synapse/scripts-dev/schema_versions.py

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

183 lines
6.7 KiB
Python
Raw Permalink Normal View History

#!/usr/bin/env python
# Copyright 2023 The Matrix.org Foundation C.I.C.
# Copyright 2023 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.
"""A script to calculate which versions of Synapse have backwards-compatible
database schemas. It creates a Markdown table of Synapse versions and the earliest
compatible version.
It is compatible with the mdbook protocol for preprocessors (see
https://rust-lang.github.io/mdBook/for_developers/preprocessors.html#implementing-a-preprocessor-with-a-different-language):
Exit 0 to denote support for all renderers:
./scripts-dev/schema_versions.py supports <mdbook renderer>
Parse a JSON list from stdin and add the table to the proper documetnation page:
./scripts-dev/schema_versions.py
Additionally, the script supports dumping the table to stdout for debugging:
./scripts-dev/schema_versions.py dump
"""
import io
import json
import sys
from collections import defaultdict
from typing import Any, Dict, Iterator, Optional, Tuple
import git
from packaging import version
# The schema version has moved around over the years.
SCHEMA_VERSION_FILES = (
"synapse/storage/schema/__init__.py",
"synapse/storage/prepare_database.py",
"synapse/storage/__init__.py",
"synapse/app/homeserver.py",
)
# Skip versions of Synapse < v1.0, they're old and essentially not
# compatible with today's federation.
OLDEST_SHOWN_VERSION = version.parse("v1.0")
def get_schema_versions(tag: git.Tag) -> Tuple[Optional[int], Optional[int]]:
"""Get the schema and schema compat versions for a tag."""
schema_version = None
schema_compat_version = None
for file in SCHEMA_VERSION_FILES:
try:
schema_file = tag.commit.tree / file
except KeyError:
continue
# We (usually) can't execute the code since it might have unknown imports.
if file != "synapse/storage/schema/__init__.py":
with io.BytesIO(schema_file.data_stream.read()) as f:
for line in f.readlines():
if line.startswith(b"SCHEMA_VERSION"):
schema_version = int(line.split()[2])
# Bail early.
break
else:
# SCHEMA_COMPAT_VERSION is sometimes across multiple lines, the easist
# thing to do is exec the code. Luckily it has only ever existed in
# a file which imports nothing else from Synapse.
locals: Dict[str, Any] = {}
exec(schema_file.data_stream.read().decode("utf-8"), {}, locals)
schema_version = locals["SCHEMA_VERSION"]
schema_compat_version = locals.get("SCHEMA_COMPAT_VERSION")
return schema_version, schema_compat_version
def get_tags(repo: git.Repo) -> Iterator[git.Tag]:
"""Return an iterator of tags sorted by version."""
tags = []
for tag in repo.tags:
# All "real" Synapse tags are of the form vX.Y.Z.
if not tag.name.startswith("v"):
continue
# There's a weird tag from the initial react UI.
if tag.name == "v0.1":
continue
try:
tag_version = version.parse(tag.name)
except version.InvalidVersion:
# Skip invalid versions.
continue
# Skip pre- and post-release versions.
if tag_version.is_prerelease or tag_version.is_postrelease or tag_version.local:
continue
# Skip old versions.
if tag_version < OLDEST_SHOWN_VERSION:
continue
tags.append((tag_version, tag))
# Sort based on the version number (not lexically).
return (tag for _, tag in sorted(tags, key=lambda t: t[0]))
def calculate_version_chart() -> str:
repo = git.Repo(path=".")
# Map of schema version -> Synapse versions which are at that schema version.
schema_versions = defaultdict(list)
# Map of schema version -> Synapse versions which are compatible with that
# schema version.
schema_compat_versions = defaultdict(list)
# Find ranges of versions which are compatible with a schema version.
#
# There are two modes of operation:
#
# 1. Pre-schema_compat_version (i.e. schema_compat_version of None), then
# Synapse is compatible up/downgrading to a version with
# schema_version >= its current version.
#
# 2. Post-schema_compat_version (i.e. schema_compat_version is *not* None),
# then Synapse is compatible up/downgrading to a version with
# schema version >= schema_compat_version.
#
# This is more generous and avoids versions that cannot be rolled back.
#
# See https://github.com/matrix-org/synapse/pull/9933 which was included in v1.37.0.
for tag in get_tags(repo):
schema_version, schema_compat_version = get_schema_versions(tag)
# If a schema compat version is given, prefer that over the schema version.
schema_versions[schema_version].append(tag.name)
schema_compat_versions[schema_compat_version or schema_version].append(tag.name)
# Generate a table which maps the latest Synapse version compatible with each
# schema version.
result = f"| {'Versions': ^19} | Compatible version |\n"
result += f"|{'-' * (19 + 2)}|{'-' * (18 + 2)}|\n"
for schema_version, synapse_versions in schema_compat_versions.items():
result += f"| {synapse_versions[0] + ' ' + synapse_versions[-1]: ^19} | {schema_versions[schema_version][0]: ^18} |\n"
return result
if __name__ == "__main__":
if len(sys.argv) == 3 and sys.argv[1] == "supports":
# We don't care about the renderer which is being used, which is the second argument.
sys.exit(0)
elif len(sys.argv) == 2 and sys.argv[1] == "dump":
print(calculate_version_chart())
else:
# Expect JSON data on stdin.
context, book = json.load(sys.stdin)
for section in book["sections"]:
if "Chapter" in section and section["Chapter"]["path"] == "upgrade.md":
section["Chapter"]["content"] = section["Chapter"]["content"].replace(
"<!-- REPLACE_WITH_SCHEMA_VERSIONS -->", calculate_version_chart()
)
# Print the result back out to stdout.
print(json.dumps(book))