fanlusky
a7779d2b05
`vendor/MetasequoiaImeEngine` submodule 改为 `engine/` 目录，导入的是 product-lock.json 原本钉住的提交 c810d201，逐文件核对过 blob，因此这是一次 搬迁而不是引擎升级。引擎自己的嵌套 submodule（googlepinyinime-rev、utfcpp、 miniaudio）按各自 gitlink 一并展开，来源与提交见 engine/UPSTREAM.md。 Windows 专门化裁掉了不属于本产品的部分：whisper.cpp 与本地推理 provider、 macOS 示例宿主、整个 dictionary/ 建库流水线、上游 CI 与 release-please 配置。 词库仍按 product-lock.json 在构建时取发布产物。 引擎不再进 product-lock.json：没有 gitlink 就没有可比对的东西，锁里写一个上游 commit 等于宣称这棵树仍等于上游，而第一个 Windows 专门化改动落地时这句话就假了。 verify-contracts / verify-checkout 随之移除，verify-published 保留。 CI 与工具链：引擎源码并入根 scripts/format.sh（删掉引擎自己的那份，避免两套排除 列表分叉）；引擎 voice.yml 里的 silero_vad.onnx 摘要校验移到本仓 lock job； engine-bump-triage 工作流与其测试删除；gitsubmodule 的 dependabot 频率由 daily 改为 monthly。 README、README.en、AGENTS.md、CONTRIBUTING、docs/product-release.md 及各子目录 文档同步说明新的位置与边界。
82 lines
3.8 KiB
Python
82 lines
3.8 KiB
Python
"""Shared asset inventory for builders and pinned consumers; standard library only."""
|
|
import json
|
|
import hashlib
|
|
import tempfile
|
|
import zipfile
|
|
import importlib.util
|
|
import re
|
|
from pathlib import Path, PurePosixPath
|
|
|
|
_dictionary_spec = importlib.util.spec_from_file_location(
|
|
"msime_assets_dictionary_contract", Path(__file__).parents[1] / "dictionary/product.py")
|
|
_dictionary = importlib.util.module_from_spec(_dictionary_spec)
|
|
_dictionary_spec.loader.exec_module(_dictionary)
|
|
verify_product = _dictionary.verify_product
|
|
|
|
CONTRACT = json.loads(Path(__file__).with_name('assets.json').read_text(encoding='utf-8'))
|
|
# Fail before generating code or touching an archive if the inventory itself is ambiguous.
|
|
_ids, _paths = set(), set()
|
|
for _entry in CONTRACT['assets']:
|
|
_name = _entry['path']
|
|
if (not re.fullmatch(r'[a-z][a-z0-9_]*', _entry['id']) or _entry['id'] in _ids or
|
|
_name.casefold() in _paths or PurePosixPath(_name).is_absolute() or
|
|
'..' in PurePosixPath(_name).parts or '\\' in _name or
|
|
str(PurePosixPath(_name)) != _name):
|
|
raise ValueError('Invalid or duplicate assets contract entry')
|
|
_ids.add(_entry['id'])
|
|
_paths.add(_name.casefold())
|
|
VERSION = CONTRACT['contract_version']
|
|
ARCHIVE = CONTRACT['archive']
|
|
MANIFEST = CONTRACT['manifest']
|
|
|
|
|
|
def assets(profile='desktop'):
|
|
if profile != 'desktop':
|
|
raise ValueError(f'Unsupported assets profile: {profile}')
|
|
return {entry['path']: entry for entry in CONTRACT['assets'] if profile in entry['profiles']}
|
|
|
|
|
|
STATIC_FILES = {name: entry['source'] for name, entry in assets().items()
|
|
if entry['source'] != 'dictionary'}
|
|
DICTIONARY_FILES = frozenset(assets()) - STATIC_FILES.keys()
|
|
HELPCODES = {entry['schema']: entry['path'] for entry in CONTRACT['assets'] if 'schema' in entry}
|
|
|
|
|
|
def sha256(data):
|
|
return hashlib.sha256(data).hexdigest()
|
|
|
|
|
|
def verify(archive: Path, expected_sha256: str | None = None) -> dict:
|
|
"""Verify the exact payload and optionally its digest pinned by the consuming product."""
|
|
if expected_sha256 is not None:
|
|
digest = hashlib.sha256()
|
|
with Path(archive).open('rb') as stream:
|
|
for chunk in iter(lambda: stream.read(1 << 20), b''):
|
|
digest.update(chunk)
|
|
if digest.hexdigest() != expected_sha256:
|
|
raise ValueError('Engine assets archive does not match pinned digest')
|
|
with zipfile.ZipFile(archive) as bundle:
|
|
names = bundle.namelist()
|
|
manifest = json.loads(bundle.read(MANIFEST))
|
|
expected = set(assets())
|
|
if (manifest.get('manifest_version') != 1 or manifest.get('assets_contract_version') != VERSION or manifest.get('profile') != 'desktop'
|
|
or set(manifest.get('files', {})) != expected
|
|
or len(names) != len(set(names))
|
|
or set(names) != expected | {MANIFEST}):
|
|
raise ValueError('Unexpected engine assets manifest or archive contents')
|
|
if manifest.get('source', {}).get('repository') != 'metasequoiaime/MSIME-Engine':
|
|
raise ValueError('Unexpected Engine assets source repository')
|
|
for name, entry in manifest['files'].items():
|
|
data = bundle.read(name)
|
|
if len(data) != entry['size'] or sha256(data) != entry['sha256']:
|
|
raise ValueError(f'Engine asset changed: {name}')
|
|
# Reuse the dictionary format contract, including model magic and profile checks.
|
|
with tempfile.TemporaryDirectory() as temporary:
|
|
directory = Path(temporary)
|
|
for name in expected - set(STATIC_FILES):
|
|
(directory / name).write_bytes(bundle.read(name))
|
|
product = verify_product(directory, 'desktop')
|
|
if product['source']['commit'] != manifest['source']['commit']:
|
|
raise ValueError('Dictionary and runtime assets must have the same source commit')
|
|
return manifest
|