Reorganise the database schema directories (#9932)

The hope here is that by moving all the schema files into synapse/storage/schema, it gets a bit easier for newcomers to navigate.

It certainly got easier for me to write a helpful README. There's more to do on that front, but I'll follow up with other PRs for that.

changelog.d/9932.misc

@@ -0,0 +1 @@
+Move database schema files into a common directory.

synapse/storage/databases/main/schema/full_schemas/README.md

@@ -1,21 +0,0 @@
-# Synapse Database Schemas
-
-These schemas are used as a basis to create brand new Synapse databases, on both
-SQLite3 and Postgres.
-
-## Building full schema dumps
-
-If you want to recreate these schemas, they need to be made from a database that
-has had all background updates run.
-
-To do so, use `scripts-dev/make_full_schema.sh`. This will produce new
-`full.sql.postgres ` and `full.sql.sqlite` files. 
-
-Ensure postgres is installed and your user has the ability to run bash commands
-such as `createdb`, then call
-
-    ./scripts-dev/make_full_schema.sh -p postgres_username -o output_dir/
-
-There are currently two folders with full-schema snapshots. `16` is a snapshot
-from 2015, for historical reference. The other contains the most recent full
-schema snapshot.

synapse/storage/prepare_database.py

@@ -26,16 +26,13 @@ from synapse.config.homeserver import HomeServerConfig
 from synapse.storage.database import LoggingDatabaseConnection
 from synapse.storage.engines import BaseDatabaseEngine
 from synapse.storage.engines.postgres import PostgresEngine
+from synapse.storage.schema import SCHEMA_VERSION
 from synapse.storage.types import Cursor
 
 logger = logging.getLogger(__name__)
 
 
-# Remember to update this number every time a change is made to database
+schema_path = os.path.join(os.path.abspath(os.path.dirname(__file__)), "schema")
-# schema files, so the users will be informed on server restarts.
-SCHEMA_VERSION = 59
-
-dir_path = os.path.abspath(os.path.dirname(__file__))
 
 
 class PrepareDatabaseException(Exception):
@@ -167,7 +164,14 @@ def _setup_new_database(
 
     Example directory structure:
 
-        schema/
+    schema/
+        common/
+            delta/
+                ...
+            full_schemas/
+                11/
+                    foo.sql
+        main/
             delta/
                 ...
             full_schemas/
@@ -175,15 +179,14 @@ def _setup_new_database(
                     test.sql
                     ...
                 11/
-                    foo.sql
                     bar.sql
                 ...
 
     In the example foo.sql and bar.sql would be run, and then any delta files
     for versions strictly greater than 11.
 
-    Note: we apply the full schemas and deltas from the top level `schema/`
+    Note: we apply the full schemas and deltas from the `schema/common`
-    folder as well those in the data stores specified.
+    folder as well those in the databases specified.
 
     Args:
         cur: a database cursor
@@ -195,12 +198,12 @@ def _setup_new_database(
     # configured to our liking.
     database_engine.check_new_database(cur)
 
-    current_dir = os.path.join(dir_path, "schema", "full_schemas")
+    full_schemas_dir = os.path.join(schema_path, "common", "full_schemas")
 
     # First we find the highest full schema version we have
     valid_versions = []
 
-    for filename in os.listdir(current_dir):
+    for filename in os.listdir(full_schemas_dir):
         try:
             ver = int(filename)
         except ValueError:
@@ -218,15 +221,13 @@ def _setup_new_database(
 
     logger.debug("Initialising schema v%d", max_current_ver)
 
-    # Now lets find all the full schema files, both in the global schema and
+    # Now let's find all the full schema files, both in the common schema and
-    # in data store schemas.
+    # in database schemas.
-    directories = [os.path.join(current_dir, str(max_current_ver))]
+    directories = [os.path.join(full_schemas_dir, str(max_current_ver))]
     directories.extend(
         os.path.join(
-            dir_path,
+            schema_path,
-            "databases",
             database,
-            "schema",
             "full_schemas",
             str(max_current_ver),
         )
@@ -357,6 +358,9 @@ def _upgrade_existing_database(
         check_database_before_upgrade(cur, database_engine, config)
 
     start_ver = current_version
+
+    # if we got to this schema version by running a full_schema rather than a series
+    # of deltas, we should not run the deltas for this version.
     if not upgraded:
         start_ver += 1
 
@@ -385,12 +389,10 @@ def _upgrade_existing_database(
         # directories for schema updates.
 
         # First we find the directories to search in
-        delta_dir = os.path.join(dir_path, "schema", "delta", str(v))
+        delta_dir = os.path.join(schema_path, "common", "delta", str(v))
         directories = [delta_dir]
         for database in databases:
-            directories.append(
+            directories.append(os.path.join(schema_path, database, "delta", str(v)))
-                os.path.join(dir_path, "databases", database, "schema", "delta", str(v))
-            )
 
         # Used to check if we have any duplicate file names
         file_name_counter = Counter()  # type: CounterType[str]
@@ -621,8 +623,8 @@ def _get_or_create_schema_state(
     txn: Cursor, database_engine: BaseDatabaseEngine
 ) -> Optional[Tuple[int, List[str], bool]]:
     # Bluntly try creating the schema_version tables.
-    schema_path = os.path.join(dir_path, "schema", "schema_version.sql")
+    sql_path = os.path.join(schema_path, "common", "schema_version.sql")
-    executescript(txn, schema_path)
+    executescript(txn, sql_path)
 
     txn.execute("SELECT version, upgraded FROM schema_version")
     row = txn.fetchone()

synapse/storage/schema/README.md

@@ -0,0 +1,37 @@
+# Synapse Database Schemas
+
+This directory contains the schema files used to build Synapse databases.
+
+Synapse supports splitting its datastore across multiple physical databases (which can
+be useful for large installations), and the schema files are therefore split according
+to the logical database they are apply to.
+
+At the time of writing, the following "logical" databases are supported:
+
+* `state` - used to store Matrix room state (more specifically, `state_groups`,
+  their relationships and contents.)
+* `main` - stores everything else.
+
+Addionally, the `common` directory contains schema files for tables which must be
+present on *all* physical databases.
+
+## Full schema dumps
+
+In the `full_schemas` directories, only the most recently-numbered snapshot is useful
+(`54` at the time of writing). Older snapshots (eg, `16`) are present for historical
+reference only.
+
+## Building full schema dumps
+
+If you want to recreate these schemas, they need to be made from a database that
+has had all background updates run.
+
+To do so, use `scripts-dev/make_full_schema.sh`. This will produce new
+`full.sql.postgres` and `full.sql.sqlite` files.
+
+Ensure postgres is installed, then run:
+
+    ./scripts-dev/make_full_schema.sh -p postgres_username -o output_dir/
+
+NB at the time of writing, this script predates the split into separate `state`/`main`
+databases so will require updates to handle that correctly.

synapse/storage/schema/__init__.py

@@ -0,0 +1,17 @@
+# Copyright 2021 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.
+
+# Remember to update this number every time a change is made to database
+# schema files, so the users will be informed on server restarts.
+SCHEMA_VERSION = 59