mdbq 3.9.8__py3-none-any.whl → 3.9.10__py3-none-any.whl

mdbq/__version__.py

@@ -1 +1 @@
-VERSION = '3.9.8'
+VERSION = '3.9.10'

mdbq/mysql/deduplicator.py

@@ -17,14 +17,14 @@ from collections import defaultdict
 warnings.filterwarnings('ignore')
 logger = mylogger.MyLogger(
     name='deduplicator',
-    logging_mode='none',
-    log_level='error',
+    logging_mode='both',
+    log_level='info',
     log_file='deduplicator.log',
     log_format='json',
     max_log_size=50,
     backup_count=5,
     enable_async=False,  # 是否启用异步日志
-    sample_rate=0.5,  # 采样50%的DEBUG/INFO日志
+    sample_rate=1,  # 采样50%的DEBUG/INFO日志
     sensitive_fields=[],  #  敏感字段列表
 )
 
@@ -116,10 +116,7 @@ class MySQLDeduplicator:
         self._processing_tables = set()  # 正在处理的表集合
 
         # 系统数据库列表
-        self.SYSTEM_DATABASES = {
-            'information_schema', 'mysql',
-            'performance_schema', 'sys'
-        }
+        self.SYSTEM_DATABASES = {'information_schema', 'mysql', 'performance_schema', 'sys'}
 
     def _get_connection(self):
         """从连接池获取连接"""
@@ -580,7 +577,7 @@ class MySQLDeduplicator:
 def main():
     deduplicator = MySQLDeduplicator(
         username='root',
-        password='pw',
+        password='188988yang188',
         host='localhost',
         port=3306
     )

mdbq/mysql/uploader.py

@@ -17,20 +17,26 @@ from collections import OrderedDict
 warnings.filterwarnings('ignore')
 logger = mylogger.MyLogger(
     name='uploader',
-    logging_mode='none',
-    log_level='error',
+    logging_mode='both',
+    log_level='info',
     log_file='uploader.log',
     log_format='json',
     max_log_size=50,
     backup_count=5,
     enable_async=False,  # 是否启用异步日志
-    sample_rate=0.5,  # 采样50%的DEBUG/INFO日志
+    sample_rate=1,  # 采样50%的DEBUG/INFO日志
     sensitive_fields=[],  #  敏感字段列表
 )
 
 
 def count_decimal_places(num_str):
-    """ 计算小数位数, 允许科学计数法 """
+    """
+    计算数字字符串的小数位数，支持科学计数法
+
+    :param num_str: 数字字符串
+    :return: 返回元组(整数位数, 小数位数)
+    :raises: 无显式抛出异常，但正则匹配失败时返回(0, 0)
+    """
     match = re.match(r'^[-+]?\d+(\.\d+)?([eE][-+]?\d+)?$', str(num_str))
     if match:
         # 如果是科学计数法
@@ -53,8 +59,13 @@ def count_decimal_places(num_str):
 
 
 class StatementCache(OrderedDict):
-    """LRU缓存策略"""
+    """基于OrderedDict实现的LRU缓存策略，用于缓存SQL语句"""
     def __init__(self, maxsize=100):
+        """
+        初始化缓存
+
+        :param maxsize: 最大缓存大小，默认为100
+        """
         super().__init__()
         self.maxsize = maxsize
 
@@ -88,7 +99,6 @@ class MySQLUploader:
         :param port: 数据库端口，默认为3306
         :param charset: 字符集，默认为utf8mb4
         :param collation: 排序规则，默认为utf8mb4_0900_ai_ci
-
         :param max_retries: 最大重试次数，默认为10
         :param retry_interval: 重试间隔(秒)，默认为10
         :param pool_size: 连接池大小，默认为5
@@ -111,7 +121,7 @@ class MySQLUploader:
         self.write_timeout = write_timeout
         self.ssl = ssl
         self._prepared_statements = StatementCache(maxsize=100)
-        self._max_cached_statements = 100
+        self._max_cached_statements = 100  # 用于控制 StatementCache 类中缓存的 SQL 语句数量，最多缓存 100 条 SQL 语句
         self._table_metadata_cache = {}
         self.metadata_cache_ttl = 300  # 5分钟缓存时间
 
@@ -119,7 +129,12 @@ class MySQLUploader:
         self.pool = self._create_connection_pool()
 
     def _create_connection_pool(self) -> PooledDB:
-        """创建数据库连接池"""
+        """
+        创建数据库连接池
+
+        :return: PooledDB连接池实例
+        :raises ConnectionError: 当连接池创建失败时抛出
+        """
         if hasattr(self, 'pool') and self.pool is not None and self._check_pool_health():
             return self.pool
 
@@ -157,22 +172,29 @@ class MySQLUploader:
 
         try:
             pool = PooledDB(**pool_params)
-            elapsed = time.time() - start_time
+            elapsed = round(time.time() - start_time, 2)
             logger.info("连接池创建成功", {
                 'pool_size': self.pool_size,
-                'time_elapsed': elapsed
+                '耗时': elapsed
             })
             return pool
         except Exception as e:
-            elapsed = time.time() - start_time
+            elapsed = round(time.time() - start_time, 2)
             self.pool = None
             logger.error("连接池创建失败", {
                 'error': str(e),
-                'time_elapsed': elapsed
+                '耗时': elapsed
             })
             raise ConnectionError(f"连接池创建失败: {str(e)}")
 
     def _execute_with_retry(self, func):
+        """
+        带重试机制的装饰器，用于数据库操作
+
+        :param func: 被装饰的函数
+        :return: 装饰后的函数
+        :raises: 可能抛出原始异常或最后一次重试的异常
+        """
         @wraps(func)
         def wrapper(*args, **kwargs):
             last_exception = None
@@ -187,18 +209,18 @@ class MySQLUploader:
             for attempt in range(self.max_retries):
                 try:
                     result = func(*args, **kwargs)
-                    elapsed = time.time() - start_time
+                    elapsed = round(time.time() - start_time, 2)
 
                     if attempt > 0:
                         logger.info("操作成功(重试后)", {
                             'operation': operation,
                             'attempts': attempt + 1,
-                            'time_elapsed': elapsed
+                            '耗时': elapsed
                         })
                     else:
                         logger.debug("操作成功", {
                             'operation': operation,
-                            'time_elapsed': elapsed
+                            '耗时': elapsed
                         })
 
                     return result
@@ -230,15 +252,15 @@ class MySQLUploader:
                                 'error': str(reconnect_error)
                             })
                     else:
-                        elapsed = time.time() - start_time
-                        error_details['time_elapsed'] = elapsed
+                        elapsed = round(time.time() - start_time, 2)
+                        error_details['耗时'] = elapsed
                         logger.error(f"操作最终失败 {error_details}")
 
                 except pymysql.IntegrityError as e:
-                    elapsed = time.time() - start_time
+                    elapsed = round(time.time() - start_time, 2)
                     logger.error("完整性约束错误", {
                         'operation': operation,
-                        'time_elapsed': elapsed,
+                        '耗时': elapsed,
                         'error_code': e.args[0] if e.args else None,
                         'error_message': e.args[1] if len(e.args) > 1 else None
                     })
@@ -246,10 +268,10 @@ class MySQLUploader:
 
                 except Exception as e:
                     last_exception = e
-                    elapsed = time.time() - start_time
+                    elapsed = round(time.time() - start_time, 2)
                     logger.error("发生意外错误", {
                         'operation': operation,
-                        'time_elapsed': elapsed,
+                        '耗时': elapsed,
                         'error_type': type(e).__name__,
                         'error_message': str(e),
                         'error_args': e.args if hasattr(e, 'args') else None
@@ -261,7 +283,12 @@ class MySQLUploader:
         return wrapper
 
     def _get_connection(self):
-        """从连接池获取连接"""
+        """
+        从连接池获取数据库连接
+
+        :return: 数据库连接对象
+        :raises ConnectionError: 当获取连接失败时抛出
+        """
         try:
             conn = self.pool.connection()
             logger.debug("获取数据库连接")
@@ -271,7 +298,13 @@ class MySQLUploader:
             raise ConnectionError(f"连接数据库失败: {str(e)}")
 
     def _check_database_exists(self, db_name: str) -> bool:
-        """检查数据库是否存在"""
+        """
+        检查数据库是否存在
+
+        :param db_name: 数据库名称
+        :return: 存在返回True，否则返回False
+        :raises: 可能抛出数据库相关异常
+        """
         db_name = self._validate_identifier(db_name)
         sql = "SELECT SCHEMA_NAME FROM INFORMATION_SCHEMA.SCHEMATA WHERE SCHEMA_NAME = %s"
 
@@ -287,7 +320,12 @@ class MySQLUploader:
             raise
 
     def _create_database(self, db_name: str):
-        """创建数据库"""
+        """
+        创建数据库
+
+        :param db_name: 要创建的数据库名称
+        :raises: 可能抛出数据库相关异常
+        """
         db_name = self._validate_identifier(db_name)
         sql = f"CREATE DATABASE IF NOT EXISTS `{db_name}` CHARACTER SET {self.charset} COLLATE {self.collation}"
 
@@ -329,18 +367,17 @@ class MySQLUploader:
         elif partition_by == 'month':
             return f"{table_name}_{date_obj.year}_{date_obj.month:02d}"
         else:
-            error_msg = "partition_by must be 'year' or 'month'"
+            error_msg = "分表方式必须是 'year' 或 'month'"
             logger.error(error_msg)
             raise ValueError(error_msg)
 
     def _validate_identifier(self, identifier: str) -> str:
         """
-        验证并清理数据库标识符(数据库名、表名、列名)
-        防止SQL注入和非法字符
+        验证并清理数据库标识符(表名、列名等)
 
         :param identifier: 要验证的标识符
         :return: 清理后的安全标识符
-        :raises ValueError: 如果标识符无效
+        :raises ValueError: 当标识符无效时抛出
         """
         if not identifier or not isinstance(identifier, str):
             error_msg = f"无效的标识符: {identifier}"
@@ -348,7 +385,11 @@ class MySQLUploader:
             raise ValueError(error_msg)
 
         # 移除非法字符，只保留字母、数字、下划线和美元符号
-        cleaned = re.sub(r'[^\w\u4e00-\u9fff$]', '', identifier)
+        cleaned = re.sub(r'[^\w\u4e00-\u9fff$]', '_', identifier)
+
+        # 将多个连续的下划线替换为单个下划线, 移除开头和结尾的下划线
+        cleaned = re.sub(r'_+', '_', cleaned).strip('_')
+
         if not cleaned:
             error_msg = f"无法清理异常标识符: {identifier}"
             logger.error(error_msg)
@@ -366,7 +407,14 @@ class MySQLUploader:
         return cleaned
 
     def _check_table_exists(self, db_name: str, table_name: str) -> bool:
-        """检查表是否存在"""
+        """
+        检查表是否存在
+
+        :param db_name: 数据库名
+        :param table_name: 表名
+        :return: 存在返回True，否则返回False
+        :raises: 可能抛出数据库相关异常
+        """
         cache_key = f"{db_name}.{table_name}"
         if cache_key in self._table_metadata_cache:
             cached_time, result = self._table_metadata_cache[cache_key]
@@ -410,9 +458,11 @@ class MySQLUploader:
         :param db_name: 数据库名
         :param table_name: 表名
         :param set_typ: 列名和数据类型字典 {列名: 数据类型}
-        :param primary_keys: 主键列列表
-        :param date_column: 日期列名，如果存在将设置为索引
-        :param indexes: 需要创建索引的列列表
+        :param primary_keys: 主键列列表，可选
+        :param date_column: 日期列名，可选，如果存在将设置为索引
+        :param indexes: 需要创建索引的列列表，可选
+        :param allow_null: 是否允许空值，默认为False
+        :raises: 可能抛出数据库相关异常
         """
         db_name = self._validate_identifier(db_name)
         table_name = self._validate_identifier(table_name)
@@ -501,7 +551,14 @@ class MySQLUploader:
             raise
 
     def _validate_datetime(self, value, date_type=False):
-        """date_type: 返回字符串类型或者日期类型"""
+        """
+        验证并标准化日期时间格式
+
+        :param value: 日期时间值
+        :param date_type: 是否返回日期类型(True)或字符串(False)
+        :return: 标准化后的日期时间字符串或日期对象
+        :raises ValueError: 当日期格式无效时抛出
+        """
         formats = [
             '%Y-%m-%d %H:%M:%S',
             '%Y-%m-%d',
@@ -524,26 +581,49 @@ class MySQLUploader:
                 continue
         raise ValueError(f"无效的日期格式2: {value}")
 
-    def _validate_value(self, value: Any, column_type: str) -> Any:
+    def _validate_value(self, value: Any, column_type: str, allow_null: bool) -> Any:
         """
-        验证并清理数据值，根据列类型进行适当转换
+        根据列类型验证并转换数据值
 
         :param value: 要验证的值
         :param column_type: 列的数据类型
-        :return: 清理后的值
-        :raises ValueError: 如果值转换失败
+        :param allow_null: 是否允许空值
+        :return: 转换后的值
+        :raises ValueError: 当值转换失败时抛出
         """
         if value is None:
+            if not allow_null:
+                return 'none'
             return None
 
         try:
             column_type_lower = column_type.lower()
 
-            if 'int' in column_type_lower:
-                if isinstance(value, (str, bytes)) and not value.strip().isdigit():
-                    raise ValueError("非数字字符串无法转换为整数")
-                return int(value)
+            # 处理百分比值
+            if isinstance(value, str):
+                if value.endswith('%'):
+                    try:
+                        # 移除百分号并转换为小数
+                        percent_value = float(value.strip().replace('%', ''))
+                        decimal_value = percent_value / 100
+                        return decimal_value
+                    except ValueError:
+                        pass  # 如果不是有效的百分比数字，继续正常处理
+
+            elif 'int' in column_type_lower:
+                if isinstance(value, str):
+                    # 移除可能的逗号和空格
+                    value = value.replace(',', '').strip()
+                    # 尝试转换为浮点数再转整数
+                    try:
+                        return int(float(value))
+                    except ValueError:
+                        raise ValueError(f"`{value}` 无法转为整数")
+                return int(value) if value is not None else None
             elif any(t in column_type_lower for t in ['float', 'double', 'decimal']):
+                if isinstance(value, str):
+                    # 处理可能包含逗号的数字字符串
+                    value = value.replace(',', '')
                 return float(value) if value is not None else None
             elif '日期' in column_type_lower or 'time' in column_type_lower:
                 if isinstance(value, (datetime.datetime, pd.Timestamp)):
@@ -560,17 +640,23 @@ class MySQLUploader:
                     return value.replace('\\', '\\\\').replace("'", "\\'")
                 return str(value)
             elif 'json' in column_type_lower:
-                import json
                 return json.dumps(value) if value is not None else None
             else:
                 return value
         except (ValueError, TypeError) as e:
-            error_msg = f"数据类型转换异常 {value} to type {column_type}: {str(e)}"
+            error_msg = f"转换异常 -> 无法将 `{value}` 的数据类型转为: `{column_type}` -> {str(e)}"
             logger.error(error_msg)
             raise ValueError(error_msg)
 
     def _get_table_columns(self, db_name: str, table_name: str) -> Dict[str, str]:
-        """获取表的列名和数据类型"""
+        """
+        获取表的列名和数据类型
+
+        :param db_name: 数据库名
+        :param table_name: 表名
+        :return: 列名和数据类型字典 {列名: 数据类型}
+        :raises: 可能抛出数据库相关异常
+        """
         db_name = self._validate_identifier(db_name)
         table_name = self._validate_identifier(table_name)
         sql = """
@@ -604,7 +690,8 @@ class MySQLUploader:
             auto_create: bool,
             date_column: Optional[str],
             indexes: Optional[List[str]],
-            batch_id: Optional[str] = None
+            batch_id: Optional[str] = None,
+            update_on_duplicate: bool = False
     ):
         """实际执行表上传的方法"""
         # 检查表是否存在
@@ -634,19 +721,26 @@ class MySQLUploader:
         # 插入数据
         self._insert_data(
             db_name, table_name, data, set_typ,
-            check_duplicate, duplicate_columns
+            check_duplicate, duplicate_columns,
+            batch_id=batch_id,
+            update_on_duplicate=update_on_duplicate
         )
 
     def _infer_data_type(self, value: Any) -> str:
         """
-        根据值推断合适的数据类型
+        根据值推断合适的MySQL数据类型
 
         :param value: 要推断的值
         :return: MySQL数据类型字符串
         """
-        if value is None:
+        if value is None or str(value).lower() in ['', 'none', 'nan']:
             return 'VARCHAR(255)'  # 默认字符串类型
 
+        # 检查是否是百分比字符串
+        if isinstance(value, str):
+            if value.endswith('%'):
+                return 'DECIMAL(10,4)'  # 百分比统一使用DECIMAL(10,4)
+
         if isinstance(value, bool):
             return 'TINYINT(1)'
         elif isinstance(value, int):
@@ -692,6 +786,26 @@ class MySQLUploader:
         else:
             return 'VARCHAR(255)'
 
+    def normalize_column_names(self, data: Union[pd.DataFrame, List[Dict[str, Any]]]) -> Union[
+        pd.DataFrame, List[Dict[str, Any]]]:
+        """
+        1. pandas：规范化列名
+        2. 字典列表：规范化每个字典的键
+
+        参数：
+            data: 输入数据，支持两种类型：
+                  - pandas.DataFrame：将规范化其列名
+                  - List[Dict[str, Any]]：将规范化列表中每个字典的键
+        """
+        if isinstance(data, pd.DataFrame):
+            # 处理DataFrame
+            data.columns = [self._validate_identifier(col) for col in data.columns]
+            return data
+        elif isinstance(data, list):
+            # 处理字典列表
+            return [{self._validate_identifier(k): v for k, v in item.items()} for item in data]
+        return data
+
     def _prepare_data(
             self,
             data: Union[Dict, List[Dict], pd.DataFrame],
@@ -701,11 +815,11 @@ class MySQLUploader:
         """
         准备要上传的数据，验证并转换数据类型
 
-        :param data: 输入数据
+        :param data: 输入数据，可以是字典、字典列表或DataFrame
         :param set_typ: 列名和数据类型字典 {列名: 数据类型}
         :param allow_null: 是否允许空值
-        :return: 待上传的数据列表和对应的数据类型
-        :raises ValueError: 如果数据验证失败
+        :return: 元组(准备好的数据列表, 过滤后的列类型字典)
+        :raises ValueError: 当数据验证失败时抛出
         """
         # 统一数据格式为字典列表
         if isinstance(data, pd.DataFrame):
@@ -726,6 +840,9 @@ class MySQLUploader:
             logger.error(error_msg)
             raise ValueError(error_msg)
 
+        # 统一处理原始数据中列名的特殊字符
+        data = self.normalize_column_names(data)
+
         # 将set_typ的键转为小写
         set_typ = {k.lower(): v for k, v in set_typ.items()}
 
@@ -745,11 +862,11 @@ class MySQLUploader:
                 if sample_values:
                     inferred_type = self._infer_data_type(sample_values[0])
                     filtered_set_typ[col] = inferred_type
-                    logger.debug(f"自动推断列'{col}'的数据类型为: {inferred_type}")
+                    logger.debug(f"自动推断列 `{col}` 的数据类型为: {inferred_type}")
                 else:
                     # 没有样本值，使用默认类型
                     filtered_set_typ[col] = 'VARCHAR(255)'
-                    logger.debug(f"为列'{col}'使用默认数据类型: VARCHAR(255)")
+                    logger.debug(f"列 `{col}` 使用默认数据类型: VARCHAR(255)")
 
         prepared_data = []
         for row_idx, row in enumerate(data, 1):
@@ -761,15 +878,15 @@ class MySQLUploader:
 
                 if col_name not in row:
                     if not allow_null:
-                        error_msg = f"Row {row_idx}: Missing required column '{col_name}' in data"
+                        error_msg = f"行号:{row_idx} -> 缺失列: `{col_name}`"
                         logger.error(error_msg)
                         raise ValueError(error_msg)
                     prepared_row[col_name] = None
                 else:
                     try:
-                        prepared_row[col_name] = self._validate_value(row[col_name], filtered_set_typ[col_name])
+                        prepared_row[col_name] = self._validate_value(row[col_name], filtered_set_typ[col_name], allow_null)
                     except ValueError as e:
-                        error_msg = f"Row {row_idx}, column '{col_name}': {str(e)}"
+                        error_msg = f"行号:{row_idx}, 列名:`{col_name}`-> 报错: {str(e)}"
                         logger.error(error_msg)
                         raise ValueError(error_msg)
             prepared_data.append(prepared_row)
@@ -790,10 +907,26 @@ class MySQLUploader:
             partition_by: Optional[str] = None,
             partition_date_column: str = '日期',
             auto_create: bool = True,
-            indexes: Optional[List[str]] = None
+            indexes: Optional[List[str]] = None,
+            update_on_duplicate: bool = False
     ):
         """
-        上传数据到数据库
+        上传数据到数据库的主入口方法
+
+        :param db_name: 数据库名
+        :param table_name: 表名
+        :param data: 要上传的数据
+        :param set_typ: 列名和数据类型字典 {列名: 数据类型}
+        :param primary_keys: 主键列列表，可选
+        :param check_duplicate: 是否检查重复数据，默认为False
+        :param duplicate_columns: 用于检查重复的列，可选
+        :param allow_null: 是否允许空值，默认为False
+        :param partition_by: 分表方式('year'或'month')，可选
+        :param partition_date_column: 用于分表的日期列名，默认为'日期'
+        :param auto_create: 表不存在时是否自动创建，默认为True
+        :param indexes: 需要创建索引的列列表，可选
+        :param update_on_duplicate: 遇到重复数据时是否更新旧数据(默认为False)
+        :raises: 可能抛出各种验证和数据库相关异常
         """
         upload_start = time.time()
         initial_row_count = len(data) if hasattr(data, '__len__') else 1
@@ -802,26 +935,28 @@ class MySQLUploader:
         success_flag = False
 
         logger.info("开始上传数据", {
-            'batch_id': batch_id,
-            'database': db_name,
-            'table': table_name,
-            'partition_by': partition_by,
-            'check_duplicate': check_duplicate,
-            'row_count': len(data) if hasattr(data, '__len__') else 1,
-            'auto_create': auto_create
+            '批次号': batch_id,
+            '库': db_name,
+            '表': table_name,
+            '分表方式': partition_by,
+            '排重': check_duplicate,
+            '总计行数': len(data) if hasattr(data, '__len__') else 1,
+            '自动建表': auto_create
         })
 
         try:
-            # 验证参数
-            if not set_typ:
-                error_msg = "列的数据类型缺失"
-                logger.error(error_msg)
-                raise ValueError(error_msg)
+            # # 验证参数
+            # if not set_typ:
+            #     error_msg = "列的数据类型缺失"
+            #     logger.error(error_msg)
+            #     raise ValueError(error_msg)
 
-            if partition_by and partition_by not in ['year', 'month']:
-                error_msg = "分表方式必须是 'year' 或 'month'"
-                logger.error(error_msg)
-                raise ValueError(error_msg)
+            if partition_by:
+                partition_by = str(partition_by).lower()
+                if partition_by not in ['year', 'month']:
+                    error_msg = "分表方式必须是 'year' 或 'month'"
+                    logger.error(error_msg)
+                    raise ValueError(error_msg)
 
             # 准备数据
             prepared_data, filtered_set_typ = self._prepare_data(data, set_typ, allow_null)
@@ -867,7 +1002,7 @@ class MySQLUploader:
                             db_name, part_table, part_data, filtered_set_typ,
                             primary_keys, check_duplicate, duplicate_columns,
                             allow_null, auto_create, partition_date_column,
-                            indexes, batch_id
+                            indexes, batch_id, update_on_duplicate
                         )
                     except Exception as e:
                         logger.error("分表上传失败", {
@@ -881,7 +1016,7 @@ class MySQLUploader:
                     db_name, table_name, prepared_data, filtered_set_typ,
                     primary_keys, check_duplicate, duplicate_columns,
                     allow_null, auto_create, partition_date_column,
-                    indexes, batch_id
+                    indexes, batch_id, update_on_duplicate
                 )
 
             success_flag = True
@@ -892,12 +1027,12 @@ class MySQLUploader:
                 'error_type': type(e).__name__
             })
         finally:
-            elapsed = time.time() - upload_start
+            elapsed = round(time.time() - upload_start, 2)
             logger.info("上传处理完成", {
-                'batch_id': batch_id,
+                '批次号': batch_id,
                 'success': success_flag,
-                'time_elapsed': elapsed,
-                'initial_row_count': initial_row_count
+                '耗时': elapsed,
+                '数据行': initial_row_count
             })
 
     def _insert_data(
@@ -909,20 +1044,21 @@ class MySQLUploader:
             check_duplicate: bool = False,
             duplicate_columns: Optional[List[str]] = None,
             batch_size: int = 1000,
-            batch_id: Optional[str] = None
+            batch_id: Optional[str] = None,
+            update_on_duplicate: bool = False
     ):
         """
-        插入数据到表中
-
-        参数:
-        db_name: 数据库名
-        table_name: 表名
-        data: 要插入的数据列表
-        set_typ: 列名和数据类型字典 {列名: 数据类型}
-        check_duplicate: 是否检查重复
-        duplicate_columns: 用于检查重复的列(为空时检查所有列)
-        batch_size: 批量插入大小
-        batch_id: 批次ID用于日志追踪
+        实际执行数据插入的方法
+
+        :param db_name: 数据库名
+        :param table_name: 表名
+        :param data: 要插入的数据列表
+        :param set_typ: 列名和数据类型字典 {列名: 数据类型}
+        :param check_duplicate: 是否检查重复数据，默认为False
+        :param duplicate_columns: 用于检查重复的列，可选
+        :param batch_size: 批量插入大小，默认为1000
+        :param update_on_duplicate: 遇到重复数据时是否更新旧数据(默认为False)
+        :param batch_id: 批次ID用于日志追踪，可选
         """
         if not data:
             return
@@ -954,8 +1090,21 @@ class MySQLUploader:
 
             where_clause = " AND ".join(conditions)
 
-            sql = f"""
+            if update_on_duplicate:
+                # 更新模式 - 使用ON DUPLICATE KEY UPDATE语法
+                update_clause = ", ".join([f"`{col}` = VALUES(`{col}`)" for col in all_columns])
+                sql = f"""
                     INSERT INTO `{db_name}`.`{table_name}` 
+                    (`{'`,`'.join(safe_columns)}`) 
+                    VALUES ({placeholders})
+                    ON DUPLICATE KEY UPDATE {update_clause}
+                    """
+
+                # 注意：在update_on_duplicate模式下，row_values只需要插入数据，不需要排重列值
+                def prepare_values(row):
+                    return [row.get(col) for col in all_columns]
+            else:
+                sql = f"""INSERT INTO `{db_name}`.`{table_name}` 
                     (`{'`,`'.join(safe_columns)}`) 
                     SELECT {placeholders}
                     FROM DUAL
@@ -964,6 +1113,10 @@ class MySQLUploader:
                         WHERE {where_clause}
                     )
                     """
+
+                # 在check_duplicate模式下，row_values需要插入数据+排重列值
+                def prepare_values(row):
+                    return [row.get(col) for col in all_columns] + [row.get(col) for col in duplicate_columns]
         else:
             sql = f"""
                 INSERT INTO `{db_name}`.`{table_name}` 
@@ -971,6 +1124,10 @@ class MySQLUploader:
                 VALUES ({placeholders})
                 """
 
+            # 普通模式下，row_values只需要插入数据
+            def prepare_values(row):
+                return [row.get(col) for col in all_columns]
+
         total_inserted = 0
         total_skipped = 0
         total_failed = 0  # 失败计数器
@@ -986,11 +1143,7 @@ class MySQLUploader:
                     for row in batch:
                         try:
                             # 准备参数
-                            row_values = [row.get(col) for col in all_columns]
-                            # 如果是排重检查，添加排重列值
-                            if check_duplicate:
-                                row_values += [row.get(col) for col in duplicate_columns]
-
+                            row_values = prepare_values(row)
                             cursor.execute(sql, row_values)
                             successful_rows += 1
                             conn.commit()  # 每次成功插入后提交
@@ -1001,14 +1154,14 @@ class MySQLUploader:
 
                             # 记录失败行详细信息
                             error_details = {
-                                'batch_id': batch_id,
-                                'database': db_name,
-                                'table': table_name,
+                                '批次号': batch_id,
+                                '库': db_name,
+                                '表': table_name,
                                 'error_type': type(e).__name__,
                                 'error_message': str(e),
-                                'column_types': set_typ,
-                                'duplicate_check': check_duplicate,
-                                'duplicate_columns': duplicate_columns
+                                '数据类型': set_typ,
+                                '是否排重': check_duplicate,
+                                '排重列': duplicate_columns
                             }
                             logger.error(f"单行插入失败: {error_details}")
                             continue  # 跳过当前行，继续处理下一行
@@ -1022,28 +1175,31 @@ class MySQLUploader:
                     else:
                         total_inserted += successful_rows
 
-                    batch_elapsed = time.time() - batch_start
+                    batch_elapsed = round(time.time() - batch_start, 2)
                     batch_info = {
-                        'batch_id': batch_id,
+                        '批次号': batch_id,
                         'batch_index': i // batch_size + 1,
                         'total_batches': (len(data) + batch_size - 1) // batch_size,
                         'batch_size': len(batch),
                         'successful_rows': successful_rows,
                         'failed_rows': len(batch) - successful_rows,
-                        'time_elapsed': batch_elapsed,
+                        '耗时': batch_elapsed,
                         'rows_per_second': successful_rows / batch_elapsed if batch_elapsed > 0 else 0
                     }
                     logger.debug(f"批次处理完成 {batch_info}")
 
         logger.info("数据插入完成", {
-            'total_rows': len(data),
-            'inserted_rows': total_inserted,
-            'skipped_rows': total_skipped,
-            'failed_rows': total_failed
+            '总数据行': len(data),
+            '插入行数': total_inserted,
+            '跳过行数': total_skipped,
+            '失败行数': total_failed
         })
 
     def close(self):
-        """关闭连接池并记录最终指标"""
+        """
+        关闭连接池并清理资源
+        :raises: 可能抛出关闭连接时的异常
+        """
         close_start = time.time()
 
         try:
@@ -1060,18 +1216,22 @@ class MySQLUploader:
 
                 elapsed = round(time.time() - close_start, 2)
                 logger.info("连接池已关闭", {
-                    'close_time_elapsed': elapsed
+                    '耗时': elapsed
                 })
         except Exception as e:
             elapsed = round(time.time() - close_start, 2)
             logger.error("关闭连接池失败", {
                 'error': str(e),
-                'close_time_elapsed': elapsed
+                '耗时': elapsed
             })
             raise
 
     def _check_pool_health(self):
-        """定期检查连接池健康状态"""
+        """
+        检查连接池健康状态
+
+        :return: 连接池健康返回True，否则返回False
+        """
         try:
             conn = self.pool.connection()
             conn.ping(reconnect=True)

{mdbq-3.9.8.dist-info → mdbq-3.9.10.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: mdbq
-Version: 3.9.8
+Version: 3.9.10
 Home-page: https://pypi.org/project/mdbq
 Author: xigua, 
 Author-email: 2587125111@qq.com

{mdbq-3.9.8.dist-info → mdbq-3.9.10.dist-info}/RECORD

@@ -1,5 +1,5 @@
 mdbq/__init__.py,sha256=Il5Q9ATdX8yXqVxtP_nYqUhExzxPC_qk_WXQ_4h0exg,16
-mdbq/__version__.py,sha256=cWQRevRu-qUG7cakaKmnu0uw_arEQikDebOc5jW6PWs,17
+mdbq/__version__.py,sha256=83jYP6xnYylgp029cctX2BP7k_exd-phUiwATgIjhH0,18
 mdbq/aggregation/__init__.py,sha256=EeDqX2Aml6SPx8363J-v1lz0EcZtgwIBYyCJV6CcEDU,40
 mdbq/aggregation/optimize.py,sha256=2oalzD9weZhDclUC22OLxYa8Zj7KnmsGUoUau_Jlyc4,19796
 mdbq/aggregation/query_data.py,sha256=5_OzjGR5Sq00q-EgAYmSE5V9i4Solw9y4hkldl4mvt8,179808
@@ -9,10 +9,10 @@ mdbq/log/__init__.py,sha256=Mpbrav0s0ifLL7lVDAuePEi1hJKiSHhxcv1byBKDl5E,15
 mdbq/log/mylogger.py,sha256=jHCVO7KPQrg2kcCaIrakHivZmFBJyy-24sIn2rsbK4Y,24440
 mdbq/log/spider_logging.py,sha256=-ozWWEGm3HVv604ozs_OOvVwumjokmUPwbaodesUrPY,1664
 mdbq/mysql/__init__.py,sha256=A_DPJyAoEvTSFojiI2e94zP0FKtCkkwKP1kYUCSyQzo,11
-mdbq/mysql/deduplicator.py,sha256=do1OqZtGwdf_KrRU_3LK4ZlKpQFsH6Pjef8eguoY4Xo,22009
+mdbq/mysql/deduplicator.py,sha256=brhX3eyE8-kn3nAYweKfBbAkXiNcyw_pL4CTyPqmPBg,21983
 mdbq/mysql/mysql.py,sha256=jTcizvUtRdwMhWK2i_LA9yDPmcifLjUzVhwTbC3wfJk,119785
 mdbq/mysql/s_query.py,sha256=X055aLRAgxVvueXx4NbfNjp6MyBI02_XBb1pTKw09L0,8660
-mdbq/mysql/uploader.py,sha256=Zfp1xeCD0oNuBMnSmg2AKFqa2SToPkKstBUOH_h5vBM,45260
+mdbq/mysql/uploader.py,sha256=V23PAzT59lMUqijkUiwV6a1qNwk9T76k8HKxY8fYW9w,52140
 mdbq/other/__init__.py,sha256=jso1oHcy6cJEfa7udS_9uO5X6kZLoPBF8l3wCYmr5dM,18
 mdbq/other/download_sku_picture.py,sha256=YU8DxKMXbdeE1OOKEA848WVp62jYHw5O4tXTjUdq9H0,44832
 mdbq/other/otk.py,sha256=iclBIFbQbhlqzUbcMMoePXBpcP1eZ06ZtjnhcA_EbmE,7241
@@ -25,7 +25,7 @@ mdbq/redis/__init__.py,sha256=YtgBlVSMDphtpwYX248wGge1x-Ex_mMufz4-8W0XRmA,12
 mdbq/redis/getredis.py,sha256=Uk8-cOWT0JU1qRyIVqdbYokSLvkDIAfcokmYj1ebw8k,24104
 mdbq/spider/__init__.py,sha256=RBMFXGy_jd1HXZhngB2T2XTvJqki8P_Fr-pBcwijnew,18
 mdbq/spider/aikucun.py,sha256=OhyEv1VyAKTOHjLDM37iNDQeRg5OnrNoKODoG2VxHes,19806
-mdbq-3.9.8.dist-info/METADATA,sha256=Ebjy7F7OmVoOel2poNGxFCX7l6AniJrupqwmuqTOJpA,363
-mdbq-3.9.8.dist-info/WHEEL,sha256=jB7zZ3N9hIM9adW7qlTAyycLYW9npaWKLRzaoVcLKcM,91
-mdbq-3.9.8.dist-info/top_level.txt,sha256=2FQ-uLnCSB-OwFiWntzmwosW3X2Xqsg0ewh1axsaylA,5
-mdbq-3.9.8.dist-info/RECORD,,
+mdbq-3.9.10.dist-info/METADATA,sha256=Ln51lgeqZn0zAjgLUKXaMNJ5ZXCkX3Eyu0iao37_IQw,364
+mdbq-3.9.10.dist-info/WHEEL,sha256=jB7zZ3N9hIM9adW7qlTAyycLYW9npaWKLRzaoVcLKcM,91
+mdbq-3.9.10.dist-info/top_level.txt,sha256=2FQ-uLnCSB-OwFiWntzmwosW3X2Xqsg0ewh1axsaylA,5
+mdbq-3.9.10.dist-info/RECORD,,