognized listener type: %s", listener["type"])
+                logger.warning("Unrecognized listener type: %s", listener["type"])
 
         self.get_tcp_replication().start_replication(self)
 
@@ -180,14 +179,10 @@ def start(config_options):
 
     events.USE_FROZEN_DICTS = config.use_frozen_dicts
 
-    database_engine = create_engine(config.database_config)
-
     ss = EventCreatorServer(
         config.server_name,
-        db_config=config.database_config,
         config=config,
         version_string="Synapse/" + get_version_string(synapse),
-        database_engine=database_engine,
     )
 
     setup_logging(ss, config, use_worker_options=True)

synapse/app/federation_reader.py

@@ -46,7 +46,6 @@ from synapse.replication.slave.storage.transactions import SlavedTransactionStor
 from synapse.replication.tcp.client import ReplicationClientHandler
 from synapse.rest.key.v2 import KeyApiV2Resource
 from synapse.server import HomeServer
-from synapse.storage.engines import create_engine
 from synapse.util.httpresourcetree import create_resource_tree
 from synapse.util.manhole import manhole
 from synapse.util.versionstring import get_version_string
@@ -132,7 +131,7 @@ class FederationReaderServer(HomeServer):
                 )
             elif listener["type"] == "metrics":
                 if not self.get_config().enable_metrics:
-                    logger.warn(
+                    logger.warning(
                         (
                             "Metrics listener configured, but "
                             "enable_metrics is not True!"
@@ -141,7 +140,7 @@ class FederationReaderServer(HomeServer):
                 else:
                     _base.listen_metrics(listener["bind_addresses"], listener["port"])
             else:
-                logger.warn("Unrecognized listener type: %s", listener["type"])
+                logger.warning("Unrecognized listener type: %s", listener["type"])
 
         self.get_tcp_replication().start_replication(self)
 
@@ -162,14 +161,10 @@ def start(config_options):
 
     events.USE_FROZEN_DICTS = config.use_frozen_dicts
 
-    database_engine = create_engine(config.database_config)
-
     ss = FederationReaderServer(
         config.server_name,
-        db_config=config.database_config,
         config=config,
         version_string="Synapse/" + get_version_string(synapse),
-        database_engine=database_engine,
     )
 
     setup_logging(ss, config, use_worker_options=True)

synapse/app/federation_sender.py

@@ -40,7 +40,7 @@ from synapse.replication.slave.storage.transactions import SlavedTransactionStor
 from synapse.replication.tcp.client import ReplicationClientHandler
 from synapse.replication.tcp.streams._base import ReceiptsStream
 from synapse.server import HomeServer
-from synapse.storage.engines import create_engine
+from synapse.storage.database import Database
 from synapse.types import ReadReceipt
 from synapse.util.async_helpers import Linearizer
 from synapse.util.httpresourcetree import create_resource_tree
@@ -59,8 +59,8 @@ class FederationSenderSlaveStore(
     SlavedDeviceStore,
     SlavedPresenceStore,
 ):
-    def __init__(self, db_conn, hs):
-        super(FederationSenderSlaveStore, self).__init__(db_conn, hs)
+    def __init__(self, database: Database, db_conn, hs):
+        super(FederationSenderSlaveStore, self).__init__(database, db_conn, hs)
 
         # We pull out the current federation stream position now so that we
         # always have a known value for the federation position in memory so
@@ -69,7 +69,7 @@ class FederationSenderSlaveStore(
         self.federation_out_pos_startup = self._get_federation_out_pos(db_conn)
 
     def _get_federation_out_pos(self, db_conn):
-        sql = "SELECT stream_id FROM federation_stream_position" " WHERE type = ?"
+        sql = "SELECT stream_id FROM federation_stream_position WHERE type = ?"
         sql = self.database_engine.convert_param_style(sql)
 
         txn = db_conn.cursor()
@@ -123,7 +123,7 @@ class FederationSenderServer(HomeServer):
                 )
             elif listener["type"] == "metrics":
                 if not self.get_config().enable_metrics:
-                    logger.warn(
+                    logger.warning(
                         (
                             "Metrics listener configured, but "
                             "enable_metrics is not True!"
@@ -132,7 +132,7 @@ class FederationSenderServer(HomeServer):
                 else:
                     _base.listen_metrics(listener["bind_addresses"], listener["port"])
             else:
-                logger.warn("Unrecognized listener type: %s", listener["type"])
+                logger.warning("Unrecognized listener type: %s", listener["type"])
 
         self.get_tcp_replication().start_replication(self)
 
@@ -173,8 +173,6 @@ def start(config_options):
 
     events.USE_FROZEN_DICTS = config.use_frozen_dicts
 
-    database_engine = create_engine(config.database_config)
-
     if config.send_federation:
         sys.stderr.write(
             "\nThe send_federation must be disabled in the main synapse process"
@@ -189,10 +187,8 @@ def start(config_options):
 
     ss = FederationSenderServer(
         config.server_name,
-        db_config=config.database_config,
         config=config,
         version_string="Synapse/" + get_version_string(synapse),
-        database_engine=database_engine,
     )
 
     setup_logging(ss, config, use_worker_options=True)

synapse/app/frontend_proxy.py

@@ -39,7 +39,6 @@ from synapse.replication.slave.storage.registration import SlavedRegistrationSto
 from synapse.replication.tcp.client import ReplicationClientHandler
 from synapse.rest.client.v2_alpha._base import client_patterns
 from synapse.server import HomeServer
-from synapse.storage.engines import create_engine
 from synapse.util.httpresourcetree import create_resource_tree
 from synapse.util.manhole import manhole
 from synapse.util.versionstring import get_version_string
@@ -204,7 +203,7 @@ class FrontendProxyServer(HomeServer):
                 )
             elif listener["type"] == "metrics":
                 if not self.get_config().enable_metrics:
-                    logger.warn(
+                    logger.warning(
                         (
                             "Metrics listener configured, but "
                             "enable_metrics is not True!"
@@ -213,7 +212,7 @@ class FrontendProxyServer(HomeServer):
                 else:
                     _base.listen_metrics(listener["bind_addresses"], listener["port"])
             else:
-                logger.warn("Unrecognized listener type: %s", listener["type"])
+                logger.warning("Unrecognized listener type: %s", listener["type"])
 
         self.get_tcp_replication().start_replication(self)
 
@@ -234,14 +233,10 @@ def start(config_options):
 
     events.USE_FROZEN_DICTS = config.use_frozen_dicts
 
-    database_engine = create_engine(config.database_config)
-
     ss = FrontendProxyServer(
         config.server_name,
-        db_config=config.database_config,
         config=config,
         version_string="Synapse/" + get_version_string(synapse),
-        database_engine=database_engine,
     )
 
     setup_logging(ss, config, use_worker_options=True)

synapse/app/homeserver.py

@@ -19,12 +19,13 @@ from __future__ import print_function
 
 import gc
 import logging
+import math
 import os
+import resource
 import sys
 
 from six import iteritems
 
-import psutil
 from prometheus_client import Gauge
 
 from twisted.application import service
@@ -67,9 +68,9 @@ from synapse.rest.key.v2 import KeyApiV2Resource
 from synapse.rest.media.v0.content_repository import ContentRepoResource
 from synapse.rest.well_known import WellKnownResource
 from synapse.server import HomeServer
-from synapse.storage import DataStore, are_all_users_on_domain
-from synapse.storage.engines import IncorrectDatabaseSetup, create_engine
-from synapse.storage.prepare_database import UpgradeDatabaseException, prepare_database
+from synapse.storage import DataStore
+from synapse.storage.engines import IncorrectDatabaseSetup
+from synapse.storage.prepare_database import UpgradeDatabaseException
 from synapse.util.caches import CACHE_SIZE_FACTOR
 from synapse.util.httpresourcetree import create_resource_tree
 from synapse.util.manhole import manhole
@@ -282,7 +283,7 @@ class SynapseHomeServer(HomeServer):
                     reactor.addSystemEventTrigger("before", "shutdown", s.stopListening)
             elif listener["type"] == "metrics":
                 if not self.get_config().enable_metrics:
-                    logger.warn(
+                    logger.warning(
                         (
                             "Metrics listener configured, but "
                             "enable_metrics is not True!"
@@ -291,23 +292,7 @@ class SynapseHomeServer(HomeServer):
                 else:
                     _base.listen_metrics(listener["bind_addresses"], listener["port"])
             else:
-                logger.warn("Unrecognized listener type: %s", listener["type"])
-
-    def run_startup_checks(self, db_conn, database_engine):
-        all_users_native = are_all_users_on_domain(
-            db_conn.cursor(), database_engine, self.hostname
-        )
-        if not all_users_native:
-            quit_with_error(
-                "Found users in database not native to %s!\n"
-                "You cannot changed a synapse server_name after it's been configured"
-                % (self.hostname,)
-            )
-
-        try:
-            database_engine.check_database(db_conn.cursor())
-        except IncorrectDatabaseSetup as e:
-            quit_with_error(str(e))
+                logger.warning("Unrecognized listener type: %s", listener["type"])
 
 
 # Gauges to expose monthly active user control metrics
@@ -343,29 +328,20 @@ def setup(config_options):
 
     events.USE_FROZEN_DICTS = config.use_frozen_dicts
 
-    database_engine = create_engine(config.database_config)
-    config.database_config["args"]["cp_openfun"] = database_engine.on_new_connection
-
     hs = SynapseHomeServer(
         config.server_name,
-        db_config=config.database_config,
         config=config,
         version_string="Synapse/" + get_version_string(synapse),
-        database_engine=database_engine,
     )
 
     synapse.config.logger.setup_logging(hs, config, use_worker_options=False)
 
-    logger.info("Preparing database: %s...", config.database_config["name"])
+    logger.info("Setting up server")
 
     try:
-        with hs.get_db_conn(run_new_connection=False) as db_conn:
-            prepare_database(db_conn, database_engine, config=config)
-            database_engine.on_new_connection(db_conn)
-
-            hs.run_startup_checks(db_conn, database_engine)
-
-            db_conn.commit()
+        hs.setup()
+    except IncorrectDatabaseSetup as e:
+        quit_with_error(str(e))
     except UpgradeDatabaseException:
         sys.stderr.write(
             "\nFailed to upgrade database.\n"
@@ -374,9 +350,6 @@ def setup(config_options):
         )
         sys.exit(1)
 
-    logger.info("Database prepared in %s.", config.database_config["name"])
-
-    hs.setup()
     hs.setup_master()
 
     @defer.inlineCallbacks
@@ -435,7 +408,7 @@ def setup(config_options):
             _base.start(hs, config.listeners)
 
             hs.get_pusherpool().start()
-            hs.get_datastore().start_doing_background_updates()
+            hs.get_datastore().db.updates.start_doing_background_updates()
         except Exception:
             # Print the exception and bail out.
             print("Error during startup:", file=sys.stderr)
@@ -471,6 +444,89 @@ class SynapseService(service.Service):
         return self._port.stopListening()
 
 
+# Contains the list of processes we will be monitoring
+# currently either 0 or 1
+_stats_process = []
+
+
+@defer.inlineCallbacks
+def phone_stats_home(hs, stats, stats_process=_stats_process):
+    logger.info("Gathering stats for reporting")
+    now = int(hs.get_clock().time())
+    uptime = int(now - hs.start_time)
+    if uptime < 0:
+        uptime = 0
+
+    stats["homeserver"] = hs.config.server_name
+    stats["server_context"] = hs.config.server_context
+    stats["timestamp"] = now
+    stats["uptime_seconds"] = uptime
+    version = sys.version_info
+    stats["python_version"] = "{}.{}.{}".format(
+        version.major, version.minor, version.micro
+    )
+    stats["total_users"] = yield hs.get_datastore().count_all_users()
+
+    total_nonbridged_users = yield hs.get_datastore().count_nonbridged_users()
+    stats["total_nonbridged_users"] = total_nonbridged_users
+
+    daily_user_type_results = yield hs.get_datastore().count_daily_user_type()
+    for name, count in iteritems(daily_user_type_results):
+        stats["daily_user_type_" + name] = count
+
+    room_count = yield hs.get_datastore().get_room_count()
+    stats["total_room_count"] = room_count
+
+    stats["daily_active_users"] = yield hs.get_datastore().count_daily_users()
+    stats["monthly_active_users"] = yield hs.get_datastore().count_monthly_users()
+    stats["daily_active_rooms"] = yield hs.get_datastore().count_daily_active_rooms()
+    stats["daily_messages"] = yield hs.get_datastore().count_daily_messages()
+
+    r30_results = yield hs.get_datastore().count_r30_users()
+    for name, count in iteritems(r30_results):
+        stats["r30_users_" + name] = count
+
+    daily_sent_messages = yield hs.get_datastore().count_daily_sent_messages()
+    stats["daily_sent_messages"] = daily_sent_messages
+    stats["cache_factor"] = CACHE_SIZE_FACTOR
+    stats["event_cache_size"] = hs.config.event_cache_size
+
+    #
+    # Performance statistics
+    #
+    old = stats_process[0]
+    new = (now, resource.getrusage(resource.RUSAGE_SELF))
+    stats_process[0] = new
+
+    # Get RSS in bytes
+    stats["memory_rss"] = new[1].ru_maxrss
+
+    # Get CPU time in % of a single core, not % of all cores
+    used_cpu_time = (new[1].ru_utime + new[1].ru_stime) - (
+        old[1].ru_utime + old[1].ru_stime
+    )
+    if used_cpu_time == 0 or new[0] == old[0]:
+        stats["cpu_average"] = 0
+    else:
+        stats["cpu_average"] = math.floor(used_cpu_time / (new[0] - old[0]) * 100)
+
+    #
+    # Database version
+    #
+
+    # This only reports info about the *main* database.
+    stats["database_engine"] = hs.get_datastore().db.engine.module.__name__
+    stats["database_server_version"] = hs.get_datastore().db.engine.server_version
+
+    logger.info("Reporting stats to %s: %s" % (hs.config.report_stats_endpoint, stats))
+    try:
+        yield hs.get_proxied_http_client().put_json(
+            hs.config.report_stats_endpoint, stats
+        )
+    except Exception as e:
+        logger.warning("Error reporting stats: %s", e)
+
+
 def run(hs):
     PROFILE_SYNAPSE = False
     if PROFILE_SYNAPSE:
@@ -497,91 +553,19 @@ def run(hs):
         reactor.run = profile(reactor.run)
 
     clock = hs.get_clock()
-    start_time = clock.time()
 
     stats = {}
 
-    # Contains the list of processes we will be monitoring
-    # currently either 0 or 1
-    stats_process = []
+    def performance_stats_init():
+        _stats_process.clear()
+        _stats_process.append(
+            (int(hs.get_clock().time()), resource.getrusage(resource.RUSAGE_SELF))
+        )
 
     def start_phone_stats_home():
-        return run_as_background_process("phone_stats_home", phone_stats_home)
-
-    @defer.inlineCallbacks
-    def phone_stats_home():
-        logger.info("Gathering stats for reporting")
-        now = int(hs.get_clock().time())
-        uptime = int(now - start_time)
-        if uptime < 0:
-            uptime = 0
-
-        stats["homeserver"] = hs.config.server_name
-        stats["server_context"] = hs.config.server_context
-        stats["timestamp"] = now
-        stats["uptime_seconds"] = uptime
-        version = sys.version_info
-        stats["python_version"] = "{}.{}.{}".format(
-            version.major, version.minor, version.micro
+        return run_as_background_process(
+            "phone_stats_home", phone_stats_home, hs, stats
         )
-        stats["total_users"] = yield hs.get_datastore().count_all_users()
-
-        total_nonbridged_users = yield hs.get_datastore().count_nonbridged_users()
-        stats["total_nonbridged_users"] = total_nonbridged_users
-
-        daily_user_type_results = yield hs.get_datastore().count_daily_user_type()
-        for name, count in iteritems(daily_user_type_results):
-            stats["daily_user_type_" + name] = count
-
-        room_count = yield hs.get_datastore().get_room_count()
-        stats["total_room_count"] = room_count
-
-        stats["daily_active_users"] = yield hs.get_datastore().count_daily_users()
-        stats["monthly_active_users"] = yield hs.get_datastore().count_monthly_users()
-        stats[
-            "daily_active_rooms"
-        ] = yield hs.get_datastore().count_daily_active_rooms()
-        stats["daily_messages"] = yield hs.get_datastore().count_daily_messages()
-
-        r30_results = yield hs.get_datastore().count_r30_users()
-        for name, count in iteritems(r30_results):
-            stats["r30_users_" + name] = count
-
-        daily_sent_messages = yield hs.get_datastore().count_daily_sent_messages()
-        stats["daily_sent_messages"] = daily_sent_messages
-        stats["cache_factor"] = CACHE_SIZE_FACTOR
-        stats["event_cache_size"] = hs.config.event_cache_size
-
-        if len(stats_process) > 0:
-            stats["memory_rss"] = 0
-            stats["cpu_average"] = 0
-            for process in stats_process:
-                stats["memory_rss"] += process.memory_info().rss
-                stats["cpu_average"] += int(process.cpu_percent(interval=None))
-
-        stats["database_engine"] = hs.get_datastore().database_engine_name
-        stats["database_server_version"] = hs.get_datastore().get_server_version()
-        logger.info(
-            "Reporting stats to %s: %s" % (hs.config.report_stats_endpoint, stats)
-        )
-        try:
-            yield hs.get_simple_http_client().put_json(
-                hs.config.report_stats_endpoint, stats
-            )
-        except Exception as e:
-            logger.warn("Error reporting stats: %s", e)
-
-    def performance_stats_init():
-        try:
-            process = psutil.Process()
-            # Ensure we can fetch both, and make the initial request for cpu_percent
-            # so the next request will use this as the initial point.
-            process.memory_info().rss
-            process.cpu_percent(interval=None)
-            logger.info("report_stats can use psutil")
-            stats_process.append(process)
-        except (AttributeError):
-            logger.warning("Unable to read memory/cpu stats. Disabling reporting.")
 
     def generate_user_daily_visit_stats():
         return run_as_background_process(

synapse/app/media_repository.py

@@ -40,7 +40,6 @@ from synapse.rest.admin import register_servlets_for_media_repo
 from synapse.rest.media.v0.content_repository import ContentRepoResource
 from synapse.server import HomeServer
 from synapse.storage.data_stores.main.media_repository import MediaRepositoryStore
-from synapse.storage.engines import create_engine
 from synapse.util.httpresourcetree import create_resource_tree
 from synapse.util.manhole import manhole
 from synapse.util.versionstring import get_version_string
@@ -120,7 +119,7 @@ class MediaRepositoryServer(HomeServer):
                 )
             elif listener["type"] == "metrics":
                 if not self.get_config().enable_metrics:
-                    logger.warn(
+                    logger.warning(
                         (
                             "Metrics listener configured, but "
                             "enable_metrics is not True!"
@@ -129,7 +128,7 @@ class MediaRepositoryServer(HomeServer):
                 else:
                     _base.listen_metrics(listener["bind_addresses"], listener["port"])
             else:
-                logger.warn("Unrecognized listener type: %s", listener["type"])
+                logger.warning("Unrecognized listener type: %s", listener["type"])
 
         self.get_tcp_replication().start_replication(self)
 
@@ -157,14 +156,10 @@ def start(config_options):
 
     events.USE_FROZEN_DICTS = config.use_frozen_dicts
 
-    database_engine = create_engine(config.database_config)
-
     ss = MediaRepositoryServer(
         config.server_name,
-        db_config=config.database_config,
         config=config,
         version_string="Synapse/" + get_version_string(synapse),
-        database_engine=database_engine,
     )
 
     setup_logging(ss, config, use_worker_options=True)

synapse/app/pusher.py

@@ -33,10 +33,10 @@ from synapse.replication.slave.storage.account_data import SlavedAccountDataStor
 from synapse.replication.slave.storage.events import SlavedEventStore
 from synapse.replication.slave.storage.pushers import SlavedPusherStore
 from synapse.replication.slave.storage.receipts import SlavedReceiptsStore
+from synapse.replication.slave.storage.room import RoomStore
 from synapse.replication.tcp.client import ReplicationClientHandler
 from synapse.server import HomeServer
 from synapse.storage import DataStore
-from synapse.storage.engines import create_engine
 from synapse.util.httpresourcetree import create_resource_tree
 from synapse.util.manhole import manhole
 from synapse.util.versionstring import get_version_string
@@ -45,7 +45,11 @@ logger = logging.getLogger("synapse.app.pusher")
 
 
 class PusherSlaveStore(
-    SlavedEventStore, SlavedPusherStore, SlavedReceiptsStore, SlavedAccountDataStore
+    SlavedEventStore,
+    SlavedPusherStore,
+    SlavedReceiptsStore,
+    SlavedAccountDataStore,
+    RoomStore,
 ):
     update_pusher_last_stream_ordering_and_success = __func__(
         DataStore.update_pusher_last_stream_ordering_and_success
@@ -114,7 +118,7 @@ class PusherServer(HomeServer):
                 )
             elif listener["type"] == "metrics":
                 if not self.get_config().enable_metrics:
-                    logger.warn(
+                    logger.warning(
                         (
                             "Metrics listener configured, but "
                             "enable_metrics is not True!"
@@ -123,7 +127,7 @@ class PusherServer(HomeServer):
                 else:
                     _base.listen_metrics(listener["bind_addresses"], listener["port"])
             else:
-                logger.warn("Unrecognized listener type: %s", listener["type"])
+                logger.warning("Unrecognized listener type: %s", listener["type"])
 
         self.get_tcp_replication().start_replication(self)
 
@@ -198,14 +202,10 @@ def start(config_options):
     # Force the pushers to start since they will be disabled in the main config
     config.start_pushers = True
 
-    database_engine = create_engine(config.database_config)
-
     ps = PusherServer(
         config.server_name,
-        db_config=config.database_config,
         config=config,
         version_string="Synapse/" + get_version_string(synapse),
-        database_engine=database_engine,
     )
 
     setup_logging(ps, config, use_worker_options=True)

synapse/app/synchrotron.py

@@ -55,7 +55,6 @@ from synapse.rest.client.v1.room import RoomInitialSyncRestServlet
 from synapse.rest.client.v2_alpha import sync
 from synapse.server import HomeServer
 from synapse.storage.data_stores.main.presence import UserPresenceState
-from synapse.storage.engines import create_engine
 from synapse.util.httpresourcetree import create_resource_tree
 from synapse.util.manhole import manhole
 from synapse.util.stringutils import random_string
@@ -151,7 +150,7 @@ class SynchrotronPresence(object):
 
     def set_state(self, user, state, ignore_status_msg=False):
         # TODO Hows this supposed to work?
-        pass
+        return defer.succeed(None)
 
     get_states = __func__(PresenceHandler.get_states)
     get_state = __func__(PresenceHandler.get_state)
@@ -326,7 +325,7 @@ class SynchrotronServer(HomeServer):
                 )
             elif listener["type"] == "metrics":
                 if not self.get_config().enable_metrics:
-                    logger.warn(
+                    logger.warning(
                         (
                             "Metrics listener configured, but "
                             "enable_metrics is not True!"
@@ -335,7 +334,7 @@ class SynchrotronServer(HomeServer):
                 else:
                     _base.listen_metrics(listener["bind_addresses"], listener["port"])
             else:
-                logger.warn("Unrecognized listener type: %s", listener["type"])
+                logger.warning("Unrecognized listener type: %s", listener["type"])
 
         self.get_tcp_replication().start_replication(self)
 
@@ -437,14 +436,10 @@ def start(config_options):
 
     synapse.events.USE_FROZEN_DICTS = config.use_frozen_dicts
 
-    database_engine = create_engine(config.database_config)
-
     ss = SynchrotronServer(
         config.server_name,
-        db_config=config.database_config,
         config=config,
         version_string="Synapse/" + get_version_string(synapse),
-        database_engine=database_engine,
         application_service_handler=SynchrotronApplicationService(),
     )
 

synapse/app/user_dir.py

@@ -43,7 +43,7 @@ from synapse.replication.tcp.streams.events import (
 from synapse.rest.client.v2_alpha import user_directory
 from synapse.server import HomeServer
 from synapse.storage.data_stores.main.user_directory import UserDirectoryStore
-from synapse.storage.engines import create_engine
+from synapse.storage.database import Database
 from synapse.util.caches.stream_change_cache import StreamChangeCache
 from synapse.util.httpresourcetree import create_resource_tree
 from synapse.util.manhole import manhole
@@ -60,11 +60,11 @@ class UserDirectorySlaveStore(
     UserDirectoryStore,
     BaseSlavedStore,
 ):
-    def __init__(self, db_conn, hs):
-        super(UserDirectorySlaveStore, self).__init__(db_conn, hs)
+    def __init__(self, database: Database, db_conn, hs):
+        super(UserDirectorySlaveStore, self).__init__(database, db_conn, hs)
 
         events_max = self._stream_id_gen.get_current_token()
-        curr_state_delta_prefill, min_curr_state_delta_id = self._get_cache_dict(
+        curr_state_delta_prefill, min_curr_state_delta_id = self.db.get_cache_dict(
             db_conn,
             "current_state_delta_stream",
             entity_column="room_id",
@@ -150,7 +150,7 @@ class UserDirectoryServer(HomeServer):
                 )
             elif listener["type"] == "metrics":
                 if not self.get_config().enable_metrics:
-                    logger.warn(
+                    logger.warning(
                         (
                             "Metrics listener configured, but "
                             "enable_metrics is not True!"
@@ -159,7 +159,7 @@ class UserDirectoryServer(HomeServer):
                 else:
                     _base.listen_metrics(listener["bind_addresses"], listener["port"])
             else:
-                logger.warn("Unrecognized listener type: %s", listener["type"])
+                logger.warning("Unrecognized listener type: %s", listener["type"])
 
         self.get_tcp_replication().start_replication(self)
 
@@ -199,8 +199,6 @@ def start(config_options):
 
     events.USE_FROZEN_DICTS = config.use_frozen_dicts
 
-    database_engine = create_engine(config.database_config)
-
     if config.update_user_directory:
         sys.stderr.write(
             "\nThe update_user_directory must be disabled in the main synapse process"
@@ -215,10 +213,8 @@ def start(config_options):
 
     ss = UserDirectoryServer(
         config.server_name,
-        db_config=config.database_config,
         config=config,
         version_string="Synapse/" + get_version_string(synapse),
-        database_engine=database_engine,
     )
 
     setup_logging(ss, config, use_worker_options=True)

synapse/appservice/__init__.py

@@ -94,7 +94,9 @@ class ApplicationService(object):
         ip_range_whitelist=None,
     ):
         self.token = token
-        self.url = url
+        self.url = (
+            url.rstrip("/") if isinstance(url, str) else None
+        )  # url must not end with a slash
         self.hs_token = hs_token
         self.sender = sender
         self.server_name = hostname

synapse/appservice/api.py

@@ -185,7 +185,7 @@ class ApplicationServiceApi(SimpleHttpClient):
 
                 if not _is_valid_3pe_metadata(info):
                     logger.warning(
-                        "query_3pe_protocol to %s did not return a" " valid result", uri
+                        "query_3pe_protocol to %s did not return a valid result", uri
                     )
                     return None
 

synapse/config/appservice.py

@@ -134,7 +134,7 @@ def _load_appservice(hostname, as_info, config_filename):
             for regex_obj in as_info["namespaces"][ns]:
                 if not isinstance(regex_obj, dict):
                     raise ValueError(
-                        "Expected namespace entry in %s to be an object," " but got %s",
+                        "Expected namespace entry in %s to be an object, but got %s",
                         ns,
                         regex_obj,
                     )

synapse/config/captcha.py

@@ -35,11 +35,11 @@ class CaptchaConfig(Config):
         ## Captcha ##
         # See docs/CAPTCHA_SETUP for full details of configuring this.
 
-        # This Home Server's ReCAPTCHA public key.
+        # This homeserver's ReCAPTCHA public key.
         #
         #recaptcha_public_key: "YOUR_PUBLIC_KEY"
 
-        # This Home Server's ReCAPTCHA private key.
+        # This homeserver's ReCAPTCHA private key.
         #
         #recaptcha_private_key: "YOUR_PRIVATE_KEY"
 

synapse/config/emailconfig.py

@@ -146,6 +146,8 @@ class EmailConfig(Config):
                 if k not in email_config:
                     missing.append("email." + k)
 
+            # public_baseurl is required to build password reset and validation links that
+            # will be emailed to users
             if config.get("public_baseurl") is None:
                 missing.append("public_baseurl")
 
@@ -305,8 +307,23 @@ class EmailConfig(Config):
         #   smtp_user: "exampleusername"
         #   smtp_pass: "examplepassword"
         #   require_transport_security: false
-        #   notif_from: "Your Friendly %(app)s Home Server <noreply@example.com>"
-        #   app_name: Matrix
+        #
+        #   # notif_from defines the "From" address to use when sending emails.
+        #   # It must be set if email sending is enabled.
+        #   #
+        #   # The placeholder '%(app)s' will be replaced by the application name,
+        #   # which is normally 'app_name' (below), but may be overridden by the
+        #   # Matrix client application.
+        #   #
+        #   # Note that the placeholder must be written '%(app)s', including the
+        #   # trailing 's'.
+        #   #
+        #   notif_from: "Your Friendly %(app)s homeserver <noreply@example.com>"
+        #
+        #   # app_name defines the default value for '%(app)s' in notif_from. It
+        #   # defaults to 'Matrix'.
+        #   #
+        #   #app_name: my_branded_matrix_server
         #
         #   # Enable email notifications by default
         #   #

synapse/config/key.py

@@ -125,7 +125,7 @@ class KeyConfig(Config):
 
         # if neither trusted_key_servers nor perspectives are given, use the default.
         if "perspectives" not in config and "trusted_key_servers" not in config:
-            logger.warn(TRUSTED_KEY_SERVER_NOT_CONFIGURED_WARN)
+            logger.warning(TRUSTED_KEY_SERVER_NOT_CONFIGURED_WARN)
             key_servers = [{"server_name": "matrix.org"}]
         else:
             key_servers = config.get("trusted_key_servers", [])
@@ -156,7 +156,7 @@ class KeyConfig(Config):
         if not self.macaroon_secret_key:
             # Unfortunately, there are people out there that don't have this
             # set. Lets just be "nice" and derive one from their secret key.
-            logger.warn("Config is missing macaroon_secret_key")
+            logger.warning("Config is missing macaroon_secret_key")
             seed = bytes(self.signing_key[0])
             self.macaroon_secret_key = hashlib.sha256(seed).digest()
 

synapse/config/logger.py

@@ -182,7 +182,7 @@ def _reload_stdlib_logging(*args, log_config=None):
     logger = logging.getLogger("")
 
     if not log_config:
-        logger.warn("Reloaded a blank config?")
+        logger.warning("Reloaded a blank config?")
 
     logging.config.dictConfig(log_config)
 
@@ -234,8 +234,8 @@ def setup_logging(
 
     # make sure that the first thing we log is a thing we can grep backwards
     # for
-    logging.warn("***** STARTING SERVER *****")
-    logging.warn("Server %s version %s", sys.argv[0], get_version_string(synapse))
+    logging.warning("***** STARTING SERVER *****")
+    logging.warning("Server %s version %s", sys.argv[0], get_version_string(synapse))
     logging.info("Server hostname: %s", config.server_name)
 
     return logger

synapse/config/registration.py

@@ -106,6 +106,13 @@ class RegistrationConfig(Config):
         account_threepid_delegates = config.get("account_threepid_delegates") or {}
         self.account_threepid_delegate_email = account_threepid_delegates.get("email")
         self.account_threepid_delegate_msisdn = account_threepid_delegates.get("msisdn")
+        if self.account_threepid_delegate_msisdn and not self.public_baseurl:
+            raise ConfigError(
+                "The configuration option `public_baseurl` is required if "
+                "`account_threepid_delegate.msisdn` is set, such that "
+                "clients know where to submit validation tokens to. Please "
+                "configure `public_baseurl`."
+            )
 
         self.default_identity_server = config.get("default_identity_server")
         self.allow_guest_access = config.get("allow_guest_access", False)
@@ -300,7 +307,7 @@ class RegistrationConfig(Config):
         # If a delegate is specified, the config option public_baseurl must also be filled out.
         #
         account_threepid_delegates:
-            #email: https://example.com     # Delegate email sending to example.org
+            #email: https://example.com     # Delegate email sending to example.com
             #msisdn: http://localhost:8090  # Delegate SMS sending to this local process
 
         # Users who register on this homeserver will automatically be joined

synapse/config/room_directory.py

@@ -170,7 +170,7 @@ class _RoomDirectoryRule(object):
             self.action = action
         else:
             raise ConfigError(
-                "%s rules can only have action of 'allow'" " or 'deny'" % (option_name,)
+                "%s rules can only have action of 'allow' or 'deny'" % (option_name,)
             )
 
         self._alias_matches_all = alias == "*"

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, required_attributes: set, optional_attributes: set
+    ):
+        """Generate a configuration dictionary with required and optional attributes that
+        will be needed to process new user registration
 
-    def _default_saml_config_dict(self):
+        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
+          user_mapping_provider:
+            # The custom module's class. Uncomment to use a custom module.
+            #
+            #module: mapping_provider.SamlMappingProvider
 
-          # 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
+            # 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
 
-          # 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.
+              # 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,
-}

synapse/config/server.py

@@ -19,7 +19,7 @@ import logging
 import os.path
 import re
 from textwrap import indent
-from typing import List
+from typing import Dict, List, Optional
 
 import attr
 import yaml
@@ -41,7 +41,7 @@ logger = logging.Logger(__name__)
 # in the list.
 DEFAULT_BIND_ADDRESSES = ["::", "0.0.0.0"]
 
-DEFAULT_ROOM_VERSION = "4"
+DEFAULT_ROOM_VERSION = "5"
 
 ROOM_COMPLEXITY_TOO_GREAT = (
     "Your homeserver is unable to join rooms this large or complex. "
@@ -118,15 +118,16 @@ class ServerConfig(Config):
             self.allow_public_rooms_without_auth = False
             self.allow_public_rooms_over_federation = False
         else:
-            # If set to 'False', requires authentication to access the server's public
-            # rooms directory through the client API. Defaults to 'True'.
+            # If set to 'true', removes the need for authentication to access the server's
+            # public rooms directory through the client API, meaning that anyone can
+            # query the room directory. Defaults to 'false'.
             self.allow_public_rooms_without_auth = config.get(
-                "allow_public_rooms_without_auth", True
+                "allow_public_rooms_without_auth", False
             )
-            # If set to 'False', forbids any other homeserver to fetch the server's public
-            # rooms directory via federation. Defaults to 'True'.
+            # If set to 'true', allows any other homeserver to fetch the server's public
+            # rooms directory via federation. Defaults to 'false'.
             self.allow_public_rooms_over_federation = config.get(
-                "allow_public_rooms_over_federation", True
+                "allow_public_rooms_over_federation", False
             )
 
         default_room_version = config.get("default_room_version", DEFAULT_ROOM_VERSION)
@@ -223,7 +224,7 @@ class ServerConfig(Config):
             self.federation_ip_range_blacklist.update(["0.0.0.0", "::"])
         except Exception as e:
             raise ConfigError(
-                "Invalid range(s) provided in " "federation_ip_range_blacklist: %s" % e
+                "Invalid range(s) provided in federation_ip_range_blacklist: %s" % e
             )
 
         if self.public_baseurl is not None:
@@ -246,6 +247,124 @@ class ServerConfig(Config):
         # events with profile information that differ from the target's global profile.
         self.allow_per_room_profiles = config.get("allow_per_room_profiles", True)
 
+        retention_config = config.get("retention")
+        if retention_config is None:
+            retention_config = {}
+
+        self.retention_enabled = retention_config.get("enabled", False)
+
+        retention_default_policy = retention_config.get("default_policy")
+
+        if retention_default_policy is not None:
+            self.retention_default_min_lifetime = retention_default_policy.get(
+                "min_lifetime"
+            )
+            if self.retention_default_min_lifetime is not None:
+                self.retention_default_min_lifetime = self.parse_duration(
+                    self.retention_default_min_lifetime
+                )
+
+            self.retention_default_max_lifetime = retention_default_policy.get(
+                "max_lifetime"
+            )
+            if self.retention_default_max_lifetime is not None:
+                self.retention_default_max_lifetime = self.parse_duration(
+                    self.retention_default_max_lifetime
+                )
+
+            if (
+                self.retention_default_min_lifetime is not None
+                and self.retention_default_max_lifetime is not None
+                and (
+                    self.retention_default_min_lifetime
+                    > self.retention_default_max_lifetime
+                )
+            ):
+                raise ConfigError(
+                    "The default retention policy's 'min_lifetime' can not be greater"
+                    " than its 'max_lifetime'"
+                )
+        else:
+            self.retention_default_min_lifetime = None
+            self.retention_default_max_lifetime = None
+
+        self.retention_allowed_lifetime_min = retention_config.get(
+            "allowed_lifetime_min"
+        )
+        if self.retention_allowed_lifetime_min is not None:
+            self.retention_allowed_lifetime_min = self.parse_duration(
+                self.retention_allowed_lifetime_min
+            )
+
+        self.retention_allowed_lifetime_max = retention_config.get(
+            "allowed_lifetime_max"
+        )
+        if self.retention_allowed_lifetime_max is not None:
+            self.retention_allowed_lifetime_max = self.parse_duration(
+                self.retention_allowed_lifetime_max
+            )
+
+        if (
+            self.retention_allowed_lifetime_min is not None
+            and self.retention_allowed_lifetime_max is not None
+            and self.retention_allowed_lifetime_min
+            > self.retention_allowed_lifetime_max
+        ):
+            raise ConfigError(
+                "Invalid retention policy limits: 'allowed_lifetime_min' can not be"
+                " greater than 'allowed_lifetime_max'"
+            )
+
+        self.retention_purge_jobs = []  # type: List[Dict[str, Optional[int]]]
+        for purge_job_config in retention_config.get("purge_jobs", []):
+            interval_config = purge_job_config.get("interval")
+
+            if interval_config is None:
+                raise ConfigError(
+                    "A retention policy's purge jobs configuration must have the"
+                    " 'interval' key set."
+                )
+
+            interval = self.parse_duration(interval_config)
+
+            shortest_max_lifetime = purge_job_config.get("shortest_max_lifetime")
+
+            if shortest_max_lifetime is not None:
+                shortest_max_lifetime = self.parse_duration(shortest_max_lifetime)
+
+            longest_max_lifetime = purge_job_config.get("longest_max_lifetime")
+
+            if longest_max_lifetime is not None:
+                longest_max_lifetime = self.parse_duration(longest_max_lifetime)
+
+            if (
+                shortest_max_lifetime is not None
+                and longest_max_lifetime is not None
+                and shortest_max_lifetime > longest_max_lifetime
+            ):
+                raise ConfigError(
+                    "A retention policy's purge jobs configuration's"
+                    " 'shortest_max_lifetime' value can not be greater than its"
+                    " 'longest_max_lifetime' value."
+                )
+
+            self.retention_purge_jobs.append(
+                {
+                    "interval": interval,
+                    "shortest_max_lifetime": shortest_max_lifetime,
+                    "longest_max_lifetime": longest_max_lifetime,
+                }
+            )
+
+        if not self.retention_purge_jobs:
+            self.retention_purge_jobs = [
+                {
+                    "interval": self.parse_duration("1d"),
+                    "shortest_max_lifetime": None,
+                    "longest_max_lifetime": None,
+                }
+            ]
+
         self.listeners = []  # type: List[dict]
         for listener in config.get("listeners", []):
             if not isinstance(listener.get("port", None), int):
@@ -372,6 +491,8 @@ class ServerConfig(Config):
             "cleanup_extremities_with_dummy_events", True
         )
 
+        self.enable_ephemeral_messages = config.get("enable_ephemeral_messages", False)
+
     def has_tls_listener(self) -> bool:
         return any(l["tls"] for l in self.listeners)
 
@@ -500,15 +621,16 @@ class ServerConfig(Config):
         #
         #require_auth_for_profile_requests: true
 
-        # If set to 'false', requires authentication to access the server's public rooms
-        # directory through the client API. Defaults to 'true'.
+        # If set to 'true', removes the need for authentication to access the server's
+        # public rooms directory through the client API, meaning that anyone can
+        # query the room directory. Defaults to 'false'.
         #
-        #allow_public_rooms_without_auth: false
+        #allow_public_rooms_without_auth: true
 
-        # If set to 'false', forbids any other homeserver to fetch the server's public
-        # rooms directory via federation. Defaults to 'true'.
+        # If set to 'true', allows any other homeserver to fetch the server's public
+        # rooms directory via federation. Defaults to 'false'.
         #
-        #allow_public_rooms_over_federation: false
+        #allow_public_rooms_over_federation: true
 
         # The default room version for newly created rooms.
         #
@@ -721,7 +843,7 @@ class ServerConfig(Config):
         # Used by phonehome stats to group together related servers.
         #server_context: context
 
-        # Resource-constrained Homeserver Settings
+        # Resource-constrained homeserver Settings
         #
         # If limit_remote_rooms.enabled is True, the room complexity will be
         # checked before a user joins a new remote room. If it is above
@@ -761,6 +883,69 @@ class ServerConfig(Config):
         # Defaults to `28d`. Set to `null` to disable clearing out of old rows.
         #
         #user_ips_max_age: 14d
+
+        # Message retention policy at the server level.
+        #
+        # Room admins and mods can define a retention period for their rooms using the
+        # 'm.room.retention' state event, and server admins can cap this period by setting
+        # the 'allowed_lifetime_min' and 'allowed_lifetime_max' config options.
+        #
+        # If this feature is enabled, Synapse will regularly look for and purge events
+        # which are older than the room's maximum retention period. Synapse will also
+        # filter events received over federation so that events that should have been
+        # purged are ignored and not stored again.
+        #
+        retention:
+          # The message retention policies feature is disabled by default. Uncomment the
+          # following line to enable it.
+          #
+          #enabled: true
+
+          # Default retention policy. If set, Synapse will apply it to rooms that lack the
+          # 'm.room.retention' state event. Currently, the value of 'min_lifetime' doesn't
+          # matter much because Synapse doesn't take it into account yet.
+          #
+          #default_policy:
+          #  min_lifetime: 1d
+          #  max_lifetime: 1y
+
+          # Retention policy limits. If set, a user won't be able to send a
+          # 'm.room.retention' event which features a 'min_lifetime' or a 'max_lifetime'
+          # that's not within this range. This is especially useful in closed federations,
+          # in which server admins can make sure every federating server applies the same
+          # rules.
+          #
+          #allowed_lifetime_min: 1d
+          #allowed_lifetime_max: 1y
+
+          # Server admins can define the settings of the background jobs purging the
+          # events which lifetime has expired under the 'purge_jobs' section.
+          #
+          # If no configuration is provided, a single job will be set up to delete expired
+          # events in every room daily.
+          #
+          # Each job's configuration defines which range of message lifetimes the job
+          # takes care of. For example, if 'shortest_max_lifetime' is '2d' and
+          # 'longest_max_lifetime' is '3d', the job will handle purging expired events in
+          # rooms whose state defines a 'max_lifetime' that's both higher than 2 days, and
+          # lower than or equal to 3 days. Both the minimum and the maximum value of a
+          # range are optional, e.g. a job with no 'shortest_max_lifetime' and a
+          # 'longest_max_lifetime' of '3d' will handle every room with a retention policy
+          # which 'max_lifetime' is lower than or equal to three days.
+          #
+          # The rationale for this per-job configuration is that some rooms might have a
+          # retention policy with a low 'max_lifetime', where history needs to be purged
+          # of outdated messages on a very frequent basis (e.g. every 5min), but not want
+          # that purge to be performed by a job that's iterating over every room it knows,
+          # which would be quite heavy on the server.
+          #
+          #purge_jobs:
+          #  - shortest_max_lifetime: 1d
+          #    longest_max_lifetime: 3d
+          #    interval: 5m:
+          #  - shortest_max_lifetime: 3d
+          #    longest_max_lifetime: 1y
+          #    interval: 24h
         """
             % locals()
         )
@@ -781,20 +966,20 @@ class ServerConfig(Config):
             "--daemonize",
             action="store_true",
             default=None,
-            help="Daemonize the home server",
+            help="Daemonize the homeserver",
         )
         server_group.add_argument(
             "--print-pidfile",
             action="store_true",
             default=None,
-            help="Print the path to the pidfile just" " before daemonizing",
+            help="Print the path to the pidfile just before daemonizing",
         )
         server_group.add_argument(
             "--manhole",
             metavar="PORT",
             dest="manhole",
             type=int,
-            help="Turn on the twisted telnet manhole" " service on the given port.",
+            help="Turn on the twisted telnet manhole service on the given port.",
         )
 
 

synapse/crypto/event_signing.py

@@ -125,9 +125,11 @@ def compute_event_signature(event_dict, signature_name, signing_key):
     redact_json = prune_event_dict(event_dict)
     redact_json.pop("age_ts", None)
     redact_json.pop("unsigned", None)
-    logger.debug("Signing event: %s", encode_canonical_json(redact_json))
+    if logger.isEnabledFor(logging.DEBUG):
+        logger.debug("Signing event: %s", encode_canonical_json(redact_json))
     redact_json = sign_json(redact_json, signature_name, signing_key)
-    logger.debug("Signed event: %s", encode_canonical_json(redact_json))
+    if logger.isEnabledFor(logging.DEBUG):
+        logger.debug("Signed event: %s", encode_canonical_json(redact_json))
     return redact_json["signatures"]
 
 

synapse/event_auth.py

@@ -42,6 +42,8 @@ def check(room_version, event, auth_events, do_sig_check=True, do_size_check=Tru
     Returns:
          if the auth checks pass.
     """
+    assert isinstance(auth_events, dict)
+
     if do_size_check:
         _check_size_limits(event)
 
@@ -74,12 +76,6 @@ def check(room_version, event, auth_events, do_sig_check=True, do_size_check=Tru
             if not event.signatures.get(event_id_domain):
                 raise AuthError(403, "Event not signed by sending server")
 
-    if auth_events is None:
-        # Oh, we don't know what the state of the room was, so we
-        # are trusting that this is allowed (at least for now)
-        logger.warn("Trusting event: %s", event.event_id)
-        return
-
     if event.type == EventTypes.Create:
         sender_domain = get_domain_from_id(event.sender)
         room_id_domain = get_domain_from_id(event.room_id)

synapse/events/snapshot.py

@@ -12,104 +12,125 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
+from typing import Dict, Optional, Tuple, Union
 
 from six import iteritems
 
+import attr
 from frozendict import frozendict
 
 from twisted.internet import defer
 
+from synapse.appservice import ApplicationService
 from synapse.logging.context import make_deferred_yieldable, run_in_background
 
 
-class EventContext(object):
+@attr.s(slots=True)
+class EventContext:
     """
+    Holds information relevant to persisting an event
+
     Attributes:
-        state_group (int|None): state group id, if the state has been stored
-            as a state group. This is usually only None if e.g. the event is
-            an outlier.
-        rejected (bool|str): A rejection reason if the event was rejected, else
-            False
+        rejected: A rejection reason if the event was rejected, else False
 
-        push_actions (list[(str, list[object])]): list of (user_id, actions)
-            tuples
+        _state_group: The ID of the state group for this event. Note that state events
+            are persisted with a state group which includes the new event, so this is
+            effectively the state *after* the event in question.
 
-        prev_group (int): Previously persisted state group. ``None`` for an
-            outlier.
-        delta_ids (dict[(str, str), str]): Delta from ``prev_group``.
-            (type, state_key) -> event_id. ``None`` for an outlier.
+            For a *rejected* state event, where the state of the rejected event is
+            ignored, this state_group should never make it into the
+            event_to_state_groups table. Indeed, inspecting this value for a rejected
+            state event is almost certainly incorrect.
 
-        prev_state_events (?): XXX: is this ever set to anything other than
-            the empty list?
+            For an outlier, where we don't have the state at the event, this will be
+            None.
+
+            Note that this is a private attribute: it should be accessed via
+            the ``state_group`` property.
+
+        state_group_before_event: The ID of the state group representing the state
+            of the room before this event.
+
+            If this is a non-state event, this will be the same as ``state_group``. If
+            it's a state event, it will be the same as ``prev_group``.
+
+            If ``state_group`` is None (ie, the event is an outlier),
+            ``state_group_before_event`` will always also be ``None``.
+
+        prev_group: If it is known, ``state_group``'s prev_group. Note that this being
+            None does not necessarily mean that ``state_group`` does not have
+            a prev_group!
+
+            If the event is a state event, this is normally the same as ``prev_group``.
+
+            If ``state_group`` is None (ie, the event is an outlier), ``prev_group``
+            will always also be ``None``.
+
+            Note that this *not* (necessarily) the state group associated with
+            ``_prev_state_ids``.
+
+        delta_ids: If ``prev_group`` is not None, the state delta between ``prev_group``
+            and ``state_group``.
+
+        app_service: If this event is being sent by a (local) application service, that
+            app service.
+
+        _current_state_ids: The room state map, including this event - ie, the state
+            in ``state_group``.
 
-        _current_state_ids (dict[(str, str), str]|None):
-            The current state map including the current event. None if outlier
-            or we haven't fetched the state from DB yet.
             (type, state_key) -> event_id
 
-        _prev_state_ids (dict[(str, str), str]|None):
-            The current state map excluding the current event. None if outlier
-            or we haven't fetched the state from DB yet.
+            FIXME: what is this for an outlier? it seems ill-defined. It seems like
+            it could be either {}, or the state we were given by the remote
+            server, depending on $THINGS
+
+            Note that this is a private attribute: it should be accessed via
+            ``get_current_state_ids``. _AsyncEventContext impl calculates this
+            on-demand: it will be None until that happens.
+
+        _prev_state_ids: The room state map, excluding this event - ie, the state
+            in ``state_group_before_event``. For a non-state
+            event, this will be the same as _current_state_events.
+
+            Note that it is a completely different thing to prev_group!
+
             (type, state_key) -> event_id
 
-        _fetching_state_deferred (Deferred|None): Resolves when *_state_ids have
-            been calculated. None if we haven't started calculating yet
+            FIXME: again, what is this for an outlier?
 
-        _event_type (str): The type of the event the context is associated with.
-            Only set when state has not been fetched yet.
-
-        _event_state_key (str|None): The state_key of the event the context is
-            associated with. Only set when state has not been fetched yet.
-
-        _prev_state_id (str|None): If the event associated with the context is
-            a state event, then `_prev_state_id` is the event_id of the state
-            that was replaced.
-            Only set when state has not been fetched yet.
+            As with _current_state_ids, this is a private attribute. It should be
+            accessed via get_prev_state_ids.
     """
 
-    __slots__ = [
-        "state_group",
-        "rejected",
-        "prev_group",
-        "delta_ids",
-        "prev_state_events",
-        "app_service",
-        "_current_state_ids",
-        "_prev_state_ids",
-        "_prev_state_id",
-        "_event_type",
-        "_event_state_key",
-        "_fetching_state_deferred",
-    ]
+    rejected = attr.ib(default=False, type=Union[bool, str])
+    _state_group = attr.ib(default=None, type=Optional[int])
+    state_group_before_event = attr.ib(default=None, type=Optional[int])
+    prev_group = attr.ib(default=None, type=Optional[int])
+    delta_ids = attr.ib(default=None, type=Optional[Dict[Tuple[str, str], str]])
+    app_service = attr.ib(default=None, type=Optional[ApplicationService])
 
-    def __init__(self):
-        self.prev_state_events = []
-        self.rejected = False
-        self.app_service = None
+    _current_state_ids = attr.ib(
+        default=None, type=Optional[Dict[Tuple[str, str], str]]
+    )
+    _prev_state_ids = attr.ib(default=None, type=Optional[Dict[Tuple[str, str], str]])
 
     @staticmethod
     def with_state(
-        state_group, current_state_ids, prev_state_ids, prev_group=None, delta_ids=None
+        state_group,
+        state_group_before_event,
+        current_state_ids,
+        prev_state_ids,
+        prev_group=None,
+        delta_ids=None,
     ):
-        context = EventContext()
-
-        # The current state including the current event
-        context._current_state_ids = current_state_ids
-        # The current state excluding the current event
-        context._prev_state_ids = prev_state_ids
-        context.state_group = state_group
-
-        context._prev_state_id = None
-        context._event_type = None
-        context._event_state_key = None
-        context._fetching_state_deferred = defer.succeed(None)
-
-        # A previously persisted state group and a delta between that
-        # and this state.
-        context.prev_group = prev_group
-        context.delta_ids = delta_ids
-
-        return context
+        return EventContext(
+            current_state_ids=current_state_ids,
+            prev_state_ids=prev_state_ids,
+            state_group=state_group,
+            state_group_before_event=state_group_before_event,
+            prev_group=prev_group,
+            delta_ids=delta_ids,
+        )
 
     @defer.inlineCallbacks
     def serialize(self, event, store):
@@ -137,11 +158,11 @@ class EventContext(object):
             "prev_state_id": prev_state_id,
             "event_type": event.type,
             "event_state_key": event.state_key if event.is_state() else None,
-            "state_group": self.state_group,
+            "state_group": self._state_group,
+            "state_group_before_event": self.state_group_before_event,
             "rejected": self.rejected,
             "prev_group": self.prev_group,
             "delta_ids": _encode_state_dict(self.delta_ids),
-            "prev_state_events": self.prev_state_events,
             "app_service_id": self.app_service.id if self.app_service else None,
         }
 
@@ -157,24 +178,18 @@ class EventContext(object):
         Returns:
             EventContext
         """
-        context = EventContext()
-
-        # We use the state_group and prev_state_id stuff to pull the
-        # current_state_ids out of the DB and construct prev_state_ids.
-        context._prev_state_id = input["prev_state_id"]
-        context._event_type = input["event_type"]
-        context._event_state_key = input["event_state_key"]
-
-        context._current_state_ids = None
-        context._prev_state_ids = None
-        context._fetching_state_deferred = None
-
-        context.state_group = input["state_group"]
-        context.prev_group = input["prev_group"]
-        context.delta_ids = _decode_state_dict(input["delta_ids"])
-
-        context.rejected = input["rejected"]
-        context.prev_state_events = input["prev_state_events"]
+        context = _AsyncEventContextImpl(
+            # We use the state_group and prev_state_id stuff to pull the
+            # current_state_ids out of the DB and construct prev_state_ids.
+            prev_state_id=input["prev_state_id"],
+            event_type=input["event_type"],
+            event_state_key=input["event_state_key"],
+            state_group=input["state_group"],
+            state_group_before_event=input["state_group_before_event"],
+            prev_group=input["prev_group"],
+            delta_ids=_decode_state_dict(input["delta_ids"]),
+            rejected=input["rejected"],
+        )
 
         app_service_id = input["app_service_id"]
         if app_service_id:
@@ -182,29 +197,52 @@ class EventContext(object):
 
         return context
 
+    @property
+    def state_group(self) -> Optional[int]:
+        """The ID of the state group for this event.
+
+        Note that state events are persisted with a state group which includes the new
+        event, so this is effectively the state *after* the event in question.
+
+        For an outlier, where we don't have the state at the event, this will be None.
+
+        It is an error to access this for a rejected event, since rejected state should
+        not make it into the room state. Accessing this property will raise an exception
+        if ``rejected`` is set.
+        """
+        if self.rejected:
+            raise RuntimeError("Attempt to access state_group of rejected event")
+
+        return self._state_group
+
     @defer.inlineCallbacks
     def get_current_state_ids(self, store):
-        """Gets the current state IDs
+        """
+        Gets the room state map, including this event - ie, the state in ``state_group``
+
+        It is an error to access this for a rejected event, since rejected state should
+        not make it into the room state. This method will raise an exception if
+        ``rejected`` is set.
 
         Returns:
             Deferred[dict[(str, str), str]|None]: Returns None if state_group
                 is None, which happens when the associated event is an outlier.
+
                 Maps a (type, state_key) to the event ID of the state event matching
                 this tuple.
         """
+        if self.rejected:
+            raise RuntimeError("Attempt to access state_ids of rejected event")
 
-        if not self._fetching_state_deferred:
-            self._fetching_state_deferred = run_in_background(
-                self._fill_out_state, store
-            )
-
-        yield make_deferred_yieldable(self._fetching_state_deferred)
-
+        yield self._ensure_fetched(store)
         return self._current_state_ids
 
     @defer.inlineCallbacks
     def get_prev_state_ids(self, store):
-        """Gets the prev state IDs
+        """
+        Gets the room state map, excluding this event.
+
+        For a non-state event, this will be the same as get_current_state_ids().
 
         Returns:
             Deferred[dict[(str, str), str]|None]: Returns None if state_group
@@ -212,27 +250,64 @@ class EventContext(object):
                 Maps a (type, state_key) to the event ID of the state event matching
                 this tuple.
         """
-
-        if not self._fetching_state_deferred:
-            self._fetching_state_deferred = run_in_background(
-                self._fill_out_state, store
-            )
-
-        yield make_deferred_yieldable(self._fetching_state_deferred)
-
+        yield self._ensure_fetched(store)
         return self._prev_state_ids
 
     def get_cached_current_state_ids(self):
         """Gets the current state IDs if we have them already cached.
 
+        It is an error to access this for a rejected event, since rejected state should
+        not make it into the room state. This method will raise an exception if
+        ``rejected`` is set.
+
         Returns:
             dict[(str, str), str]|None: Returns None if we haven't cached the
             state or if state_group is None, which happens when the associated
             event is an outlier.
         """
+        if self.rejected:
+            raise RuntimeError("Attempt to access state_ids of rejected event")
 
         return self._current_state_ids
 
+    def _ensure_fetched(self, store):
+        return defer.succeed(None)
+
+
+@attr.s(slots=True)
+class _AsyncEventContextImpl(EventContext):
+    """
+    An implementation of EventContext which fetches _current_state_ids and
+    _prev_state_ids from the database on demand.
+
+    Attributes:
+
+        _fetching_state_deferred (Deferred|None): Resolves when *_state_ids have
+            been calculated. None if we haven't started calculating yet
+
+        _event_type (str): The type of the event the context is associated with.
+
+        _event_state_key (str): The state_key of the event the context is
+            associated with.
+
+        _prev_state_id (str|None): If the event associated with the context is
+            a state event, then `_prev_state_id` is the event_id of the state
+            that was replaced.
+    """
+
+    _prev_state_id = attr.ib(default=None)
+    _event_type = attr.ib(default=None)
+    _event_state_key = attr.ib(default=None)
+    _fetching_state_deferred = attr.ib(default=None)
+
+    def _ensure_fetched(self, store):
+        if not self._fetching_state_deferred:
+            self._fetching_state_deferred = run_in_background(
+                self._fill_out_state, store
+            )
+
+        return make_deferred_yieldable(self._fetching_state_deferred)
+
     @defer.inlineCallbacks
     def _fill_out_state(self, store):
         """Called to populate the _current_state_ids and _prev_state_ids
@@ -250,27 +325,6 @@ class EventContext(object):
         else:
             self._prev_state_ids = self._current_state_ids
 
-    @defer.inlineCallbacks
-    def update_state(
-        self, state_group, prev_state_ids, current_state_ids, prev_group, delta_ids
-    ):
-        """Replace the state in the context
-        """
-
-        # We need to make sure we wait for any ongoing fetching of state
-        # to complete so that the updated state doesn't get clobbered
-        if self._fetching_state_deferred:
-            yield make_deferred_yieldable(self._fetching_state_deferred)
-
-        self.state_group = state_group
-        self._prev_state_ids = prev_state_ids
-        self.prev_group = prev_group
-        self._current_state_ids = current_state_ids
-        self.delta_ids = delta_ids
-
-        # We need to ensure that that we've marked as having fetched the state
-        self._fetching_state_deferred = defer.succeed(None)
-
 
 def _encode_state_dict(state_dict):
     """Since dicts of (type, state_key) -> event_id cannot be serialized in

synapse/events/spamcheck.py

@@ -1,5 +1,6 @@
 # -*- coding: utf-8 -*-
 # Copyright 2017 New Vector Ltd
+# Copyright 2019 The Matrix.org Foundation C.I.C.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -13,6 +14,10 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+import inspect
+
+from synapse.spam_checker_api import SpamCheckerApi
+
 
 class SpamChecker(object):
     def __init__(self, hs):
@@ -26,7 +31,14 @@ class SpamChecker(object):
             pass
 
         if module is not None:
-            self.spam_checker = module(config=config)
+            # Older spam checkers don't accept the `api` argument, so we
+            # try and detect support.
+            spam_args = inspect.getfullargspec(module)
+            if "api" in spam_args.args:
+                api = SpamCheckerApi(hs)
+                self.spam_checker = module(config=config, api=api)
+            else:
+                self.spam_checker = module(config=config)
 
     def check_event_for_spam(self, event):
         """Checks if a given event is considered "spammy" by this server.

synapse/events/validator.py

@@ -13,7 +13,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-from six import string_types
+from six import integer_types, string_types
 
 from synapse.api.constants import MAX_ALIAS_LENGTH, EventTypes, Membership
 from synapse.api.errors import Codes, SynapseError
@@ -22,11 +22,12 @@ from synapse.types import EventID, RoomID, UserID
 
 
 class EventValidator(object):
-    def validate_new(self, event):
+    def validate_new(self, event, config):
         """Validates the event has roughly the right format
 
         Args:
-            event (FrozenEvent)
+            event (FrozenEvent): The event to validate.
+            config (Config): The homeserver's configuration.
         """
         self.validate_builder(event)
 
@@ -67,6 +68,99 @@ class EventValidator(object):
                             Codes.INVALID_PARAM,
                         )
 
+        if event.type == EventTypes.Retention:
+            self._validate_retention(event, config)
+
+    def _validate_retention(self, event, config):
+        """Checks that an event that defines the retention policy for a room respects the
+        boundaries imposed by the server's administrator.
+
+        Args:
+            event (FrozenEvent): The event to validate.
+            config (Config): The homeserver's configuration.
+        """
+        min_lifetime = event.content.get("min_lifetime")
+        max_lifetime = event.content.get("max_lifetime")
+
+        if min_lifetime is not None:
+            if not isinstance(min_lifetime, integer_types):
+                raise SynapseError(
+                    code=400,
+                    msg="'min_lifetime' must be an integer",
+                    errcode=Codes.BAD_JSON,
+                )
+
+            if (
+                config.retention_allowed_lifetime_min is not None
+                and min_lifetime < config.retention_allowed_lifetime_min
+            ):
+                raise SynapseError(
+                    code=400,
+                    msg=(
+                        "'min_lifetime' can't be lower than the minimum allowed"
+                        " value enforced by the server's administrator"
+                    ),
+                    errcode=Codes.BAD_JSON,
+                )
+
+            if (
+                config.retention_allowed_lifetime_max is not None
+                and min_lifetime > config.retention_allowed_lifetime_max
+            ):
+                raise SynapseError(
+                    code=400,
+                    msg=(
+                        "'min_lifetime' can't be greater than the maximum allowed"
+                        " value enforced by the server's administrator"
+                    ),
+                    errcode=Codes.BAD_JSON,
+                )
+
+        if max_lifetime is not None:
+            if not isinstance(max_lifetime, integer_types):
+                raise SynapseError(
+                    code=400,
+                    msg="'max_lifetime' must be an integer",
+                    errcode=Codes.BAD_JSON,
+                )
+
+            if (
+                config.retention_allowed_lifetime_min is not None
+                and max_lifetime < config.retention_allowed_lifetime_min
+            ):
+                raise SynapseError(
+                    code=400,
+                    msg=(
+                        "'max_lifetime' can't be lower than the minimum allowed value"
+                        " enforced by the server's administrator"
+                    ),
+                    errcode=Codes.BAD_JSON,
+                )
+
+            if (
+                config.retention_allowed_lifetime_max is not None
+                and max_lifetime > config.retention_allowed_lifetime_max
+            ):
+                raise SynapseError(
+                    code=400,
+                    msg=(
+                        "'max_lifetime' can't be greater than the maximum allowed"
+                        " value enforced by the server's administrator"
+                    ),
+                    errcode=Codes.BAD_JSON,
+                )
+
+        if (
+            min_lifetime is not None
+            and max_lifetime is not None
+            and min_lifetime > max_lifetime
+        ):
+            raise SynapseError(
+                code=400,
+                msg="'min_lifetime' can't be greater than 'max_lifetime",
+                errcode=Codes.BAD_JSON,
+            )
+
     def validate_builder(self, event):
         """Validates that the builder/event has roughly the right format. Only
         checks values that we expect a proto event to have, rather than all the

synapse/federation/federation_base.py

@@ -102,7 +102,7 @@ class FederationBase(object):
                     pass
 
             if not res:
-                logger.warn(
+                logger.warning(
                     "Failed to find copy of %s with valid signature", pdu.event_id
                 )
 
@@ -173,7 +173,7 @@ class FederationBase(object):
                     return redacted_event
 
                 if self.spam_checker.check_event_for_spam(pdu):
-                    logger.warn(
+                    logger.warning(
                         "Event contains spam, redacting %s: %s",
                         pdu.event_id,
                         pdu.get_pdu_json(),
@@ -185,7 +185,7 @@ class FederationBase(object):
         def errback(failure, pdu):
             failure.trap(SynapseError)
             with PreserveLoggingContext(ctx):
-                logger.warn(
+                logger.warning(
                     "Signature check failed for %s: %s",
                     pdu.event_id,
                     failure.getErrorMessage(),

synapse/federation/federation_client.py

@@ -18,8 +18,6 @@ import copy
 import itertools
 import logging
 
-from six.moves import range
-
 from prometheus_client import Counter
 
 from twisted.internet import defer
@@ -39,7 +37,7 @@ from synapse.api.room_versions import (
 )
 from synapse.events import builder, room_version_to_event_format
 from synapse.federation.federation_base import FederationBase, event_from_pdu_json
-from synapse.logging.context import make_deferred_yieldable, run_in_background
+from synapse.logging.context import make_deferred_yieldable
 from synapse.logging.utils import log_function
 from synapse.util import unwrapFirstError
 from synapse.util.caches.expiringcache import ExpiringCache
@@ -177,7 +175,7 @@ class FederationClient(FederationBase):
         given destination server.
 
         Args:
-            dest (str): The remote home server to ask.
+            dest (str): The remote homeserver to ask.
             room_id (str): The room_id to backfill.
             limit (int): The maximum number of PDUs to return.
             extremities (list): List of PDU id and origins of the first pdus
@@ -196,7 +194,7 @@ class FederationClient(FederationBase):
             dest, room_id, extremities, limit
         )
 
-        logger.debug("backfill transaction_data=%s", repr(transaction_data))
+        logger.debug("backfill transaction_data=%r", transaction_data)
 
         room_version = yield self.store.get_room_version(room_id)
         format_ver = room_version_to_event_format(room_version)
@@ -227,7 +225,7 @@ class FederationClient(FederationBase):
         one succeeds.
 
         Args:
-            destinations (list): Which home servers to query
+            destinations (list): Which homeservers to query
             event_id (str): event to fetch
             room_version (str): version of the room
             outlier (bool): Indicates whether the PDU is an `outlier`, i.e. if
@@ -310,162 +308,26 @@ class FederationClient(FederationBase):
         return signed_pdu
 
     @defer.inlineCallbacks
-    @log_function
-    def get_state_for_room(self, destination, room_id, event_id):
-        """Requests all of the room state at a given event from a remote home server.
-
-        Args:
-            destination (str): The remote homeserver to query for the state.
-            room_id (str): The id of the room we're interested in.
-            event_id (str): The id of the event we want the state at.
+    def get_room_state_ids(self, destination: str, room_id: str, event_id: str):
+        """Calls the /state_ids endpoint to fetch the state at a particular point
+        in the room, and the auth events for the given event
 
         Returns:
-            Deferred[Tuple[List[EventBase], List[EventBase]]]:
-                A list of events in the state, and a list of events in the auth chain
-                for the given event.
+            Tuple[List[str], List[str]]:  a tuple of (state event_ids, auth event_ids)
         """
-        try:
-            # First we try and ask for just the IDs, as thats far quicker if
-            # we have most of the state and auth_chain already.
-            # However, this may 404 if the other side has an old synapse.
-            result = yield self.transport_layer.get_room_state_ids(
-                destination, room_id, event_id=event_id
-            )
-
-            state_event_ids = result["pdu_ids"]
-            auth_event_ids = result.get("auth_chain_ids", [])
-
-            fetched_events, failed_to_fetch = yield self.get_events_from_store_or_dest(
-                destination, room_id, set(state_event_ids + auth_event_ids)
-            )
-
-            if failed_to_fetch:
-                logger.warning(
-                    "Failed to fetch missing state/auth events for %s: %s",
-                    room_id,
-                    failed_to_fetch,
-                )
-
-            event_map = {ev.event_id: ev for ev in fetched_events}
-
-            pdus = [event_map[e_id] for e_id in state_event_ids if e_id in event_map]
-            auth_chain = [
-                event_map[e_id] for e_id in auth_event_ids if e_id in event_map
-            ]
-
-            auth_chain.sort(key=lambda e: e.depth)
-
-            return pdus, auth_chain
-        except HttpResponseException as e:
-            if e.code == 400 or e.code == 404:
-                logger.info("Failed to use get_room_state_ids API, falling back")
-            else:
-                raise e
-
-        result = yield self.transport_layer.get_room_state(
+        result = yield self.transport_layer.get_room_state_ids(
             destination, room_id, event_id=event_id
         )
 
-        room_version = yield self.store.get_room_version(room_id)
-        format_ver = room_version_to_event_format(room_version)
+        state_event_ids = result["pdu_ids"]
+        auth_event_ids = result.get("auth_chain_ids", [])
 
-        pdus = [
-            event_from_pdu_json(p, format_ver, outlier=True) for p in result["pdus"]
-        ]
+        if not isinstance(state_event_ids, list) or not isinstance(
+            auth_event_ids, list
+        ):
+            raise Exception("invalid response from /state_ids")
 
-        auth_chain = [
-            event_from_pdu_json(p, format_ver, outlier=True)
-            for p in result.get("auth_chain", [])
-        ]
-
-        seen_events = yield self.store.get_events(
-            [ev.event_id for ev in itertools.chain(pdus, auth_chain)]
-        )
-
-        signed_pdus = yield self._check_sigs_and_hash_and_fetch(
-            destination,
-            [p for p in pdus if p.event_id not in seen_events],
-            outlier=True,
-            room_version=room_version,
-        )
-        signed_pdus.extend(
-            seen_events[p.event_id] for p in pdus if p.event_id in seen_events
-        )
-
-        signed_auth = yield self._check_sigs_and_hash_and_fetch(
-            destination,
-            [p for p in auth_chain if p.event_id not in seen_events],
-            outlier=True,
-            room_version=room_version,
-        )
-        signed_auth.extend(
-            seen_events[p.event_id] for p in auth_chain if p.event_id in seen_events
-        )
-
-        signed_auth.sort(key=lambda e: e.depth)
-
-        return signed_pdus, signed_auth
-
-    @defer.inlineCallbacks
-    def get_events_from_store_or_dest(self, destination, room_id, event_ids):
-        """Fetch events from a remote destination, checking if we already have them.
-
-        Args:
-            destination (str)
-            room_id (str)
-            event_ids (list)
-
-        Returns:
-            Deferred: A deferred resolving to a 2-tuple where the first is a list of
-            events and the second is a list of event ids that we failed to fetch.
-        """
-        seen_events = yield self.store.get_events(event_ids, allow_rejected=True)
-        signed_events = list(seen_events.values())
-
-        failed_to_fetch = set()
-
-        missing_events = set(event_ids)
-        for k in seen_events:
-            missing_events.discard(k)
-
-        if not missing_events:
-            return signed_events, failed_to_fetch
-
-        logger.debug(
-            "Fetching unknown state/auth events %s for room %s",
-            missing_events,
-            event_ids,
-        )
-
-        room_version = yield self.store.get_room_version(room_id)
-
-        batch_size = 20
-        missing_events = list(missing_events)
-        for i in range(0, len(missing_events), batch_size):
-            batch = set(missing_events[i : i + batch_size])
-
-            deferreds = [
-                run_in_background(
-                    self.get_pdu,
-                    destinations=[destination],
-                    event_id=e_id,
-                    room_version=room_version,
-                )
-                for e_id in batch
-            ]
-
-            res = yield make_deferred_yieldable(
-                defer.DeferredList(deferreds, consumeErrors=True)
-            )
-            for success, result in res:
-                if success and result:
-                    signed_events.append(result)
-                    batch.discard(result.event_id)
-
-            # We removed all events we successfully fetched from `batch`
-            failed_to_fetch.update(batch)
-
-        return signed_events, failed_to_fetch
+        return state_event_ids, auth_event_ids
 
     @defer.inlineCallbacks
     @log_function
@@ -522,12 +384,12 @@ class FederationClient(FederationBase):
                 res = yield callback(destination)
                 return res
             except InvalidResponseError as e:
-                logger.warn("Failed to %s via %s: %s", description, destination, e)
+                logger.warning("Failed to %s via %s: %s", description, destination, e)
             except HttpResponseException as e:
                 if not 500 <= e.code < 600:
                     raise e.to_synapse_error()
                 else:
-                    logger.warn(
+                    logger.warning(
                         "Failed to %s via %s: %i %s",
                         description,
                         destination,
@@ -535,7 +397,9 @@ class FederationClient(FederationBase):
                         e.args[0],
                     )
             except Exception:
-                logger.warn("Failed to %s via %s", description, destination, exc_info=1)
+                logger.warning(
+                    "Failed to %s via %s", description, destination, exc_info=1
+                )
 
         raise SynapseError(502, "Failed to %s via any server" % (description,))
 
@@ -553,7 +417,7 @@ class FederationClient(FederationBase):
         Note that this does not append any events to any graphs.
 
         Args:
-            destinations (str): Candidate homeservers which are probably
+            destinations (Iterable[str]): Candidate homeservers which are probably
                 participating in the room.
             room_id (str): The room in which the event will happen.
             user_id (str): The user whose membership is being evented.
@@ -662,13 +526,7 @@ class FederationClient(FederationBase):
 
         @defer.inlineCallbacks
         def send_request(destination):
-            time_now = self._clock.time_msec()
-            _, content = yield self.transport_layer.send_join(
-                destination=destination,
-                room_id=pdu.room_id,
-                event_id=pdu.event_id,
-                content=pdu.get_pdu_json(time_now),
-            )
+            content = yield self._do_send_join(destination, pdu)
 
             logger.debug("Got content: %s", content)
 
@@ -735,6 +593,44 @@ class FederationClient(FederationBase):
 
         return self._try_destination_list("send_join", destinations, send_request)
 
+    @defer.inlineCallbacks
+    def _do_send_join(self, destination, pdu):
+        time_now = self._clock.time_msec()
+
+        try:
+            content = yield self.transport_layer.send_join_v2(
+                destination=destination,
+                room_id=pdu.room_id,
+                event_id=pdu.event_id,
+                content=pdu.get_pdu_json(time_now),
+            )
+
+            return content
+        except HttpResponseException as e:
+            if e.code in [400, 404]:
+                err = e.to_synapse_error()
+
+                # If we receive an error response that isn't a generic error, or an
+                # unrecognised endpoint error, we  assume that the remote understands
+                # the v2 invite API and this is a legitimate error.
+                if err.errcode not in [Codes.UNKNOWN, Codes.UNRECOGNIZED]:
+                    raise err
+            else:
+                raise e.to_synapse_error()
+
+        logger.debug("Couldn't send_join with the v2 API, falling back to the v1 API")
+
+        resp = yield self.transport_layer.send_join_v1(
+            destination=destination,
+            room_id=pdu.room_id,
+            event_id=pdu.event_id,
+            content=pdu.get_pdu_json(time_now),
+        )
+
+        # We expect the v1 API to respond with [200, content], so we only return the
+        # content.
+        return resp[1]
+
     @defer.inlineCallbacks
     def send_invite(self, destination, room_id, event_id, pdu):
         room_version = yield self.store.get_room_version(room_id)
@@ -844,18 +740,50 @@ class FederationClient(FederationBase):
 
         @defer.inlineCallbacks
         def send_request(destination):
-            time_now = self._clock.time_msec()
-            _, content = yield self.transport_layer.send_leave(
+            content = yield self._do_send_leave(destination, pdu)
+
+            logger.debug("Got content: %s", content)
+            return None
+
+        return self._try_destination_list("send_leave", destinations, send_request)
+
+    @defer.inlineCallbacks
+    def _do_send_leave(self, destination, pdu):
+        time_now = self._clock.time_msec()
+
+        try:
+            content = yield self.transport_layer.send_leave_v2(
                 destination=destination,
                 room_id=pdu.room_id,
                 event_id=pdu.event_id,
                 content=pdu.get_pdu_json(time_now),
             )
 
-            logger.debug("Got content: %s", content)
-            return None
+            return content
+        except HttpResponseException as e:
+            if e.code in [400, 404]:
+                err = e.to_synapse_error()
 
-        return self._try_destination_list("send_leave", destinations, send_request)
+                # If we receive an error response that isn't a generic error, or an
+                # unrecognised endpoint error, we  assume that the remote understands
+                # the v2 invite API and this is a legitimate error.
+                if err.errcode not in [Codes.UNKNOWN, Codes.UNRECOGNIZED]:
+                    raise err
+            else:
+                raise e.to_synapse_error()
+
+        logger.debug("Couldn't send_leave with the v2 API, falling back to the v1 API")
+
+        resp = yield self.transport_layer.send_leave_v1(
+            destination=destination,
+            room_id=pdu.room_id,
+            event_id=pdu.event_id,
+            content=pdu.get_pdu_json(time_now),
+        )
+
+        # We expect the v1 API to respond with [200, content], so we only return the
+        # content.
+        return resp[1]
 
     def get_public_rooms(
         self,

synapse/federation/federation_server.py

@@ -1,6 +1,7 @@
 # -*- coding: utf-8 -*-
 # Copyright 2015, 2016 OpenMarket Ltd
 # Copyright 2018 New Vector Ltd
+# Copyright 2019 Matrix.org Federation C.I.C
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -21,7 +22,6 @@ from six import iteritems
 from canonicaljson import json
 from prometheus_client import Counter
 
-from twisted.internet import defer
 from twisted.internet.abstract import isIPAddress
 from twisted.python import failure
 
@@ -74,6 +74,7 @@ class FederationServer(FederationBase):
 
         self.auth = hs.get_auth()
         self.handler = hs.get_handlers().federation_handler
+        self.state = hs.get_state_handler()
 
         self._server_linearizer = Linearizer("fed_server")
         self._transaction_linearizer = Linearizer("fed_txn_handler")
@@ -86,14 +87,12 @@ class FederationServer(FederationBase):
         # come in waves.
         self._state_resp_cache = ResponseCache(hs, "state_resp", timeout_ms=30000)
 
-    @defer.inlineCallbacks
-    @log_function
-    def on_backfill_request(self, origin, room_id, versions, limit):
-        with (yield self._server_linearizer.queue((origin, room_id))):
+    async def on_backfill_request(self, origin, room_id, versions, limit):
+        with (await self._server_linearizer.queue((origin, room_id))):
             origin_host, _ = parse_server_name(origin)
-            yield self.check_server_matches_acl(origin_host, room_id)
+            await self.check_server_matches_acl(origin_host, room_id)
 
-            pdus = yield self.handler.on_backfill_request(
+            pdus = await self.handler.on_backfill_request(
                 origin, room_id, versions, limit
             )
 
@@ -101,9 +100,7 @@ class FederationServer(FederationBase):
 
         return 200, res
 
-    @defer.inlineCallbacks
-    @log_function
-    def on_incoming_transaction(self, origin, transaction_data):
+    async def on_incoming_transaction(self, origin, transaction_data):
         # keep this as early as possible to make the calculated origin ts as
         # accurate as possible.
         request_time = self._clock.time_msec()
@@ -118,18 +115,17 @@ class FederationServer(FederationBase):
         # use a linearizer to ensure that we don't process the same transaction
         # multiple times in parallel.
         with (
-            yield self._transaction_linearizer.queue(
+            await self._transaction_linearizer.queue(
                 (origin, transaction.transaction_id)
             )
         ):
-            result = yield self._handle_incoming_transaction(
+            result = await self._handle_incoming_transaction(
                 origin, transaction, request_time
             )
 
         return result
 
-    @defer.inlineCallbacks
-    def _handle_incoming_transaction(self, origin, transaction, request_time):
+    async def _handle_incoming_transaction(self, origin, transaction, request_time):
         """ Process an incoming transaction and return the HTTP response
 
         Args:
@@ -140,7 +136,7 @@ class FederationServer(FederationBase):
         Returns:
             Deferred[(int, object)]: http response code and body
         """
-        response = yield self.transaction_actions.have_responded(origin, transaction)
+        response = await self.transaction_actions.have_responded(origin, transaction)
 
         if response:
             logger.debug(
@@ -151,7 +147,7 @@ class FederationServer(FederationBase):
 
         logger.debug("[%s] Transaction is new", transaction.transaction_id)
 
-        # Reject if PDU count > 50 and EDU count > 100
+        # Reject if PDU count > 50 or EDU count > 100
         if len(transaction.pdus) > 50 or (
             hasattr(transaction, "edus") and len(transaction.edus) > 100
         ):
@@ -159,7 +155,7 @@ class FederationServer(FederationBase):
             logger.info("Transaction PDU or EDU count too large. Returning 400")
 
             response = {}
-            yield self.transaction_actions.set_response(
+            await self.transaction_actions.set_response(
                 origin, transaction, 400, response
             )
             return 400, response
@@ -195,7 +191,7 @@ class FederationServer(FederationBase):
                 continue
 
             try:
-                room_version = yield self.store.get_room_version(room_id)
+                room_version = await self.store.get_room_version(room_id)
             except NotFoundError:
                 logger.info("Ignoring PDU for unknown room_id: %s", room_id)
                 continue
@@ -221,13 +217,12 @@ class FederationServer(FederationBase):
         # require callouts to other servers to fetch missing events), but
         # impose a limit to avoid going too crazy with ram/cpu.
 
-        @defer.inlineCallbacks
-        def process_pdus_for_room(room_id):
+        async def process_pdus_for_room(room_id):
             logger.debug("Processing PDUs for %s", room_id)
             try:
-                yield self.check_server_matches_acl(origin_host, room_id)
+                await self.check_server_matches_acl(origin_host, room_id)
             except AuthError as e:
-                logger.warn("Ignoring PDUs for room %s from banned server", room_id)
+                logger.warning("Ignoring PDUs for room %s from banned server", room_id)
                 for pdu in pdus_by_room[room_id]:
                     event_id = pdu.event_id
                     pdu_results[event_id] = e.error_dict()
@@ -237,10 +232,10 @@ class FederationServer(FederationBase):
                 event_id = pdu.event_id
                 with nested_logging_context(event_id):
                     try:
-                        yield self._handle_received_pdu(origin, pdu)
+                        await self._handle_received_pdu(origin, pdu)
                         pdu_results[event_id] = {}
                     except FederationError as e:
-                        logger.warn("Error handling PDU %s: %s", event_id, e)
+                        logger.warning("Error handling PDU %s: %s", event_id, e)
                         pdu_results[event_id] = {"error": str(e)}
                     except Exception as e:
                         f = failure.Failure()
@@ -251,36 +246,30 @@ class FederationServer(FederationBase):
                             exc_info=(f.type, f.value, f.getTracebackObject()),
                         )
 
-        yield concurrently_execute(
+        await concurrently_execute(
             process_pdus_for_room, pdus_by_room.keys(), TRANSACTION_CONCURRENCY_LIMIT
         )
 
         if hasattr(transaction, "edus"):
             for edu in (Edu(**x) for x in transaction.edus):
-                yield self.received_edu(origin, edu.edu_type, edu.content)
+                await self.received_edu(origin, edu.edu_type, edu.content)
 
         response = {"pdus": pdu_results}
 
         logger.debug("Returning: %s", str(response))
 
-        yield self.transaction_actions.set_response(origin, transaction, 200, response)
+        await self.transaction_actions.set_response(origin, transaction, 200, response)
         return 200, response
 
-    @defer.inlineCallbacks
-    def received_edu(self, origin, edu_type, content):
+    async def received_edu(self, origin, edu_type, content):
         received_edus_counter.inc()
-        yield self.registry.on_edu(edu_type, origin, content)
-
-    @defer.inlineCallbacks
-    @log_function
-    def on_context_state_request(self, origin, room_id, event_id):
-        if not event_id:
-            raise NotImplementedError("Specify an event")
+        await self.registry.on_edu(edu_type, origin, content)
 
+    async def on_context_state_request(self, origin, room_id, event_id):
         origin_host, _ = parse_server_name(origin)
-        yield self.check_server_matches_acl(origin_host, room_id)
+        await self.check_server_matches_acl(origin_host, room_id)
 
-        in_room = yield self.auth.check_host_in_room(room_id, origin)
+        in_room = await self.auth.check_host_in_room(room_id, origin)
         if not in_room:
             raise AuthError(403, "Host not in room.")
 
@@ -289,75 +278,79 @@ class FederationServer(FederationBase):
         # in the cache so we could return it without waiting for the linearizer
         # - but that's non-trivial to get right, and anyway somewhat defeats
         # the point of the linearizer.
-        with (yield self._server_linearizer.queue((origin, room_id))):
-            resp = yield self._state_resp_cache.wrap(
-                (room_id, event_id),
-                self._on_context_state_request_compute,
-                room_id,
-                event_id,
+        with (await self._server_linearizer.queue((origin, room_id))):
+            resp = dict(
+                await self._state_resp_cache.wrap(
+                    (room_id, event_id),
+                    self._on_context_state_request_compute,
+                    room_id,
+                    event_id,
+                )
             )
 
+        room_version = await self.store.get_room_version(room_id)
+        resp["room_version"] = room_version
+
         return 200, resp
 
-    @defer.inlineCallbacks
-    def on_state_ids_request(self, origin, room_id, event_id):
+    async def on_state_ids_request(self, origin, room_id, event_id):
         if not event_id:
             raise NotImplementedError("Specify an event")
 
         origin_host, _ = parse_server_name(origin)
-        yield self.check_server_matches_acl(origin_host, room_id)
+        await self.check_server_matches_acl(origin_host, room_id)
 
-        in_room = yield self.auth.check_host_in_room(room_id, origin)
+        in_room = await self.auth.check_host_in_room(room_id, origin)
         if not in_room:
             raise AuthError(403, "Host not in room.")
 
-        state_ids = yield self.handler.get_state_ids_for_pdu(room_id, event_id)
-        auth_chain_ids = yield self.store.get_auth_chain_ids(state_ids)
+        state_ids = await self.handler.get_state_ids_for_pdu(room_id, event_id)
+        auth_chain_ids = await self.store.get_auth_chain_ids(state_ids)
 
         return 200, {"pdu_ids": state_ids, "auth_chain_ids": auth_chain_ids}
 
-    @defer.inlineCallbacks
-    def _on_context_state_request_compute(self, room_id, event_id):
-        pdus = yield self.handler.get_state_for_pdu(room_id, event_id)
-        auth_chain = yield self.store.get_auth_chain([pdu.event_id for pdu in pdus])
+    async def _on_context_state_request_compute(self, room_id, event_id):
+        if event_id:
+            pdus = await self.handler.get_state_for_pdu(room_id, event_id)
+        else:
+            pdus = (await self.state.get_current_state(room_id)).values()
+
+        auth_chain = await self.store.get_auth_chain([pdu.event_id for pdu in pdus])
 
         return {
             "pdus": [pdu.get_pdu_json() for pdu in pdus],
             "auth_chain": [pdu.get_pdu_json() for pdu in auth_chain],
         }
 
-    @defer.inlineCallbacks
-    @log_function
-    def on_pdu_request(self, origin, event_id):
-        pdu = yield self.handler.get_persisted_pdu(origin, event_id)
+    async def on_pdu_request(self, origin, event_id):
+        pdu = await self.handler.get_persisted_pdu(origin, event_id)
 
         if pdu:
             return 200, self._transaction_from_pdus([pdu]).get_dict()
         else:
             return 404, ""
 
-    @defer.inlineCallbacks
-    def on_query_request(self, query_type, args):
+    async def on_query_request(self, query_type, args):
         received_queries_counter.labels(query_type).inc()
-        resp = yield self.registry.on_query(query_type, args)
+        resp = await self.registry.on_query(query_type, args)
         return 200, resp
 
-    @defer.inlineCallbacks
-    def on_make_join_request(self, origin, room_id, user_id, supported_versions):
+    async def on_make_join_request(self, origin, room_id, user_id, supported_versions):
         origin_host, _ = parse_server_name(origin)
-        yield self.check_server_matches_acl(origin_host, room_id)
+        await self.check_server_matches_acl(origin_host, room_id)
 
-        room_version = yield self.store.get_room_version(room_id)
+        room_version = await self.store.get_room_version(room_id)
         if room_version not in supported_versions:
-            logger.warn("Room version %s not in %s", room_version, supported_versions)
+            logger.warning(
+                "Room version %s not in %s", room_version, supported_versions
+            )
             raise IncompatibleRoomVersionError(room_version=room_version)
 
-        pdu = yield self.handler.on_make_join_request(origin, room_id, user_id)
+        pdu = await self.handler.on_make_join_request(origin, room_id, user_id)
         time_now = self._clock.time_msec()
         return {"event": pdu.get_pdu_json(time_now), "room_version": room_version}
 
-    @defer.inlineCallbacks
-    def on_invite_request(self, origin, content, room_version):
+    async def on_invite_request(self, origin, content, room_version):
         if room_version not in KNOWN_ROOM_VERSIONS:
             raise SynapseError(
                 400,
@@ -369,81 +362,71 @@ class FederationServer(FederationBase):
 
         pdu = event_from_pdu_json(content, format_ver)
         origin_host, _ = parse_server_name(origin)
-        yield self.check_server_matches_acl(origin_host, pdu.room_id)
-        pdu = yield self._check_sigs_and_hash(room_version, pdu)
-        ret_pdu = yield self.handler.on_invite_request(origin, pdu)
+        await self.check_server_matches_acl(origin_host, pdu.room_id)
+        pdu = await self._check_sigs_and_hash(room_version, pdu)
+        ret_pdu = await self.handler.on_invite_request(origin, pdu)
         time_now = self._clock.time_msec()
         return {"event": ret_pdu.get_pdu_json(time_now)}
 
-    @defer.inlineCallbacks
-    def on_send_join_request(self, origin, content, room_id):
+    async def on_send_join_request(self, origin, content, room_id):
         logger.debug("on_send_join_request: content: %s", content)
 
-        room_version = yield self.store.get_room_version(room_id)
+        room_version = await self.store.get_room_version(room_id)
         format_ver = room_version_to_event_format(room_version)
         pdu = event_from_pdu_json(content, format_ver)
 
         origin_host, _ = parse_server_name(origin)
-        yield self.check_server_matches_acl(origin_host, pdu.room_id)
+        await self.check_server_matches_acl(origin_host, pdu.room_id)
 
         logger.debug("on_send_join_request: pdu sigs: %s", pdu.signatures)
 
-        pdu = yield self._check_sigs_and_hash(room_version, pdu)
+        pdu = await self._check_sigs_and_hash(room_version, pdu)
 
-        res_pdus = yield self.handler.on_send_join_request(origin, pdu)
+        res_pdus = await self.handler.on_send_join_request(origin, pdu)
         time_now = self._clock.time_msec()
-        return (
-            200,
-            {
-                "state": [p.get_pdu_json(time_now) for p in res_pdus["state"]],
-                "auth_chain": [
-                    p.get_pdu_json(time_now) for p in res_pdus["auth_chain"]
-                ],
-            },
-        )
+        return {
+            "state": [p.get_pdu_json(time_now) for p in res_pdus["state"]],
+            "auth_chain": [p.get_pdu_json(time_now) for p in res_pdus["auth_chain"]],
+        }
 
-    @defer.inlineCallbacks
-    def on_make_leave_request(self, origin, room_id, user_id):
+    async def on_make_leave_request(self, origin, room_id, user_id):
         origin_host, _ = parse_server_name(origin)
-        yield self.check_server_matches_acl(origin_host, room_id)
-        pdu = yield self.handler.on_make_leave_request(origin, room_id, user_id)
+        await self.check_server_matches_acl(origin_host, room_id)
+        pdu = await self.handler.on_make_leave_request(origin, room_id, user_id)
 
-        room_version = yield self.store.get_room_version(room_id)
+        room_version = await self.store.get_room_version(room_id)
 
         time_now = self._clock.time_msec()
         return {"event": pdu.get_pdu_json(time_now), "room_version": room_version}
 
-    @defer.inlineCallbacks
-    def on_send_leave_request(self, origin, content, room_id):
+    async def on_send_leave_request(self, origin, content, room_id):
         logger.debug("on_send_leave_request: content: %s", content)
 
-        room_version = yield self.store.get_room_version(room_id)
+        room_version = await self.store.get_room_version(room_id)
         format_ver = room_version_to_event_format(room_version)
         pdu = event_from_pdu_json(content, format_ver)
 
         origin_host, _ = parse_server_name(origin)
-        yield self.check_server_matches_acl(origin_host, pdu.room_id)
+        await self.check_server_matches_acl(origin_host, pdu.room_id)
 
         logger.debug("on_send_leave_request: pdu sigs: %s", pdu.signatures)
 
-        pdu = yield self._check_sigs_and_hash(room_version, pdu)
+        pdu = await self._check_sigs_and_hash(room_version, pdu)
 
-        yield self.handler.on_send_leave_request(origin, pdu)
-        return 200, {}
+        await self.handler.on_send_leave_request(origin, pdu)
+        return {}
 
-    @defer.inlineCallbacks
-    def on_event_auth(self, origin, room_id, event_id):
-        with (yield self._server_linearizer.queue((origin, room_id))):
+    async def on_event_auth(self, origin, room_id, event_id):
+        with (await self._server_linearizer.queue((origin, room_id))):
             origin_host, _ = parse_server_name(origin)
-            yield self.check_server_matches_acl(origin_host, room_id)
+            await self.check_server_matches_acl(origin_host, room_id)
 
             time_now = self._clock.time_msec()
-            auth_pdus = yield self.handler.on_event_auth(event_id)
+            auth_pdus = await self.handler.on_event_auth(event_id)
             res = {"auth_chain": [a.get_pdu_json(time_now) for a in auth_pdus]}
         return 200, res
 
-    @defer.inlineCallbacks
-    def on_query_auth_request(self, origin, content, room_id, event_id):
+    async def on_query_auth_request(self, origin, content, room_id, event_id):
         """
         Content is a dict with keys::
             auth_chain (list): A list of events that give the auth chain.
@@ -462,22 +445,22 @@ class FederationServer(FederationBase):
         Returns:
             Deferred: Results in `dict` with the same format as `content`
         """
-        with (yield self._server_linearizer.queue((origin, room_id))):
+        with (await self._server_linearizer.queue((origin, room_id))):
             origin_host, _ = parse_server_name(origin)
-            yield self.check_server_matches_acl(origin_host, room_id)
+            await self.check_server_matches_acl(origin_host, room_id)
 
-            room_version = yield self.store.get_room_version(room_id)
+            room_version = await self.store.get_room_version(room_id)
             format_ver = room_version_to_event_format(room_version)
 
             auth_chain = [
                 event_from_pdu_json(e, format_ver) for e in content["auth_chain"]
             ]
 
-            signed_auth = yield self._check_sigs_and_hash_and_fetch(
+            signed_auth = await self._check_sigs_and_hash_and_fetch(
                 origin, auth_chain, outlier=True, room_version=room_version
             )
 
-            ret = yield self.handler.on_query_auth(
+            ret = await self.handler.on_query_auth(
                 origin,
                 event_id,
                 room_id,
@@ -503,16 +486,14 @@ class FederationServer(FederationBase):
         return self.on_query_request("user_devices", user_id)
 
     @trace
-    @defer.inlineCallbacks
-    @log_function
-    def on_claim_client_keys(self, origin, content):
+    async def on_claim_client_keys(self, origin, content):
         query = []
         for user_id, device_keys in content.get("one_time_keys", {}).items():
             for device_id, algorithm in device_keys.items():
                 query.append((user_id, device_id, algorithm))
 
         log_kv({"message": "Claiming one time keys.", "user, device pairs": query})
-        results = yield self.store.claim_e2e_one_time_keys(query)
+        results = await self.store.claim_e2e_one_time_keys(query)
 
         json_result = {}
         for user_id, device_keys in results.items():
@@ -536,14 +517,12 @@ class FederationServer(FederationBase):
 
         return {"one_time_keys": json_result}
 
-    @defer.inlineCallbacks
-    @log_function
-    def on_get_missing_events(
+    async def on_get_missing_events(
         self, origin, room_id, earliest_events, latest_events, limit
     ):
-        with (yield self._server_linearizer.queue((origin, room_id))):
+        with (await self._server_linearizer.queue((origin, room_id))):
             origin_host, _ = parse_server_name(origin)
-            yield self.check_server_matches_acl(origin_host, room_id)
+            await self.check_server_matches_acl(origin_host, room_id)
 
             logger.info(
                 "on_get_missing_events: earliest_events: %r, latest_events: %r,"
@@ -553,7 +532,7 @@ class FederationServer(FederationBase):
                 limit,
             )
 
-            missing_events = yield self.handler.on_get_missing_events(
+            missing_events = await self.handler.on_get_missing_events(
                 origin, room_id, earliest_events, latest_events, limit
             )
 
@@ -586,8 +565,7 @@ class FederationServer(FederationBase):
             destination=None,
         )
 
-    @defer.inlineCallbacks
-    def _handle_received_pdu(self, origin, pdu):
+    async def _handle_received_pdu(self, origin, pdu):
         """ Process a PDU received in a federation /send/ transaction.
 
         If the event is invalid, then this method throws a FederationError.
@@ -640,37 +618,34 @@ class FederationServer(FederationBase):
                 logger.info("Accepting join PDU %s from %s", pdu.event_id, origin)
 
         # We've already checked that we know the room version by this point
-        room_version = yield self.store.get_room_version(pdu.room_id)
+        room_version = await self.store.get_room_version(pdu.room_id)
 
         # Check signature.
         try:
-            pdu = yield self._check_sigs_and_hash(room_version, pdu)
+            pdu = await self._check_sigs_and_hash(room_version, pdu)
         except SynapseError as e:
             raise FederationError("ERROR", e.code, e.msg, affected=pdu.event_id)
 
-        yield self.handler.on_receive_pdu(origin, pdu, sent_to_us_directly=True)
+        await self.handler.on_receive_pdu(origin, pdu, sent_to_us_directly=True)
 
     def __str__(self):
         return "<ReplicationLayer(%s)>" % self.server_name
 
-    @defer.inlineCallbacks
-    def exchange_third_party_invite(
+    async def exchange_third_party_invite(
         self, sender_user_id, target_user_id, room_id, signed
     ):
-        ret = yield self.handler.exchange_third_party_invite(
+        ret = await self.handler.exchange_third_party_invite(
             sender_user_id, target_user_id, room_id, signed
         )
         return ret
 
-    @defer.inlineCallbacks
-    def on_exchange_third_party_invite_request(self, room_id, event_dict):
-        ret = yield self.handler.on_exchange_third_party_invite_request(
+    async def on_exchange_third_party_invite_request(self, room_id, event_dict):
+        ret = await self.handler.on_exchange_third_party_invite_request(
             room_id, event_dict
         )
         return ret
 
-    @defer.inlineCallbacks
-    def check_server_matches_acl(self, server_name, room_id):
+    async def check_server_matches_acl(self, server_name, room_id):
         """Check if the given server is allowed by the server ACLs in the room
 
         Args:
@@ -680,13 +655,13 @@ class FederationServer(FederationBase):
         Raises:
             AuthError if the server does not match the ACL
         """
-        state_ids = yield self.store.get_current_state_ids(room_id)
+        state_ids = await self.store.get_current_state_ids(room_id)
         acl_event_id = state_ids.get((EventTypes.ServerACL, ""))
 
         if not acl_event_id:
             return
 
-        acl_event = yield self.store.get_event(acl_event_id)
+        acl_event = await self.store.get_event(acl_event_id)
         if server_matches_acl_event(server_name, acl_event):
             return
 
@@ -709,7 +684,7 @@ def server_matches_acl_event(server_name, acl_event):
     # server name is a literal IP
     allow_ip_literals = acl_event.content.get("allow_ip_literals", True)
     if not isinstance(allow_ip_literals, bool):
-        logger.warn("Ignorning non-bool allow_ip_literals flag")
+        logger.warning("Ignorning non-bool allow_ip_literals flag")
         allow_ip_literals = True
     if not allow_ip_literals:
         # check for ipv6 literals. These start with '['.
@@ -723,7 +698,7 @@ def server_matches_acl_event(server_name, acl_event):
     # next,  check the deny list
     deny = acl_event.content.get("deny", [])
     if not isinstance(deny, (list, tuple)):
-        logger.warn("Ignorning non-list deny ACL %s", deny)
+        logger.warning("Ignorning non-list deny ACL %s", deny)
         deny = []
     for e in deny:
         if _acl_entry_matches(server_name, e):
@@ -733,7 +708,7 @@ def server_matches_acl_event(server_name, acl_event):
     # then the allow list.
     allow = acl_event.content.get("allow", [])
     if not isinstance(allow, (list, tuple)):
-        logger.warn("Ignorning non-list allow ACL %s", allow)
+        logger.warning("Ignorning non-list allow ACL %s", allow)
         allow = []
     for e in allow:
         if _acl_entry_matches(server_name, e):
@@ -747,7 +722,7 @@ def server_matches_acl_event(server_name, acl_event):
 
 def _acl_entry_matches(server_name, acl_entry):
     if not isinstance(acl_entry, six.string_types):
-        logger.warn(
+        logger.warning(
             "Ignoring non-str ACL entry '%s' (is %s)", acl_entry, type(acl_entry)
         )
         return False
@@ -799,15 +774,14 @@ class FederationHandlerRegistry(object):
 
         self.query_handlers[query_type] = handler
 
-    @defer.inlineCallbacks
-    def on_edu(self, edu_type, origin, content):
+    async def on_edu(self, edu_type, origin, content):
         handler = self.edu_handlers.get(edu_type)
         if not handler:
-            logger.warn("No handler registered for EDU type %s", edu_type)
+            logger.warning("No handler registered for EDU type %s", edu_type)
 
         with start_active_span_from_edu(content, "handle_edu"):
             try:
-                yield handler(origin, content)
+                await handler(origin, content)
             except SynapseError as e:
                 logger.info("Failed to handle edu %r: %r", edu_type, e)
             except Exception:
@@ -816,7 +790,7 @@ class FederationHandlerRegistry(object):
     def on_query(self, query_type, args):
         handler = self.query_handlers.get(query_type)
         if not handler:
-            logger.warn("No handler registered for query type %s", query_type)
+            logger.warning("No handler registered for query type %s", query_type)
             raise NotFoundError("No handler for Query type '%s'" % (query_type,))
 
         return handler(args)
@@ -840,7 +814,7 @@ class ReplicationFederationHandlerRegistry(FederationHandlerRegistry):
 
         super(ReplicationFederationHandlerRegistry, self).__init__()
 
-    def on_edu(self, edu_type, origin, content):
+    async def on_edu(self, edu_type, origin, content):
         """Overrides FederationHandlerRegistry
         """
         if not self.config.use_presence and edu_type == "m.presence":
@@ -848,17 +822,17 @@ class ReplicationFederationHandlerRegistry(FederationHandlerRegistry):
 
         handler = self.edu_handlers.get(edu_type)
         if handler:
-            return super(ReplicationFederationHandlerRegistry, self).on_edu(
+            return await super(ReplicationFederationHandlerRegistry, self).on_edu(
                 edu_type, origin, content
             )
 
-        return self._send_edu(edu_type=edu_type, origin=origin, content=content)
+        return await self._send_edu(edu_type=edu_type, origin=origin, content=content)
 
-    def on_query(self, query_type, args):
+    async def on_query(self, query_type, args):
         """Overrides FederationHandlerRegistry
         """
         handler = self.query_handlers.get(query_type)
         if handler:
-            return handler(args)
+            return await handler(args)
 
-        return self._get_query_client(query_type=query_type, args=args)
+        return await self._get_query_client(query_type=query_type, args=args)

synapse/federation/persistence.py

@@ -44,7 +44,7 @@ class TransactionActions(object):
             response code and response body.
         """
         if not transaction.transaction_id:
-            raise RuntimeError("Cannot persist a transaction with no " "transaction_id")
+            raise RuntimeError("Cannot persist a transaction with no transaction_id")
 
         return self.store.get_received_txn_response(transaction.transaction_id, origin)
 
@@ -56,7 +56,7 @@ class TransactionActions(object):
             Deferred
         """
         if not transaction.transaction_id:
-            raise RuntimeError("Cannot persist a transaction with no " "transaction_id")
+            raise RuntimeError("Cannot persist a transaction with no transaction_id")
 
         return self.store.set_received_txn_response(
             transaction.transaction_id, origin, code, response

synapse/federation/send_queue.py

@@ -36,6 +36,8 @@ from six import iteritems
 
 from sortedcontainers import SortedDict
 
+from twisted.internet import defer
+
 from synapse.metrics import LaterGauge
 from synapse.storage.presence import UserPresenceState
 from synapse.util.metrics import Measure
@@ -212,7 +214,7 @@ class FederationRemoteSendQueue(object):
             receipt (synapse.types.ReadReceipt):
         """
         # nothing to do here: the replication listener will handle it.
-        pass
+        return defer.succeed(None)
 
     def send_presence(self, states):
         """As per FederationSender

synapse/federation/sender/__init__.py

@@ -49,7 +49,7 @@ sent_pdus_destination_dist_count = Counter(
 
 sent_pdus_destination_dist_total = Counter(
     "synapse_federation_client_sent_pdu_destinations:total",
-    "" "Total number of PDUs queued for sending across all destinations",
+    "Total number of PDUs queued for sending across all destinations",
 )
 
 

synapse/federation/sender/per_destination_queue.py

@@ -192,15 +192,16 @@ class PerDestinationQueue(object):
                 # We have to keep 2 free slots for presence and rr_edus
                 limit = MAX_EDUS_PER_TRANSACTION - 2
 
-                device_update_edus, dev_list_id = (
-                    yield self._get_device_update_edus(limit)
+                device_update_edus, dev_list_id = yield self._get_device_update_edus(
+                    limit
                 )
 
                 limit -= len(device_update_edus)
 
-                to_device_edus, device_stream_id = (
-                    yield self._get_to_device_message_edus(limit)
-                )
+                (
+                    to_device_edus,
+                    device_stream_id,
+                ) = yield self._get_to_device_message_edus(limit)
 
                 pending_edus = device_update_edus + to_device_edus
 
@@ -359,20 +360,20 @@ class PerDestinationQueue(object):
         last_device_list = self._last_device_list_stream_id
 
         # Retrieve list of new device updates to send to the destination
-        now_stream_id, results = yield self._store.get_devices_by_remote(
+        now_stream_id, results = yield self._store.get_device_updates_by_remote(
             self._destination, last_device_list, limit=limit
         )
         edus = [
             Edu(
                 origin=self._server_name,
                 destination=self._destination,
-                edu_type="m.device_list_update",
+                edu_type=edu_type,
                 content=content,
             )
-            for content in results
+            for (edu_type, content) in results
         ]
 
-        assert len(edus) <= limit, "get_devices_by_remote returned too many EDUs"
+        assert len(edus) <= limit, "get_device_updates_by_remote returned too many EDUs"
 
         return (edus, now_stream_id)