Files
class-schedule/shiguang_warehouse/scripts/build_data.py
T
2026-09-20 13:33:09 +08:00

150 lines
5.3 KiB
Python

import sys
import yaml
import os
import datetime
from pathlib import Path
SCRIPT_DIR = Path(__file__).resolve().parent
# 确保该目录在模块搜索路径中
if str(SCRIPT_DIR) not in sys.path:
sys.path.insert(0, str(SCRIPT_DIR))
# --- 1. Protobuf 依赖导入 ---
# 注意:这依赖于 protoc 编译生成的 school_index_pb2.py 文件
try:
from school_index_pb2 import SchoolIndex, School, Adapter, AdapterCategory
except ImportError:
print("错误:无法导入 Protobuf 模块 (school_index_pb2)。请确认文件已存在。")
sys.exit(1)
# --- 2. 工程常量与版本定义 ---
# 协议版本:结构大改动时手动递增。
PROTOCOL_VERSION = 2
# YAML 文件路径常量
ROOT_INDEX_PATH = Path("index/root_index.yaml")
RESOURCES_ROOT = Path("resources")
OUTPUT_PB_FILE = "school_index.pb"
# --- 3. 辅助函数:获取版本ID (用于 version_id 字段) ---
def get_version_id():
"""
【统一标准】:只使用精确到毫秒的时间戳作为数据版本ID。
"""
# 获取当前时间(包含微秒)
now = datetime.datetime.now()
# 格式化时间部分
time_str = now.strftime("TIME_%Y%m%d%H%M%S")
micro_str = f"{now.microsecond // 1000:03d}"
return f"{time_str}_{micro_str}"
# --- 4. 辅助函数:YAML 解析逻辑 ---
def get_adapter_category_enum(category_str):
"""将 YAML 字符串类别映射到 Protobuf 枚举值。"""
enum_value = getattr(AdapterCategory, category_str, AdapterCategory.ADAPTER_CATEGORY_UNKNOWN)
return enum_value
def parse_all_yaml():
"""解析所有 YAML 文件,返回结构化数据。"""
if not ROOT_INDEX_PATH.exists():
raise FileNotFoundError(f"根索引文件未找到: {ROOT_INDEX_PATH}")
with open(ROOT_INDEX_PATH, 'r', encoding='utf-8') as f:
root_data = yaml.safe_load(f)
schools_data = root_data.get('schools', [])
parsed_schools = []
for school_entry in schools_data:
# 1. 拼接 adapters.yaml 的路径
folder_name = school_entry['resource_folder']
adapter_yaml_path = RESOURCES_ROOT / folder_name / "adapters.yaml"
if not adapter_yaml_path.exists():
print(f"警告:未找到适配器配置 {adapter_yaml_path},跳过学校 {school_entry['id']}")
continue
# 2. 读取适配器详情
with open(adapter_yaml_path, 'r', encoding='utf-8') as f:
adapter_data = yaml.safe_load(f)
# 将适配器列表添加到学校数据中
school_entry['adapters'] = adapter_data.get('adapters', [])
parsed_schools.append(school_entry)
return parsed_schools
# --- 5. 核心构建函数:数据映射与填充 ---
def build_protobuf_index(parsed_schools_data):
"""将解析后的数据填充到 Protobuf 消息中。"""
index = SchoolIndex()
# 设置顶层版本信息
index.protocol_version = PROTOCOL_VERSION
index.version_id = get_version_id()
for school_data in parsed_schools_data:
# 1. 填充 School 消息
school_msg = index.schools.add()
school_msg.id = school_data.get('id', '')
school_msg.name = school_data.get('name', '')
school_msg.initial = school_data.get('initial', '')
school_msg.resource_folder = school_data.get('resource_folder', '')
# 2. 填充 Adapter 列表
for adapter_data in school_data.get('adapters', []):
adapter_msg = school_msg.adapters.add()
# 填充普通 string 字段
adapter_msg.adapter_id = adapter_data.get('adapter_id', '')
adapter_msg.adapter_name = adapter_data.get('adapter_name', '')
adapter_msg.asset_js_path = adapter_data.get('asset_js_path', '')
adapter_msg.description = adapter_data.get('description', '')
adapter_msg.maintainer = adapter_data.get('maintainer', '')
# 填充 optional 字段:只有当字段不为 None 时才设置
import_url = adapter_data.get('import_url')
if import_url is not None:
adapter_msg.import_url = import_url
# 填充枚举字段
category_str = adapter_data.get('category', 'ADAPTER_CATEGORY_UNKNOWN')
adapter_msg.category = get_adapter_category_enum(category_str)
return index
# --- 6. 主执行流程 ---
if __name__ == "__main__":
OUTPUT_FILE = OUTPUT_PB_FILE
print(f"目标输出文件: {OUTPUT_FILE}")
try:
print("--- 阶段一:解析 YAML 源文件 ---")
parsed_data = parse_all_yaml()
print("\n--- 阶段二:构建 Protobuf 消息 ---")
index_message = build_protobuf_index(parsed_data)
print(f"Protobuf 协议版本: {index_message.protocol_version}")
print(f"数据版本ID: {index_message.version_id}")
print("\n--- 阶段三:序列化并写入磁盘 ---")
# 序列化并写入二进制文件
with open(OUTPUT_FILE, "wb") as f:
# Protobuf 核心序列化方法
f.write(index_message.SerializeToString())
print(f"\n构建成功!二进制文件已保存到: {OUTPUT_FILE}")
print(f"文件大小: {os.path.getsize(OUTPUT_FILE) / 1024:.2f} KB")
except Exception as e:
print(f"构建失败!致命错误: {e}", file=sys.stderr)
sys.exit(1)