From: Mike Bayer Date: Thu, 25 Feb 2010 23:15:39 +0000 (+0000) Subject: - merge -r6823:6841 of branches/chrisw_mixin X-Git-Tag: rel_0_6beta2~119 X-Git-Url: http://git.ipfire.org/gitweb.cgi?a=commitdiff_plain;h=f60a5fbd5d52c74a33441ea834ede839584595fc;p=thirdparty%2Fsqlalchemy%2Fsqlalchemy.git - merge -r6823:6841 of branches/chrisw_mixin - declarative now accepts mixin classes directly, as a means to provide common functional and column-based elements on all subclasses, as well as a means to propagate a fixed set of __table_args__ or __mapper_args__ to subclasses. For custom combinations of __table_args__/__mapper_args__ from an inherited mixin to local, descriptors can now be used. New details are all up in the Declarative documentation. Thanks to Chris Withers for putting up with my strife on this. [ticket:1707] --- diff --git a/CHANGES b/CHANGES index 82a15639eb..5e0d3662dc 100644 --- a/CHANGES +++ b/CHANGES @@ -201,11 +201,21 @@ CHANGES in theory make it easier for custom metaclasses to modify the state passed into _as_declarative. - - the __mapper_args__ dict is copied when propagating to a subclass, - and is taken straight off the class __dict__ to avoid any - propagation from the parent. mapper inheritance already - propagates the things you want from the parent mapper. - [ticket:1393] + - declarative now accepts mixin classes directly, as a means + to provide common functional and column-based elements on + all subclasses, as well as a means to propagate a fixed + set of __table_args__ or __mapper_args__ to subclasses. + For custom combinations of __table_args__/__mapper_args__ from + an inherited mixin to local, descriptors can now be used. + New details are all up in the Declarative documentation. + Thanks to Chris Withers for putting up with my strife + on this. [ticket:1707] + + - the __mapper_args__ dict is copied when propagating to a subclass, + and is taken straight off the class __dict__ to avoid any + propagation from the parent. mapper inheritance already + propagates the things you want from the parent mapper. + [ticket:1393] - mysql - Fixed reflection bug whereby when COLLATE was present, diff --git a/doc/build/conf.py b/doc/build/conf.py index 8b7650e14f..bee23d3086 100644 --- a/doc/build/conf.py +++ b/doc/build/conf.py @@ -67,7 +67,7 @@ release = sqlalchemy.__version__ #today_fmt = '%B %d, %Y' # List of documents that shouldn't be included in the build. -unused_docs = ['output.txt'] +unused_docs = ['output.txt','copyright'] # List of directories, relative to source directory, that shouldn't be searched # for source files. diff --git a/doc/build/intro.rst b/doc/build/intro.rst index d4fdfccdfd..dd4a975076 100644 --- a/doc/build/intro.rst +++ b/doc/build/intro.rst @@ -52,7 +52,7 @@ Installing SQLAlchemy from scratch is most easily achieved with `setuptools `_ and install it to your system. -* `setuptools `_ +* setuptools_ * `install setuptools `_ * `pypi `_ diff --git a/doc/build/ormtutorial.rst b/doc/build/ormtutorial.rst index 4b7eaffeb0..e93364387f 100644 --- a/doc/build/ormtutorial.rst +++ b/doc/build/ormtutorial.rst @@ -37,7 +37,7 @@ Next we want to tell SQLAlchemy about our tables. We will start with just a sin ... Column('password', String) ... ) -All about how to define :class:`~sqlalchemy.schema.Table` objects, as well as how to load their definition from an existing database (known as **reflection**), is described in :ref:`metadata_toplevel`. +:ref:`metadata_toplevel` covers all about how to define :class:`~sqlalchemy.schema.Table` objects, as well as how to load their definition from an existing database (known as **reflection**). Next, we can issue CREATE TABLE statements derived from our table metadata, by calling ``create_all()`` and passing it the ``engine`` instance which points to our database. This will check for the presence of a table first before creating, so it's safe to call multiple times: @@ -124,7 +124,15 @@ Since we have not yet told SQLAlchemy to persist ``Ed Jones`` within the databas Creating Table, Class and Mapper All at Once Declaratively =========================================================== -The preceding approach to configuration involving a :class:`~sqlalchemy.schema.Table`, user-defined class, and ``mapper()`` call illustrate classical SQLAlchemy usage, which values the highest separation of concerns possible. A large number of applications don't require this degree of separation, and for those SQLAlchemy offers an alternate "shorthand" configurational style called **declarative**. For many applications, this is the only style of configuration needed. Our above example using this style is as follows:: +The preceding approach to configuration involved a +:class:`~sqlalchemy.schema.Table`, a user-defined class, and +a call to``mapper()``. This illustrates classical SQLAlchemy usage, which values +the highest separation of concerns possible. +A large number of applications don't require this degree of +separation, and for those SQLAlchemy offers an alternate "shorthand" +configurational style called :mod:`~sqlalchemy.ext.declarative`. +For many applications, this is the only style of configuration needed. +Our above example using this style is as follows:: >>> from sqlalchemy.ext.declarative import declarative_base @@ -145,16 +153,25 @@ The preceding approach to configuration involving a :class:`~sqlalchemy.schema.T ... def __repr__(self): ... return "" % (self.name, self.fullname, self.password) -Above, the ``declarative_base()`` function defines a new class which we name ``Base``, from which all of our ORM-enabled classes will derive. Note that we define :class:`~sqlalchemy.schema.Column` objects with no "name" field, since it's inferred from the given attribute name. +Above, the ``declarative_base()`` function defines a new class which +we name ``Base``, from which all of our ORM-enabled classes will +derive. Note that we define :class:`~sqlalchemy.schema.Column` +objects with no "name" field, since it's inferred from the given +attribute name. -The underlying :class:`~sqlalchemy.schema.Table` object created by our ``declarative_base()`` version of ``User`` is accessible via the ``__table__`` attribute:: +The underlying :class:`~sqlalchemy.schema.Table` object created by our +``declarative_base()`` version of ``User`` is accessible via the +``__table__`` attribute:: >>> users_table = User.__table__ -and the owning :class:`~sqlalchemy.schema.MetaData` object is available as well:: +The owning :class:`~sqlalchemy.schema.MetaData` object is available as well:: >>> metadata = Base.metadata +Full documentation for :mod:`~sqlalchemy.ext.declarative` can be found +in the :doc:`reference/index` section for :doc:`reference/ext/declarative`. + Yet another "declarative" method is available for SQLAlchemy as a third party library called `Elixir `_. This is a full-featured configurational product which also includes many higher level mapping configurations built in. Like declarative, once classes and mappings are defined, ORM usage is the same as with a classical SQLAlchemy configuration. Creating a Session diff --git a/doc/build/reference/ext/declarative.rst b/doc/build/reference/ext/declarative.rst index 81c713df72..5762d03783 100644 --- a/doc/build/reference/ext/declarative.rst +++ b/doc/build/reference/ext/declarative.rst @@ -2,4 +2,16 @@ declarative =========== .. automodule:: sqlalchemy.ext.declarative - :members: \ No newline at end of file + +API Reference +------------- + +.. autofunction:: declarative_base + +.. autofunction:: _declarative_constructor + +.. autofunction:: synonym_for + +.. autofunction:: comparable_using + +.. autofunction:: instrument_declarative diff --git a/lib/sqlalchemy/ext/declarative.py b/lib/sqlalchemy/ext/declarative.py index 6d39418583..65c9df1333 100644 --- a/lib/sqlalchemy/ext/declarative.py +++ b/lib/sqlalchemy/ext/declarative.py @@ -1,13 +1,16 @@ -"""A simple declarative layer for SQLAlchemy ORM. - +""" Synopsis ======== -SQLAlchemy object-relational configuration involves the usage of :class:`~sqlalchemy.schema.Table`, -:func:`~sqlalchemy.orm.mapper`, and class objects to define the three areas of configuration. -``declarative`` moves these three types of configuration underneath the individual -mapped class. Regular SQLAlchemy schema and ORM constructs are used in most -cases:: +SQLAlchemy object-relational configuration involves the use of +:class:`~sqlalchemy.schema.Table`, :func:`~sqlalchemy.orm.mapper`, and +class objects to define the three areas of configuration. +:mod:`~sqlalchemy.ext.declarative` allows all three types of +configuration to be expressed declaratively on an individual +mapped class. Regular SQLAlchemy schema elements and ORM constructs +are used in most cases. + +As a simple example:: from sqlalchemy.ext.declarative import declarative_base @@ -18,16 +21,21 @@ cases:: id = Column(Integer, primary_key=True) name = Column(String(50)) -Above, the :func:`declarative_base` callable produces a new base class from which -all mapped classes inherit from. When the class definition is completed, a -new :class:`~sqlalchemy.schema.Table` and :class:`~sqlalchemy.orm.mapper` have been generated, accessible via the -``__table__`` and ``__mapper__`` attributes on the ``SomeClass`` class. +Above, the :func:`declarative_base` callable returns a new base class from which +all mapped classes should inherit. When the class definition is completed, a +new :class:`~sqlalchemy.schema.Table` and +:class:`~sqlalchemy.orm.mapper` will have been generated, accessible +via the ``__table__`` and ``__mapper__`` attributes on the ``SomeClass`` class. Defining Attributes =================== -:class:`~sqlalchemy.schema.Column` objects may be explicitly named, -including using a different name than the attribute in which they are associated. +In the above example, the :class:`~sqlalchemy.schema.Column` objects are +automatically named with the name of the attribute to which they are +assigned. + +They can also be explicitly named, and that name does not have to be +the same as name assigned on the class. The column will be assigned to the :class:`~sqlalchemy.schema.Table` using the given name, and mapped to the class using the attribute name:: @@ -36,52 +44,54 @@ given name, and mapped to the class using the attribute name:: id = Column("some_table_id", Integer, primary_key=True) name = Column("name", String(50)) -Otherwise, you may omit the names from the Column definitions. -Declarative will set the ``name`` attribute on the column when the class -is initialized:: - - class SomeClass(Base): - __tablename__ = 'some_table' - id = Column(Integer, primary_key=True) - name = Column(String(50)) - Attributes may be added to the class after its construction, and they will be -added to the underlying :class:`~sqlalchemy.schema.Table` and :func:`~sqlalchemy.orm.mapper()` definitions as -appropriate:: +added to the underlying :class:`~sqlalchemy.schema.Table` and +:func:`~sqlalchemy.orm.mapper()` definitions as appropriate:: SomeClass.data = Column('data', Unicode) SomeClass.related = relation(RelatedInfo) -Classes which are mapped explicitly using :func:`~sqlalchemy.orm.mapper()` can interact freely -with declarative classes. It is recommended, though not required, that all tables -share the same underlying :class:`~sqlalchemy.schema.MetaData` object, so that -string-configured :class:`~sqlalchemy.schema.ForeignKey` references can be resolved without issue. +Classes which are mapped explicitly using +:func:`~sqlalchemy.orm.mapper()` can interact freely with declarative +classes. + +It is recommended, though not required, that all tables +share the same underlying :class:`~sqlalchemy.schema.MetaData` object, +so that string-configured :class:`~sqlalchemy.schema.ForeignKey` +references can be resolved without issue. Association of Metadata and Engine ================================== -The :func:`declarative_base` base class contains a :class:`~sqlalchemy.schema.MetaData` object where newly -defined :class:`~sqlalchemy.schema.Table` objects are collected. This is accessed via the -:class:`~sqlalchemy.schema.MetaData` class level accessor, so to create tables we can say:: +The :func:`declarative_base` base class contains a +:class:`~sqlalchemy.schema.MetaData` object where newly +defined :class:`~sqlalchemy.schema.Table` objects are collected. This +is accessed via the :class:`~sqlalchemy.schema.MetaData` class level +accessor, so to create tables we can say:: engine = create_engine('sqlite://') Base.metadata.create_all(engine) -The :class:`~sqlalchemy.engine.base.Engine` created above may also be directly associated with the -declarative base class using the ``bind`` keyword argument, where it will be -associated with the underlying :class:`~sqlalchemy.schema.MetaData` object and allow SQL operations +The :class:`~sqlalchemy.engine.base.Engine` created above may also be +directly associated with the declarative base class using the ``bind`` +keyword argument, where it will be associated with the underlying +:class:`~sqlalchemy.schema.MetaData` object and allow SQL operations involving that metadata and its tables to make use of that engine automatically:: Base = declarative_base(bind=create_engine('sqlite://')) -Or, as :class:`~sqlalchemy.schema.MetaData` allows, at any time using the ``bind`` attribute:: +Alternatively, by way of the normal +:class:`~sqlalchemy.schema.MetaData` behaviour, the ``bind`` attribute +of the class level accessor can be assigned at any time as follows:: Base.metadata.bind = create_engine('sqlite://') -The :func:`declarative_base` can also receive a pre-created :class:`~sqlalchemy.schema.MetaData` object, -which allows a declarative setup to be associated with an already existing -traditional collection of :class:`~sqlalchemy.schema.Table` objects:: +The :func:`declarative_base` can also receive a pre-created +:class:`~sqlalchemy.schema.MetaData` object, which allows a +declarative setup to be associated with an already +existing traditional collection of :class:`~sqlalchemy.schema.Table` +objects:: mymetadata = MetaData() Base = declarative_base(metadata=mymetadata) @@ -89,11 +99,12 @@ traditional collection of :class:`~sqlalchemy.schema.Table` objects:: Configuring Relations ===================== -Relations to other classes are done in the usual way, with the added feature -that the class specified to :func:`~sqlalchemy.orm.relation()` may be a string name. The "class -registry" associated with ``Base`` is used at mapper compilation time to -resolve the name into the actual class object, which is expected to have been -defined once the mapper configuration is used:: +Relations to other classes are done in the usual way, with the added +feature that the class specified to :func:`~sqlalchemy.orm.relation()` +may be a string name. The "class registry" associated with ``Base`` +is used at mapper compilation time to resolve the name into the actual +class object, which is expected to have been defined once the mapper +configuration is used:: class User(Base): __tablename__ = 'users' @@ -109,8 +120,9 @@ defined once the mapper configuration is used:: email = Column(String(50)) user_id = Column(Integer, ForeignKey('users.id')) -Column constructs, since they are just that, are immediately usable, as below -where we define a primary join condition on the ``Address`` class using them:: +Column constructs, since they are just that, are immediately usable, +as below where we define a primary join condition on the ``Address`` +class using them:: class Address(Base): __tablename__ = 'addresses' @@ -120,52 +132,60 @@ where we define a primary join condition on the ``Address`` class using them:: user_id = Column(Integer, ForeignKey('users.id')) user = relation(User, primaryjoin=user_id == User.id) -In addition to the main argument for :func:`~sqlalchemy.orm.relation`, other arguments -which depend upon the columns present on an as-yet undefined class -may also be specified as strings. These strings are evaluated as -Python expressions. The full namespace available within this -evaluation includes all classes mapped for this declarative base, -as well as the contents of the ``sqlalchemy`` package, including -expression functions like :func:`~sqlalchemy.sql.expression.desc` and :attr:`~sqlalchemy.sql.expression.func`:: +In addition to the main argument for :func:`~sqlalchemy.orm.relation`, +other arguments which depend upon the columns present on an as-yet +undefined class may also be specified as strings. These strings are +evaluated as Python expressions. The full namespace available within +this evaluation includes all classes mapped for this declarative base, +as well as the contents of the ``sqlalchemy`` package, including +expression functions like :func:`~sqlalchemy.sql.expression.desc` and +:attr:`~sqlalchemy.sql.expression.func`:: class User(Base): # .... - addresses = relation("Address", order_by="desc(Address.email)", - primaryjoin="Address.user_id==User.id") + addresses = relation("Address", + order_by="desc(Address.email)", + primaryjoin="Address.user_id==User.id") As an alternative to string-based attributes, attributes may also be defined after all classes have been created. Just add them to the target class after the fact:: - User.addresses = relation(Address, primaryjoin=Address.user_id == User.id) + User.addresses = relation(Address, + primaryjoin=Address.user_id==User.id) Configuring Many-to-Many Relations ================================== -There's nothing special about many-to-many with declarative. The ``secondary`` -argument to :func:`~sqlalchemy.orm.relation` still requires a :class:`~sqlalchemy.schema.Table` object, not a declarative class. -The :class:`~sqlalchemy.schema.Table` should share the same :class:`~sqlalchemy.schema.MetaData` object used by the declarative base:: - - keywords = Table('keywords', Base.metadata, - Column('author_id', Integer, ForeignKey('authors.id')), - Column('keyword_id', Integer, ForeignKey('keywords.id')) - ) +There's nothing special about many-to-many with declarative. The +``secondary`` argument to :func:`~sqlalchemy.orm.relation` still +requires a :class:`~sqlalchemy.schema.Table` object, not a declarative +class. The :class:`~sqlalchemy.schema.Table` should share the same +:class:`~sqlalchemy.schema.MetaData` object used by the declarative +base:: + + keywords = Table( + 'keywords', Base.metadata, + Column('author_id', Integer, ForeignKey('authors.id')), + Column('keyword_id', Integer, ForeignKey('keywords.id')) + ) class Author(Base): __tablename__ = 'authors' id = Column(Integer, primary_key=True) keywords = relation("Keyword", secondary=keywords) -You should generally **not** map a class and also specify its table in a many-to-many -relation, since the ORM may issue duplicate INSERT and DELETE statements. +You should generally **not** map a class and also specify its table in +a many-to-many relation, since the ORM may issue duplicate INSERT and +DELETE statements. Defining Synonyms ================= -Synonyms are introduced in :ref:`synonyms`. To define a getter/setter which -proxies to an underlying attribute, use :func:`~sqlalchemy.orm.synonym` with the -``descriptor`` argument:: +Synonyms are introduced in :ref:`synonyms`. To define a getter/setter +which proxies to an underlying attribute, use +:func:`~sqlalchemy.orm.synonym` with the ``descriptor`` argument:: class MyClass(Base): __tablename__ = 'sometable' @@ -185,8 +205,8 @@ class-level expression construct:: x.attr = "some value" session.query(MyClass).filter(MyClass.attr == 'some other value').all() -For simple getters, the :func:`synonym_for` decorator can be used in conjunction -with ``@property``:: +For simple getters, the :func:`synonym_for` decorator can be used in +conjunction with ``@property``:: class MyClass(Base): __tablename__ = 'sometable' @@ -198,8 +218,8 @@ with ``@property``:: def attr(self): return self._some_attr -Similarly, :func:`comparable_using` is a front end for the :func:`~sqlalchemy.orm.comparable_property` -ORM function:: +Similarly, :func:`comparable_using` is a front end for the +:func:`~sqlalchemy.orm.comparable_property` ORM function:: class MyClass(Base): __tablename__ = 'sometable' @@ -214,18 +234,21 @@ ORM function:: Table Configuration =================== -Table arguments other than the name, metadata, and mapped Column arguments -are specified using the ``__table_args__`` class attribute. This attribute -accommodates both positional as well as keyword arguments that are normally -sent to the :class:`~sqlalchemy.schema.Table` constructor. The attribute can be specified -in one of two forms. One is as a dictionary:: +Table arguments other than the name, metadata, and mapped Column +arguments are specified using the ``__table_args__`` class attribute. +This attribute accommodates both positional as well as keyword +arguments that are normally sent to the +:class:`~sqlalchemy.schema.Table` constructor. +The attribute can be specified in one of two forms. One is as a +dictionary:: class MyClass(Base): __tablename__ = 'sometable' __table_args__ = {'mysql_engine':'InnoDB'} -The other, a tuple of the form ``(arg1, arg2, ..., {kwarg1:value, ...})``, which -allows positional arguments to be specified as well (usually constraints):: +The other, a tuple of the form +``(arg1, arg2, ..., {kwarg1:value, ...})``, which allows positional +arguments to be specified as well (usually constraints):: class MyClass(Base): __tablename__ = 'sometable' @@ -235,12 +258,14 @@ allows positional arguments to be specified as well (usually constraints):: {'autoload':True} ) -Note that the dictionary is required in the tuple form even if empty. +Note that the keyword parameters dictionary is required in the tuple +form even if empty. -As an alternative to ``__tablename__``, a direct :class:`~sqlalchemy.schema.Table` -construct may be used. The :class:`~sqlalchemy.schema.Column` objects, which -in this case require their names, will be -added to the mapping just like a regular mapping to a table:: +As an alternative to ``__tablename__``, a direct +:class:`~sqlalchemy.schema.Table` construct may be used. The +:class:`~sqlalchemy.schema.Column` objects, which in this case require +their names, will be added to the mapping just like a regular mapping +to a table:: class MyClass(Base): __table__ = Table('my_table', Base.metadata, @@ -251,9 +276,14 @@ added to the mapping just like a regular mapping to a table:: Mapper Configuration ==================== -Mapper arguments are specified using the ``__mapper_args__`` class variable, -which is a dictionary that accepts the same names as the :class:`~sqlalchemy.orm.mapper` -function accepts as keywords:: +Configuration of mappers is done with the +:func:`~sqlalchemy.orm.mapper` function and all the possible mapper +configuration parameters can be found in the documentation for that +function. + +:func:`~sqlalchemy.orm.mapper` is still used by declaratively mapped +classes and keyword parameters to the function can be passed by +placing them in the ``__mapper_args__`` class variable:: class Widget(Base): __tablename__ = 'widgets' @@ -265,7 +295,7 @@ Inheritance Configuration ========================= Declarative supports all three forms of inheritance as intuitively -as possible. The ``inherits`` mapper keyword argument is not needed, +as possible. The ``inherits`` mapper keyword argument is not needed as declarative will determine this from the class itself. The various "polymorphic" keyword arguments are specified using ``__mapper_args__``. @@ -287,11 +317,12 @@ table:: id = Column(Integer, ForeignKey('people.id'), primary_key=True) primary_language = Column(String(50)) -Note that above, the ``Engineer.id`` attribute, since it shares the same -attribute name as the ``Person.id`` attribute, will in fact represent the ``people.id`` -and ``engineers.id`` columns together, and will render inside a query as ``"people.id"``. -To provide the ``Engineer`` class with an attribute that represents only the -``engineers.id`` column, give it a different attribute name:: +Note that above, the ``Engineer.id`` attribute, since it shares the +same attribute name as the ``Person.id`` attribute, will in fact +represent the ``people.id`` and ``engineers.id`` columns together, and +will render inside a query as ``"people.id"``. +To provide the ``Engineer`` class with an attribute that represents +only the ``engineers.id`` column, give it a different attribute name:: class Engineer(Person): __tablename__ = 'engineers' @@ -302,8 +333,9 @@ To provide the ``Engineer`` class with an attribute that represents only the Single Table Inheritance ~~~~~~~~~~~~~~~~~~~~~~~~ -Single table inheritance is defined as a subclass that does not have its -own table; you just leave out the ``__table__`` and ``__tablename__`` attributes:: +Single table inheritance is defined as a subclass that does not have +its own table; you just leave out the ``__table__`` and ``__tablename__`` +attributes:: class Person(Base): __tablename__ = 'people' @@ -315,28 +347,31 @@ own table; you just leave out the ``__table__`` and ``__tablename__`` attributes __mapper_args__ = {'polymorphic_identity': 'engineer'} primary_language = Column(String(50)) -When the above mappers are configured, the ``Person`` class is mapped to the ``people`` -table *before* the ``primary_language`` column is defined, and this column will not be included -in its own mapping. When ``Engineer`` then defines the ``primary_language`` -column, the column is added to the ``people`` table so that it is included in the mapping -for ``Engineer`` and is also part of the table's full set of columns. -Columns which are not mapped to ``Person`` are also excluded from any other -single or joined inheriting classes using the ``exclude_properties`` mapper argument. -Below, ``Manager`` will have all the attributes of ``Person`` and ``Manager`` but *not* -the ``primary_language`` attribute of ``Engineer``:: +When the above mappers are configured, the ``Person`` class is mapped +to the ``people`` table *before* the ``primary_language`` column is +defined, and this column will not be included in its own mapping. +When ``Engineer`` then defines the ``primary_language`` column, the +column is added to the ``people`` table so that it is included in the +mapping for ``Engineer`` and is also part of the table's full set of +columns. Columns which are not mapped to ``Person`` are also excluded +from any other single or joined inheriting classes using the +``exclude_properties`` mapper argument. Below, ``Manager`` will have +all the attributes of ``Person`` and ``Manager`` but *not* the +``primary_language`` attribute of ``Engineer``:: class Manager(Person): __mapper_args__ = {'polymorphic_identity': 'manager'} golf_swing = Column(String(50)) -The attribute exclusion logic is provided by the ``exclude_properties`` mapper argument, -and declarative's default behavior can be disabled by passing an explicit -``exclude_properties`` collection (empty or otherwise) to the ``__mapper_args__``. +The attribute exclusion logic is provided by the +``exclude_properties`` mapper argument, and declarative's default +behavior can be disabled by passing an explicit ``exclude_properties`` +collection (empty or otherwise) to the ``__mapper_args__``. Concrete Table Inheritance ~~~~~~~~~~~~~~~~~~~~~~~~~~ -Concrete is defined as a subclass which has its own table and sets the +Concrete is defined as a subclass which has its own table and sets the ``concrete`` keyword argument to ``True``:: class Person(Base): @@ -351,8 +386,8 @@ Concrete is defined as a subclass which has its own table and sets the primary_language = Column(String(50)) name = Column(String(50)) -Usage of an abstract base class is a little less straightforward as it requires -usage of :func:`~sqlalchemy.orm.util.polymorphic_union`:: +Usage of an abstract base class is a little less straightforward as it +requires usage of :func:`~sqlalchemy.orm.util.polymorphic_union`:: engineers = Table('engineers', Base.metadata, Column('id', Integer, primary_key=True), @@ -382,19 +417,78 @@ usage of :func:`~sqlalchemy.orm.util.polymorphic_union`:: __table__ = managers __mapper_args__ = {'polymorphic_identity':'manager', 'concrete':True} -Class Usage -=========== + +Mix-in Classes +============== + +A common need when using :mod:`~sqlalchemy.ext.declarative` is to +share some functionality, often a set of columns, across many +classes. The normal python idiom would be to put this common code into +a base class and have all the other classes subclass this class. + +When using :mod:`~sqlalchemy.ext.declarative`, this need is met by +using a "mix-in class". A mix-in class is one that isn't mapped to a +table and doesn't subclass the declarative :class:`Base`. For example:: + + class MyMixin(object): + + __table_args__ = {'mysql_engine':'InnoDB'} + __mapper_args__=dict(always_refresh=True) + id = Column(Integer, primary_key=True) + + def foo(self): + return 'bar'+str(self.id) + + class MyModel(Base,MyMixin): + __tablename__='test' + name = Column(String(1000), nullable=False, index=True) + +As the above example shows, ``__table_args__`` and ``__mapper_args__`` +can both be abstracted out into a mix-in if you use common values for +these across many classes. + +However, particularly in the case of ``__table_args__``, you may want +to combine some parameters from several mix-ins with those you wish to +define on the class iteself. To help with this, a +:func:`~sqlalchemy.util.classproperty` decorator is provided that lets +you implement a class property with a function. For example:: + + from sqlalchemy.util import classproperty + + class MySQLSettings: + __table_args__ = {'mysql_engine':'InnoDB'} + + class MyOtherMixin: + __table_args__ = {'info':'foo'} + + class MyModel(Base,MySQLSettings,MyOtherMixin): + __tablename__='my_model' + + @classproperty + def __table_args__(self): + args = dict() + args.update(MySQLSettings.__table_args__) + args.update(MyOtherMixin.__table_args__) + return args + + id = Column(Integer, primary_key=True) + +Class Constructor +================= As a convenience feature, the :func:`declarative_base` sets a default -constructor on classes which takes keyword arguments, and assigns them to the -named attributes:: +constructor on classes which takes keyword arguments, and assigns them +to the named attributes:: e = Engineer(primary_language='python') -Note that ``declarative`` has no integration built in with sessions, and is -only intended as an optional syntax for the regular usage of mappers and Table -objects. A typical application setup using :func:`~sqlalchemy.orm.scoped_session` might look -like:: +Sessions +======== + +Note that ``declarative`` does nothing special with sessions, and is +only intended as an easier way to configure mappers and +:class:`~sqlalchemy.schema.Table` objects. A typical application +setup using :func:`~sqlalchemy.orm.scoped_session` might look like:: engine = create_engine('postgresql://scott:tiger@localhost/test') Session = scoped_session(sessionmaker(autocommit=False, @@ -402,7 +496,8 @@ like:: bind=engine)) Base = declarative_base() -Mapped instances then make usage of :class:`~sqlalchemy.orm.session.Session` in the usual way. +Mapped instances then make usage of +:class:`~sqlalchemy.orm.session.Session` in the usual way. """ @@ -419,8 +514,8 @@ __all__ = 'declarative_base', 'synonym_for', 'comparable_using', 'instrument_dec def instrument_declarative(cls, registry, metadata): """Given a class, configure the class declaratively, - using the given registry (any dictionary) and MetaData object. - This operation does not assume any kind of class hierarchy. + using the given registry, which can be any dictionary, and + MetaData object. """ if '_decl_class_registry' in cls.__dict__: @@ -432,6 +527,31 @@ def instrument_declarative(cls, registry, metadata): _as_declarative(cls, cls.__name__, cls.__dict__) def _as_declarative(cls, classname, dict_): + + # this spelling enables these attributes to be descriptors + mapper_args = '__mapper_args__' in dict_ and cls.__mapper_args__ or {} + table_args = '__table_args__' in dict_ and cls.__table_args__ or None + + # dict_ will be a dictproxy, which we can't write to, and we need to! + dict_ = dict(dict_) + + column_copies = dict() + + for base in cls.__bases__: + names = dir(base) + if not _is_mapped_class(base): + for name in names: + obj = getattr(base,name) + if isinstance(obj, Column): + dict_[name]=column_copies[obj]=obj.copy() + mapper_args = mapper_args or getattr(base,'__mapper_args__',mapper_args) + table_args = table_args or getattr(base,'__table_args__',None) + + # make sure that column copies are used rather than the original columns + # from any mixins + for k, v in mapper_args.iteritems(): + mapper_args[k] = column_copies.get(v,v) + cls._decl_class_registry[classname] = cls our_stuff = util.OrderedDict() for k in dict_: @@ -474,7 +594,6 @@ def _as_declarative(cls, classname, dict_): if '__tablename__' in dict_: tablename = cls.__tablename__ - table_args = dict_.get('__table_args__') if isinstance(table_args, dict): args, table_kw = (), table_args elif isinstance(table_args, tuple): @@ -500,12 +619,9 @@ def _as_declarative(cls, classname, dict_): for c in cols: if not table.c.contains_column(c): raise exceptions.ArgumentError( - "Can't add additional column %r when specifying __table__" % key) + "Can't add additional column %r when specifying __table__" % key + ) - if '__mapper_args__' in dict_: - mapper_args = dict(dict_['__mapper_args__']) - else: - mapper_args = {} if 'inherits' not in mapper_args: for c in cls.__bases__: if _is_mapped_class(c): @@ -518,8 +634,10 @@ def _as_declarative(cls, classname, dict_): mapper_cls = mapper if table is None and 'inherits' not in mapper_args: - raise exceptions.InvalidRequestError("Class %r does not have a __table__ or __tablename__ " - "specified and does not inherit from an existing table-mapped class." % cls) + raise exceptions.InvalidRequestError( + "Class %r does not have a __table__ or __tablename__ " + "specified and does not inherit from an existing table-mapped class." % cls + ) elif 'inherits' in mapper_args and not mapper_args.get('concrete', False): inherited_mapper = class_mapper(mapper_args['inherits'], compile=False) @@ -536,25 +654,30 @@ def _as_declarative(cls, classname, dict_): if table is None: # single table inheritance. # ensure no table args - table_args = cls.__dict__.get('__table_args__') if table_args is not None: - raise exceptions.ArgumentError("Can't place __table_args__ on an inherited class with no table.") + raise exceptions.ArgumentError( + "Can't place __table_args__ on an inherited class with no table." + ) # add any columns declared here to the inherited table. for c in cols: if c.primary_key: - raise exceptions.ArgumentError("Can't place primary key columns on an inherited class with no table.") + raise exceptions.ArgumentError( + "Can't place primary key columns on an inherited class with no table." + ) inherited_table.append_column(c) # single or joined inheritance - # exclude any cols on the inherited table which are not mapped on the parent class, to avoid + # exclude any cols on the inherited table which are not mapped on the + # parent class, to avoid # mapping columns specific to sibling/nephew classes inherited_mapper = class_mapper(mapper_args['inherits'], compile=False) inherited_table = inherited_mapper.local_table if 'exclude_properties' not in mapper_args: mapper_args['exclude_properties'] = exclude_properties = \ - set([c.key for c in inherited_table.c if c not in inherited_mapper._columntoproperty]) + set([c.key for c in inherited_table.c + if c not in inherited_mapper._columntoproperty]) exclude_properties.difference_update([c.key for c in cols]) cls.__mapper__ = mapper_cls(cls, table, properties=our_stuff, **mapper_args) @@ -633,14 +756,16 @@ def _deferred_relation(cls, prop): return return_cls if isinstance(prop, PropertyLoader): - for attr in ('argument', 'order_by', 'primaryjoin', 'secondaryjoin', 'secondary', '_foreign_keys', 'remote_side'): + for attr in ('argument', 'order_by', 'primaryjoin', 'secondaryjoin', + 'secondary', '_foreign_keys', 'remote_side'): v = getattr(prop, attr) if isinstance(v, basestring): setattr(prop, attr, resolve_arg(v)) if prop.backref and isinstance(prop.backref, tuple): key, kwargs = prop.backref - for attr in ('primaryjoin', 'secondaryjoin', 'secondary', 'foreign_keys', 'remote_side', 'order_by'): + for attr in ('primaryjoin', 'secondaryjoin', 'secondary', + 'foreign_keys', 'remote_side', 'order_by'): if attr in kwargs and isinstance(kwargs[attr], basestring): kwargs[attr] = resolve_arg(kwargs[attr]) @@ -672,7 +797,8 @@ def synonym_for(name, map_column=False): def comparable_using(comparator_factory): """Decorator, allow a Python @property to be used in query criteria. - A decorator front end to :func:`~sqlalchemy.orm.comparable_property`, passes + This is a decorator front end to + :func:`~sqlalchemy.orm.comparable_property` that passes through the comparator_factory and the function being decorated:: @comparable_using(MyComparatorType) @@ -693,10 +819,12 @@ def comparable_using(comparator_factory): def _declarative_constructor(self, **kwargs): """A simple constructor that allows initialization from kwargs. - Sets kwargs on the constructed instance. Only keys that are present as - attributes of type(self) are allowed (for example, any mapped column or - relation). - + Sets attributes on the constructed instance using the names and + values in ``kwargs``. + + Only keys that are present as + attributes of the instance's class are allowed. These could be, + for example, any mapped columns or relations. """ for k in kwargs: if not hasattr(type(self), k): @@ -711,22 +839,25 @@ def declarative_base(bind=None, metadata=None, mapper=None, cls=object, metaclass=DeclarativeMeta): """Construct a base class for declarative class definitions. - The new base class will be given a metaclass that invokes - :func:`instrument_declarative()` upon each subclass definition, and routes - later Column- and Mapper-related attribute assignments made on the class - into Table and Mapper assignments. + The new base class will be given a metaclass that produces + appropriate :class:`~sqlalchemy.schema.Table` objects and makes + the appropriate :func:`~sqlalchemy.orm.mapper` calls based on the + information provided declaratively in the class and any subclasses + of the class. - :param bind: An optional :class:`~sqlalchemy.engine.base.Connectable`, will be assigned - the ``bind`` attribute on the :class:`~sqlalchemy.MetaData` instance. + :param bind: An optional + :class:`~sqlalchemy.engine.base.Connectable`, will be assigned + the ``bind`` attribute on the :class:`~sqlalchemy.MetaData` + instance. :param metadata: - An optional :class:`~sqlalchemy.MetaData` instance. All :class:`~sqlalchemy.schema.Table` - objects implicitly declared by + An optional :class:`~sqlalchemy.MetaData` instance. All + :class:`~sqlalchemy.schema.Table` objects implicitly declared by subclasses of the base will share this MetaData. A MetaData instance - will be create if none is provided. The MetaData instance will be - available via the `metadata` attribute of the generated declarative - base class. + will be created if none is provided. The + :class:`~sqlalchemy.MetaData` instance will be available via the + `metadata` attribute of the generated declarative base class. :param mapper: An optional callable, defaults to :func:`~sqlalchemy.orm.mapper`. Will be @@ -734,7 +865,7 @@ def declarative_base(bind=None, metadata=None, mapper=None, cls=object, :param cls: Defaults to :class:`object`. A type to use as the base for the generated - declarative base class. May be a type or tuple of types. + declarative base class. May be a class or tuple of classes. :param name: Defaults to ``Base``. The display name for the generated @@ -742,11 +873,12 @@ def declarative_base(bind=None, metadata=None, mapper=None, cls=object, tracebacks and debugging. :param constructor: - Defaults to declarative._declarative_constructor, an __init__ - implementation that assigns \**kwargs for declared fields and relations - to an instance. If ``None`` is supplied, no __init__ will be installed - and construction will fall back to cls.__init__ with normal Python - semantics. + Defaults to + :func:`~sqlalchemy.ext.declarative._declarative_constructor`, an + __init__ implementation that assigns \**kwargs for declared + fields and relations to an instance. If ``None`` is supplied, + no __init__ will be provided and construction will fall back to + cls.__init__ by way of the normal Python semantics. :param metaclass: Defaults to :class:`DeclarativeMeta`. A metaclass or __metaclass__ diff --git a/test/ext/test_declarative.py b/test/ext/test_declarative.py index 4722427d50..457936d78f 100644 --- a/test/ext/test_declarative.py +++ b/test/ext/test_declarative.py @@ -8,7 +8,7 @@ from sqlalchemy import MetaData, Integer, String, ForeignKey, ForeignKeyConstrai from sqlalchemy.test.schema import Table, Column from sqlalchemy.orm import relation, create_session, class_mapper, eagerload, compile_mappers, backref, clear_mappers, polymorphic_union, deferred from sqlalchemy.test.testing import eq_ - +from sqlalchemy.util import classproperty from test.orm._base import ComparableEntity, MappedTest @@ -1473,7 +1473,10 @@ class DeclarativeInheritanceTest(DeclarativeTestBase): class Engineer(Person): __mapper_args__ = {'polymorphic_identity':'engineer'} primary_language = Column('primary_language', String(50)) - __table_args__ = () + # this should be on the Person class, as this is single + # table inheritance, which is why we test that this + # throws an exception! + __table_args__ = {'mysql_engine':'InnoDB'} assert_raises_message(sa.exc.ArgumentError, "place __table_args__", go) def test_concrete(self): @@ -1789,3 +1792,171 @@ class DeclarativeReflectionTest(testing.TestBase): eq_(a1, IMHandle(network='lol', handle='zomg')) eq_(a1.user, User(name='u1')) +class DeclarativeMixinTest(DeclarativeTestBase): + + def test_simple(self): + + class MyMixin(object): + id = Column(Integer, primary_key=True) + + def foo(self): + return 'bar'+str(self.id) + + class MyModel(Base,MyMixin): + __tablename__='test' + name = Column(String(1000), nullable=False, index=True) + + Base.metadata.create_all() + + session = create_session() + session.add(MyModel(name='testing')) + session.flush() + session.expunge_all() + + obj = session.query(MyModel).one() + eq_(obj.id,1) + eq_(obj.name,'testing') + eq_(obj.foo(),'bar1') + + def test_hierarchical_bases(self): + + class MyMixinParent: + id = Column(Integer, primary_key=True) + + def foo(self): + return 'bar'+str(self.id) + + class MyMixin(MyMixinParent): + baz = Column(String(1000), nullable=False, index=True) + + class MyModel(Base,MyMixin): + __tablename__='test' + name = Column(String(1000), nullable=False, index=True) + + Base.metadata.create_all() + + session = create_session() + session.add(MyModel(name='testing',baz='fu')) + session.flush() + session.expunge_all() + + obj = session.query(MyModel).one() + eq_(obj.id,1) + eq_(obj.name,'testing') + eq_(obj.foo(),'bar1') + eq_(obj.baz,'fu') + + def test_table_args_inherited(self): + + class MyMixin: + __table_args__ = {'mysql_engine':'InnoDB'} + + class MyModel(Base,MyMixin): + __tablename__='test' + id = Column(Integer, primary_key=True) + + eq_(MyModel.__table__.kwargs,{'mysql_engine': 'InnoDB'}) + + def test_table_args_overridden(self): + + class MyMixin: + __table_args__ = {'mysql_engine':'Foo'} + + class MyModel(Base,MyMixin): + __tablename__='test' + __table_args__ = {'mysql_engine':'InnoDB'} + id = Column(Integer, primary_key=True) + + eq_(MyModel.__table__.kwargs,{'mysql_engine': 'InnoDB'}) + + def test_table_args_composite(self): + + class MyMixin1: + __table_args__ = {'info':{'baz':'bob'}} + + class MyMixin2: + __table_args__ = {'info':{'foo':'bar'}} + + class MyModel(Base,MyMixin1,MyMixin2): + __tablename__='test' + + @classproperty + def __table_args__(self): + info = {} + args = dict(info=info) + info.update(MyMixin1.__table_args__['info']) + info.update(MyMixin2.__table_args__['info']) + return args + + id = Column(Integer, primary_key=True) + + eq_(MyModel.__table__.info,{ + 'foo': 'bar', + 'baz': 'bob', + }) + + def test_mapper_args_inherited(self): + + class MyMixin: + __mapper_args__=dict(always_refresh=True) + + class MyModel(Base,MyMixin): + __tablename__='test' + id = Column(Integer, primary_key=True) + + eq_(MyModel.__mapper__.always_refresh,True) + + + def test_mapper_args_polymorphic_on_inherited(self): + + class MyMixin: + type_ = Column(String(50)) + __mapper_args__=dict(polymorphic_on=type_) + + class MyModel(Base,MyMixin): + __tablename__='test' + id = Column(Integer, primary_key=True) + + col = MyModel.__mapper__.polymorphic_on + eq_(col.name,'type_') + assert col.table is not None + + + def test_mapper_args_overridden(self): + + class MyMixin: + __mapper_args__=dict(always_refresh=True) + + class MyModel(Base,MyMixin): + __tablename__='test' + __mapper_args__=dict(always_refresh=False) + id = Column(Integer, primary_key=True) + + eq_(MyModel.__mapper__.always_refresh,False) + + def test_mapper_args_composite(self): + + class MyMixin1: + type_ = Column(String(50)) + __mapper_args__=dict(polymorphic_on=type_) + + class MyMixin2: + __mapper_args__=dict(always_refresh=True) + + class MyModel(Base,MyMixin1,MyMixin2): + __tablename__='test' + + @classproperty + def __mapper_args__(self): + args = {} + args.update(MyMixin1.__mapper_args__) + args.update(MyMixin2.__mapper_args__) + return args + + id = Column(Integer, primary_key=True) + + col = MyModel.__mapper__.polymorphic_on + eq_(col.name,'type_') + assert col.table is not None + + eq_(MyModel.__mapper__.always_refresh,True)