DataJoint:删除表时的IntegrityError

Question

我正在设计我们的数据库，并且在common_exp模式中有一个名为Session的表，其定义如下：

@schema
class Session(dj.Manual):
    definition = """ # Information about the session and experimental setup
    -> common_mice.Mouse
    day             : date           # Date of the experimental session (YYYY-MM-DD)
    trial           : tinyint        # Counter of experimental sessions on the same day (base 1)
    ---
    id              : varchar(128)   # Unique identifier
    path            : varchar(256)   # Relative path of this session on the server
    counter         : smallint       # Overall counter of all sessions across mice (base 0)
    experimenter    : varchar(128)   # Who actually performed the experiment, must be a username from Investigator
    -> Anesthesia
    -> Setup
    -> Task
    notes           : varchar(2048)  # description of important things that happened
    """

我想要更改一些属性的名称，因此想要删除该表。然而，我遇到了这个错误：

common_exp.Session().drop()
`common_exp`.`session` (0 tuples)
Proceed? [yes, No]: >? yes

Traceback (most recent call last):
  File "C:\Anaconda3\envs\datajoint_wahl\lib\site-packages\IPython\core\interactiveshell.py", line 3441, in run_code
    exec(code_obj, self.user_global_ns, self.user_ns)
  File "<ipython-input-4-bce682713228>", line 1, in <module>
    common_exp.Session().drop()
  File "C:\Anaconda3\envs\datajoint_wahl\lib\site-packages\datajoint\table.py", line 474, in drop
    FreeTable(self.connection, table).drop_quick()
  File "C:\Anaconda3\envs\datajoint_wahl\lib\site-packages\datajoint\table.py", line 450, in drop_quick
    self.connection.query(query)
  File "C:\Anaconda3\envs\datajoint_wahl\lib\site-packages\datajoint\connection.py", line 302, in query
    self._execute_query(cursor, query, args, suppress_warnings)
  File "C:\Anaconda3\envs\datajoint_wahl\lib\site-packages\datajoint\connection.py", line 268, in _execute_query
    raise translate_query_error(err, query)
  File "C:\Anaconda3\envs\datajoint_wahl\lib\site-packages\datajoint\connection.py", line 266, in _execute_query
    cursor.execute(query, args)
  File "C:\Anaconda3\envs\datajoint_wahl\lib\site-packages\pymysql\cursors.py", line 148, in execute
    result = self._query(query)
  File "C:\Anaconda3\envs\datajoint_wahl\lib\site-packages\pymysql\cursors.py", line 310, in _query
    conn.query(q)
  File "C:\Anaconda3\envs\datajoint_wahl\lib\site-packages\pymysql\connections.py", line 548, in query
    self._affected_rows = self._read_query_result(unbuffered=unbuffered)
  File "C:\Anaconda3\envs\datajoint_wahl\lib\site-packages\pymysql\connections.py", line 775, in _read_query_result
    result.read()
  File "C:\Anaconda3\envs\datajoint_wahl\lib\site-packages\pymysql\connections.py", line 1156, in read
    first_packet = self.connection._read_packet()
  File "C:\Anaconda3\envs\datajoint_wahl\lib\site-packages\pymysql\connections.py", line 725, in _read_packet
    packet.raise_for_error()
  File "C:\Anaconda3\envs\datajoint_wahl\lib\site-packages\pymysql\protocol.py", line 221, in raise_for_error
    err.raise_mysql_exception(self._data)
  File "C:\Anaconda3\envs\datajoint_wahl\lib\site-packages\pymysql\err.py", line 143, in raise_mysql_exception
    raise errorclass(errno, errval)
pymysql.err.IntegrityError: (1217, 'Cannot delete or update a parent row: a foreign key constraint fails')

正如您所看到的，该表是空的，并且没有进一步的依赖关系。错误消息也没有告诉我哪个键产生了问题，或者哪个表产生了问题，所以我有点困惑问题可能出在哪里。

我使用root帐户访问数据库，所以权限应该不是问题。从其他模式中删除表是可行的，只是这个模式会产生这个错误。

Answer 1

谢谢你的问题-

根据表中没有记录，但删除失败并出现此错误这一事实，似乎确实存在使用外键定义回会话表的下游表，因此删除会话表将使这些表成为“孤立”表，从而创建完整性错误。从示例中看不出事实并非如此。

以下SQL查询在模式级别检查正向/反向依赖关系(回答问题：“哪个模式取决于哪些其他模式”)：

    SELECT distinct(UNIQUE_CONSTRAINT_SCHEMA)
    FROM information_schema.REFERENTIAL_CONSTRAINTS
    where constraint_schema='my_schema'; -- forward dependencies for 'my_schema'

    SELECT distinct(CONSTRAINT_SCHEMA)
    FROM information_schema.REFERENTIAL_CONSTRAINTS
    where unique_constraint_schema='my_schema'; -- reverse dependencies for 'my_schema'

不幸的是，我记不起确切的列了，但是信息模式中也提供了表级信息。

Answer 2

此错误表示在删除Session之前，需要删除common_exp.Session下游的一个依赖表。

通常，DataJoint将级联并删除所有下游表。但是，有时DataJoint看不到依赖表，因为它们位于尚未加载的不同模式中。您将需要加载进行此引用的架构，以便允许DataJoint级联drop。您可以使用dj.list_schemas()查看所有可用的模式。