Oracle¶

Support for the Oracle Database database.

The following table summarizes current support levels for database release versions.

**Supported Oracle Database versions**¶
Support type	Versions
Supported version	11+
Best effort	9+

DBAPI Support¶

The following dialect/DBAPI options are available. Please refer to individual DBAPI sections for connect information.

python-oracledb
cx-Oracle

自动增量行为¶

Auto Increment Behavior

中文

SQLAlchemy 表对象通常假定包含整数主键的表具有“自动递增”行为，意味着它们可以在插入时自动生成主键值。在 Oracle 数据库中，有两种可选方式可用，分别是使用 IDENTITY 列（仅适用于 Oracle 数据库 12 及以上版本）或将 SEQUENCE 与列关联。

英文

SQLAlchemy Table objects which include integer primary keys are usually assumed to have “autoincrementing” behavior, meaning they can generate their own primary key values upon INSERT. For use within Oracle Database, two options are available, which are the use of IDENTITY columns (Oracle Database 12 and above only) or the association of a SEQUENCE with the column.

指定 GENERATED AS IDENTITY（Oracle Database 12 及以上版本）¶

Specifying GENERATED AS IDENTITY (Oracle Database 12 and above)

中文

从版本 12 开始，Oracle 数据库可以使用 Identity 来指定自动递增行为，如下所示:

t = Table(
    "mytable",
    metadata,
    Column("id", Integer, Identity(start=3), primary_key=True),
    Column(...),
    ...,
)

上面 Table 对象的 CREATE TABLE 语句如下：

CREATE TABLE mytable (
    id INTEGER GENERATED BY DEFAULT AS IDENTITY (START WITH 3),
    ...,
    PRIMARY KEY (id)
)

Identity 对象支持多种选项来控制列的“自动递增”行为，例如起始值、递增值等。除了标准选项，Oracle 数据库还支持将 Identity.always 设置为 None，以使用默认生成模式，这将在 DDL 中渲染为 GENERATED AS IDENTITY。Oracle 数据库还支持通过方言关键字参数指定两个自定义选项：

oracle_on_null：设置为 True 时，与 ‘BY DEFAULT’ 自动递增列一起渲染 ON NULL。
oracle_order：设置为 True 时，渲染 ORDER 关键字，表示该自动递增列是有序的。对于使用 Oracle Real Application Clusters (RAC) 的环境，可能需要提供确定性排序。

英文

Starting from version 12, Oracle Database can make use of identity columns using the Identity to specify the autoincrementing behavior:

t = Table(
    "mytable",
    metadata,
    Column("id", Integer, Identity(start=3), primary_key=True),
    Column(...),
    ...,
)

The CREATE TABLE for the above Table object would be:

CREATE TABLE mytable (
    id INTEGER GENERATED BY DEFAULT AS IDENTITY (START WITH 3),
    ...,
    PRIMARY KEY (id)
)

The Identity object support many options to control the “autoincrementing” behavior of the column, like the starting value, the incrementing value, etc. In addition to the standard options, Oracle Database supports setting Identity.always to None to use the default generated mode, rendering GENERATED AS IDENTITY in the DDL. Oracle Database also supports two custom options specified using dialect kwargs:

oracle_on_null: when set to True renders ON NULL in conjunction with a ‘BY DEFAULT’ identity column.
oracle_order: when True, renders the ORDER keyword, indicating the identity is definitively ordered. May be necessary to provide deterministic ordering using Oracle Real Application Clusters (RAC).

使用 SEQUENCE（所有 Oracle Database 版本）¶

Using a SEQUENCE (all Oracle Database versions)

中文

Oracle 数据库的旧版本没有“自动递增”特性：SQLAlchemy 依赖序列来生成这些值。在较早版本的 Oracle 数据库中，必须明确指定序列以启用自动递增。这与大多数文档示例的假设（使用支持自动递增的数据库）有所不同。要指定序列，可以使用 sqlalchemy.schema.Sequence 对象并将其传递给 Column 构造函数:

t = Table(
    "mytable",
    metadata,
    Column("id", Integer, Sequence("id_seq", start=1), primary_key=True),
    Column(...),
    ...,
)

使用表反射时也需要进行此步骤，即 autoload_with=engine:

t = Table(
    "mytable",
    metadata,
    Column("id", Integer, Sequence("id_seq", start=1), primary_key=True),
    autoload_with=engine,
)

除了标准选项，Oracle 数据库还支持通过方言关键字参数指定以下自定义选项：

oracle_order：设置为 True 时，渲染 ORDER 关键字，表示该序列是有序的。对于使用 Oracle RAC 的环境，可能需要提供确定性排序。

在 1.4 版本发生变更: 增加了 Identity 构造函数用于 Column 中，指定自动递增列的选项。

英文

Older version of Oracle Database had no “autoincrement” feature: SQLAlchemy relies upon sequences to produce these values. With the older Oracle Database versions, a sequence must always be explicitly specified to enable autoincrement. This is divergent with the majority of documentation examples which assume the usage of an autoincrement-capable database. To specify sequences, use the sqlalchemy.schema.Sequence object which is passed to a Column construct:

t = Table(
    "mytable",
    metadata,
    Column("id", Integer, Sequence("id_seq", start=1), primary_key=True),
    Column(...),
    ...,
)

This step is also required when using table reflection, i.e. autoload_with=engine:

t = Table(
    "mytable",
    metadata,
    Column("id", Integer, Sequence("id_seq", start=1), primary_key=True),
    autoload_with=engine,
)

In addition to the standard options, Oracle Database supports the following custom option specified using dialect kwargs:

oracle_order: when True, renders the ORDER keyword, indicating the sequence is definitively ordered. May be necessary to provide deterministic ordering using Oracle RAC.

在 1.4 版本发生变更: Added Identity construct in a Column to specify the option of an autoincrementing column.

事务隔离级别/自动提交¶

Transaction Isolation Level / Autocommit

中文

Oracle 数据库支持“READ COMMITTED”和“SERIALIZABLE”隔离级别。AUTOCOMMIT 隔离级别也被 python-oracledb 和 cx_Oracle 方言支持。

可以通过每连接执行选项来设置隔离级别:

connection = engine.connect()
connection = connection.execution_options(isolation_level="AUTOCOMMIT")

对于 READ COMMITTED 和 SERIALIZABLE，Oracle 数据库方言通过 ALTER SESSION 设置会话级别的隔离级别，当连接返回连接池时会恢复为默认设置。

isolation_level 的有效值包括：

READ COMMITTED
AUTOCOMMIT
SERIALIZABLE

备注

由 Oracle 数据库方言实现的 Connection.get_isolation_level() 方法必须强制开始事务，使用 Oracle 数据库 DBMS_TRANSACTION.LOCAL_TRANSACTION_ID 函数；否则，通常无法读取任何隔离级别。

此外，若由于权限或其他原因无法访问 v$transaction 视图， Connection.get_isolation_level() 方法将引发异常，这在 Oracle 数据库安装中是常见的情况。

python-oracledb 和 cx_Oracle 方言会在方言首次连接数据库时调用 Connection.get_isolation_level() 方法，以获取“默认”的隔离级别。该默认级别对于在通过 Connection.execution_options() 方法临时修改后重置连接上的级别非常重要。若 Connection.get_isolation_level() 方法因 v$transaction 无法读取或其他数据库相关失败而引发异常，则假定该级别为 “READ COMMITTED”。此初次连接时的条件不会发出警告，因为这是 Oracle 数据库中常见的限制。

参见

设置事务隔离级别（包括 DBAPI 自动提交）

英文

Oracle Database supports “READ COMMITTED” and “SERIALIZABLE” modes of isolation. The AUTOCOMMIT isolation level is also supported by the python-oracledb and cx_Oracle dialects.

To set using per-connection execution options:

connection = engine.connect()
connection = connection.execution_options(isolation_level="AUTOCOMMIT")

For READ COMMITTED and SERIALIZABLE, the Oracle Database dialects sets the level at the session level using ALTER SESSION, which is reverted back to its default setting when the connection is returned to the connection pool.

Valid values for isolation_level include:

READ COMMITTED
AUTOCOMMIT
SERIALIZABLE

备注

The implementation for the Connection.get_isolation_level() method as implemented by the Oracle Database dialects necessarily force the start of a transaction using the Oracle Database DBMS_TRANSACTION.LOCAL_TRANSACTION_ID function; otherwise no level is normally readable.

Additionally, the Connection.get_isolation_level() method will raise an exception if the v$transaction view is not available due to permissions or other reasons, which is a common occurrence in Oracle Database installations.

The python-oracledb and cx_Oracle dialects attempt to call the Connection.get_isolation_level() method when the dialect makes its first connection to the database in order to acquire the “default”isolation level. This default level is necessary so that the level can be reset on a connection after it has been temporarily modified using Connection.execution_options() method. In the common event that the Connection.get_isolation_level() method raises an exception due to v$transaction not being readable as well as any other database-related failure, the level is assumed to be “READ COMMITTED”. No warning is emitted for this initial first-connect condition as it is expected to be a common restriction on Oracle databases.

参见

设置事务隔离级别（包括 DBAPI 自动提交）

标识符大小写¶

Identifier Casing

中文

在 Oracle 数据库中，数据字典使用大写字母表示所有不区分大小写的标识符名称。这与 SQLAlchemy 的预期相矛盾，SQLAlchemy 假设不区分大小写的名称表示为小写字母。

作为不区分大小写的标识符名称的示例，考虑以下表：

CREATE TABLE MyTable (Identifier INTEGER PRIMARY KEY)

如果你查询 Oracle 数据库关于此表的信息，表名将报告为 MYTABLE，列名将报告为 IDENTIFIER。与 PostgreSQL 和 MySQL 等大多数其他数据库相比，它们会将这些名称报告为 mytable 和 identifier。这些名称 没有被引号引用，因此是不区分大小写的 。仅当它们在表定义中用引号括起来时，MyTable 和 Identifier 的特殊大小写才会被保留：

CREATE TABLE "MyTable" ("Identifier" INTEGER PRIMARY KEY)

在构建 SQLAlchemy Table 对象时， 所有小写的名称被视为不区分大小写 。因此，以下表假定名称不区分大小写:

Table("mytable", metadata, Column("identifier", Integer, primary_key=True))

而当使用混合大小写或大写名称时，假定名称是区分大小写的:

Table("MyTable", metadata, Column("Identifier", Integer, primary_key=True))

当发出文本 SQL SELECT 语句并查看 DBAPI cursor.description 属性中的列名时，数据库驱动程序层也会出现类似的情况。像 PostgreSQL 这样的数据库会将不区分大小写的名称标准化为小写：

>>> pg_engine = create_engine("postgresql://scott:tiger@localhost/test")
>>> pg_connection = pg_engine.connect()
>>> result = pg_connection.exec_driver_sql("SELECT 1 AS SomeName")
>>> result.cursor.description
(Column(name='somename', type_code=23),)

而 Oracle 会将其标准化为大写：

>>> oracle_engine = create_engine("oracle+oracledb://scott:tiger@oracle18c/xe")
>>> oracle_connection = oracle_engine.connect()
>>> result = oracle_connection.exec_driver_sql(
...     "SELECT 1 AS SomeName FROM DUAL"
... )
>>> result.cursor.description
[('SOMENAME', <DbType DB_TYPE_NUMBER>, 127, None, 0, -127, True)]

为了实现表反射和仅文本 SQL 语句往返的跨数据库一致性，SQLAlchemy 在使用 Oracle 方言时执行一个步骤，称为 名称规范化。这个过程也可能适用于其他具有类似大写字母处理不区分大小写名称的第三方方言。

在使用名称规范化时，SQLAlchemy 会通过检查名称是否全部为大写字母来检测是否为不区分大小写的名称；如果是，它会假定这是一个不区分大小写的名称，并将其转换为小写名称。

对于表反射，在 Oracle 数据库的目录表中看到的全部大写的表名将被假定为不区分大小写的名称。这使得 Table 定义可以使用小写名称，并且在跨数据库视角下与 PostgreSQL 和 MySQL 等其他数据库兼容：

# 匹配使用 CREATE TABLE mytable 创建的表 Table(“mytable”, metadata, autoload_with=some_engine)

如上所示，所有小写的名称 "mytable" 是不区分大小写的；它将与 PostgreSQL 中报告的 "mytable" 和 Oracle 中报告的 "MYTABLE" 匹配。如果没有名称规范化，上述 Table 定义将无法在跨数据库的方式中进行自省，因为我们处理的是一个不区分大小写的名称，而不同的数据库报告该名称的方式不同。

在这种情况下，可以强制区分大小写，例如，如果我们想要表示带有精确大小写的引号表名 "MYTABLE"，最简单的方式就是直接使用该大小写，这将被视为区分大小写的名称：

# 匹配使用 CREATE TABLE “MYTABLE” 创建的表 Table(“MYTABLE”, metadata, autoload_with=some_engine)

对于引号全部小写的名称的特殊情况，可以使用 quoted_name 构造函数：

from sqlalchemy import quoted_name

# 匹配使用 CREATE TABLE “mytable” 创建的表 Table(

quoted_name(“mytable”, quote=True), metadata, autoload_with=some_engine

)

名称规范化还适用于处理 纯文本 SQL 字符串 的结果集，这些 SQL 字符串没有其他 Table 或 Column 元数据与之关联。这包括使用 Connection.exec_driver_sql() 执行的 SQL 字符串和使用 text() 构造函数执行的 SQL 字符串，这些 SQL 字符串不包含 Column 元数据。

回到 Oracle 数据库的 SELECT 语句，尽管 cursor.description 报告列名为 SOMENAME，SQLAlchemy 会将其名称规范化为 somename：

>>> oracle_engine = create_engine("oracle+oracledb://scott:tiger@oracle18c/xe")
>>> oracle_connection = oracle_engine.connect()
>>> result = oracle_connection.exec_driver_sql(
...     "SELECT 1 AS SomeName FROM DUAL"
... )
>>> result.cursor.description
[('SOMENAME', <DbType DB_TYPE_NUMBER>, 127, None, 0, -127, True)]
>>> result.keys()
RMKeyView(['somename'])

唯一的情况是，当使用全大写的引号名称时，上述行为会产生不准确的结果。SQLAlchemy 无法确定 cursor.description 中的某个特定名称是否为引号名称，因此无法判断它是否应当区分大小写，或是否应该进行名称规范化：

>>> result = oracle_connection.exec_driver_sql(
...     'SELECT 1 AS "SOMENAME" FROM DUAL'
... )
>>> result.cursor.description
[('SOMENAME', <DbType DB_TYPE_NUMBER>, 127, None, 0, -127, True)]
>>> result.keys()
RMKeyView(['somename'])

对于这个特殊的情况，SQLAlchemy 提供了 Connection.execution_options.driver_column_names 执行选项，它会关闭结果集的名称规范化：

>>> result = oracle_connection.exec_driver_sql(
...     'SELECT 1 AS "SOMENAME" FROM DUAL',
...     execution_options={"driver_column_names": True},
... )
>>> result.keys()
RMKeyView(['SOMENAME'])

在 2.1 版本加入: 添加了 Connection.execution_options.driver_column_names 执行选项

英文

In Oracle Database, the data dictionary represents all case insensitive identifier names using UPPERCASE text. This is in contradiction to the expectations of SQLAlchemy, which assume a case insensitive name is represented as lowercase text.

As an example of case insensitive identifier names, consider the following table:

CREATE TABLE MyTable (Identifier INTEGER PRIMARY KEY)

If you were to ask Oracle Database for information about this table, the table name would be reported as MYTABLE and the column name would be reported as IDENTIFIER. Compare to most other databases such as PostgreSQL and MySQL which would report these names as mytable and identifier. The names are not quoted, therefore are case insensitive. The special casing of MyTable and Identifier would only be maintained if they were quoted in the table definition:

CREATE TABLE "MyTable" ("Identifier" INTEGER PRIMARY KEY)

When constructing a SQLAlchemy Table object, an all lowercase name is considered to be case insensitive. So the following table assumes case insensitive names:

Table("mytable", metadata, Column("identifier", Integer, primary_key=True))

Whereas when mixed case or UPPERCASE names are used, case sensitivity is assumed:

Table("MyTable", metadata, Column("Identifier", Integer, primary_key=True))

A similar situation occurs at the database driver level when emitting a textual SQL SELECT statement and looking at column names in the DBAPI cursor.description attribute. A database like PostgreSQL will normalize case insensitive names to be lowercase:

>>> pg_engine = create_engine("postgresql://scott:tiger@localhost/test")
>>> pg_connection = pg_engine.connect()
>>> result = pg_connection.exec_driver_sql("SELECT 1 AS SomeName")
>>> result.cursor.description
(Column(name='somename', type_code=23),)

Whereas Oracle normalizes them to UPPERCASE:

>>> oracle_engine = create_engine("oracle+oracledb://scott:tiger@oracle18c/xe")
>>> oracle_connection = oracle_engine.connect()
>>> result = oracle_connection.exec_driver_sql(
...     "SELECT 1 AS SomeName FROM DUAL"
... )
>>> result.cursor.description
[('SOMENAME', <DbType DB_TYPE_NUMBER>, 127, None, 0, -127, True)]

In order to achieve cross-database parity for the two cases of a. table reflection and b. textual-only SQL statement round trips, SQLAlchemy performs a step called name normalization when using the Oracle dialect. This process may also apply to other third party dialects that have similar UPPERCASE handling of case insensitive names.

When using name normalization, SQLAlchemy attempts to detect if a name is case insensitive by checking if all characters are UPPERCASE letters only; if so, then it assumes this is a case insensitive name and is delivered as a lowercase name.

For table reflection, a tablename that is seen represented as all UPPERCASE in Oracle Database’s catalog tables will be assumed to have a case insensitive name. This is what allows the Table definition to use lower case names and be equally compatible from a reflection point of view on Oracle Database and all other databases such as PostgreSQL and MySQL:

# matches a table created with CREATE TABLE mytable
Table("mytable", metadata, autoload_with=some_engine)

Above, the all lowercase name "mytable" is case insensitive; it will match a table reported by PostgreSQL as "mytable" and a table reported by Oracle as "MYTABLE". If name normalization were not present, it would not be possible for the above Table definition to be introspectable in a cross-database way, since we are dealing with a case insensitive name that is not reported by each database in the same way.

Case sensitivity can be forced on in this case, such as if we wanted to represent the quoted tablename "MYTABLE" with that exact casing, most simply by using that casing directly, which will be seen as a case sensitive name:

# matches a table created with CREATE TABLE "MYTABLE"
Table("MYTABLE", metadata, autoload_with=some_engine)

For the unusual case of a quoted all-lowercase name, the quoted_name construct may be used:

from sqlalchemy import quoted_name

# matches a table created with CREATE TABLE "mytable"
Table(
    quoted_name("mytable", quote=True), metadata, autoload_with=some_engine
)

Name normalization also takes place when handling result sets from purely textual SQL strings, that have no other Table or Column metadata associated with them. This includes SQL strings executed using Connection.exec_driver_sql() and SQL strings executed using the text() construct which do not include Column metadata.

Returning to the Oracle Database SELECT statement, we see that even though cursor.description reports the column name as SOMENAME, SQLAlchemy name normalizes this to somename:

>>> oracle_engine = create_engine("oracle+oracledb://scott:tiger@oracle18c/xe")
>>> oracle_connection = oracle_engine.connect()
>>> result = oracle_connection.exec_driver_sql(
...     "SELECT 1 AS SomeName FROM DUAL"
... )
>>> result.cursor.description
[('SOMENAME', <DbType DB_TYPE_NUMBER>, 127, None, 0, -127, True)]
>>> result.keys()
RMKeyView(['somename'])

The single scenario where the above behavior produces inaccurate results is when using an all-uppercase, quoted name. SQLAlchemy has no way to determine that a particular name in cursor.description was quoted, and is therefore case sensitive, or was not quoted, and should be name normalized:

>>> result = oracle_connection.exec_driver_sql(
...     'SELECT 1 AS "SOMENAME" FROM DUAL'
... )
>>> result.cursor.description
[('SOMENAME', <DbType DB_TYPE_NUMBER>, 127, None, 0, -127, True)]
>>> result.keys()
RMKeyView(['somename'])

For this exact scenario, SQLAlchemy offers the Connection.execution_options.driver_column_names execution options, which turns off name normalize for result sets:

>>> result = oracle_connection.exec_driver_sql(
...     'SELECT 1 AS "SOMENAME" FROM DUAL',
...     execution_options={"driver_column_names": True},
... )
>>> result.keys()
RMKeyView(['SOMENAME'])

在 2.1 版本加入: Added the Connection.execution_options.driver_column_names execution option

最大标识符长度¶

Maximum Identifier Lengths

中文

SQLAlchemy 会注意到 Oracle 数据库支持的最大标识符长度。这会影响生成的 SQL 标签名称以及约束名称的生成，尤其是在使用配置约束命名约定中描述的约束命名约定特性时。

Oracle 数据库 12.2 将默认的最大标识符长度从 30 增加到 128。自 SQLAlchemy 1.4 起，Oracle 方言的默认最大标识符长度为 128 个字符。首次连接时，会获取数据库实际支持的最大长度。在所有情况下，设置 create_engine.max_identifier_length 参数会绕过此更改，所提供的值将直接使用：

engine = create_engine(
“oracle+oracledb://scott:tiger@localhost:1521?service_name=freepdb1”, max_identifier_length=30,

)

如果未设置 create_engine.max_identifier_length，oracledb 方言会在内部使用自 python-oracledb 2.5 版本以来在驱动程序连接上可用的 max_identifier_length 属性。当使用较旧的驱动程序版本或使用 cx_Oracle 方言时，SQLAlchemy 会尝试在首次连接时使用查询 SELECT value FROM v$parameter WHERE name = 'compatible' 来确定数据库的有效兼容版本。兼容版本是一个独立于实际数据库版本的版本号。它用于协助数据库迁移，并由 Oracle 数据库初始化参数配置。兼容版本随后确定数据库的最大标识符长度。如果无法访问 V$ 视图，则会改用数据库版本信息。

最大标识符长度在生成 SELECT 语句中的匿名 SQL 标签时也会发挥作用，但更关键的是在从命名约定生成约束名称时发挥作用。正是这一领域导致 SQLAlchemy 必须谨慎地更改此默认值。例如，以下命名约定会根据标识符长度生成两个非常不同的约束名称：

from sqlalchemy import Column from sqlalchemy import Index from sqlalchemy import Integer from sqlalchemy import MetaData from sqlalchemy import Table from sqlalchemy.dialects import oracle from sqlalchemy.schema import CreateIndex

m = MetaData(naming_convention={“ix”: “ix_%(column_0N_name)s”})

t = Table(
“t”, m, Column(“some_column_name_1”,

Integer),

Column(“some_column_name_2”, Integer), Column(“some_column_name_3”, Integer),

)

ix = Index(
None, t.c.some_column_name_1, t.c.some_column_name_2, t.c.some_column_name_3,

)

oracle_dialect = oracle.dialect(max_identifier_length=30) print(CreateIndex(ix).compile(dialect=oracle_dialect))

当标识符长度为 30 时，以上 CREATE INDEX 看起来如下：

CREATE INDEX ix_some_column_name_1s_70cd ON t
(some_column_name_1, some_column_name_2, some_column_name_3)

然而，当长度为 128 时，它变成了：

CREATE INDEX ix_some_column_name_1some_column_name_2some_column_name_3 ON t
(some_column_name_1, some_column_name_2, some_column_name_3)

在 Oracle 数据库 12.2 或更高版本上运行 SQLAlchemy 1.4 之前的版本的应用程序因此可能面临数据库迁移场景，在这种场景下，想要“DROP CONSTRAINT”一个先前使用较短长度生成的名称。当标识符长度更改时，如果索引或约束的名称没有首先调整，这个迁移将失败。强烈建议这些应用程序使用 create_engine.max_identifier_length 来维护截断名称的生成控制，并在更改此值时在分阶段环境中充分审查和测试所有数据库迁移，以确保减轻此更改的影响。

在 1.4 版本发生变更: Oracle 数据库的默认 max_identifier_length 为 128 个字符，若 Oracle 数据库或其兼容性设置低于 12.2 版本，首次连接时会调整为 30。

英文

SQLAlchemy is sensitive to the maximum identifier length supported by Oracle Database. This affects generated SQL label names as well as the generation of constraint names, particularly in the case where the constraint naming convention feature described at 配置约束命名约定 is being used.

Oracle Database 12.2 increased the default maximum identifier length from 30 to 128. As of SQLAlchemy 1.4, the default maximum identifier length for the Oracle dialects is 128 characters. Upon first connection, the maximum length actually supported by the database is obtained. In all cases, setting the create_engine.max_identifier_length parameter will bypass this change and the value given will be used as is:

engine = create_engine(
    "oracle+oracledb://scott:tiger@localhost:1521?service_name=freepdb1",
    max_identifier_length=30,
)

If create_engine.max_identifier_length is not set, the oracledb dialect internally uses the max_identifier_length attribute available on driver connections since python-oracledb version 2.5. When using an older driver version, or using the cx_Oracle dialect, SQLAlchemy will instead attempt to use the query SELECT value FROM v$parameter WHERE name = 'compatible' upon first connect in order to determine the effective compatibility version of the database. The “compatibility” version is a version number that is independent of the actual database version. It is used to assist database migration. It is configured by an Oracle Database initialization parameter. The compatibility version then determines the maximum allowed identifier length for the database. If the V$ view is not available, the database version information is used instead.

The maximum identifier length comes into play both when generating anonymized SQL labels in SELECT statements, but more crucially when generating constraint names from a naming convention. It is this area that has created the need for SQLAlchemy to change this default conservatively. For example, the following naming convention produces two very different constraint names based on the identifier length:

from sqlalchemy import Column
from sqlalchemy import Index
from sqlalchemy import Integer
from sqlalchemy import MetaData
from sqlalchemy import Table
from sqlalchemy.dialects import oracle
from sqlalchemy.schema import CreateIndex

m = MetaData(naming_convention={"ix": "ix_%(column_0N_name)s"})

t = Table(
    "t",
    m,
    Column("some_column_name_1", Integer),
    Column("some_column_name_2", Integer),
    Column("some_column_name_3", Integer),
)

ix = Index(
    None,
    t.c.some_column_name_1,
    t.c.some_column_name_2,
    t.c.some_column_name_3,
)

oracle_dialect = oracle.dialect(max_identifier_length=30)
print(CreateIndex(ix).compile(dialect=oracle_dialect))

With an identifier length of 30, the above CREATE INDEX looks like:

CREATE INDEX ix_some_column_name_1s_70cd ON t
(some_column_name_1, some_column_name_2, some_column_name_3)

However with length of 128, it becomes:

.. sourcecode:: sql

CREATE INDEX ix_some_column_name_1some_column_name_2some_column_name_3 ON t (some_column_name_1, some_column_name_2, some_column_name_3)

Applications which have run versions of SQLAlchemy prior to 1.4 on Oracle Database version 12.2 or greater are therefore subject to the scenario of a database migration that wishes to “DROP CONSTRAINT” on a name that was previously generated with the shorter length. This migration will fail when the identifier length is changed without the name of the index or constraint first being adjusted. Such applications are strongly advised to make use of create_engine.max_identifier_length in order to maintain control of the generation of truncated names, and to fully review and test all database migrations in a staging environment when changing this value to ensure that the impact of this change has been mitigated.

在 1.4 版本发生变更: the default max_identifier_length for Oracle Database is 128 characters, which is adjusted down to 30 upon first connect if the Oracle Database, or its compatibility setting, are lower than version 12.2.

LIMIT/OFFSET/FETCH 支持¶

LIMIT/OFFSET/FETCH Support

中文

像 Select.limit() 和 Select.offset() 这样的方法使用了 FETCH FIRST N ROW / OFFSET N ROWS 语法，假设使用的是 Oracle 数据库 12c 或更高版本，并且假设 SELECT 语句没有嵌套在联合语句（如 UNION）中。通过使用 Select.fetch() 方法，也可以直接使用这种语法。

在 2.0 版本发生变更: Oracle 数据库方言现在对所有 Select.limit() 和 Select.offset() 的使用，包括 ORM 和遗留的 Query，都使用 FETCH FIRST N ROW / OFFSET N ROWS。要强制使用旧版行为（使用窗口函数），请在 create_engine() 中指定 enable_offset_fetch=False 方言参数。

通过将 enable_offset_fetch=False 传递给 create_engine()，可以禁用 FETCH FIRST / OFFSET，从而强制使用基于窗口函数的“旧版”模式。对于 12c 之前的 Oracle 数据库版本，系统也会自动选择此模式。

在使用旧版模式时，或者当包含限制/偏移的 Select 语句嵌套在复合语句中时，会使用基于窗口函数的模拟方法来实现 LIMIT / OFFSET，这涉及到使用 ROW_NUMBER 创建子查询，这可能会导致性能问题以及复杂语句的 SQL 构造问题。然而，这种方法在所有 Oracle 数据库版本中都受到支持。请参阅下面的注释。

英文

Methods like Select.limit() and Select.offset() make use of FETCH FIRST N ROW / OFFSET N ROWS syntax assuming Oracle Database 12c or above, and assuming the SELECT statement is not embedded within a compound statement like UNION. This syntax is also available directly by using the Select.fetch() method.

在 2.0 版本发生变更: the Oracle Database dialects now use FETCH FIRST N ROW / OFFSET N ROWS for all Select.limit() and Select.offset() usage including within the ORM and legacy Query. To force the legacy behavior using window functions, specify the enable_offset_fetch=False dialect parameter to create_engine().

The use of FETCH FIRST / OFFSET may be disabled on any Oracle Database version by passing enable_offset_fetch=False to create_engine(), which will force the use of “legacy” mode that makes use of window functions. This mode is also selected automatically when using a version of Oracle Database prior to 12c.

When using legacy mode, or when a Select statement with limit/offset is embedded in a compound statement, an emulated approach for LIMIT / OFFSET based on window functions is used, which involves creation of a subquery using ROW_NUMBER that is prone to performance issues as well as SQL construction issues for complex statements. However, this approach is supported by all Oracle Database versions. See notes below.

LIMIT/OFFSET 模拟注意事项（无法使用 fetch() 方法时）¶

Notes on LIMIT / OFFSET emulation (when fetch() method cannot be used)

中文

如果在 Oracle 数据库 12c 之前的版本上使用 Select.limit() 和 Select.offset()，或在 ORM 中使用 Query.limit() 和 Query.offset() 方法，以下注意事项适用：

SQLAlchemy 当前使用 ROWNUM 来实现 LIMIT/OFFSET；具体方法可以参考 https://blogs.oracle.com/oraclemagazine/on-rownum-and-limiting-results 。
默认情况下不会使用 “FIRST_ROWS()” 优化关键字。要启用此优化指令，请在 create_engine() 中指定 optimize_limits=True。

在 1.4 版本发生变更: Oracle 数据库方言使用“后编译”方案呈现 limit/offset 整数值，在将语句传递给游标执行之前，直接呈现整数值。use_binds_for_limits 标志不再生效。

参见

New “post compile” bound parameters used for LIMIT/OFFSET in Oracle, SQL Server。

英文

If using Select.limit() and Select.offset(), or with the ORM the Query.limit() and Query.offset() methods on an Oracle Database version prior to 12c, the following notes apply:

SQLAlchemy currently makes use of ROWNUM to achieve LIMIT/OFFSET; the exact methodology is taken from https://blogs.oracle.com/oraclemagazine/on-rownum-and-limiting-results .
the “FIRST_ROWS()” optimization keyword is not used by default. To enable the usage of this optimization directive, specify optimize_limits=True to create_engine().

在 1.4 版本发生变更: The Oracle Database dialect renders limit/offset integer values using a “post compile” scheme which renders the integer directly before passing the statement to the cursor for execution. The use_binds_for_limits flag no longer has an effect.

参见

New “post compile” bound parameters used for LIMIT/OFFSET in Oracle, SQL Server.

RETURNING 支持¶

RETURNING Support

中文

Oracle 数据库完全支持在 INSERT、UPDATE 和 DELETE 语句中使用 RETURNING，并且这些语句仅使用一组绑定参数（即 cursor.execute() 风格的语句；SQLAlchemy 通常不支持与 executemany 语句一起使用 RETURNING）。多个行也可以返回。

在 2.0 版本发生变更: Oracle 数据库后端对 RETURNING 的支持与其他后端保持一致。

英文

Oracle Database supports RETURNING fully for INSERT, UPDATE and DELETE statements that are invoked with a single collection of bound parameters (that is, a cursor.execute() style statement; SQLAlchemy does not generally support RETURNING with executemany statements). Multiple rows may be returned as well.

在 2.0 版本发生变更: the Oracle Database backend has full support for RETURNING on parity with other backends.

ON UPDATE CASCADE¶

ON UPDATE CASCADE

中文

Oracle 数据库没有原生的 ON UPDATE CASCADE 功能。可以参考基于触发器的解决方案，地址为 https://web.archive.org/web/20090317041251/https://asktom.oracle.com/tkyte/update_cascade/index.html

在使用 SQLAlchemy ORM 时，ORM 具有有限的手动发出级联更新的能力 - 使用 “deferrable=True, initially=’deferred’” 关键字参数指定 ForeignKey 对象，并在每个 relationship() 上指定 “passive_updates=False”。

英文

Oracle Database doesn’t have native ON UPDATE CASCADE functionality. A trigger based solution is available at https://web.archive.org/web/20090317041251/https://asktom.oracle.com/tkyte/update_cascade/index.html

When using the SQLAlchemy ORM, the ORM has limited ability to manually issue cascading updates - specify ForeignKey objects using the “deferrable=True, initially=’deferred’” keyword arguments, and specify “passive_updates=False” on each relationship().

Oracle Database 8 兼容性¶

Oracle Database 8 Compatibility

中文

警告

不知道 Oracle 数据库 8 的兼容性状态在 SQLAlchemy 2.0 中的表现。

当检测到 Oracle 数据库 8 时，方言会自动配置为以下行为：

将 use_ansi 标志设置为 False。这会将所有 JOIN 子句转换为 WHERE 子句，并在 LEFT OUTER JOIN 的情况下使用 Oracle 的 (+) 运算符。
当使用 Unicode 时，不再生成 NVARCHAR2 和 NCLOB 数据类型的 DDL，而是生成 VARCHAR2 和 CLOB。这是因为即使这些类型可用，它们在 Oracle 8 上似乎不能正常工作。 NVARCHAR 和 NCLOB 类型将始终生成 NVARCHAR2 和 NCLOB。

英文

警告

The status of Oracle Database 8 compatibility is not known for SQLAlchemy 2.0.

When Oracle Database 8 is detected, the dialect internally configures itself to the following behaviors:

the use_ansi flag is set to False. This has the effect of converting all JOIN phrases into the WHERE clause, and in the case of LEFT OUTER JOIN makes use of Oracle’s (+) operator.
the NVARCHAR2 and NCLOB datatypes are no longer generated as DDL when the Unicode is used - VARCHAR2 and CLOB are issued instead. This because these types don’t seem to work correctly on Oracle 8 even though they are available. The NVARCHAR and NCLOB types will always generate NVARCHAR2 and NCLOB.

同义词/DBLINK 反射¶

Synonym/DBLINK Reflection

中文

在使用反射与 Table 对象时，方言可以通过将 oracle_resolve_synonyms=True 作为关键字参数传递给 Table 构造函数，选择性地查找由同义词指示的表，无论是位于本地或远程模式中，还是通过 DBLINK 访问:

some_table = Table(
    "some_table", autoload_with=some_engine, oracle_resolve_synonyms=True
)

设置此标志时，给定的名称（例如上面的 some_table）不仅会在 ALL_TABLES 视图中查找，还会在 ALL_SYNONYMS 视图中查找，看看这个名称是否实际上是另一个名称的同义词。如果找到了同义词并且指向 DBLINK，Oracle 数据库方言会知道如何使用 DBLINK 语法（例如 @dblink）来定位该表的信息。

oracle_resolve_synonyms 可以在接受反射参数的任何地方使用，包括 MetaData.reflect() 和 Inspector.get_columns() 等方法。

如果不使用同义词，应该禁用此标志。

英文

When using reflection with Table objects, the dialect can optionally search for tables indicated by synonyms, either in local or remote schemas or accessed over DBLINK, by passing the flag oracle_resolve_synonyms=True as a keyword argument to the Table construct:

some_table = Table(
    "some_table", autoload_with=some_engine, oracle_resolve_synonyms=True
)

When this flag is set, the given name (such as some_table above) will be searched not just in the ALL_TABLES view, but also within the ALL_SYNONYMS view to see if this name is actually a synonym to another name. If the synonym is located and refers to a DBLINK, the Oracle Database dialects know how to locate the table’s information using DBLINK syntax(e.g. @dblink).

oracle_resolve_synonyms is accepted wherever reflection arguments are accepted, including methods such as MetaData.reflect() and Inspector.get_columns().

If synonyms are not in use, this flag should be left disabled.

约束反射¶

Constraint Reflection

中文

Oracle 数据库方言可以返回关于外键、唯一约束、CHECK 约束以及表上索引的信息。

有关这些约束的原始信息可以通过以下方法获取： Inspector.get_foreign_keys()， Inspector.get_unique_constraints()， Inspector.get_check_constraints()， Inspector.get_indexes()。

在 Table 层级使用反射时， Table 也将包括这些约束。

请注意以下几点：

使用 Inspector.get_check_constraints() 方法时， Oracle 数据库会为指定 “NOT NULL” 的列构建一个特殊的 “IS NOT NULL” 约束。默认情况下，此约束不会返回；要包括 “IS NOT NULL” 约束，请传递 include_all=True 标志:

from sqlalchemy import create_engine, inspect

engine = create_engine(
    "oracle+oracledb://scott:tiger@localhost:1521?service_name=freepdb1"
)
inspector = inspect(engine)
all_check_constraints = inspector.get_check_constraints(
    "some_table", include_all=True
)

在大多数情况下，当反射 Table 时，唯一约束* 不会* 作为 UniqueConstraint 对象提供，因为 Oracle 数据库通常用 UNIQUE 索引来镜像唯一约束（例外情况似乎是当两个或更多的唯一约束表示相同的列时）；因此，Table 将使用 Index 并设置 unique=True 标志来表示这些约束。
Oracle 数据库会为表的主键创建一个隐式索引；该索引 会从所有索引结果中排除 。
反射时，索引列的列表不会包括以 SYS_NC 开头的列名。

英文

The Oracle Database dialects can return information about foreign key, unique, and CHECK constraints, as well as indexes on tables.

Raw information regarding these constraints can be acquired using Inspector.get_foreign_keys(), Inspector.get_unique_constraints(), Inspector.get_check_constraints(), and Inspector.get_indexes().

When using reflection at the Table level, the Table will also include these constraints.

Note the following caveats:

When using the Inspector.get_check_constraints() method, Oracle Database builds a special “IS NOT NULL” constraint for columns that specify “NOT NULL”. This constraint is not returned by default; to include the “IS NOT NULL” constraints, pass the flag include_all=True:
```
from sqlalchemy import create_engine, inspect

engine = create_engine(
    "oracle+oracledb://scott:tiger@localhost:1521?service_name=freepdb1"
)
inspector = inspect(engine)
all_check_constraints = inspector.get_check_constraints(
    "some_table", include_all=True
)
```
in most cases, when reflecting a Table, a UNIQUE constraint will not be available as a UniqueConstraint object, as Oracle Database mirrors unique constraints with a UNIQUE index in most cases (the exception seems to be when two or more unique constraints represent the same columns); the Table will instead represent these using Index with the unique=True flag set.
Oracle Database creates an implicit index for the primary key of a table; this index is excluded from all index results.
the list of columns reflected for an index will not include column names that start with SYS_NC.

SYSTEM/SYSAUX 表空间的表名¶

Table names with SYSTEM/SYSAUX tablespaces

中文

Inspector.get_table_names() 和 Inspector.get_temp_table_names() 方法会返回当前引擎的表名列表。这些方法也是反射操作的一部分，例如在 MetaData.reflect() 中执行的操作。默认情况下，这些操作会排除 SYSTEM 和 SYSAUX 表空间。如果要更改此行为，可以使用 exclude_tablespaces 参数在引擎级别更改默认排除的表空间列表:

# 排除 SYSAUX 和 SOME_TABLESPACE，但不排除 SYSTEM
e = create_engine(
    "oracle+oracledb://scott:tiger@localhost:1521/?service_name=freepdb1",
    exclude_tablespaces=["SYSAUX", "SOME_TABLESPACE"],
)

英文

The Inspector.get_table_names() and Inspector.get_temp_table_names() methods each return a list of table names for the current engine. These methods are also part of the reflection which occurs within an operation such as MetaData.reflect(). By default, these operations exclude the SYSTEM and SYSAUX tablespaces from the operation. In order to change this, the default list of tablespaces excluded can be changed at the engine level using the exclude_tablespaces parameter:

# exclude SYSAUX and SOME_TABLESPACE, but not SYSTEM
e = create_engine(
    "oracle+oracledb://scott:tiger@localhost:1521/?service_name=freepdb1",
    exclude_tablespaces=["SYSAUX", "SOME_TABLESPACE"],
)

FLOAT/DOUBLE 支持和行为¶

FLOAT / DOUBLE Support and Behaviors

中文

SQLAlchemy 的 Float 和 Double 数据类型是通用数据类型，会解析为给定后端的“最符合预期”的数据类型。对于 Oracle 数据库，这意味着它们解析为 FLOAT 和 DOUBLE 类型:

>>> from sqlalchemy import cast, literal, Float
>>> from sqlalchemy.dialects import oracle
>>> float_datatype = Float()
>>> print(cast(literal(5.0), float_datatype).compile(dialect=oracle.dialect()))
CAST(:param_1 AS FLOAT)

Oracle 的 FLOAT / DOUBLE 数据类型是 NUMBER 的别名。Oracle 数据库存储 NUMBER 值时具有完全精度，而不是浮动点精度，这意味着 FLOAT / DOUBLE 实际上并不表现得像原生的浮点数值。Oracle 数据库则提供了特殊的数据类型 BINARY_FLOAT 和 BINARY_DOUBLE 来提供真正的 4 字节和 8 字节浮点数值。

SQLAlchemy 通过 BINARY_FLOAT 和 BINARY_DOUBLE 直接支持这些数据类型。要以数据库无关的方式使用 Float 或 Double 数据类型，同时允许 Oracle 后端使用其中之一，请使用 TypeEngine.with_variant() 方法来设置变体:

>>> from sqlalchemy import cast, literal, Float
>>> from sqlalchemy.dialects import oracle
>>> float_datatype = Float().with_variant(oracle.BINARY_FLOAT(), "oracle")
>>> print(cast(literal(5.0), float_datatype).compile(dialect=oracle.dialect()))
CAST(:param_1 AS BINARY_FLOAT)

例如，在 Table 定义中使用此数据类型:

my_table = Table(
    "my_table",
    metadata,
    Column(
        "fp_data", Float().with_variant(oracle.BINARY_FLOAT(), "oracle")
    ),
)

英文

The SQLAlchemy Float and Double datatypes are generic datatypes that resolve to the “least surprising” datatype for a given backend. For Oracle Database, this means they resolve to the FLOAT and DOUBLE types:

>>> from sqlalchemy import cast, literal, Float
>>> from sqlalchemy.dialects import oracle
>>> float_datatype = Float()
>>> print(cast(literal(5.0), float_datatype).compile(dialect=oracle.dialect()))
CAST(:param_1 AS FLOAT)

Oracle’s FLOAT / DOUBLE datatypes are aliases for NUMBER. Oracle Database stores NUMBER values with full precision, not floating point precision, which means that FLOAT / DOUBLE do not actually behave like native FP values. Oracle Database instead offers special datatypes BINARY_FLOAT and BINARY_DOUBLE to deliver real 4- and 8- byte FP values.

SQLAlchemy supports these datatypes directly using BINARY_FLOAT and BINARY_DOUBLE. To use the Float or Double datatypes in a database agnostic way, while allowing Oracle backends to utilize one of these types, use the TypeEngine.with_variant() method to set up a variant:

>>> from sqlalchemy import cast, literal, Float
>>> from sqlalchemy.dialects import oracle
>>> float_datatype = Float().with_variant(oracle.BINARY_FLOAT(), "oracle")
>>> print(cast(literal(5.0), float_datatype).compile(dialect=oracle.dialect()))
CAST(:param_1 AS BINARY_FLOAT)

E.g. to use this datatype in a Table definition:

my_table = Table(
    "my_table",
    metadata,
    Column(
        "fp_data", Float().with_variant(oracle.BINARY_FLOAT(), "oracle")
    ),
)

DateTime 兼容性¶

DateTime Compatibility

中文

Oracle 数据库没有名为 DATETIME 的数据类型，只有 DATE 类型，它实际上可以存储日期和时间值。因此，Oracle 数据库方言提供了一个类型 DATE，它是 DateTime 的子类。此类型没有特殊行为，仅作为该类型的“标记”存在；此外，当数据库列被反射并且类型报告为 DATE 时，将使用支持时间的 DATE 类型。

英文

Oracle Database has no datatype known as DATETIME, it instead has only DATE, which can actually store a date and time value. For this reason, the Oracle Database dialects provide a type DATE which is a subclass of DateTime. This type has no special behavior, and is only present as a “marker” for this type; additionally, when a database column is reflected and the type is reported as DATE, the time-supporting DATE type is used.

Oracle 数据库表选项¶

Oracle Database Table Options

中文

在与 Table 构造结合使用时，CREATE TABLE 短语支持以下选项与 Oracle 数据库方言：

ON COMMIT:

Table(
    "some_table",
    metadata,
    ...,
    prefixes=["GLOBAL TEMPORARY"],
    oracle_on_commit="PRESERVE ROWS",
)

COMPRESS:

Table(
    "mytable", metadata, Column("data", String(32)), oracle_compress=True
)

Table("mytable", metadata, Column("data", String(32)), oracle_compress=6)

oracle_compress 参数接受一个整数压缩级别，或者 True 来使用默认压缩级别。

TABLESPACE:
```
Table("mytable", metadata, ..., oracle_tablespace="EXAMPLE_TABLESPACE")
```
oracle_tablespace 参数指定表要创建的表空间。当你希望在用户的默认表空间以外的表空间中创建表时，这非常有用。

在 2.0.37 版本加入.

英文

The CREATE TABLE phrase supports the following options with Oracle Database dialects in conjunction with the Table construct:

ON COMMIT:

Table(
    "some_table",
    metadata,
    ...,
    prefixes=["GLOBAL TEMPORARY"],
    oracle_on_commit="PRESERVE ROWS",
)

COMPRESS:

Table(
    "mytable", metadata, Column("data", String(32)), oracle_compress=True
)

Table("mytable", metadata, Column("data", String(32)), oracle_compress=6)

The oracle_compress parameter accepts either an integer compression level, or True to use the default compression level.

TABLESPACE:
```
Table("mytable", metadata, ..., oracle_tablespace="EXAMPLE_TABLESPACE")
```
The oracle_tablespace parameter specifies the tablespace in which the table is to be created. This is useful when you want to create a table in a tablespace other than the default tablespace of the user.

在 2.0.37 版本加入.

Oracle 数据库特定索引选项¶

Oracle Database Specific Index Options

位图索引¶

Bitmap Indexes

中文

你可以指定 oracle_bitmap 参数来创建位图索引，而不是 B 树索引:

Index("my_index", my_table.c.data, oracle_bitmap=True)

位图索引不能是唯一的，也不能被压缩。SQLAlchemy 不会检查这些限制，只有数据库会检查。

英文

You can specify the oracle_bitmap parameter to create a bitmap index instead of a B-tree index:

Index("my_index", my_table.c.data, oracle_bitmap=True)

Bitmap indexes cannot be unique and cannot be compressed. SQLAlchemy will not check for such limitations, only the database will.

索引压缩¶

Index compression

中文

Oracle 数据库对包含大量重复值的索引有更高效的存储模式。使用 oracle_compress 参数启用键压缩:

Index("my_index", my_table.c.data, oracle_compress=True)

Index(
    "my_index",
    my_table.c.data1,
    my_table.c.data2,
    unique=True,
    oracle_compress=1,
)

oracle_compress 参数接受一个整数，表示要压缩的前缀列的数量，或者 True 来使用默认值（对于非唯一索引压缩所有列，对于唯一索引压缩所有列但不包括最后一列）。

英文

Oracle Database has a more efficient storage mode for indexes containing lots of repeated values. Use the oracle_compress parameter to turn on key compression:

Index("my_index", my_table.c.data, oracle_compress=True)

Index(
    "my_index",
    my_table.c.data1,
    my_table.c.data2,
    unique=True,
    oracle_compress=1,
)

The oracle_compress parameter accepts either an integer specifying the number of prefix columns to compress, or True to use the default (all columns for non-unique indexes, all but the last column for unique indexes).

Oracle 数据库数据类型¶

Oracle Database Data Types

中文

与所有SQLAlchemy方言一样，所有已知对Oracle数据库有效的UPPERCASE类型都可以从顶级方言中导入，无论它们是来自 sqlalchemy.types 还是来自本地方言:

from sqlalchemy.dialects.oracle import (
    BFILE,
    BLOB,
    CHAR,
    CLOB,
    DATE,
    DOUBLE_PRECISION,
    FLOAT,
    INTERVAL,
    LONG,
    NCLOB,
    NCHAR,
    NUMBER,
    NVARCHAR,
    NVARCHAR2,
    RAW,
    TIMESTAMP,
    VARCHAR,
    VARCHAR2,
)

特定于Oracle数据库或具有Oracle特定构造参数的类型如下：

英文

As with all SQLAlchemy dialects, all UPPERCASE types that are known to be valid with Oracle Database are importable from the top level dialect, whether they originate from sqlalchemy.types or from the local dialect:

from sqlalchemy.dialects.oracle import (
    BFILE,
    BLOB,
    CHAR,
    CLOB,
    DATE,
    DOUBLE_PRECISION,
    FLOAT,
    INTERVAL,
    LONG,
    NCLOB,
    NCHAR,
    NUMBER,
    NVARCHAR,
    NVARCHAR2,
    RAW,
    TIMESTAMP,
    VARCHAR,
    VARCHAR2,
)

Types which are specific to Oracle Database, or have Oracle-specific construction arguments, are as follows:

Object Name	Description
BFILE
BINARY_DOUBLE	Implement the Oracle `BINARY_DOUBLE` datatype.
BINARY_FLOAT	Implement the Oracle `BINARY_FLOAT` datatype.
DATE	Provide the Oracle Database DATE type.
FLOAT	Oracle Database FLOAT.
INTERVAL
LONG
NCLOB
NUMBER
NVARCHAR2	`NVARCHAR` 的别名
RAW
ROWID	Oracle Database ROWID type.
TIMESTAMP	Oracle Database implementation of `TIMESTAMP`, which supports additional Oracle Database-specific modes

class sqlalchemy.dialects.oracle.BFILE¶

Members

__init__()

Class signature

class sqlalchemy.dialects.oracle.BFILE (sqlalchemy.types.LargeBinary)

method sqlalchemy.dialects.oracle.BFILE.__init__(length: int | None = None)¶

inherited from the sqlalchemy.types.LargeBinary.__init__ method of LargeBinary

Construct a LargeBinary type.

参数:: length¶ – optional, a length for the column for use in DDL statements, for those binary types that accept a length, such as the MySQL BLOB type.

class sqlalchemy.dialects.oracle.BINARY_DOUBLE¶

Implement the Oracle BINARY_DOUBLE datatype.

This datatype differs from the Oracle DOUBLE datatype in that it delivers a true 8-byte FP value. The datatype may be combined with a generic Double datatype using TypeEngine.with_variant().

参见

FLOAT/DOUBLE 支持和行为

Members

__init__()

Class signature

class sqlalchemy.dialects.oracle.BINARY_DOUBLE (sqlalchemy.types.Double)

method sqlalchemy.dialects.oracle.BINARY_DOUBLE.__init__(precision: int | None = None, asdecimal: bool = False, decimal_return_scale: int | None = None)¶

inherited from the sqlalchemy.types.Float.__init__ method of Float

Construct a Float.

参数:

precision¶ –
the numeric precision for use in DDL CREATE TABLE. Backends should attempt to ensure this precision indicates a number of digits for the generic Float datatype.
备注

For the Oracle Database backend, the Float.precision parameter is not accepted when rendering DDL, as Oracle Database does not support float precision specified as a number of decimal places. Instead, use the Oracle Database-specific FLOAT datatype and specify the FLOAT.binary_precision parameter. This is new in version 2.0 of SQLAlchemy.

To create a database agnostic Float that separately specifies binary precision for Oracle Database, use TypeEngine.with_variant() as follows:
```
from sqlalchemy import Column
from sqlalchemy import Float
from sqlalchemy.dialects import oracle

Column(
    "float_data",
    Float(5).with_variant(oracle.FLOAT(binary_precision=16), "oracle"),
)
```
asdecimal¶ – the same flag as that of Numeric, but defaults to False. Note that setting this flag to True results in floating point conversion.
decimal_return_scale¶ – Default scale to use when converting from floats to Python decimals. Floating point values will typically be much longer due to decimal inaccuracy, and most floating point database types don’t have a notion of “scale”, so by default the float type looks for the first ten decimal places when converting. Specifying this value will override that length. Note that the MySQL float types, which do include “scale”, will use “scale” as the default for decimal_return_scale, if not otherwise specified.

class sqlalchemy.dialects.oracle.BINARY_FLOAT¶

Implement the Oracle BINARY_FLOAT datatype.

This datatype differs from the Oracle FLOAT datatype in that it delivers a true 4-byte FP value. The datatype may be combined with a generic Float datatype using TypeEngine.with_variant().

参见

FLOAT/DOUBLE 支持和行为

Members

__init__()

Class signature

class sqlalchemy.dialects.oracle.BINARY_FLOAT (sqlalchemy.types.Float)

method sqlalchemy.dialects.oracle.BINARY_FLOAT.__init__(precision: int | None = None, asdecimal: bool = False, decimal_return_scale: int | None = None)¶

inherited from the sqlalchemy.types.Float.__init__ method of Float

Construct a Float.

参数:

precision¶ –
the numeric precision for use in DDL CREATE TABLE. Backends should attempt to ensure this precision indicates a number of digits for the generic Float datatype.
备注

For the Oracle Database backend, the Float.precision parameter is not accepted when rendering DDL, as Oracle Database does not support float precision specified as a number of decimal places. Instead, use the Oracle Database-specific FLOAT datatype and specify the FLOAT.binary_precision parameter. This is new in version 2.0 of SQLAlchemy.

To create a database agnostic Float that separately specifies binary precision for Oracle Database, use TypeEngine.with_variant() as follows:
```
from sqlalchemy import Column
from sqlalchemy import Float
from sqlalchemy.dialects import oracle

Column(
    "float_data",
    Float(5).with_variant(oracle.FLOAT(binary_precision=16), "oracle"),
)
```
asdecimal¶ – the same flag as that of Numeric, but defaults to False. Note that setting this flag to True results in floating point conversion.
decimal_return_scale¶ – Default scale to use when converting from floats to Python decimals. Floating point values will typically be much longer due to decimal inaccuracy, and most floating point database types don’t have a notion of “scale”, so by default the float type looks for the first ten decimal places when converting. Specifying this value will override that length. Note that the MySQL float types, which do include “scale”, will use “scale” as the default for decimal_return_scale, if not otherwise specified.

class sqlalchemy.dialects.oracle.DATE¶

Provide the Oracle Database DATE type.

This type has no special Python behavior, except that it subclasses DateTime; this is to suit the fact that the Oracle Database DATE type supports a time value.

Members

__init__()

Class signature

class sqlalchemy.dialects.oracle.DATE (sqlalchemy.dialects.oracle.types._OracleDateLiteralRender, sqlalchemy.types.DateTime)

method sqlalchemy.dialects.oracle.DATE.__init__(timezone: bool = False)¶

inherited from the sqlalchemy.types.DateTime.__init__ method of DateTime

Construct a new DateTime.

参数:: timezone¶ – boolean. Indicates that the datetime type should enable timezone support, if available on the base date/time-holding type only. It is recommended to make use of the TIMESTAMP datatype directly when using this flag, as some databases include separate generic date/time-holding types distinct from the timezone-capable TIMESTAMP datatype, such as Oracle Database.

class sqlalchemy.dialects.oracle.FLOAT¶

Oracle Database FLOAT.

This is the same as FLOAT except that an Oracle Database -specific FLOAT.binary_precision parameter is accepted, and the Float.precision parameter is not accepted.

Oracle Database FLOAT types indicate precision in terms of “binary precision”, which defaults to 126. For a REAL type, the value is 63. This parameter does not cleanly map to a specific number of decimal places but is roughly equivalent to the desired number of decimal places divided by 0.3103.

在 2.0 版本加入.

Members

__init__()

Class signature

class sqlalchemy.dialects.oracle.FLOAT (sqlalchemy.types.FLOAT)

method sqlalchemy.dialects.oracle.FLOAT.__init__(binary_precision=None, asdecimal=False, decimal_return_scale=None)¶

Construct a FLOAT

参数:

binary_precision¶ – Oracle Database binary precision value to be rendered in DDL. This may be approximated to the number of decimal characters using the formula “decimal precision = 0.30103 * binary precision”. The default value used by Oracle Database for FLOAT / DOUBLE PRECISION is 126.
asdecimal¶ – See Float.asdecimal
decimal_return_scale¶ – See Float.decimal_return_scale

class sqlalchemy.dialects.oracle.INTERVAL¶

Members

__init__()

Class signature

class sqlalchemy.dialects.oracle.INTERVAL (sqlalchemy.types.NativeForEmulated, sqlalchemy.types._AbstractInterval)

method sqlalchemy.dialects.oracle.INTERVAL.__init__(day_precision=None, second_precision=None)¶

Construct an INTERVAL.

Note that only DAY TO SECOND intervals are currently supported. This is due to a lack of support for YEAR TO MONTH intervals within available DBAPIs.

参数:

day_precision¶ – the day precision value. this is the number of digits to store for the day field. Defaults to “2”
second_precision¶ – the second precision value. this is the number of digits to store for the fractional seconds field. Defaults to “6”.

class sqlalchemy.dialects.oracle.NCLOB¶

Members

__init__()

Class signature

class sqlalchemy.dialects.oracle.NCLOB (sqlalchemy.types.Text)

method sqlalchemy.dialects.oracle.NCLOB.__init__(length: int | None = None, collation: str | None = None)¶

inherited from the sqlalchemy.types.String.__init__ method of String

Create a string-holding type.

参数:

length¶ – optional, a length for the column for use in DDL and CAST expressions. May be safely omitted if no CREATE TABLE will be issued. Certain databases may require a length for use in DDL, and will raise an exception when the CREATE TABLE DDL is issued if a VARCHAR with no length is included. Whether the value is interpreted as bytes or characters is database specific.
collation¶ –
Optional, a column-level collation for use in DDL and CAST expressions. Renders using the COLLATE keyword supported by SQLite, MySQL, and PostgreSQL. E.g.:
```
>>> from sqlalchemy import cast, select, String
>>> print(select(cast("some string", String(collation="utf8"))))
SELECT CAST(:param_1 AS VARCHAR COLLATE utf8) AS anon_1
```
备注

In most cases, the Unicode or UnicodeText datatypes should be used for a Column that expects to store non-ascii data. These datatypes will ensure that the correct types are used on the database.

attribute sqlalchemy.dialects.oracle..sqlalchemy.dialects.oracle.NVARCHAR2¶: NVARCHAR 的别名

class sqlalchemy.dialects.oracle.NUMBER¶: Class signature

class sqlalchemy.dialects.oracle.NUMBER (sqlalchemy.types.Numeric, sqlalchemy.types.Integer)

class sqlalchemy.dialects.oracle.LONG¶

Members

__init__()

Class signature

class sqlalchemy.dialects.oracle.LONG (sqlalchemy.types.Text)

method sqlalchemy.dialects.oracle.LONG.__init__(length: int | None = None, collation: str | None = None)¶

inherited from the sqlalchemy.types.String.__init__ method of String

Create a string-holding type.

参数:

length¶ – optional, a length for the column for use in DDL and CAST expressions. May be safely omitted if no CREATE TABLE will be issued. Certain databases may require a length for use in DDL, and will raise an exception when the CREATE TABLE DDL is issued if a VARCHAR with no length is included. Whether the value is interpreted as bytes or characters is database specific.
collation¶ –
Optional, a column-level collation for use in DDL and CAST expressions. Renders using the COLLATE keyword supported by SQLite, MySQL, and PostgreSQL. E.g.:
```
>>> from sqlalchemy import cast, select, String
>>> print(select(cast("some string", String(collation="utf8"))))
SELECT CAST(:param_1 AS VARCHAR COLLATE utf8) AS anon_1
```
备注

In most cases, the Unicode or UnicodeText datatypes should be used for a Column that expects to store non-ascii data. These datatypes will ensure that the correct types are used on the database.

class sqlalchemy.dialects.oracle.RAW¶: Class signature

class sqlalchemy.dialects.oracle.RAW (sqlalchemy.types._Binary)

class sqlalchemy.dialects.oracle.ROWID¶

Oracle Database ROWID type.

When used in a cast() or similar, generates ROWID.

Class signature

class sqlalchemy.dialects.oracle.ROWID (sqlalchemy.types.TypeEngine)

class sqlalchemy.dialects.oracle.TIMESTAMP¶

Oracle Database implementation of TIMESTAMP, which supports additional Oracle Database-specific modes

在 2.0 版本加入.

Members

__init__()

Class signature

class sqlalchemy.dialects.oracle.TIMESTAMP (sqlalchemy.types.TIMESTAMP)

method sqlalchemy.dialects.oracle.TIMESTAMP.__init__(timezone: bool = False, local_timezone: bool = False)¶

Construct a new TIMESTAMP.

参数:

timezone¶ – boolean. Indicates that the TIMESTAMP type should use Oracle Database’s TIMESTAMP WITH TIME ZONE datatype.
local_timezone¶ – boolean. Indicates that the TIMESTAMP type should use Oracle Database’s TIMESTAMP WITH LOCAL TIME ZONE datatype.

Oracle¶

DBAPI Support¶

自动增量行为¶

指定 GENERATED AS IDENTITY（Oracle Database 12 及以上版本）¶

使用 SEQUENCE（所有 Oracle Database 版本）¶

事务隔离级别/自动提交¶

标识符大小写¶

最大标识符长度¶

LIMIT/OFFSET/FETCH 支持¶

LIMIT/OFFSET 模拟注意事项（无法使用 fetch() 方法时）¶

RETURNING 支持¶

ON UPDATE CASCADE¶

Oracle Database 8 兼容性¶

同义词/DBLINK 反射¶

约束反射¶

SYSTEM/SYSAUX 表空间的表名¶

FLOAT/DOUBLE 支持和行为¶

DateTime 兼容性¶

Oracle 数据库表选项¶

Oracle 数据库特定索引选项¶

位图索引¶

索引压缩¶

Oracle 数据库数据类型¶

python-oracledb¶

DBAPI¶

Connecting¶

描述¶

Thick模式支持¶

连接到 Oracle 数据库¶

使用 Easy Connect 字符串连接¶

传递 python-oracledb 连接参数¶

使用 tnsnames.ora TNS 别名连接¶

连接到 Oracle 自治数据库¶

连接池¶

使用 Oracle 数据库驻留连接池 (DRCP)¶

驱动程序外部 SQLAlchemy oracledb 方言使用的引擎选项¶

Unicode¶

确保客户端编码正确¶

Unicode 特定的列数据类型¶

编码错误¶

使用 setinputsizes 对 python-oracledb 数据绑定进行细粒度控制¶

示例 1 - 记录所有 setinputsizes 调用¶

示例 2 - 移除所有与 CLOB 的绑定¶

LOB 数据类型¶

返回支持¶

两阶段事务支持¶

精度数值¶

cx_Oracle¶

DBAPI¶

Connecting¶

描述¶

DSN 与主机名连接¶

使用 Easy Connect 语法的主机名连接¶

使用 tnsnames.ora 或 Oracle 自治数据库的连接¶

SID 连接¶

传递 cx_Oracle 连接参数¶

驱动程序外部 SQLAlchemy cx_Oracle 方言使用的选项¶

使用 cx_Oracle 会话池¶

使用 Oracle 数据库驻留连接池 (DRCP)¶

Unicode¶

确保正确的客户端编码¶

Unicode 特定的列数据类型¶

编码错误¶

使用 setinputsizes 对 cx_Oracle 数据绑定性能进行细粒度控制¶

示例 1 - 记录所有 setinputsizes 调用¶

示例 2 - 移除所有与 CLOB 的绑定¶

LOB 数据类型¶

RETURNING 支持¶

不支持两阶段事务¶

精度数值¶