SELECT and Related Constructs

The term “selectable” refers to any object that represents database rows. In SQLAlchemy, these objects descend from Selectable, the most prominent being Select, which represents a SQL SELECT statement. A subset of Selectable is FromClause, which represents objects that can be within the FROM clause of a Select statement. A distinguishing feature of FromClause is the FromClause.c attribute, which is a namespace of all the columns contained within the FROM clause (these elements are themselves ColumnElement subclasses).

Selectable Foundational Constructors

Top level “FROM clause” and “SELECT” constructors.

Object NameDescription

except_(selects)

Return an EXCEPT of multiple selectables.

except_all(selects)

Return an EXCEPT ALL of multiple selectables.

exists([argument])

Construct a new Exists construct.

intersect(selects)

Return an INTERSECT of multiple selectables.

intersect_all(selects)

Return an INTERSECT ALL of multiple selectables.

select(entities, *kw)

Construct a new Select.

table(name, columns, **kw)

Produce a new TableClause.

union(selects)

Return a UNION of multiple selectables.

union_all(selects)

Return a UNION ALL of multiple selectables.

values(columns, [name, literal_binds])

Construct a Values construct.

function sqlalchemy.sql.expression.except_(*selects: _SelectStatementForCompoundArgument) → CompoundSelect

Return an EXCEPT of multiple selectables.

The returned object is an instance of CompoundSelect.

  • Parameters:

    *selects – a list of Select instances.

function sqlalchemy.sql.expression.except_all(*selects: _SelectStatementForCompoundArgument) → CompoundSelect

Return an EXCEPT ALL of multiple selectables.

The returned object is an instance of CompoundSelect.

  • Parameters:

    *selects – a list of Select instances.

function sqlalchemy.sql.expression.exists(\_argument: Optional[Union[_ColumnsClauseArgument[Any], SelectBase, ScalarSelect[Any]]] = None_) → Exists

Construct a new Exists construct.

The exists() can be invoked by itself to produce an Exists construct, which will accept simple WHERE criteria:

  1. exists_criteria = exists().where(table1.c.col1 == table2.c.col2)

However, for greater flexibility in constructing the SELECT, an existing Select construct may be converted to an Exists, most conveniently by making use of the SelectBase.exists() method:

  1. exists_criteria = (
  2.     select(table2.c.col2).
  3.     where(table1.c.col1 == table2.c.col2).
  4.     exists()
  5. )

The EXISTS criteria is then used inside of an enclosing SELECT:

  1. stmt = select(table1.c.col1).where(exists_criteria)

The above statement will then be of the form:

  1. SELECT col1 FROM table1 WHERE EXISTS
  2. (SELECT table2.col2 FROM table2 WHERE table2.col2 = table1.col1)

See also

EXISTS subqueries - in the 2.0 style tutorial.

SelectBase.exists() - method to transform a SELECT to an EXISTS clause.

function sqlalchemy.sql.expression.intersect(*selects: _SelectStatementForCompoundArgument) → CompoundSelect

Return an INTERSECT of multiple selectables.

The returned object is an instance of CompoundSelect.

  • Parameters:

    *selects – a list of Select instances.

function sqlalchemy.sql.expression.intersect_all(*selects: _SelectStatementForCompoundArgument) → CompoundSelect

Return an INTERSECT ALL of multiple selectables.

The returned object is an instance of CompoundSelect.

  • Parameters:

    *selects – a list of Select instances.

function sqlalchemy.sql.expression.select(*entities: _ColumnsClauseArgument[Any], **\_kw: Any_) → Select[Any]

Construct a new Select.

New in version 1.4: - The select() function now accepts column arguments positionally. The top-level select() function will automatically use the 1.x or 2.x style API based on the incoming arguments; using select() from the sqlalchemy.future module will enforce that only the 2.x style constructor is used.

Similar functionality is also available via the FromClause.select() method on any FromClause.

See also

Using SELECT Statements - in the SQLAlchemy Unified Tutorial

  • Parameters:

    *entities –

    Entities to SELECT from. For Core usage, this is typically a series of ColumnElement and / or FromClause objects which will form the columns clause of the resulting statement. For those objects that are instances of FromClause (typically Table or Alias objects), the FromClause.c collection is extracted to form a collection of ColumnElement objects.

    This parameter will also accept TextClause constructs as given, as well as ORM-mapped classes.

function sqlalchemy.sql.expression.table(name: str, *columns: ColumnClause[Any], **kw: Any) → TableClause

Produce a new TableClause.

The object returned is an instance of TableClause, which represents the “syntactical” portion of the schema-level Table object. It may be used to construct lightweight table constructs.

Changed in version 1.0.0: table() can now be imported from the plain sqlalchemy namespace like any other SQL element.

  • Parameters:

    • name – Name of the table.

    • columns – A collection of column() constructs.

    • schema –

      The schema name for this table.

      New in version 1.3.18: table() can now accept a schema argument.

function sqlalchemy.sql.expression.union(*selects: _SelectStatementForCompoundArgument) → CompoundSelect

Return a UNION of multiple selectables.

The returned object is an instance of CompoundSelect.

A similar union() method is available on all FromClause subclasses.

  • Parameters:

    • *selects – a list of Select instances.

    • **kwargs – available keyword arguments are the same as those of select().

function sqlalchemy.sql.expression.union_all(*selects: _SelectStatementForCompoundArgument) → CompoundSelect

Return a UNION ALL of multiple selectables.

The returned object is an instance of CompoundSelect.

A similar union_all() method is available on all FromClause subclasses.

  • Parameters:

    *selects – a list of Select instances.

function sqlalchemy.sql.expression.values(*columns: ColumnClause[Any], name: Optional[str] = None, literal_binds: bool = False) → Values

Construct a Values construct.

The column expressions and the actual data for Values are given in two separate steps. The constructor receives the column expressions typically as column() constructs, and the data is then passed via the Values.data() method as a list, which can be called multiple times to add more data, e.g.:

  1. from sqlalchemy import column
  2. from sqlalchemy import values
  4. value_expr = values(
  5.     column('id', Integer),
  6.     column('name', String),
  7.     name="my_values"
  8. ).data(
  9.     [(1, 'name1'), (2, 'name2'), (3, 'name3')]
  10. )

  • Parameters:

    • *columns – column expressions, typically composed using column() objects.

    • name – the name for this VALUES construct. If omitted, the VALUES construct will be unnamed in a SQL expression. Different backends may have different requirements here.

    • literal_binds – Defaults to False. Whether or not to render the data values inline in the SQL output, rather than using bound parameters.

Selectable Modifier Constructors

Functions listed here are more commonly available as methods from FromClause and Selectable elements, for example, the alias() function is usually invoked via the FromClause.alias() method.

Object NameDescription

alias(selectable[, name, flat])

Return a named alias of the given FromClause.

cte(selectable[, name, recursive])

Return a new CTE, or Common Table Expression instance.

join(left, right[, onclause, isouter, …])

Produce a Join object, given two FromClause expressions.

lateral(selectable[, name])

Return a Lateral object.

outerjoin(left, right[, onclause, full])

Return an OUTER JOIN clause element.

tablesample(selectable, sampling[, name, seed])

Return a TableSample object.

function sqlalchemy.sql.expression.alias(selectable: FromClause, name: Optional[str] = None, flat: bool = False) → NamedFromClause

Return a named alias of the given FromClause.

For Table and Join objects, the return type is the Alias object. Other kinds of NamedFromClause objects may be returned for other kinds of FromClause objects.

The named alias represents any FromClause with an alternate name assigned within SQL, typically using the AS clause when generated, e.g. SELECT * FROM table AS aliasname.

Equivalent functionality is available via the FromClause.alias() method available on all FromClause objects.

  • Parameters:

    • selectable – any FromClause subclass, such as a table, select statement, etc.

    • name – string name to be assigned as the alias. If None, a name will be deterministically generated at compile time. Deterministic means the name is guaranteed to be unique against other constructs used in the same statement, and will also be the same name for each successive compilation of the same statement object.

    • flat – Will be passed through to if the given selectable is an instance of Join - see Join.alias() for details.

function sqlalchemy.sql.expression.cte(selectable: HasCTE, name: Optional[str] = None, recursive: bool = False) → CTE

Return a new CTE, or Common Table Expression instance.

Please see HasCTE.cte() for detail on CTE usage.

function sqlalchemy.sql.expression.join(left: _FromClauseArgument, right: _FromClauseArgument, onclause: Optional[_OnClauseArgument] = None, isouter: bool = False, full: bool = False) → Join

Produce a Join object, given two FromClause expressions.

E.g.:

  1. j = join(user_table, address_table,
  2.          user_table.c.id == address_table.c.user_id)
  3. stmt = select(user_table).select_from(j)

would emit SQL along the lines of:

  1. SELECT user.id, user.name FROM user
  2. JOIN address ON user.id = address.user_id

Similar functionality is available given any FromClause object (e.g. such as a Table) using the FromClause.join() method.

  • Parameters:

    • left – The left side of the join.

    • right – the right side of the join; this is any FromClause object such as a Table object, and may also be a selectable-compatible object such as an ORM-mapped class.

    • onclause – a SQL expression representing the ON clause of the join. If left at None, FromClause.join() will attempt to join the two tables based on a foreign key relationship.

    • isouter – if True, render a LEFT OUTER JOIN, instead of JOIN.

    • full –

      if True, render a FULL OUTER JOIN, instead of JOIN.

      New in version 1.1.

See also

FromClause.join() - method form, based on a given left side.

Join - the type of object produced.

function sqlalchemy.sql.expression.lateral(selectable: Union[SelectBase, _FromClauseArgument], name: Optional[str] = None) → LateralFromClause

Return a Lateral object.

Lateral is an Alias subclass that represents a subquery with the LATERAL keyword applied to it.

The special behavior of a LATERAL subquery is that it appears in the FROM clause of an enclosing SELECT, but may correlate to other FROM clauses of that SELECT. It is a special case of subquery only supported by a small number of backends, currently more recent PostgreSQL versions.

New in version 1.1.

See also

LATERAL correlation - overview of usage.

function sqlalchemy.sql.expression.outerjoin(left: _FromClauseArgument, right: _FromClauseArgument, onclause: Optional[_OnClauseArgument] = None, full: bool = False) → Join

Return an OUTER JOIN clause element.

The returned object is an instance of Join.

Similar functionality is also available via the FromClause.outerjoin() method on any FromClause.

  • Parameters:

    • left – The left side of the join.

    • right – The right side of the join.

    • onclause – Optional criterion for the ON clause, is derived from foreign key relationships established between left and right otherwise.

To chain joins together, use the FromClause.join() or FromClause.outerjoin() methods on the resulting Join object.

function sqlalchemy.sql.expression.tablesample(selectable: _FromClauseArgument, sampling: Union[float, Function[Any]], name: Optional[str] = None, seed: Optional[roles.ExpressionElementRole[Any]] = None) → TableSample

Return a TableSample object.

TableSample is an Alias subclass that represents a table with the TABLESAMPLE clause applied to it. tablesample() is also available from the FromClause class via the FromClause.tablesample() method.

The TABLESAMPLE clause allows selecting a randomly selected approximate percentage of rows from a table. It supports multiple sampling methods, most commonly BERNOULLI and SYSTEM.

e.g.:

  1. from sqlalchemy import func
  3. selectable = people.tablesample(
  4.             func.bernoulli(1),
  5.             name='alias',
  6.             seed=func.random())
  7. stmt = select(selectable.c.people_id)

Assuming people with a column people_id, the above statement would render as:

  1. SELECT alias.people_id FROM
  2. people AS alias TABLESAMPLE bernoulli(:bernoulli_1)
  3. REPEATABLE (random())

New in version 1.1.

  • Parameters:

    • sampling – a float percentage between 0 and 100 or Function.

    • name – optional alias name

    • seed – any real-valued SQL expression. When specified, the REPEATABLE sub-clause is also rendered.

Selectable Class Documentation

The classes here are generated using the constructors listed at Selectable Foundational Constructors and Selectable Modifier Constructors.

Object NameDescription

Alias

Represents an table or selectable alias (AS).

AliasedReturnsRows

Base class of aliases against tables, subqueries, and other selectables.

CompoundSelect

Forms the basis of UNION, UNION ALL, and other SELECT-based set operations.

CTE

Represent a Common Table Expression.

Executable

Mark a ClauseElement as supporting execution.

Exists

Represent an EXISTS clause.

FromClause

Represent an element that can be used within the FROM clause of a SELECT statement.

GenerativeSelect

Base class for SELECT statements where additional elements can be added.

HasCTE

Mixin that declares a class to include CTE support.

HasPrefixes

HasSuffixes

Join

Represent a JOIN construct between two FromClause elements.

Lateral

Represent a LATERAL subquery.

ReturnsRows

The base-most class for Core constructs that have some concept of columns that can represent rows.

ScalarSelect

Represent a scalar subquery.

ScalarValues

Represent a scalar VALUES construct that can be used as a COLUMN element in a statement.

Select

Represents a SELECT statement.

Selectable

Mark a class as being selectable.

SelectBase

Base class for SELECT statements.

Subquery

Represent a subquery of a SELECT.

TableClause

Represents a minimal “table” construct.

TableSample

Represent a TABLESAMPLE clause.

TableValuedAlias

An alias against a “table valued” SQL function.

TextualSelect

Wrap a TextClause construct within a SelectBase interface.

Values

Represent a VALUES construct that can be used as a FROM element in a statement.

class sqlalchemy.sql.expression.Alias

Represents an table or selectable alias (AS).

Represents an alias, as typically applied to any table or sub-select within a SQL statement using the AS keyword (or without the keyword on certain databases such as Oracle).

This object is constructed from the alias() module level function as well as the FromClause.alias() method available on all FromClause subclasses.

See also

FromClause.alias()

Members

inherit_cache

Class signature

class sqlalchemy.sql.expression.Alias (sqlalchemy.sql.roles.DMLTableRole, sqlalchemy.sql.expression.FromClauseAlias)

  • attribute sqlalchemy.sql.expression.Alias.inherit_cache: Optional[bool] = True

    Indicate if this HasCacheKey instance should make use of the cache key generation scheme used by its immediate superclass.

    The attribute defaults to None, which indicates that a construct has not yet taken into account whether or not its appropriate for it to participate in caching; this is functionally equivalent to setting the value to False, except that a warning is also emitted.

    This flag can be set to True on a particular class, if the SQL that corresponds to the object does not change based on attributes which are local to this class, and not its superclass.

    See also

    Enabling Caching Support for Custom Constructs - General guideslines for setting the HasCacheKey.inherit_cache attribute for third-party or user defined SQL constructs.

class sqlalchemy.sql.expression.AliasedReturnsRows

Base class of aliases against tables, subqueries, and other selectables.

Members

description, is_derived_from(), original

Class signature

class sqlalchemy.sql.expression.AliasedReturnsRows (sqlalchemy.sql.expression.NoInit, sqlalchemy.sql.expression.NamedFromClause)

class sqlalchemy.sql.expression.CompoundSelect

Forms the basis of UNION, UNION ALL, and other SELECT-based set operations.

See also

union()

union_all()

intersect()

intersect_all()

except()

except_all()

Members

add_cte(), alias(), as_scalar(), c, corresponding_column(), cte(), execution_options(), exists(), exported_columns, fetch(), get_execution_options(), get_label_style(), group_by(), is_derived_from(), label(), lateral(), limit(), offset(), options(), order_by(), replace_selectable(), scalar_subquery(), select(), selected_columns, self_group(), set_label_style(), slice(), subquery(), with_for_update()

Class signature

class sqlalchemy.sql.expression.CompoundSelect (sqlalchemy.sql.expression.HasCompileState, sqlalchemy.sql.expression.GenerativeSelect, sqlalchemy.sql.expression.ExecutableReturnsRows)

  • method sqlalchemy.sql.expression.CompoundSelect.add_cte(*ctes: CTE, nest_here: bool = False) → SelfHasCTE

    inherited from the HasCTE.add_cte() method of HasCTE

    Add one or more CTE constructs to this statement.

    This method will associate the given CTE constructs with the parent statement such that they will each be unconditionally rendered in the WITH clause of the final statement, even if not referenced elsewhere within the statement or any sub-selects.

    The optional HasCTE.add_cte.nest_here parameter when set to True will have the effect that each given CTE will render in a WITH clause rendered directly along with this statement, rather than being moved to the top of the ultimate rendered statement, even if this statement is rendered as a subquery within a larger statement.

    This method has two general uses. One is to embed CTE statements that serve some purpose without being referenced explicitly, such as the use case of embedding a DML statement such as an INSERT or UPDATE as a CTE inline with a primary statement that may draw from its results indirectly. The other is to provide control over the exact placement of a particular series of CTE constructs that should remain rendered directly in terms of a particular statement that may be nested in a larger statement.

    E.g.:

    1. from sqlalchemy import table, column, select
    2. t = table('t', column('c1'), column('c2'))
    4. ins = t.insert().values({"c1": "x", "c2": "y"}).cte()
    6. stmt = select(t).add_cte(ins)

    Would render:

    1. WITH anon_1 AS
    2. (INSERT INTO t (c1, c2) VALUES (:param_1, :param_2))
    3. SELECT t.c1, t.c2
    4. FROM t

    Above, the “anon_1” CTE is not referred towards in the SELECT statement, however still accomplishes the task of running an INSERT statement.

    Similarly in a DML-related context, using the PostgreSQL Insert construct to generate an “upsert”:

    1. from sqlalchemy import table, column
    2. from sqlalchemy.dialects.postgresql import insert
    4. t = table("t", column("c1"), column("c2"))
    6. delete_statement_cte = (
    7.     t.delete().where(t.c.c1 < 1).cte("deletions")
    8. )
    10. insert_stmt = insert(t).values({"c1": 1, "c2": 2})
    11. update_statement = insert_stmt.on_conflict_do_update(
    12.     index_elements=[t.c.c1],
    13.     set_={
    14.         "c1": insert_stmt.excluded.c1,
    15.         "c2": insert_stmt.excluded.c2,
    16.     },
    17. ).add_cte(delete_statement_cte)
    19. print(update_statement)

    The above statement renders as:

    1. WITH deletions AS
    2. (DELETE FROM t WHERE t.c1 < %(c1_1)s)
    3. INSERT INTO t (c1, c2) VALUES (%(c1)s, %(c2)s)
    4. ON CONFLICT (c1) DO UPDATE SET c1 = excluded.c1, c2 = excluded.c2

    New in version 1.4.21.

    • Parameters:

      • *ctes –

        zero or more CTE constructs.

        Changed in version 2.0: Multiple CTE instances are accepted

      • nest_here –

        if True, the given CTE or CTEs will be rendered as though they specified the HasCTE.cte.nesting flag to True when they were added to this HasCTE. Assuming the given CTEs are not referenced in an outer-enclosing statement as well, the CTEs given should render at the level of this statement when this flag is given.

        New in version 2.0.

        See also

        HasCTE.cte.nesting

  1. See also
  3. [Selectable.exported\_columns](#sqlalchemy.sql.expression.Selectable.exported_columns "sqlalchemy.sql.expression.Selectable.exported_columns") - the [ColumnCollection]($aafca12b71ff5dd3.md#sqlalchemy.sql.expression.ColumnCollection "sqlalchemy.sql.expression.ColumnCollection") that is used for the operation.
  5. [ColumnCollection.corresponding\_column()]($aafca12b71ff5dd3.md#sqlalchemy.sql.expression.ColumnCollection.corresponding_column "sqlalchemy.sql.expression.ColumnCollection.corresponding_column") - implementation method.

  • method sqlalchemy.sql.expression.CompoundSelect.cte(name: Optional[str] = None, recursive: bool = False, nesting: bool = False) → CTE

    inherited from the HasCTE.cte() method of HasCTE

    Return a new CTE, or Common Table Expression instance.

    Common table expressions are a SQL standard whereby SELECT statements can draw upon secondary statements specified along with the primary statement, using a clause called “WITH”. Special semantics regarding UNION can also be employed to allow “recursive” queries, where a SELECT statement can draw upon the set of rows that have previously been selected.

    CTEs can also be applied to DML constructs UPDATE, INSERT and DELETE on some databases, both as a source of CTE rows when combined with RETURNING, as well as a consumer of CTE rows.

    Changed in version 1.1: Added support for UPDATE/INSERT/DELETE as CTE, CTEs added to UPDATE/INSERT/DELETE.

    SQLAlchemy detects CTE objects, which are treated similarly to Alias objects, as special elements to be delivered to the FROM clause of the statement as well as to a WITH clause at the top of the statement.

    For special prefixes such as PostgreSQL “MATERIALIZED” and “NOT MATERIALIZED”, the CTE.prefix_with() method may be used to establish these.

    Changed in version 1.3.13: Added support for prefixes. In particular - MATERIALIZED and NOT MATERIALIZED.

    • Parameters:

      • name – name given to the common table expression. Like FromClause.alias(), the name can be left as None in which case an anonymous symbol will be used at query compile time.

      • recursive – if True, will render WITH RECURSIVE. A recursive common table expression is intended to be used in conjunction with UNION ALL in order to derive rows from those already selected.

      • nesting –

        if True, will render the CTE locally to the statement in which it is referenced. For more complex scenarios, the HasCTE.add_cte() method using the HasCTE.add_cte.nest_here parameter may also be used to more carefully control the exact placement of a particular CTE.

        New in version 1.4.24.

        See also

        HasCTE.add_cte()

  1. The following examples include two from PostgreSQL’s documentation at [https://www.postgresql.org/docs/current/static/queries-with.html](https://www.postgresql.org/docs/current/static/queries-with.html), as well as additional examples.
  3. Example 1, non recursive:
  5. ```
  6. from sqlalchemy import (Table, Column, String, Integer,
  7.                         MetaData, select, func)
  9. metadata = MetaData()
  11. orders = Table('orders', metadata,
  12.     Column('region', String),
  13.     Column('amount', Integer),
  14.     Column('product', String),
  15.     Column('quantity', Integer)
  16. )
  18. regional_sales = select(
  19.                     orders.c.region,
  20.                     func.sum(orders.c.amount).label('total_sales')
  21.                 ).group_by(orders.c.region).cte("regional_sales")
  24. top_regions = select(regional_sales.c.region).\
  25.         where(
  26.             regional_sales.c.total_sales >
  27.             select(
  28.                 func.sum(regional_sales.c.total_sales) / 10
  29.             )
  30.         ).cte("top_regions")
  32. statement = select(
  33.             orders.c.region,
  34.             orders.c.product,
  35.             func.sum(orders.c.quantity).label("product_units"),
  36.             func.sum(orders.c.amount).label("product_sales")
  37.     ).where(orders.c.region.in_(
  38.         select(top_regions.c.region)
  39.     )).group_by(orders.c.region, orders.c.product)
  41. result = conn.execute(statement).fetchall()
  42. ```
  44. Example 2, WITH RECURSIVE:
  46. ```
  47. from sqlalchemy import (Table, Column, String, Integer,
  48.                         MetaData, select, func)
  50. metadata = MetaData()
  52. parts = Table('parts', metadata,
  53.     Column('part', String),
  54.     Column('sub_part', String),
  55.     Column('quantity', Integer),
  56. )
  58. included_parts = select(\
  59.     parts.c.sub_part, parts.c.part, parts.c.quantity\
  60.     ).\
  61.     where(parts.c.part=='our part').\
  62.     cte(recursive=True)
  65. incl_alias = included_parts.alias()
  66. parts_alias = parts.alias()
  67. included_parts = included_parts.union_all(
  68.     select(
  69.         parts_alias.c.sub_part,
  70.         parts_alias.c.part,
  71.         parts_alias.c.quantity
  72.     ).\
  73.     where(parts_alias.c.part==incl_alias.c.sub_part)
  74. )
  76. statement = select(
  77.             included_parts.c.sub_part,
  78.             func.sum(included_parts.c.quantity).
  79.               label('total_quantity')
  80.         ).\
  81.         group_by(included_parts.c.sub_part)
  83. result = conn.execute(statement).fetchall()
  84. ```
  86. Example 3, an upsert using UPDATE and INSERT with CTEs:
  88. ```
  89. from datetime import date
  90. from sqlalchemy import (MetaData, Table, Column, Integer,
  91.                         Date, select, literal, and_, exists)
  93. metadata = MetaData()
  95. visitors = Table('visitors', metadata,
  96.     Column('product_id', Integer, primary_key=True),
  97.     Column('date', Date, primary_key=True),
  98.     Column('count', Integer),
  99. )
  101. # add 5 visitors for the product_id == 1
  102. product_id = 1
  103. day = date.today()
  104. count = 5
  106. update_cte = (
  107.     visitors.update()
  108.     .where(and_(visitors.c.product_id == product_id,
  109.                 visitors.c.date == day))
  110.     .values(count=visitors.c.count + count)
  111.     .returning(literal(1))
  112.     .cte('update_cte')
  113. )
  115. upsert = visitors.insert().from_select(
  116.     [visitors.c.product_id, visitors.c.date, visitors.c.count],
  117.     select(literal(product_id), literal(day), literal(count))
  118.         .where(~exists(update_cte.select()))
  119. )
  121. connection.execute(upsert)
  122. ```
  124. Example 4, Nesting CTE (SQLAlchemy 1.4.24 and above):
  126. ```
  127. value_a = select(
  128.     literal("root").label("n")
  129. ).cte("value_a")
  131. # A nested CTE with the same name as the root one
  132. value_a_nested = select(
  133.     literal("nesting").label("n")
  134. ).cte("value_a", nesting=True)
  136. # Nesting CTEs takes ascendency locally
  137. # over the CTEs at a higher level
  138. value_b = select(value_a_nested.c.n).cte("value_b")
  140. value_ab = select(value_a.c.n.label("a"), value_b.c.n.label("b"))
  141. ```
  143. The above query will render the second CTE nested inside the first, shown with inline parameters below as:
  145. ```
  146. WITH
  147.     value_a AS
  148.         (SELECT 'root' AS n),
  149.     value_b AS
  150.         (WITH value_a AS
  151.             (SELECT 'nesting' AS n)
  152.         SELECT value_a.n AS n FROM value_a)
  153. SELECT value_a.n AS a, value_b.n AS b
  154. FROM value_a, value_b
  155. ```
  157. The same CTE can be set up using the [HasCTE.add\_cte()](#sqlalchemy.sql.expression.HasCTE.add_cte "sqlalchemy.sql.expression.HasCTE.add_cte") method as follows (SQLAlchemy 2.0 and above):
  159. ```
  160. value_a = select(
  161.     literal("root").label("n")
  162. ).cte("value_a")
  164. # A nested CTE with the same name as the root one
  165. value_a_nested = select(
  166.     literal("nesting").label("n")
  167. ).cte("value_a")
  169. # Nesting CTEs takes ascendency locally
  170. # over the CTEs at a higher level
  171. value_b = (
  172.     select(value_a_nested.c.n).
  173.     add_cte(value_a_nested, nest_here=True).
  174.     cte("value_b")
  175. )
  177. value_ab = select(value_a.c.n.label("a"), value_b.c.n.label("b"))
  178. ```
  180. Example 5, Non-Linear CTE (SQLAlchemy 1.4.28 and above):
  182. ```
  183. edge = Table(
  184.     "edge",
  185.     metadata,
  186.     Column("id", Integer, primary_key=True),
  187.     Column("left", Integer),
  188.     Column("right", Integer),
  189. )
  191. root_node = select(literal(1).label("node")).cte(
  192.     "nodes", recursive=True
  193. )
  195. left_edge = select(edge.c.left).join(
  196.     root_node, edge.c.right == root_node.c.node
  197. )
  198. right_edge = select(edge.c.right).join(
  199.     root_node, edge.c.left == root_node.c.node
  200. )
  202. subgraph_cte = root_node.union(left_edge, right_edge)
  204. subgraph = select(subgraph_cte)
  205. ```
  207. The above query will render 2 UNIONs inside the recursive CTE:
  209. ```
  210. WITH RECURSIVE nodes(node) AS (
  211.         SELECT 1 AS node
  212.     UNION
  213.         SELECT edge."left" AS "left"
  214.         FROM edge JOIN nodes ON edge."right" = nodes.node
  215.     UNION
  216.         SELECT edge."right" AS "right"
  217.         FROM edge JOIN nodes ON edge."left" = nodes.node
  218. )
  219. SELECT nodes.node FROM nodes
  220. ```
  222. See also
  224. [Query.cte()]($3d0cc000ec6c7150.md#sqlalchemy.orm.Query.cte "sqlalchemy.orm.Query.cte") - ORM version of [HasCTE.cte()](#sqlalchemy.sql.expression.HasCTE.cte "sqlalchemy.sql.expression.HasCTE.cte").
  1. See also
  3. [GenerativeSelect.limit()](#sqlalchemy.sql.expression.GenerativeSelect.limit "sqlalchemy.sql.expression.GenerativeSelect.limit")
  5. [GenerativeSelect.offset()](#sqlalchemy.sql.expression.GenerativeSelect.offset "sqlalchemy.sql.expression.GenerativeSelect.offset")

class sqlalchemy.sql.expression.CTE

Represent a Common Table Expression.

The CTE object is obtained using the SelectBase.cte() method from any SELECT statement. A less often available syntax also allows use of the HasCTE.cte() method present on DML constructs such as Insert, Update and Delete. See the HasCTE.cte() method for usage details on CTEs.

See also

Subqueries and CTEs - in the 2.0 tutorial

HasCTE.cte() - examples of calling styles

Members

alias(), union(), union_all()

Class signature

class sqlalchemy.sql.expression.CTE (sqlalchemy.sql.roles.DMLTableRole, sqlalchemy.sql.roles.IsCTERole, sqlalchemy.sql.expression.Generative, sqlalchemy.sql.expression.HasPrefixes, sqlalchemy.sql.expression.HasSuffixes, sqlalchemy.sql.expression.AliasedReturnsRows)

  • method sqlalchemy.sql.expression.CTE.alias(name: Optional[str] = None, flat: bool = False) → CTE

    Return an Alias of this CTE.

    This method is a CTE-specific specialization of the FromClause.alias() method.

    See also

    Using Aliases

    alias()

  • method sqlalchemy.sql.expression.CTE.union(*other: _SelectStatementForCompoundArgument) → CTE

    Return a new CTE with a SQL UNION of the original CTE against the given selectables provided as positional arguments.

    • Parameters:

      *other –

      one or more elements with which to create a UNION.

      Changed in version 1.4.28: multiple elements are now accepted.

    See also

    HasCTE.cte() - examples of calling styles

  • method sqlalchemy.sql.expression.CTE.union_all(*other: _SelectStatementForCompoundArgument) → CTE

    Return a new CTE with a SQL UNION ALL of the original CTE against the given selectables provided as positional arguments.

    • Parameters:

      *other –

      one or more elements with which to create a UNION.

      Changed in version 1.4.28: multiple elements are now accepted.

    See also

    HasCTE.cte() - examples of calling styles

class sqlalchemy.sql.expression.Executable

Mark a ClauseElement as supporting execution.

Executable is a superclass for all “statement” types of objects, including select(), delete(), update(), insert(), text().

Members

execution_options(), get_execution_options(), options()

Class signature

class sqlalchemy.sql.expression.Executable (sqlalchemy.sql.roles.StatementRole)

class sqlalchemy.sql.expression.Exists

Represent an EXISTS clause.

See exists() for a description of usage.

An EXISTS clause can also be constructed from a select() instance by calling SelectBase.exists().

Members

correlate(), correlate_except(), inherit_cache, select(), select_from(), where()

Class signature

class sqlalchemy.sql.expression.Exists (sqlalchemy.sql.expression.UnaryExpression)

  • method sqlalchemy.sql.expression.Exists.correlate(*fromclauses: Union[Literal[None, False], _FromClauseArgument]) → SelfExists

    Apply correlation to the subquery noted by this Exists.

    See also

    ScalarSelect.correlate()

  • method sqlalchemy.sql.expression.Exists.correlate_except(*fromclauses: Union[Literal[None, False], _FromClauseArgument]) → SelfExists

    Apply correlation to the subquery noted by this Exists.

    See also

    ScalarSelect.correlate_except()

  • attribute sqlalchemy.sql.expression.Exists.inherit_cache: Optional[bool] = True

    Indicate if this HasCacheKey instance should make use of the cache key generation scheme used by its immediate superclass.

    The attribute defaults to None, which indicates that a construct has not yet taken into account whether or not its appropriate for it to participate in caching; this is functionally equivalent to setting the value to False, except that a warning is also emitted.

    This flag can be set to True on a particular class, if the SQL that corresponds to the object does not change based on attributes which are local to this class, and not its superclass.

    See also

    Enabling Caching Support for Custom Constructs - General guideslines for setting the HasCacheKey.inherit_cache attribute for third-party or user defined SQL constructs.

  • method sqlalchemy.sql.expression.Exists.select() → Select

    Return a SELECT of this Exists.

    e.g.:

    1. stmt = exists(some_table.c.id).where(some_table.c.id == 5).select()

    This will produce a statement resembling:

    1. SELECT EXISTS (SELECT id FROM some_table WHERE some_table = :param) AS anon_1

    See also

    select() - general purpose method which allows for arbitrary column lists.

  • method sqlalchemy.sql.expression.Exists.select_from(*froms: FromClause) → SelfExists

    Return a new Exists construct, applying the given expression to the Select.select_from() method of the select statement contained.

    Note

    it is typically preferable to build a Select statement first, including the desired WHERE clause, then use the SelectBase.exists() method to produce an Exists object at once.

  • method sqlalchemy.sql.expression.Exists.where(*clause: _ColumnExpressionArgument[bool]) → SelfExists

    Return a new exists() construct with the given expression added to its WHERE clause, joined to the existing clause via AND, if any.

    Note

    it is typically preferable to build a Select statement first, including the desired WHERE clause, then use the SelectBase.exists() method to produce an Exists object at once.

class sqlalchemy.sql.expression.FromClause

Represent an element that can be used within the FROM clause of a SELECT statement.

The most common forms of FromClause are the Table and the select() constructs. Key features common to all FromClause objects include:

Members

alias(), c, columns, description, entity_namespace, exported_columns, foreign_keys, is_derived_from(), join(), outerjoin(), primary_key, schema, select(), tablesample()

Class signature

class sqlalchemy.sql.expression.FromClause (sqlalchemy.sql.roles.AnonymizedFromClauseRole, sqlalchemy.sql.expression.Selectable)

  1. See also
  3. [join()](#sqlalchemy.sql.expression.join "sqlalchemy.sql.expression.join") - standalone function
  5. [Join](#sqlalchemy.sql.expression.Join "sqlalchemy.sql.expression.Join") - the type of object produced

  • method sqlalchemy.sql.expression.FromClause.outerjoin(right: _FromClauseArgument, onclause: Optional[_ColumnExpressionArgument[bool]] = None, full: bool = False) → Join

    Return a Join from this FromClause to another FromClause, with the “isouter” flag set to True.

    E.g.:

    1. from sqlalchemy import outerjoin
    3. j = user_table.outerjoin(address_table,
    4.                 user_table.c.id == address_table.c.user_id)

    The above is equivalent to:

    1. j = user_table.join(
    2.     address_table,
    3.     user_table.c.id == address_table.c.user_id,
    4.     isouter=True)

    • Parameters:

      • right – the right side of the join; this is any FromClause object such as a Table object, and may also be a selectable-compatible object such as an ORM-mapped class.

      • onclause – a SQL expression representing the ON clause of the join. If left at None, FromClause.join() will attempt to join the two tables based on a foreign key relationship.

      • full –

        if True, render a FULL OUTER JOIN, instead of LEFT OUTER JOIN.

        New in version 1.1.

  1. See also
  3. [FromClause.join()](#sqlalchemy.sql.expression.FromClause.join "sqlalchemy.sql.expression.FromClause.join")
  5. [Join](#sqlalchemy.sql.expression.Join "sqlalchemy.sql.expression.Join")

class sqlalchemy.sql.expression.GenerativeSelect

Base class for SELECT statements where additional elements can be added.

This serves as the base for Select and CompoundSelect where elements such as ORDER BY, GROUP BY can be added and column rendering can be controlled. Compare to TextualSelect, which, while it subclasses SelectBase and is also a SELECT construct, represents a fixed textual string which cannot be altered at this level, only wrapped as a subquery.

Members

fetch(), get_label_style(), group_by(), limit(), offset(), order_by(), set_label_style(), slice(), with_for_update()

Class signature

class sqlalchemy.sql.expression.GenerativeSelect (sqlalchemy.sql.expression.SelectBase, sqlalchemy.sql.expression.Generative)

  • method sqlalchemy.sql.expression.GenerativeSelect.fetch(count: Union[int, _ColumnExpressionArgument[int]], with_ties: bool = False, percent: bool = False) → SelfGenerativeSelect

    Return a new selectable with the given FETCH FIRST criterion applied.

    This is a numeric value which usually renders as FETCH {FIRST | NEXT} [ count ] {ROW | ROWS} {ONLY | WITH TIES} expression in the resulting select. This functionality is is currently implemented for Oracle, PostgreSQL, MSSQL.

    Use GenerativeSelect.offset() to specify the offset.

    Note

    The GenerativeSelect.fetch() method will replace any clause applied with GenerativeSelect.limit().

    New in version 1.4.

    • Parameters:

      • count – an integer COUNT parameter, or a SQL expression that provides an integer result. When percent=True this will represent the percentage of rows to return, not the absolute value. Pass None to reset it.

      • with_ties – When True, the WITH TIES option is used to return any additional rows that tie for the last place in the result set according to the ORDER BY clause. The ORDER BY may be mandatory in this case. Defaults to False

      • percent – When True, count represents the percentage of the total number of selected rows to return. Defaults to False

  1. See also
  3. [GenerativeSelect.limit()](#sqlalchemy.sql.expression.GenerativeSelect.limit "sqlalchemy.sql.expression.GenerativeSelect.limit")
  5. [GenerativeSelect.offset()](#sqlalchemy.sql.expression.GenerativeSelect.offset "sqlalchemy.sql.expression.GenerativeSelect.offset")

  • method sqlalchemy.sql.expression.GenerativeSelect.get_label_style() → SelectLabelStyle

    Retrieve the current label style.

    New in version 1.4.

  • method sqlalchemy.sql.expression.GenerativeSelect.group_by(_GenerativeSelect\_first: Union[Literal[None, _NoArg.NO_ARG], _ColumnExpressionOrStrLabelArgument[Any]] = _NoArg.NO_ARG, *clauses: _ColumnExpressionOrStrLabelArgument[Any]_) → SelfGenerativeSelect

    Return a new selectable with the given list of GROUP BY criterion applied.

    All existing GROUP BY settings can be suppressed by passing None.

    e.g.:

    1. stmt = select(table.c.name, func.max(table.c.stat)).\
    2. group_by(table.c.name)

    • Parameters:

      *clauses – a series of ColumnElement constructs which will be used to generate an GROUP BY clause.

    See also

    Aggregate functions with GROUP BY / HAVING - in the SQLAlchemy Unified Tutorial

    Ordering or Grouping by a Label - in the SQLAlchemy Unified Tutorial

  • method sqlalchemy.sql.expression.GenerativeSelect.limit(limit: Union[int, _ColumnExpressionArgument[int]]) → SelfGenerativeSelect

    Return a new selectable with the given LIMIT criterion applied.

    This is a numerical value which usually renders as a LIMIT expression in the resulting select. Backends that don’t support LIMIT will attempt to provide similar functionality.

    Note

    The GenerativeSelect.limit() method will replace any clause applied with GenerativeSelect.fetch().

    Changed in version 1.0.0: - Select.limit() can now accept arbitrary SQL expressions as well as integer values.

    • Parameters:

      limit – an integer LIMIT parameter, or a SQL expression that provides an integer result. Pass None to reset it.

    See also

    GenerativeSelect.fetch()

    GenerativeSelect.offset()

  • method sqlalchemy.sql.expression.GenerativeSelect.offset(offset: Union[int, _ColumnExpressionArgument[int]]) → SelfGenerativeSelect

    Return a new selectable with the given OFFSET criterion applied.

    This is a numeric value which usually renders as an OFFSET expression in the resulting select. Backends that don’t support OFFSET will attempt to provide similar functionality.

    Changed in version 1.0.0: - Select.offset() can now accept arbitrary SQL expressions as well as integer values.

    • Parameters:

      offset – an integer OFFSET parameter, or a SQL expression that provides an integer result. Pass None to reset it.

    See also

    GenerativeSelect.limit()

    GenerativeSelect.fetch()

  • method sqlalchemy.sql.expression.GenerativeSelect.order_by(_GenerativeSelect\_first: Union[Literal[None, _NoArg.NO_ARG], _ColumnExpressionOrStrLabelArgument[Any]] = _NoArg.NO_ARG, *clauses: _ColumnExpressionOrStrLabelArgument[Any]_) → SelfGenerativeSelect

    Return a new selectable with the given list of ORDER BY criteria applied.

    e.g.:

    1. stmt = select(table).order_by(table.c.id, table.c.name)

    Calling this method multiple times is equivalent to calling it once with all the clauses concatenated. All existing ORDER BY criteria may be cancelled by passing None by itself. New ORDER BY criteria may then be added by invoking Query.order_by() again, e.g.:

    1. # will erase all ORDER BY and ORDER BY new_col alone
    2. stmt = stmt.order_by(None).order_by(new_col)

    • Parameters:

      *clauses – a series of ColumnElement constructs which will be used to generate an ORDER BY clause.

    See also

    ORDER BY - in the SQLAlchemy Unified Tutorial

    Ordering or Grouping by a Label - in the SQLAlchemy Unified Tutorial

  • method sqlalchemy.sql.expression.GenerativeSelect.set_label_style(style: SelectLabelStyle) → SelfGenerativeSelect

    Return a new selectable with the specified label style.

    There are three “label styles” available, SelectLabelStyle.LABEL_STYLE_DISAMBIGUATE_ONLY, SelectLabelStyle.LABEL_STYLE_TABLENAME_PLUS_COL, and SelectLabelStyle.LABEL_STYLE_NONE. The default style is SelectLabelStyle.LABEL_STYLE_TABLENAME_PLUS_COL.

    In modern SQLAlchemy, there is not generally a need to change the labeling style, as per-expression labels are more effectively used by making use of the ColumnElement.label() method. In past versions, LABEL_STYLE_TABLENAME_PLUS_COL was used to disambiguate same-named columns from different tables, aliases, or subqueries; the newer LABEL_STYLE_DISAMBIGUATE_ONLY now applies labels only to names that conflict with an existing name so that the impact of this labeling is minimal.

    The rationale for disambiguation is mostly so that all column expressions are available from a given FromClause.c collection when a subquery is created.

    New in version 1.4: - the GenerativeSelect.set_label_style() method replaces the previous combination of .apply_labels(), .with_labels() and use_labels=True methods and/or parameters.

    See also

    LABEL_STYLE_DISAMBIGUATE_ONLY

    LABEL_STYLE_TABLENAME_PLUS_COL

    LABEL_STYLE_NONE

    LABEL_STYLE_DEFAULT

  • method sqlalchemy.sql.expression.GenerativeSelect.slice(start: int, stop: int) → SelfGenerativeSelect

    Apply LIMIT / OFFSET to this statement based on a slice.

    The start and stop indices behave like the argument to Python’s built-in range() function. This method provides an alternative to using LIMIT/OFFSET to get a slice of the query.

    For example,

    1. stmt = select(User).order_by(User).id.slice(1, 3)

    renders as

    1. SELECT users.id AS users_id,
    2.        users.name AS users_name
    3. FROM users ORDER BY users.id
    4. LIMIT ? OFFSET ?
    5. (2, 1)

    Note

    The GenerativeSelect.slice() method will replace any clause applied with GenerativeSelect.fetch().

    New in version 1.4: Added the GenerativeSelect.slice() method generalized from the ORM.

    See also

    GenerativeSelect.limit()

    GenerativeSelect.offset()

    GenerativeSelect.fetch()

  • method sqlalchemy.sql.expression.GenerativeSelect.with_for_update(*, nowait: bool = False, read: bool = False, of: Optional[_ForUpdateOfArgument] = None, skip_locked: bool = False, key_share: bool = False) → SelfGenerativeSelect

    Specify a FOR UPDATE clause for this GenerativeSelect.

    E.g.:

    1. stmt = select(table).with_for_update(nowait=True)

    On a database like PostgreSQL or Oracle, the above would render a statement like:

    1. SELECT table.a, table.b FROM table FOR UPDATE NOWAIT

    on other backends, the nowait option is ignored and instead would produce:

    1. SELECT table.a, table.b FROM table FOR UPDATE

    When called with no arguments, the statement will render with the suffix FOR UPDATE. Additional arguments can then be provided which allow for common database-specific variants.

    • Parameters:

      • nowait – boolean; will render FOR UPDATE NOWAIT on Oracle and PostgreSQL dialects.

      • read – boolean; will render LOCK IN SHARE MODE on MySQL, FOR SHARE on PostgreSQL. On PostgreSQL, when combined with nowait, will render FOR SHARE NOWAIT.

      • of – SQL expression or list of SQL expression elements, (typically Column objects or a compatible expression, for some backends may also be a table expression) which will render into a FOR UPDATE OF clause; supported by PostgreSQL, Oracle, some MySQL versions and possibly others. May render as a table or as a column depending on backend.

      • skip_locked – boolean, will render FOR UPDATE SKIP LOCKED on Oracle and PostgreSQL dialects or FOR SHARE SKIP LOCKED if read=True is also specified.

      • key_share – boolean, will render FOR NO KEY UPDATE, or if combined with read=True will render FOR KEY SHARE, on the PostgreSQL dialect.

class sqlalchemy.sql.expression.HasCTE

Mixin that declares a class to include CTE support.

New in version 1.1.

Members

add_cte(), cte()

Class signature

class sqlalchemy.sql.expression.HasCTE (sqlalchemy.sql.roles.HasCTERole, sqlalchemy.sql.expression.SelectsRows)

  • method sqlalchemy.sql.expression.HasCTE.add_cte(*ctes: CTE, nest_here: bool = False) → SelfHasCTE

    Add one or more CTE constructs to this statement.

    This method will associate the given CTE constructs with the parent statement such that they will each be unconditionally rendered in the WITH clause of the final statement, even if not referenced elsewhere within the statement or any sub-selects.

    The optional HasCTE.add_cte.nest_here parameter when set to True will have the effect that each given CTE will render in a WITH clause rendered directly along with this statement, rather than being moved to the top of the ultimate rendered statement, even if this statement is rendered as a subquery within a larger statement.

    This method has two general uses. One is to embed CTE statements that serve some purpose without being referenced explicitly, such as the use case of embedding a DML statement such as an INSERT or UPDATE as a CTE inline with a primary statement that may draw from its results indirectly. The other is to provide control over the exact placement of a particular series of CTE constructs that should remain rendered directly in terms of a particular statement that may be nested in a larger statement.

    E.g.:

    1. from sqlalchemy import table, column, select
    2. t = table('t', column('c1'), column('c2'))
    4. ins = t.insert().values({"c1": "x", "c2": "y"}).cte()
    6. stmt = select(t).add_cte(ins)

    Would render:

    1. WITH anon_1 AS
    2. (INSERT INTO t (c1, c2) VALUES (:param_1, :param_2))
    3. SELECT t.c1, t.c2
    4. FROM t

    Above, the “anon_1” CTE is not referred towards in the SELECT statement, however still accomplishes the task of running an INSERT statement.

    Similarly in a DML-related context, using the PostgreSQL Insert construct to generate an “upsert”:

    1. from sqlalchemy import table, column
    2. from sqlalchemy.dialects.postgresql import insert
    4. t = table("t", column("c1"), column("c2"))
    6. delete_statement_cte = (
    7.     t.delete().where(t.c.c1 < 1).cte("deletions")
    8. )
    10. insert_stmt = insert(t).values({"c1": 1, "c2": 2})
    11. update_statement = insert_stmt.on_conflict_do_update(
    12.     index_elements=[t.c.c1],
    13.     set_={
    14.         "c1": insert_stmt.excluded.c1,
    15.         "c2": insert_stmt.excluded.c2,
    16.     },
    17. ).add_cte(delete_statement_cte)
    19. print(update_statement)

    The above statement renders as:

    1. WITH deletions AS
    2. (DELETE FROM t WHERE t.c1 < %(c1_1)s)
    3. INSERT INTO t (c1, c2) VALUES (%(c1)s, %(c2)s)
    4. ON CONFLICT (c1) DO UPDATE SET c1 = excluded.c1, c2 = excluded.c2

    New in version 1.4.21.

    • Parameters:

      • *ctes –

        zero or more CTE constructs.

        Changed in version 2.0: Multiple CTE instances are accepted

      • nest_here –

        if True, the given CTE or CTEs will be rendered as though they specified the HasCTE.cte.nesting flag to True when they were added to this HasCTE. Assuming the given CTEs are not referenced in an outer-enclosing statement as well, the CTEs given should render at the level of this statement when this flag is given.

        New in version 2.0.

        See also

        HasCTE.cte.nesting

  • method sqlalchemy.sql.expression.HasCTE.cte(name: Optional[str] = None, recursive: bool = False, nesting: bool = False) → CTE

    Return a new CTE, or Common Table Expression instance.

    Common table expressions are a SQL standard whereby SELECT statements can draw upon secondary statements specified along with the primary statement, using a clause called “WITH”. Special semantics regarding UNION can also be employed to allow “recursive” queries, where a SELECT statement can draw upon the set of rows that have previously been selected.

    CTEs can also be applied to DML constructs UPDATE, INSERT and DELETE on some databases, both as a source of CTE rows when combined with RETURNING, as well as a consumer of CTE rows.

    Changed in version 1.1: Added support for UPDATE/INSERT/DELETE as CTE, CTEs added to UPDATE/INSERT/DELETE.

    SQLAlchemy detects CTE objects, which are treated similarly to Alias objects, as special elements to be delivered to the FROM clause of the statement as well as to a WITH clause at the top of the statement.

    For special prefixes such as PostgreSQL “MATERIALIZED” and “NOT MATERIALIZED”, the CTE.prefix_with() method may be used to establish these.

    Changed in version 1.3.13: Added support for prefixes. In particular - MATERIALIZED and NOT MATERIALIZED.

    • Parameters:

      • name – name given to the common table expression. Like FromClause.alias(), the name can be left as None in which case an anonymous symbol will be used at query compile time.

      • recursive – if True, will render WITH RECURSIVE. A recursive common table expression is intended to be used in conjunction with UNION ALL in order to derive rows from those already selected.

      • nesting –

        if True, will render the CTE locally to the statement in which it is referenced. For more complex scenarios, the HasCTE.add_cte() method using the HasCTE.add_cte.nest_here parameter may also be used to more carefully control the exact placement of a particular CTE.

        New in version 1.4.24.

        See also

        HasCTE.add_cte()

  1. The following examples include two from PostgreSQL’s documentation at [https://www.postgresql.org/docs/current/static/queries-with.html](https://www.postgresql.org/docs/current/static/queries-with.html), as well as additional examples.
  3. Example 1, non recursive:
  5. ```
  6. from sqlalchemy import (Table, Column, String, Integer,
  7.                         MetaData, select, func)
  9. metadata = MetaData()
  11. orders = Table('orders', metadata,
  12.     Column('region', String),
  13.     Column('amount', Integer),
  14.     Column('product', String),
  15.     Column('quantity', Integer)
  16. )
  18. regional_sales = select(
  19.                     orders.c.region,
  20.                     func.sum(orders.c.amount).label('total_sales')
  21.                 ).group_by(orders.c.region).cte("regional_sales")
  24. top_regions = select(regional_sales.c.region).\
  25.         where(
  26.             regional_sales.c.total_sales >
  27.             select(
  28.                 func.sum(regional_sales.c.total_sales) / 10
  29.             )
  30.         ).cte("top_regions")
  32. statement = select(
  33.             orders.c.region,
  34.             orders.c.product,
  35.             func.sum(orders.c.quantity).label("product_units"),
  36.             func.sum(orders.c.amount).label("product_sales")
  37.     ).where(orders.c.region.in_(
  38.         select(top_regions.c.region)
  39.     )).group_by(orders.c.region, orders.c.product)
  41. result = conn.execute(statement).fetchall()
  42. ```
  44. Example 2, WITH RECURSIVE:
  46. ```
  47. from sqlalchemy import (Table, Column, String, Integer,
  48.                         MetaData, select, func)
  50. metadata = MetaData()
  52. parts = Table('parts', metadata,
  53.     Column('part', String),
  54.     Column('sub_part', String),
  55.     Column('quantity', Integer),
  56. )
  58. included_parts = select(\
  59.     parts.c.sub_part, parts.c.part, parts.c.quantity\
  60.     ).\
  61.     where(parts.c.part=='our part').\
  62.     cte(recursive=True)
  65. incl_alias = included_parts.alias()
  66. parts_alias = parts.alias()
  67. included_parts = included_parts.union_all(
  68.     select(
  69.         parts_alias.c.sub_part,
  70.         parts_alias.c.part,
  71.         parts_alias.c.quantity
  72.     ).\
  73.     where(parts_alias.c.part==incl_alias.c.sub_part)
  74. )
  76. statement = select(
  77.             included_parts.c.sub_part,
  78.             func.sum(included_parts.c.quantity).
  79.               label('total_quantity')
  80.         ).\
  81.         group_by(included_parts.c.sub_part)
  83. result = conn.execute(statement).fetchall()
  84. ```
  86. Example 3, an upsert using UPDATE and INSERT with CTEs:
  88. ```
  89. from datetime import date
  90. from sqlalchemy import (MetaData, Table, Column, Integer,
  91.                         Date, select, literal, and_, exists)
  93. metadata = MetaData()
  95. visitors = Table('visitors', metadata,
  96.     Column('product_id', Integer, primary_key=True),
  97.     Column('date', Date, primary_key=True),
  98.     Column('count', Integer),
  99. )
  101. # add 5 visitors for the product_id == 1
  102. product_id = 1
  103. day = date.today()
  104. count = 5
  106. update_cte = (
  107.     visitors.update()
  108.     .where(and_(visitors.c.product_id == product_id,
  109.                 visitors.c.date == day))
  110.     .values(count=visitors.c.count + count)
  111.     .returning(literal(1))
  112.     .cte('update_cte')
  113. )
  115. upsert = visitors.insert().from_select(
  116.     [visitors.c.product_id, visitors.c.date, visitors.c.count],
  117.     select(literal(product_id), literal(day), literal(count))
  118.         .where(~exists(update_cte.select()))
  119. )
  121. connection.execute(upsert)
  122. ```
  124. Example 4, Nesting CTE (SQLAlchemy 1.4.24 and above):
  126. ```
  127. value_a = select(
  128.     literal("root").label("n")
  129. ).cte("value_a")
  131. # A nested CTE with the same name as the root one
  132. value_a_nested = select(
  133.     literal("nesting").label("n")
  134. ).cte("value_a", nesting=True)
  136. # Nesting CTEs takes ascendency locally
  137. # over the CTEs at a higher level
  138. value_b = select(value_a_nested.c.n).cte("value_b")
  140. value_ab = select(value_a.c.n.label("a"), value_b.c.n.label("b"))
  141. ```
  143. The above query will render the second CTE nested inside the first, shown with inline parameters below as:
  145. ```
  146. WITH
  147.     value_a AS
  148.         (SELECT 'root' AS n),
  149.     value_b AS
  150.         (WITH value_a AS
  151.             (SELECT 'nesting' AS n)
  152.         SELECT value_a.n AS n FROM value_a)
  153. SELECT value_a.n AS a, value_b.n AS b
  154. FROM value_a, value_b
  155. ```
  157. The same CTE can be set up using the [HasCTE.add\_cte()](#sqlalchemy.sql.expression.HasCTE.add_cte "sqlalchemy.sql.expression.HasCTE.add_cte") method as follows (SQLAlchemy 2.0 and above):
  159. ```
  160. value_a = select(
  161.     literal("root").label("n")
  162. ).cte("value_a")
  164. # A nested CTE with the same name as the root one
  165. value_a_nested = select(
  166.     literal("nesting").label("n")
  167. ).cte("value_a")
  169. # Nesting CTEs takes ascendency locally
  170. # over the CTEs at a higher level
  171. value_b = (
  172.     select(value_a_nested.c.n).
  173.     add_cte(value_a_nested, nest_here=True).
  174.     cte("value_b")
  175. )
  177. value_ab = select(value_a.c.n.label("a"), value_b.c.n.label("b"))
  178. ```
  180. Example 5, Non-Linear CTE (SQLAlchemy 1.4.28 and above):
  182. ```
  183. edge = Table(
  184.     "edge",
  185.     metadata,
  186.     Column("id", Integer, primary_key=True),
  187.     Column("left", Integer),
  188.     Column("right", Integer),
  189. )
  191. root_node = select(literal(1).label("node")).cte(
  192.     "nodes", recursive=True
  193. )
  195. left_edge = select(edge.c.left).join(
  196.     root_node, edge.c.right == root_node.c.node
  197. )
  198. right_edge = select(edge.c.right).join(
  199.     root_node, edge.c.left == root_node.c.node
  200. )
  202. subgraph_cte = root_node.union(left_edge, right_edge)
  204. subgraph = select(subgraph_cte)
  205. ```
  207. The above query will render 2 UNIONs inside the recursive CTE:
  209. ```
  210. WITH RECURSIVE nodes(node) AS (
  211.         SELECT 1 AS node
  212.     UNION
  213.         SELECT edge."left" AS "left"
  214.         FROM edge JOIN nodes ON edge."right" = nodes.node
  215.     UNION
  216.         SELECT edge."right" AS "right"
  217.         FROM edge JOIN nodes ON edge."left" = nodes.node
  218. )
  219. SELECT nodes.node FROM nodes
  220. ```
  222. See also
  224. [Query.cte()]($3d0cc000ec6c7150.md#sqlalchemy.orm.Query.cte "sqlalchemy.orm.Query.cte") - ORM version of [HasCTE.cte()](#sqlalchemy.sql.expression.HasCTE.cte "sqlalchemy.sql.expression.HasCTE.cte").

class sqlalchemy.sql.expression.HasPrefixes

Members

prefix_with()

  • method sqlalchemy.sql.expression.HasPrefixes.prefix_with(*prefixes: _TextCoercedExpressionArgument[Any], dialect: str = ‘*‘) → SelfHasPrefixes

    Add one or more expressions following the statement keyword, i.e. SELECT, INSERT, UPDATE, or DELETE. Generative.

    This is used to support backend-specific prefix keywords such as those provided by MySQL.

    E.g.:

    1. stmt = table.insert().prefix_with("LOW_PRIORITY", dialect="mysql")
    3. # MySQL 5.7 optimizer hints
    4. stmt = select(table).prefix_with(
    5.     "/*+ BKA(t1) */", dialect="mysql")

    Multiple prefixes can be specified by multiple calls to HasPrefixes.prefix_with().

    • Parameters:

      • *prefixes – textual or ClauseElement construct which will be rendered following the INSERT, UPDATE, or DELETE keyword.

      • dialect – optional string dialect name which will limit rendering of this prefix to only that dialect.

class sqlalchemy.sql.expression.HasSuffixes

Members

suffix_with()

  • method sqlalchemy.sql.expression.HasSuffixes.suffix_with(*suffixes: _TextCoercedExpressionArgument[Any], dialect: str = ‘*‘) → SelfHasSuffixes

    Add one or more expressions following the statement as a whole.

    This is used to support backend-specific suffix keywords on certain constructs.

    E.g.:

    1. stmt = select(col1, col2).cte().suffix_with(
    2.     "cycle empno set y_cycle to 1 default 0", dialect="oracle")

    Multiple suffixes can be specified by multiple calls to HasSuffixes.suffix_with().

    • Parameters:

      • *suffixes – textual or ClauseElement construct which will be rendered following the target clause.

      • dialect – Optional string dialect name which will limit rendering of this suffix to only that dialect.

class sqlalchemy.sql.expression.Join

Represent a JOIN construct between two FromClause elements.

The public constructor function for Join is the module-level join() function, as well as the FromClause.join() method of any FromClause (e.g. such as Table).

See also

join()

FromClause.join()

Members

__init__(), description, is_derived_from(), select(), self_group()

Class signature

class sqlalchemy.sql.expression.Join (sqlalchemy.sql.roles.DMLTableRole, sqlalchemy.sql.expression.FromClause)

  • method sqlalchemy.sql.expression.Join.__init__(left: _FromClauseArgument, right: _FromClauseArgument, onclause: Optional[_OnClauseArgument] = None, isouter: bool = False, full: bool = False)

    Construct a new Join.

    The usual entrypoint here is the join() function or the FromClause.join() method of any FromClause object.

  • attribute sqlalchemy.sql.expression.Join.description

  • method sqlalchemy.sql.expression.Join.is_derived_from(fromclause: Optional[FromClause]) → bool

    Return True if this FromClause is ‘derived’ from the given FromClause.

    An example would be an Alias of a Table is derived from that Table.

  • method sqlalchemy.sql.expression.Join.select() → Select

    Create a Select from this Join.

    E.g.:

    1. stmt = table_a.join(table_b, table_a.c.id == table_b.c.a_id)
    3. stmt = stmt.select()

    The above will produce a SQL string resembling:

    1. SELECT table_a.id, table_a.col, table_b.id, table_b.a_id
    2. FROM table_a JOIN table_b ON table_a.id = table_b.a_id

  • method sqlalchemy.sql.expression.Join.self_group(against: Optional[OperatorType] = None) → FromGrouping

    Apply a ‘grouping’ to this ClauseElement.

    This method is overridden by subclasses to return a “grouping” construct, i.e. parenthesis. In particular it’s used by “binary” expressions to provide a grouping around themselves when placed into a larger expression, as well as by select() constructs when placed into the FROM clause of another select(). (Note that subqueries should be normally created using the Select.alias() method, as many platforms require nested SELECT statements to be named).

    As expressions are composed together, the application of self_group() is automatic - end-user code should never need to use this method directly. Note that SQLAlchemy’s clause constructs take operator precedence into account - so parenthesis might not be needed, for example, in an expression like x OR (y AND z) - AND takes precedence over OR.

    The base self_group() method of ClauseElement just returns self.

class sqlalchemy.sql.expression.Lateral

Represent a LATERAL subquery.

This object is constructed from the lateral() module level function as well as the FromClause.lateral() method available on all FromClause subclasses.

While LATERAL is part of the SQL standard, currently only more recent PostgreSQL versions provide support for this keyword.

New in version 1.1.

See also

LATERAL correlation - overview of usage.

Members

inherit_cache

Class signature

class sqlalchemy.sql.expression.Lateral (sqlalchemy.sql.expression.FromClauseAlias, sqlalchemy.sql.expression.LateralFromClause)

  • attribute sqlalchemy.sql.expression.Lateral.inherit_cache: Optional[bool] = True

    Indicate if this HasCacheKey instance should make use of the cache key generation scheme used by its immediate superclass.

    The attribute defaults to None, which indicates that a construct has not yet taken into account whether or not its appropriate for it to participate in caching; this is functionally equivalent to setting the value to False, except that a warning is also emitted.

    This flag can be set to True on a particular class, if the SQL that corresponds to the object does not change based on attributes which are local to this class, and not its superclass.

    See also

    Enabling Caching Support for Custom Constructs - General guideslines for setting the HasCacheKey.inherit_cache attribute for third-party or user defined SQL constructs.

class sqlalchemy.sql.expression.ReturnsRows

The base-most class for Core constructs that have some concept of columns that can represent rows.

While the SELECT statement and TABLE are the primary things we think of in this category, DML like INSERT, UPDATE and DELETE can also specify RETURNING which means they can be used in CTEs and other forms, and PostgreSQL has functions that return rows also.

New in version 1.4.

Members

exported_columns, is_derived_from()

Class signature

class sqlalchemy.sql.expression.ReturnsRows (sqlalchemy.sql.roles.ReturnsRowsRole, sqlalchemy.sql.expression.DQLDMLClauseElement)

class sqlalchemy.sql.expression.ScalarSelect

Represent a scalar subquery.

A ScalarSelect is created by invoking the SelectBase.scalar_subquery() method. The object then participates in other SQL expressions as a SQL column expression within the ColumnElement hierarchy.

See also

SelectBase.scalar_subquery()

Scalar and Correlated Subqueries - in the 2.0 tutorial

Members

correlate(), correlate_except(), inherit_cache, self_group(), where()

Class signature

class sqlalchemy.sql.expression.ScalarSelect (sqlalchemy.sql.roles.InElementRole, sqlalchemy.sql.expression.Generative, sqlalchemy.sql.expression.GroupedElement, sqlalchemy.sql.expression.ColumnElement)

  • method sqlalchemy.sql.expression.ScalarSelect.correlate(*fromclauses: Union[Literal[None, False], _FromClauseArgument]) → SelfScalarSelect

    Return a new ScalarSelect which will correlate the given FROM clauses to that of an enclosing Select.

    This method is mirrored from the Select.correlate() method of the underlying Select. The method applies the :meth:_sql.Select.correlate` method, then returns a new ScalarSelect against that statement.

    New in version 1.4: Previously, the ScalarSelect.correlate() method was only available from Select.

    • Parameters:

      *fromclauses – a list of one or more FromClause constructs, or other compatible constructs (i.e. ORM-mapped classes) to become part of the correlate collection.

    See also

    ScalarSelect.correlate_except()

    Scalar and Correlated Subqueries - in the 2.0 tutorial

  • method sqlalchemy.sql.expression.ScalarSelect.correlate_except(*fromclauses: Union[Literal[None, False], _FromClauseArgument]) → SelfScalarSelect

    Return a new ScalarSelect which will omit the given FROM clauses from the auto-correlation process.

    This method is mirrored from the Select.correlate_except() method of the underlying Select. The method applies the :meth:_sql.Select.correlate_except` method, then returns a new ScalarSelect against that statement.

    New in version 1.4: Previously, the ScalarSelect.correlate_except() method was only available from Select.

    • Parameters:

      *fromclauses – a list of one or more FromClause constructs, or other compatible constructs (i.e. ORM-mapped classes) to become part of the correlate-exception collection.

    See also

    ScalarSelect.correlate()

    Scalar and Correlated Subqueries - in the 2.0 tutorial

  • attribute sqlalchemy.sql.expression.ScalarSelect.inherit_cache: Optional[bool] = True

    Indicate if this HasCacheKey instance should make use of the cache key generation scheme used by its immediate superclass.

    The attribute defaults to None, which indicates that a construct has not yet taken into account whether or not its appropriate for it to participate in caching; this is functionally equivalent to setting the value to False, except that a warning is also emitted.

    This flag can be set to True on a particular class, if the SQL that corresponds to the object does not change based on attributes which are local to this class, and not its superclass.

    See also

    Enabling Caching Support for Custom Constructs - General guideslines for setting the HasCacheKey.inherit_cache attribute for third-party or user defined SQL constructs.

  • method sqlalchemy.sql.expression.ScalarSelect.self_group(against: Optional[OperatorType] = None) → ColumnElement[Any]

    Apply a ‘grouping’ to this ClauseElement.

    This method is overridden by subclasses to return a “grouping” construct, i.e. parenthesis. In particular it’s used by “binary” expressions to provide a grouping around themselves when placed into a larger expression, as well as by select() constructs when placed into the FROM clause of another select(). (Note that subqueries should be normally created using the Select.alias() method, as many platforms require nested SELECT statements to be named).

    As expressions are composed together, the application of self_group() is automatic - end-user code should never need to use this method directly. Note that SQLAlchemy’s clause constructs take operator precedence into account - so parenthesis might not be needed, for example, in an expression like x OR (y AND z) - AND takes precedence over OR.

    The base self_group() method of ClauseElement just returns self.

  • method sqlalchemy.sql.expression.ScalarSelect.where(crit: _ColumnExpressionArgument[bool]) → SelfScalarSelect

    Apply a WHERE clause to the SELECT statement referred to by this ScalarSelect.

class sqlalchemy.sql.expression.Select

Represents a SELECT statement.

The Select object is normally constructed using the select() function. See that function for details.

See also

select()

Using SELECT Statements - in the 2.0 tutorial

Members

__init__(), add_columns(), add_cte(), alias(), as_scalar(), c, column(), column_descriptions, columns_clause_froms, correlate(), correlate_except(), corresponding_column(), cte(), distinct(), except_(), except_all(), execution_options(), exists(), exported_columns, fetch(), filter(), filter_by(), from_statement(), froms, get_children(), get_execution_options(), get_final_froms(), get_label_style(), group_by(), having(), inherit_cache, inner_columns, intersect(), intersect_all(), is_derived_from(), join(), join_from(), label(), lateral(), limit(), offset(), options(), order_by(), outerjoin(), outerjoin_from(), prefix_with(), reduce_columns(), replace_selectable(), scalar_subquery(), select(), select_from(), selected_columns, self_group(), set_label_style(), slice(), subquery(), suffix_with(), union(), union_all(), where(), whereclause, with_for_update(), with_hint(), with_only_columns(), with_statement_hint()

Class signature

class sqlalchemy.sql.expression.Select (sqlalchemy.sql.expression.HasPrefixes, sqlalchemy.sql.expression.HasSuffixes, sqlalchemy.sql.expression.HasHints, sqlalchemy.sql.expression.HasCompileState, sqlalchemy.sql.expression._SelectFromElements, sqlalchemy.sql.expression.GenerativeSelect, sqlalchemy.sql.expression.TypedReturnsRows)

  • method sqlalchemy.sql.expression.Select.__init__(*entities: _ColumnsClauseArgument[Any])

    Construct a new Select.

    The public constructor for Select is the select() function.

  • method sqlalchemy.sql.expression.Select.add_columns(*entities: _ColumnsClauseArgument[Any]) → Select[Any]

    Return a new select() construct with the given entities appended to its columns clause.

    E.g.:

    1. my_select = my_select.add_columns(table.c.new_column)

    The original expressions in the columns clause remain in place. To replace the original expressions with new ones, see the method Select.with_only_columns().

    • Parameters:

      *entities – column, table, or other entity expressions to be added to the columns clause

    See also

    Select.with_only_columns() - replaces existing expressions rather than appending.

    Selecting Multiple ORM Entities Simultaneously - ORM-centric example

  • method sqlalchemy.sql.expression.Select.add_cte(*ctes: CTE, nest_here: bool = False) → SelfHasCTE

    inherited from the HasCTE.add_cte() method of HasCTE

    Add one or more CTE constructs to this statement.

    This method will associate the given CTE constructs with the parent statement such that they will each be unconditionally rendered in the WITH clause of the final statement, even if not referenced elsewhere within the statement or any sub-selects.

    The optional HasCTE.add_cte.nest_here parameter when set to True will have the effect that each given CTE will render in a WITH clause rendered directly along with this statement, rather than being moved to the top of the ultimate rendered statement, even if this statement is rendered as a subquery within a larger statement.

    This method has two general uses. One is to embed CTE statements that serve some purpose without being referenced explicitly, such as the use case of embedding a DML statement such as an INSERT or UPDATE as a CTE inline with a primary statement that may draw from its results indirectly. The other is to provide control over the exact placement of a particular series of CTE constructs that should remain rendered directly in terms of a particular statement that may be nested in a larger statement.

    E.g.:

    1. from sqlalchemy import table, column, select
    2. t = table('t', column('c1'), column('c2'))
    4. ins = t.insert().values({"c1": "x", "c2": "y"}).cte()
    6. stmt = select(t).add_cte(ins)

    Would render:

    1. WITH anon_1 AS
    2. (INSERT INTO t (c1, c2) VALUES (:param_1, :param_2))
    3. SELECT t.c1, t.c2
    4. FROM t

    Above, the “anon_1” CTE is not referred towards in the SELECT statement, however still accomplishes the task of running an INSERT statement.

    Similarly in a DML-related context, using the PostgreSQL Insert construct to generate an “upsert”:

    1. from sqlalchemy import table, column
    2. from sqlalchemy.dialects.postgresql import insert
    4. t = table("t", column("c1"), column("c2"))
    6. delete_statement_cte = (
    7.     t.delete().where(t.c.c1 < 1).cte("deletions")
    8. )
    10. insert_stmt = insert(t).values({"c1": 1, "c2": 2})
    11. update_statement = insert_stmt.on_conflict_do_update(
    12.     index_elements=[t.c.c1],
    13.     set_={
    14.         "c1": insert_stmt.excluded.c1,
    15.         "c2": insert_stmt.excluded.c2,
    16.     },
    17. ).add_cte(delete_statement_cte)
    19. print(update_statement)

    The above statement renders as:

    1. WITH deletions AS
    2. (DELETE FROM t WHERE t.c1 < %(c1_1)s)
    3. INSERT INTO t (c1, c2) VALUES (%(c1)s, %(c2)s)
    4. ON CONFLICT (c1) DO UPDATE SET c1 = excluded.c1, c2 = excluded.c2

    New in version 1.4.21.

    • Parameters:

      • *ctes –

        zero or more CTE constructs.

        Changed in version 2.0: Multiple CTE instances are accepted

      • nest_here –

        if True, the given CTE or CTEs will be rendered as though they specified the HasCTE.cte.nesting flag to True when they were added to this HasCTE. Assuming the given CTEs are not referenced in an outer-enclosing statement as well, the CTEs given should render at the level of this statement when this flag is given.

        New in version 2.0.

        See also

        HasCTE.cte.nesting

  1. See also
  3. [Selectable.exported\_columns](#sqlalchemy.sql.expression.Selectable.exported_columns "sqlalchemy.sql.expression.Selectable.exported_columns") - the [ColumnCollection]($aafca12b71ff5dd3.md#sqlalchemy.sql.expression.ColumnCollection "sqlalchemy.sql.expression.ColumnCollection") that is used for the operation.
  5. [ColumnCollection.corresponding\_column()]($aafca12b71ff5dd3.md#sqlalchemy.sql.expression.ColumnCollection.corresponding_column "sqlalchemy.sql.expression.ColumnCollection.corresponding_column") - implementation method.

  • method sqlalchemy.sql.expression.Select.cte(name: Optional[str] = None, recursive: bool = False, nesting: bool = False) → CTE

    inherited from the HasCTE.cte() method of HasCTE

    Return a new CTE, or Common Table Expression instance.

    Common table expressions are a SQL standard whereby SELECT statements can draw upon secondary statements specified along with the primary statement, using a clause called “WITH”. Special semantics regarding UNION can also be employed to allow “recursive” queries, where a SELECT statement can draw upon the set of rows that have previously been selected.

    CTEs can also be applied to DML constructs UPDATE, INSERT and DELETE on some databases, both as a source of CTE rows when combined with RETURNING, as well as a consumer of CTE rows.

    Changed in version 1.1: Added support for UPDATE/INSERT/DELETE as CTE, CTEs added to UPDATE/INSERT/DELETE.

    SQLAlchemy detects CTE objects, which are treated similarly to Alias objects, as special elements to be delivered to the FROM clause of the statement as well as to a WITH clause at the top of the statement.

    For special prefixes such as PostgreSQL “MATERIALIZED” and “NOT MATERIALIZED”, the CTE.prefix_with() method may be used to establish these.

    Changed in version 1.3.13: Added support for prefixes. In particular - MATERIALIZED and NOT MATERIALIZED.

    • Parameters:

      • name – name given to the common table expression. Like FromClause.alias(), the name can be left as None in which case an anonymous symbol will be used at query compile time.

      • recursive – if True, will render WITH RECURSIVE. A recursive common table expression is intended to be used in conjunction with UNION ALL in order to derive rows from those already selected.

      • nesting –

        if True, will render the CTE locally to the statement in which it is referenced. For more complex scenarios, the HasCTE.add_cte() method using the HasCTE.add_cte.nest_here parameter may also be used to more carefully control the exact placement of a particular CTE.

        New in version 1.4.24.

        See also

        HasCTE.add_cte()

  1. The following examples include two from PostgreSQL’s documentation at [https://www.postgresql.org/docs/current/static/queries-with.html](https://www.postgresql.org/docs/current/static/queries-with.html), as well as additional examples.
  3. Example 1, non recursive:
  5. ```
  6. from sqlalchemy import (Table, Column, String, Integer,
  7.                         MetaData, select, func)
  9. metadata = MetaData()
  11. orders = Table('orders', metadata,
  12.     Column('region', String),
  13.     Column('amount', Integer),
  14.     Column('product', String),
  15.     Column('quantity', Integer)
  16. )
  18. regional_sales = select(
  19.                     orders.c.region,
  20.                     func.sum(orders.c.amount).label('total_sales')
  21.                 ).group_by(orders.c.region).cte("regional_sales")
  24. top_regions = select(regional_sales.c.region).\
  25.         where(
  26.             regional_sales.c.total_sales >
  27.             select(
  28.                 func.sum(regional_sales.c.total_sales) / 10
  29.             )
  30.         ).cte("top_regions")
  32. statement = select(
  33.             orders.c.region,
  34.             orders.c.product,
  35.             func.sum(orders.c.quantity).label("product_units"),
  36.             func.sum(orders.c.amount).label("product_sales")
  37.     ).where(orders.c.region.in_(
  38.         select(top_regions.c.region)
  39.     )).group_by(orders.c.region, orders.c.product)
  41. result = conn.execute(statement).fetchall()
  42. ```
  44. Example 2, WITH RECURSIVE:
  46. ```
  47. from sqlalchemy import (Table, Column, String, Integer,
  48.                         MetaData, select, func)
  50. metadata = MetaData()
  52. parts = Table('parts', metadata,
  53.     Column('part', String),
  54.     Column('sub_part', String),
  55.     Column('quantity', Integer),
  56. )
  58. included_parts = select(\
  59.     parts.c.sub_part, parts.c.part, parts.c.quantity\
  60.     ).\
  61.     where(parts.c.part=='our part').\
  62.     cte(recursive=True)
  65. incl_alias = included_parts.alias()
  66. parts_alias = parts.alias()
  67. included_parts = included_parts.union_all(
  68.     select(
  69.         parts_alias.c.sub_part,
  70.         parts_alias.c.part,
  71.         parts_alias.c.quantity
  72.     ).\
  73.     where(parts_alias.c.part==incl_alias.c.sub_part)
  74. )
  76. statement = select(
  77.             included_parts.c.sub_part,
  78.             func.sum(included_parts.c.quantity).
  79.               label('total_quantity')
  80.         ).\
  81.         group_by(included_parts.c.sub_part)
  83. result = conn.execute(statement).fetchall()
  84. ```
  86. Example 3, an upsert using UPDATE and INSERT with CTEs:
  88. ```
  89. from datetime import date
  90. from sqlalchemy import (MetaData, Table, Column, Integer,
  91.                         Date, select, literal, and_, exists)
  93. metadata = MetaData()
  95. visitors = Table('visitors', metadata,
  96.     Column('product_id', Integer, primary_key=True),
  97.     Column('date', Date, primary_key=True),
  98.     Column('count', Integer),
  99. )
  101. # add 5 visitors for the product_id == 1
  102. product_id = 1
  103. day = date.today()
  104. count = 5
  106. update_cte = (
  107.     visitors.update()
  108.     .where(and_(visitors.c.product_id == product_id,
  109.                 visitors.c.date == day))
  110.     .values(count=visitors.c.count + count)
  111.     .returning(literal(1))
  112.     .cte('update_cte')
  113. )
  115. upsert = visitors.insert().from_select(
  116.     [visitors.c.product_id, visitors.c.date, visitors.c.count],
  117.     select(literal(product_id), literal(day), literal(count))
  118.         .where(~exists(update_cte.select()))
  119. )
  121. connection.execute(upsert)
  122. ```
  124. Example 4, Nesting CTE (SQLAlchemy 1.4.24 and above):
  126. ```
  127. value_a = select(
  128.     literal("root").label("n")
  129. ).cte("value_a")
  131. # A nested CTE with the same name as the root one
  132. value_a_nested = select(
  133.     literal("nesting").label("n")
  134. ).cte("value_a", nesting=True)
  136. # Nesting CTEs takes ascendency locally
  137. # over the CTEs at a higher level
  138. value_b = select(value_a_nested.c.n).cte("value_b")
  140. value_ab = select(value_a.c.n.label("a"), value_b.c.n.label("b"))
  141. ```
  143. The above query will render the second CTE nested inside the first, shown with inline parameters below as:
  145. ```
  146. WITH
  147.     value_a AS
  148.         (SELECT 'root' AS n),
  149.     value_b AS
  150.         (WITH value_a AS
  151.             (SELECT 'nesting' AS n)
  152.         SELECT value_a.n AS n FROM value_a)
  153. SELECT value_a.n AS a, value_b.n AS b
  154. FROM value_a, value_b
  155. ```
  157. The same CTE can be set up using the [HasCTE.add\_cte()](#sqlalchemy.sql.expression.HasCTE.add_cte "sqlalchemy.sql.expression.HasCTE.add_cte") method as follows (SQLAlchemy 2.0 and above):
  159. ```
  160. value_a = select(
  161.     literal("root").label("n")
  162. ).cte("value_a")
  164. # A nested CTE with the same name as the root one
  165. value_a_nested = select(
  166.     literal("nesting").label("n")
  167. ).cte("value_a")
  169. # Nesting CTEs takes ascendency locally
  170. # over the CTEs at a higher level
  171. value_b = (
  172.     select(value_a_nested.c.n).
  173.     add_cte(value_a_nested, nest_here=True).
  174.     cte("value_b")
  175. )
  177. value_ab = select(value_a.c.n.label("a"), value_b.c.n.label("b"))
  178. ```
  180. Example 5, Non-Linear CTE (SQLAlchemy 1.4.28 and above):
  182. ```
  183. edge = Table(
  184.     "edge",
  185.     metadata,
  186.     Column("id", Integer, primary_key=True),
  187.     Column("left", Integer),
  188.     Column("right", Integer),
  189. )
  191. root_node = select(literal(1).label("node")).cte(
  192.     "nodes", recursive=True
  193. )
  195. left_edge = select(edge.c.left).join(
  196.     root_node, edge.c.right == root_node.c.node
  197. )
  198. right_edge = select(edge.c.right).join(
  199.     root_node, edge.c.left == root_node.c.node
  200. )
  202. subgraph_cte = root_node.union(left_edge, right_edge)
  204. subgraph = select(subgraph_cte)
  205. ```
  207. The above query will render 2 UNIONs inside the recursive CTE:
  209. ```
  210. WITH RECURSIVE nodes(node) AS (
  211.         SELECT 1 AS node
  212.     UNION
  213.         SELECT edge."left" AS "left"
  214.         FROM edge JOIN nodes ON edge."right" = nodes.node
  215.     UNION
  216.         SELECT edge."right" AS "right"
  217.         FROM edge JOIN nodes ON edge."left" = nodes.node
  218. )
  219. SELECT nodes.node FROM nodes
  220. ```
  222. See also
  224. [Query.cte()]($3d0cc000ec6c7150.md#sqlalchemy.orm.Query.cte "sqlalchemy.orm.Query.cte") - ORM version of [HasCTE.cte()](#sqlalchemy.sql.expression.HasCTE.cte "sqlalchemy.sql.expression.HasCTE.cte").
  1. See also
  3. [GenerativeSelect.limit()](#sqlalchemy.sql.expression.GenerativeSelect.limit "sqlalchemy.sql.expression.GenerativeSelect.limit")
  5. [GenerativeSelect.offset()](#sqlalchemy.sql.expression.GenerativeSelect.offset "sqlalchemy.sql.expression.GenerativeSelect.offset")

  • method sqlalchemy.sql.expression.Select.intersect_all(*other: _SelectStatementForCompoundArgument) → CompoundSelect

    Return a SQL INTERSECT ALL of this select() construct against the given selectables provided as positional arguments.

    • Parameters:

      • *other –

        one or more elements with which to create a UNION.

        Changed in version 1.4.28: multiple elements are now accepted.

      • **kwargs – keyword arguments are forwarded to the constructor for the newly created CompoundSelect object.

  • method sqlalchemy.sql.expression.Select.is_derived_from(fromclause: Optional[FromClause]) → bool

    Return True if this ReturnsRows is ‘derived’ from the given FromClause.

    An example would be an Alias of a Table is derived from that Table.

  • method sqlalchemy.sql.expression.Select.join(target: _JoinTargetArgument, onclause: Optional[_OnClauseArgument] = None, *, isouter: bool = False, full: bool = False) → SelfSelect

    Create a SQL JOIN against this Select object’s criterion and apply generatively, returning the newly resulting Select.

    E.g.:

    1. stmt = select(user_table).join(address_table, user_table.c.id == address_table.c.user_id)

    The above statement generates SQL similar to:

    1. SELECT user.id, user.name FROM user JOIN address ON user.id = address.user_id

    Changed in version 1.4: Select.join() now creates a Join object between a FromClause source that is within the FROM clause of the existing SELECT, and a given target FromClause, and then adds this Join to the FROM clause of the newly generated SELECT statement. This is completely reworked from the behavior in 1.3, which would instead create a subquery of the entire Select and then join that subquery to the target.

    This is a backwards incompatible change as the previous behavior was mostly useless, producing an unnamed subquery rejected by most databases in any case. The new behavior is modeled after that of the very successful Query.join() method in the ORM, in order to support the functionality of Query being available by using a Select object with an Session.

    See the notes for this change at select().join() and outerjoin() add JOIN criteria to the current query, rather than creating a subquery.

    • Parameters:

      • target – target table to join towards

      • onclause – ON clause of the join. If omitted, an ON clause is generated automatically based on the ForeignKey linkages between the two tables, if one can be unambiguously determined, otherwise an error is raised.

      • isouter – if True, generate LEFT OUTER join. Same as Select.outerjoin().

      • full – if True, generate FULL OUTER join.

  1. See also
  3. [Explicit FROM clauses and JOINs]($93605e6ef77d4344.md#tutorial-select-join) - in the [SQLAlchemy Unified Tutorial]($4406c4fa3e52f66b.md)
  5. [Joins]($b3109e359428cd80.md#orm-queryguide-joins) - in the [ORM Querying Guide]($86681c2576bdda58.md)
  7. [Select.join\_from()](#sqlalchemy.sql.expression.Select.join_from "sqlalchemy.sql.expression.Select.join_from")
  9. [Select.outerjoin()](#sqlalchemy.sql.expression.Select.outerjoin "sqlalchemy.sql.expression.Select.outerjoin")

  • method sqlalchemy.sql.expression.Select.join_from(from\: _FromClauseArgument, _target: _JoinTargetArgument, onclause: Optional[_OnClauseArgument] = None, *, isouter: bool = False, full: bool = False) → SelfSelect

    Create a SQL JOIN against this Select object’s criterion and apply generatively, returning the newly resulting Select.

    E.g.:

    1. stmt = select(user_table, address_table).join_from(
    2.     user_table, address_table, user_table.c.id == address_table.c.user_id
    3. )

    The above statement generates SQL similar to:

    1. SELECT user.id, user.name, address.id, address.email, address.user_id
    2. FROM user JOIN address ON user.id = address.user_id

    New in version 1.4.

    • Parameters:

      • from_ – the left side of the join, will be rendered in the FROM clause and is roughly equivalent to using the Select.select_from() method.

      • target – target table to join towards

      • onclause – ON clause of the join.

      • isouter – if True, generate LEFT OUTER join. Same as Select.outerjoin().

      • full – if True, generate FULL OUTER join.

  1. See also
  3. [Explicit FROM clauses and JOINs]($93605e6ef77d4344.md#tutorial-select-join) - in the [SQLAlchemy Unified Tutorial]($4406c4fa3e52f66b.md)
  5. [Joins]($b3109e359428cd80.md#orm-queryguide-joins) - in the [ORM Querying Guide]($86681c2576bdda58.md)
  7. [Select.join()](#sqlalchemy.sql.expression.Select.join "sqlalchemy.sql.expression.Select.join")

  • method sqlalchemy.sql.expression.Select.label(name: Optional[str]) → Label[Any]

    inherited from the SelectBase.label() method of SelectBase

    Return a ‘scalar’ representation of this selectable, embedded as a subquery with a label.

    See also

    SelectBase.as_scalar().

  • method sqlalchemy.sql.expression.Select.lateral(name: Optional[str] = None) → LateralFromClause

    inherited from the SelectBase.lateral() method of SelectBase

    Return a LATERAL alias of this Selectable.

    The return value is the Lateral construct also provided by the top-level lateral() function.

    New in version 1.1.

    See also

    LATERAL correlation - overview of usage.

  • method sqlalchemy.sql.expression.Select.limit(limit: Union[int, _ColumnExpressionArgument[int]]) → SelfGenerativeSelect

    inherited from the GenerativeSelect.limit() method of GenerativeSelect

    Return a new selectable with the given LIMIT criterion applied.

    This is a numerical value which usually renders as a LIMIT expression in the resulting select. Backends that don’t support LIMIT will attempt to provide similar functionality.

    Note

    The GenerativeSelect.limit() method will replace any clause applied with GenerativeSelect.fetch().

    Changed in version 1.0.0: - Select.limit() can now accept arbitrary SQL expressions as well as integer values.

    • Parameters:

      limit – an integer LIMIT parameter, or a SQL expression that provides an integer result. Pass None to reset it.

    See also

    GenerativeSelect.fetch()

    GenerativeSelect.offset()

  • method sqlalchemy.sql.expression.Select.offset(offset: Union[int, _ColumnExpressionArgument[int]]) → SelfGenerativeSelect

    inherited from the GenerativeSelect.offset() method of GenerativeSelect

    Return a new selectable with the given OFFSET criterion applied.

    This is a numeric value which usually renders as an OFFSET expression in the resulting select. Backends that don’t support OFFSET will attempt to provide similar functionality.

    Changed in version 1.0.0: - Select.offset() can now accept arbitrary SQL expressions as well as integer values.

    • Parameters:

      offset – an integer OFFSET parameter, or a SQL expression that provides an integer result. Pass None to reset it.

    See also

    GenerativeSelect.limit()

    GenerativeSelect.fetch()

  • method sqlalchemy.sql.expression.Select.options(*options: ExecutableOption) → SelfExecutable

    inherited from the Executable.options() method of Executable

    Apply options to this statement.

    In the general sense, options are any kind of Python object that can be interpreted by the SQL compiler for the statement. These options can be consumed by specific dialects or specific kinds of compilers.

    The most commonly known kind of option are the ORM level options that apply “eager load” and other loading behaviors to an ORM query. However, options can theoretically be used for many other purposes.

    For background on specific kinds of options for specific kinds of statements, refer to the documentation for those option objects.

    Changed in version 1.4: - added Executable.options() to Core statement objects towards the goal of allowing unified Core / ORM querying capabilities.

    See also

    Column Loading Options - refers to options specific to the usage of ORM queries

    Relationship Loading with Loader Options - refers to options specific to the usage of ORM queries

  • method sqlalchemy.sql.expression.Select.order_by(_GenerativeSelect\_first: Union[Literal[None, _NoArg.NO_ARG], _ColumnExpressionOrStrLabelArgument[Any]] = _NoArg.NO_ARG, *clauses: _ColumnExpressionOrStrLabelArgument[Any]_) → SelfGenerativeSelect

    inherited from the GenerativeSelect.order_by() method of GenerativeSelect

    Return a new selectable with the given list of ORDER BY criteria applied.

    e.g.:

    1. stmt = select(table).order_by(table.c.id, table.c.name)

    Calling this method multiple times is equivalent to calling it once with all the clauses concatenated. All existing ORDER BY criteria may be cancelled by passing None by itself. New ORDER BY criteria may then be added by invoking Query.order_by() again, e.g.:

    1. # will erase all ORDER BY and ORDER BY new_col alone
    2. stmt = stmt.order_by(None).order_by(new_col)

    • Parameters:

      *clauses – a series of ColumnElement constructs which will be used to generate an ORDER BY clause.

    See also

    ORDER BY - in the SQLAlchemy Unified Tutorial

    Ordering or Grouping by a Label - in the SQLAlchemy Unified Tutorial

  • method sqlalchemy.sql.expression.Select.outerjoin(target: _JoinTargetArgument, onclause: Optional[_OnClauseArgument] = None, *, full: bool = False) → SelfSelect

    Create a left outer join.

    Parameters are the same as that of Select.join().

    Changed in version 1.4: Select.outerjoin() now creates a Join object between a FromClause source that is within the FROM clause of the existing SELECT, and a given target FromClause, and then adds this Join to the FROM clause of the newly generated SELECT statement. This is completely reworked from the behavior in 1.3, which would instead create a subquery of the entire Select and then join that subquery to the target.

    This is a backwards incompatible change as the previous behavior was mostly useless, producing an unnamed subquery rejected by most databases in any case. The new behavior is modeled after that of the very successful Query.join() method in the ORM, in order to support the functionality of Query being available by using a Select object with an Session.

    See the notes for this change at select().join() and outerjoin() add JOIN criteria to the current query, rather than creating a subquery.

    See also

    Explicit FROM clauses and JOINs - in the SQLAlchemy Unified Tutorial

    Joins - in the ORM Querying Guide

    Select.join()

  • method sqlalchemy.sql.expression.Select.outerjoin_from(from\: _FromClauseArgument, _target: _JoinTargetArgument, onclause: Optional[_OnClauseArgument] = None, *, full: bool = False) → SelfSelect

    Create a SQL LEFT OUTER JOIN against this Select object’s criterion and apply generatively, returning the newly resulting Select.

    Usage is the same as that of Select.join_from().

  • method sqlalchemy.sql.expression.Select.prefix_with(*prefixes: _TextCoercedExpressionArgument[Any], dialect: str = ‘*‘) → SelfHasPrefixes

    inherited from the HasPrefixes.prefix_with() method of HasPrefixes

    Add one or more expressions following the statement keyword, i.e. SELECT, INSERT, UPDATE, or DELETE. Generative.

    This is used to support backend-specific prefix keywords such as those provided by MySQL.

    E.g.:

    1. stmt = table.insert().prefix_with("LOW_PRIORITY", dialect="mysql")
    3. # MySQL 5.7 optimizer hints
    4. stmt = select(table).prefix_with(
    5.     "/*+ BKA(t1) */", dialect="mysql")

    Multiple prefixes can be specified by multiple calls to HasPrefixes.prefix_with().

    • Parameters:

      • *prefixes – textual or ClauseElement construct which will be rendered following the INSERT, UPDATE, or DELETE keyword.

      • dialect – optional string dialect name which will limit rendering of this prefix to only that dialect.

  • method sqlalchemy.sql.expression.Select.reduce_columns(only_synonyms: bool = True) → Select

    Return a new select() construct with redundantly named, equivalently-valued columns removed from the columns clause.

    “Redundant” here means two columns where one refers to the other either based on foreign key, or via a simple equality comparison in the WHERE clause of the statement. The primary purpose of this method is to automatically construct a select statement with all uniquely-named columns, without the need to use table-qualified labels as Select.set_label_style() does.

    When columns are omitted based on foreign key, the referred-to column is the one that’s kept. When columns are omitted based on WHERE equivalence, the first column in the columns clause is the one that’s kept.

    • Parameters:

      only_synonyms – when True, limit the removal of columns to those which have the same name as the equivalent. Otherwise, all columns that are equivalent to another are removed.

  • method sqlalchemy.sql.expression.Select.replace_selectable(old: FromClause, alias: Alias) → SelfSelectable

    inherited from the Selectable.replace_selectable() method of Selectable

    Replace all occurrences of FromClause ‘old’ with the given Alias object, returning a copy of this FromClause.

    Deprecated since version 1.4: The Selectable.replace_selectable() method is deprecated, and will be removed in a future release. Similar functionality is available via the sqlalchemy.sql.visitors module.

  • method sqlalchemy.sql.expression.Select.scalar_subquery() → ScalarSelect[Any]

    inherited from the SelectBase.scalar_subquery() method of SelectBase

    Return a ‘scalar’ representation of this selectable, which can be used as a column expression.

    The returned object is an instance of ScalarSelect.

    Typically, a select statement which has only one column in its columns clause is eligible to be used as a scalar expression. The scalar subquery can then be used in the WHERE clause or columns clause of an enclosing SELECT.

    Note that the scalar subquery differentiates from the FROM-level subquery that can be produced using the SelectBase.subquery() method.

    See also

    Scalar and Correlated Subqueries - in the 2.0 tutorial

  • method sqlalchemy.sql.expression.Select.select(*arg: Any, **kw: Any) → Select

    inherited from the SelectBase.select() method of SelectBase

    Deprecated since version 1.4: The SelectBase.select() method is deprecated and will be removed in a future release; this method implicitly creates a subquery that should be explicit. Please call SelectBase.subquery() first in order to create a subquery, which then can be selected.

  • method sqlalchemy.sql.expression.Select.select_from(*froms: _FromClauseArgument) → SelfSelect

    Return a new select() construct with the given FROM expression(s) merged into its list of FROM objects.

    E.g.:

    1. table1 = table('t1', column('a'))
    2. table2 = table('t2', column('b'))
    3. s = select(table1.c.a).\
    4.     select_from(
    5.         table1.join(table2, table1.c.a==table2.c.b)
    6.     )

    The “from” list is a unique set on the identity of each element, so adding an already present Table or other selectable will have no effect. Passing a Join that refers to an already present Table or other selectable will have the effect of concealing the presence of that selectable as an individual element in the rendered FROM list, instead rendering it into a JOIN clause.

    While the typical purpose of Select.select_from() is to replace the default, derived FROM clause with a join, it can also be called with individual table elements, multiple times if desired, in the case that the FROM clause cannot be fully derived from the columns clause:

    1. select(func.count('*')).select_from(table1)

  • attribute sqlalchemy.sql.expression.Select.selected_columns

    A ColumnCollection representing the columns that this SELECT statement or similar construct returns in its result set, not including TextClause constructs.

    This collection differs from the FromClause.columns collection of a FromClause in that the columns within this collection cannot be directly nested inside another SELECT statement; a subquery must be applied first which provides for the necessary parenthesization required by SQL.

    For a select() construct, the collection here is exactly what would be rendered inside the “SELECT” statement, and the ColumnElement objects are directly present as they were given, e.g.:

    1. col1 = column('q', Integer)
    2. col2 = column('p', Integer)
    3. stmt = select(col1, col2)

    Above, stmt.selected_columns would be a collection that contains the col1 and col2 objects directly. For a statement that is against a Table or other FromClause, the collection will use the ColumnElement objects that are in the FromClause.c collection of the from element.

    Note

    The Select.selected_columns collection does not include expressions established in the columns clause using the text() construct; these are silently omitted from the collection. To use plain textual column expressions inside of a Select construct, use the literal_column() construct.

    New in version 1.4.

  • method sqlalchemy.sql.expression.Select.self_group(against: Optional[OperatorType] = None) → Union[SelectStatementGrouping, Self]

    Apply a ‘grouping’ to this ClauseElement.

    This method is overridden by subclasses to return a “grouping” construct, i.e. parenthesis. In particular it’s used by “binary” expressions to provide a grouping around themselves when placed into a larger expression, as well as by select() constructs when placed into the FROM clause of another select(). (Note that subqueries should be normally created using the Select.alias() method, as many platforms require nested SELECT statements to be named).

    As expressions are composed together, the application of self_group() is automatic - end-user code should never need to use this method directly. Note that SQLAlchemy’s clause constructs take operator precedence into account - so parenthesis might not be needed, for example, in an expression like x OR (y AND z) - AND takes precedence over OR.

    The base self_group() method of ClauseElement just returns self.

  • method sqlalchemy.sql.expression.Select.set_label_style(style: SelectLabelStyle) → SelfGenerativeSelect

    inherited from the GenerativeSelect.set_label_style() method of GenerativeSelect

    Return a new selectable with the specified label style.

    There are three “label styles” available, SelectLabelStyle.LABEL_STYLE_DISAMBIGUATE_ONLY, SelectLabelStyle.LABEL_STYLE_TABLENAME_PLUS_COL, and SelectLabelStyle.LABEL_STYLE_NONE. The default style is SelectLabelStyle.LABEL_STYLE_TABLENAME_PLUS_COL.

    In modern SQLAlchemy, there is not generally a need to change the labeling style, as per-expression labels are more effectively used by making use of the ColumnElement.label() method. In past versions, LABEL_STYLE_TABLENAME_PLUS_COL was used to disambiguate same-named columns from different tables, aliases, or subqueries; the newer LABEL_STYLE_DISAMBIGUATE_ONLY now applies labels only to names that conflict with an existing name so that the impact of this labeling is minimal.

    The rationale for disambiguation is mostly so that all column expressions are available from a given FromClause.c collection when a subquery is created.

    New in version 1.4: - the GenerativeSelect.set_label_style() method replaces the previous combination of .apply_labels(), .with_labels() and use_labels=True methods and/or parameters.

    See also

    LABEL_STYLE_DISAMBIGUATE_ONLY

    LABEL_STYLE_TABLENAME_PLUS_COL

    LABEL_STYLE_NONE

    LABEL_STYLE_DEFAULT

  • method sqlalchemy.sql.expression.Select.slice(start: int, stop: int) → SelfGenerativeSelect

    inherited from the GenerativeSelect.slice() method of GenerativeSelect

    Apply LIMIT / OFFSET to this statement based on a slice.

    The start and stop indices behave like the argument to Python’s built-in range() function. This method provides an alternative to using LIMIT/OFFSET to get a slice of the query.

    For example,

    1. stmt = select(User).order_by(User).id.slice(1, 3)

    renders as

    1. SELECT users.id AS users_id,
    2.        users.name AS users_name
    3. FROM users ORDER BY users.id
    4. LIMIT ? OFFSET ?
    5. (2, 1)

    Note

    The GenerativeSelect.slice() method will replace any clause applied with GenerativeSelect.fetch().

    New in version 1.4: Added the GenerativeSelect.slice() method generalized from the ORM.

    See also

    GenerativeSelect.limit()

    GenerativeSelect.offset()

    GenerativeSelect.fetch()

  • method sqlalchemy.sql.expression.Select.subquery(name: Optional[str] = None) → Subquery

    inherited from the SelectBase.subquery() method of SelectBase

    Return a subquery of this SelectBase.

    A subquery is from a SQL perspective a parenthesized, named construct that can be placed in the FROM clause of another SELECT statement.

    Given a SELECT statement such as:

    1. stmt = select(table.c.id, table.c.name)

    The above statement might look like:

    1. SELECT table.id, table.name FROM table

    The subquery form by itself renders the same way, however when embedded into the FROM clause of another SELECT statement, it becomes a named sub-element:

    1. subq = stmt.subquery()
    2. new_stmt = select(subq)

    The above renders as:

    1. SELECT anon_1.id, anon_1.name
    2. FROM (SELECT table.id, table.name FROM table) AS anon_1

    Historically, SelectBase.subquery() is equivalent to calling the FromClause.alias() method on a FROM object; however, as a SelectBase object is not directly FROM object, the SelectBase.subquery() method provides clearer semantics.

    New in version 1.4.

  • method sqlalchemy.sql.expression.Select.suffix_with(*suffixes: _TextCoercedExpressionArgument[Any], dialect: str = ‘*‘) → SelfHasSuffixes

    inherited from the HasSuffixes.suffix_with() method of HasSuffixes

    Add one or more expressions following the statement as a whole.

    This is used to support backend-specific suffix keywords on certain constructs.

    E.g.:

    1. stmt = select(col1, col2).cte().suffix_with(
    2.     "cycle empno set y_cycle to 1 default 0", dialect="oracle")

    Multiple suffixes can be specified by multiple calls to HasSuffixes.suffix_with().

    • Parameters:

      • *suffixes – textual or ClauseElement construct which will be rendered following the target clause.

      • dialect – Optional string dialect name which will limit rendering of this suffix to only that dialect.

  • method sqlalchemy.sql.expression.Select.union(*other: _SelectStatementForCompoundArgument) → CompoundSelect

    Return a SQL UNION of this select() construct against the given selectables provided as positional arguments.

    • Parameters:

      • *other –

        one or more elements with which to create a UNION.

        Changed in version 1.4.28: multiple elements are now accepted.

      • **kwargs – keyword arguments are forwarded to the constructor for the newly created CompoundSelect object.

  • method sqlalchemy.sql.expression.Select.union_all(*other: _SelectStatementForCompoundArgument) → CompoundSelect

    Return a SQL UNION ALL of this select() construct against the given selectables provided as positional arguments.

    • Parameters:

      • *other –

        one or more elements with which to create a UNION.

        Changed in version 1.4.28: multiple elements are now accepted.

      • **kwargs – keyword arguments are forwarded to the constructor for the newly created CompoundSelect object.

  • method sqlalchemy.sql.expression.Select.where(*whereclause: _ColumnExpressionArgument[bool]) → SelfSelect

    Return a new select() construct with the given expression added to its WHERE clause, joined to the existing clause via AND, if any.

  • attribute sqlalchemy.sql.expression.Select.whereclause

    Return the completed WHERE clause for this Select statement.

    This assembles the current collection of WHERE criteria into a single BooleanClauseList construct.

    New in version 1.4.

  • method sqlalchemy.sql.expression.Select.with_for_update(*, nowait: bool = False, read: bool = False, of: Optional[_ForUpdateOfArgument] = None, skip_locked: bool = False, key_share: bool = False) → SelfGenerativeSelect

    inherited from the GenerativeSelect.with_for_update() method of GenerativeSelect

    Specify a FOR UPDATE clause for this GenerativeSelect.

    E.g.:

    1. stmt = select(table).with_for_update(nowait=True)

    On a database like PostgreSQL or Oracle, the above would render a statement like:

    1. SELECT table.a, table.b FROM table FOR UPDATE NOWAIT

    on other backends, the nowait option is ignored and instead would produce:

    1. SELECT table.a, table.b FROM table FOR UPDATE

    When called with no arguments, the statement will render with the suffix FOR UPDATE. Additional arguments can then be provided which allow for common database-specific variants.

    • Parameters:

      • nowait – boolean; will render FOR UPDATE NOWAIT on Oracle and PostgreSQL dialects.

      • read – boolean; will render LOCK IN SHARE MODE on MySQL, FOR SHARE on PostgreSQL. On PostgreSQL, when combined with nowait, will render FOR SHARE NOWAIT.

      • of – SQL expression or list of SQL expression elements, (typically Column objects or a compatible expression, for some backends may also be a table expression) which will render into a FOR UPDATE OF clause; supported by PostgreSQL, Oracle, some MySQL versions and possibly others. May render as a table or as a column depending on backend.

      • skip_locked – boolean, will render FOR UPDATE SKIP LOCKED on Oracle and PostgreSQL dialects or FOR SHARE SKIP LOCKED if read=True is also specified.

      • key_share – boolean, will render FOR NO KEY UPDATE, or if combined with read=True will render FOR KEY SHARE, on the PostgreSQL dialect.

  • method sqlalchemy.sql.expression.Select.with_hint(selectable: _FromClauseArgument, text: str, dialect_name: str = ‘*‘) → SelfHasHints

    inherited from the HasHints.with_hint() method of HasHints

    Add an indexing or other executional context hint for the given selectable to this Select or other selectable object.

    The text of the hint is rendered in the appropriate location for the database backend in use, relative to the given Table or Alias passed as the selectable argument. The dialect implementation typically uses Python string substitution syntax with the token %(name)s to render the name of the table or alias. E.g. when using Oracle, the following:

    1. select(mytable).\
    2.     with_hint(mytable, "index(%(name)s ix_mytable)")

    Would render SQL as:

    1. select /*+ index(mytable ix_mytable) */ ... from mytable

    The dialect_name option will limit the rendering of a particular hint to a particular backend. Such as, to add hints for both Oracle and Sybase simultaneously:

    1. select(mytable).\
    2.     with_hint(mytable, "index(%(name)s ix_mytable)", 'oracle').\
    3.     with_hint(mytable, "WITH INDEX ix_mytable", 'mssql')

    See also

    Select.with_statement_hint()

  • method sqlalchemy.sql.expression.Select.with_only_columns(*entities: _ColumnsClauseArgument[Any], maintain_column_froms: bool = False, **_Select\_kw: Any_) → Select[Any]

    Return a new select() construct with its columns clause replaced with the given entities.

    By default, this method is exactly equivalent to as if the original select() had been called with the given entities. E.g. a statement:

    1. s = select(table1.c.a, table1.c.b)
    2. s = s.with_only_columns(table1.c.b)

    should be exactly equivalent to:

    1. s = select(table1.c.b)

    In this mode of operation, Select.with_only_columns() will also dynamically alter the FROM clause of the statement if it is not explicitly stated. To maintain the existing set of FROMs including those implied by the current columns clause, add the Select.with_only_columns.maintain_column_froms parameter:

    1. s = select(table1.c.a, table2.c.b)
    2. s = s.with_only_columns(table1.c.a, maintain_column_froms=True)

    The above parameter performs a transfer of the effective FROMs in the columns collection to the Select.select_from() method, as though the following were invoked:

    1. s = select(table1.c.a, table2.c.b)
    2. s = s.select_from(table1, table2).with_only_columns(table1.c.a)

    The Select.with_only_columns.maintain_column_froms parameter makes use of the Select.columns_clause_froms collection and performs an operation equivalent to the following:

    1. s = select(table1.c.a, table2.c.b)
    2. s = s.select_from(*s.columns_clause_froms).with_only_columns(table1.c.a)

    • Parameters:

      • *entities – column expressions to be used.

      • maintain_column_froms –

        boolean parameter that will ensure the FROM list implied from the current columns clause will be transferred to the Select.select_from() method first.

        New in version 1.4.23.

  • method sqlalchemy.sql.expression.Select.with_statement_hint(text: str, dialect_name: str = ‘*‘) → SelfHasHints

    inherited from the HasHints.with_statement_hint() method of HasHints

    Add a statement hint to this Select or other selectable object.

    This method is similar to Select.with_hint() except that it does not require an individual table, and instead applies to the statement as a whole.

    Hints here are specific to the backend database and may include directives such as isolation levels, file directives, fetch directives, etc.

    New in version 1.0.0.

    See also

    Select.with_hint()

    Select.prefix_with() - generic SELECT prefixing which also can suit some database-specific HINT syntaxes such as MySQL optimizer hints

class sqlalchemy.sql.expression.Selectable

Mark a class as being selectable.

Members

corresponding_column(), exported_columns, inherit_cache, is_derived_from(), lateral(), replace_selectable()

Class signature

class sqlalchemy.sql.expression.Selectable (sqlalchemy.sql.expression.ReturnsRows)

  1. See also
  3. [Selectable.exported\_columns](#sqlalchemy.sql.expression.Selectable.exported_columns "sqlalchemy.sql.expression.Selectable.exported_columns") - the [ColumnCollection]($aafca12b71ff5dd3.md#sqlalchemy.sql.expression.ColumnCollection "sqlalchemy.sql.expression.ColumnCollection") that is used for the operation.
  5. [ColumnCollection.corresponding\_column()]($aafca12b71ff5dd3.md#sqlalchemy.sql.expression.ColumnCollection.corresponding_column "sqlalchemy.sql.expression.ColumnCollection.corresponding_column") - implementation method.

class sqlalchemy.sql.expression.SelectBase

Base class for SELECT statements.

This includes Select, CompoundSelect and TextualSelect.

Members

add_cte(), alias(), as_scalar(), c, corresponding_column(), cte(), exists(), exported_columns, get_label_style(), inherit_cache, is_derived_from(), label(), lateral(), replace_selectable(), scalar_subquery(), select(), selected_columns, set_label_style(), subquery()

Class signature

class sqlalchemy.sql.expression.SelectBase (sqlalchemy.sql.roles.SelectStatementRole, sqlalchemy.sql.roles.DMLSelectRole, sqlalchemy.sql.roles.CompoundElementRole, sqlalchemy.sql.roles.InElementRole, sqlalchemy.sql.expression.HasCTE, sqlalchemy.sql.annotation.SupportsCloneAnnotations, sqlalchemy.sql.expression.Selectable)

  • method sqlalchemy.sql.expression.SelectBase.add_cte(*ctes: CTE, nest_here: bool = False) → SelfHasCTE

    inherited from the HasCTE.add_cte() method of HasCTE

    Add one or more CTE constructs to this statement.

    This method will associate the given CTE constructs with the parent statement such that they will each be unconditionally rendered in the WITH clause of the final statement, even if not referenced elsewhere within the statement or any sub-selects.

    The optional HasCTE.add_cte.nest_here parameter when set to True will have the effect that each given CTE will render in a WITH clause rendered directly along with this statement, rather than being moved to the top of the ultimate rendered statement, even if this statement is rendered as a subquery within a larger statement.

    This method has two general uses. One is to embed CTE statements that serve some purpose without being referenced explicitly, such as the use case of embedding a DML statement such as an INSERT or UPDATE as a CTE inline with a primary statement that may draw from its results indirectly. The other is to provide control over the exact placement of a particular series of CTE constructs that should remain rendered directly in terms of a particular statement that may be nested in a larger statement.

    E.g.:

    1. from sqlalchemy import table, column, select
    2. t = table('t', column('c1'), column('c2'))
    4. ins = t.insert().values({"c1": "x", "c2": "y"}).cte()
    6. stmt = select(t).add_cte(ins)

    Would render:

    1. WITH anon_1 AS
    2. (INSERT INTO t (c1, c2) VALUES (:param_1, :param_2))
    3. SELECT t.c1, t.c2
    4. FROM t

    Above, the “anon_1” CTE is not referred towards in the SELECT statement, however still accomplishes the task of running an INSERT statement.

    Similarly in a DML-related context, using the PostgreSQL Insert construct to generate an “upsert”:

    1. from sqlalchemy import table, column
    2. from sqlalchemy.dialects.postgresql import insert
    4. t = table("t", column("c1"), column("c2"))
    6. delete_statement_cte = (
    7.     t.delete().where(t.c.c1 < 1).cte("deletions")
    8. )
    10. insert_stmt = insert(t).values({"c1": 1, "c2": 2})
    11. update_statement = insert_stmt.on_conflict_do_update(
    12.     index_elements=[t.c.c1],
    13.     set_={
    14.         "c1": insert_stmt.excluded.c1,
    15.         "c2": insert_stmt.excluded.c2,
    16.     },
    17. ).add_cte(delete_statement_cte)
    19. print(update_statement)

    The above statement renders as:

    1. WITH deletions AS
    2. (DELETE FROM t WHERE t.c1 < %(c1_1)s)
    3. INSERT INTO t (c1, c2) VALUES (%(c1)s, %(c2)s)
    4. ON CONFLICT (c1) DO UPDATE SET c1 = excluded.c1, c2 = excluded.c2

    New in version 1.4.21.

    • Parameters:

      • *ctes –

        zero or more CTE constructs.

        Changed in version 2.0: Multiple CTE instances are accepted

      • nest_here –

        if True, the given CTE or CTEs will be rendered as though they specified the HasCTE.cte.nesting flag to True when they were added to this HasCTE. Assuming the given CTEs are not referenced in an outer-enclosing statement as well, the CTEs given should render at the level of this statement when this flag is given.

        New in version 2.0.

        See also

        HasCTE.cte.nesting

  1. See also
  3. [Selectable.exported\_columns](#sqlalchemy.sql.expression.Selectable.exported_columns "sqlalchemy.sql.expression.Selectable.exported_columns") - the [ColumnCollection]($aafca12b71ff5dd3.md#sqlalchemy.sql.expression.ColumnCollection "sqlalchemy.sql.expression.ColumnCollection") that is used for the operation.
  5. [ColumnCollection.corresponding\_column()]($aafca12b71ff5dd3.md#sqlalchemy.sql.expression.ColumnCollection.corresponding_column "sqlalchemy.sql.expression.ColumnCollection.corresponding_column") - implementation method.

  • method sqlalchemy.sql.expression.SelectBase.cte(name: Optional[str] = None, recursive: bool = False, nesting: bool = False) → CTE

    inherited from the HasCTE.cte() method of HasCTE

    Return a new CTE, or Common Table Expression instance.

    Common table expressions are a SQL standard whereby SELECT statements can draw upon secondary statements specified along with the primary statement, using a clause called “WITH”. Special semantics regarding UNION can also be employed to allow “recursive” queries, where a SELECT statement can draw upon the set of rows that have previously been selected.

    CTEs can also be applied to DML constructs UPDATE, INSERT and DELETE on some databases, both as a source of CTE rows when combined with RETURNING, as well as a consumer of CTE rows.

    Changed in version 1.1: Added support for UPDATE/INSERT/DELETE as CTE, CTEs added to UPDATE/INSERT/DELETE.

    SQLAlchemy detects CTE objects, which are treated similarly to Alias objects, as special elements to be delivered to the FROM clause of the statement as well as to a WITH clause at the top of the statement.

    For special prefixes such as PostgreSQL “MATERIALIZED” and “NOT MATERIALIZED”, the CTE.prefix_with() method may be used to establish these.

    Changed in version 1.3.13: Added support for prefixes. In particular - MATERIALIZED and NOT MATERIALIZED.

    • Parameters:

      • name – name given to the common table expression. Like FromClause.alias(), the name can be left as None in which case an anonymous symbol will be used at query compile time.

      • recursive – if True, will render WITH RECURSIVE. A recursive common table expression is intended to be used in conjunction with UNION ALL in order to derive rows from those already selected.

      • nesting –

        if True, will render the CTE locally to the statement in which it is referenced. For more complex scenarios, the HasCTE.add_cte() method using the HasCTE.add_cte.nest_here parameter may also be used to more carefully control the exact placement of a particular CTE.

        New in version 1.4.24.

        See also

        HasCTE.add_cte()

  1. The following examples include two from PostgreSQL’s documentation at [https://www.postgresql.org/docs/current/static/queries-with.html](https://www.postgresql.org/docs/current/static/queries-with.html), as well as additional examples.
  3. Example 1, non recursive:
  5. ```
  6. from sqlalchemy import (Table, Column, String, Integer,
  7.                         MetaData, select, func)
  9. metadata = MetaData()
  11. orders = Table('orders', metadata,
  12.     Column('region', String),
  13.     Column('amount', Integer),
  14.     Column('product', String),
  15.     Column('quantity', Integer)
  16. )
  18. regional_sales = select(
  19.                     orders.c.region,
  20.                     func.sum(orders.c.amount).label('total_sales')
  21.                 ).group_by(orders.c.region).cte("regional_sales")
  24. top_regions = select(regional_sales.c.region).\
  25.         where(
  26.             regional_sales.c.total_sales >
  27.             select(
  28.                 func.sum(regional_sales.c.total_sales) / 10
  29.             )
  30.         ).cte("top_regions")
  32. statement = select(
  33.             orders.c.region,
  34.             orders.c.product,
  35.             func.sum(orders.c.quantity).label("product_units"),
  36.             func.sum(orders.c.amount).label("product_sales")
  37.     ).where(orders.c.region.in_(
  38.         select(top_regions.c.region)
  39.     )).group_by(orders.c.region, orders.c.product)
  41. result = conn.execute(statement).fetchall()
  42. ```
  44. Example 2, WITH RECURSIVE:
  46. ```
  47. from sqlalchemy import (Table, Column, String, Integer,
  48.                         MetaData, select, func)
  50. metadata = MetaData()
  52. parts = Table('parts', metadata,
  53.     Column('part', String),
  54.     Column('sub_part', String),
  55.     Column('quantity', Integer),
  56. )
  58. included_parts = select(\
  59.     parts.c.sub_part, parts.c.part, parts.c.quantity\
  60.     ).\
  61.     where(parts.c.part=='our part').\
  62.     cte(recursive=True)
  65. incl_alias = included_parts.alias()
  66. parts_alias = parts.alias()
  67. included_parts = included_parts.union_all(
  68.     select(
  69.         parts_alias.c.sub_part,
  70.         parts_alias.c.part,
  71.         parts_alias.c.quantity
  72.     ).\
  73.     where(parts_alias.c.part==incl_alias.c.sub_part)
  74. )
  76. statement = select(
  77.             included_parts.c.sub_part,
  78.             func.sum(included_parts.c.quantity).
  79.               label('total_quantity')
  80.         ).\
  81.         group_by(included_parts.c.sub_part)
  83. result = conn.execute(statement).fetchall()
  84. ```
  86. Example 3, an upsert using UPDATE and INSERT with CTEs:
  88. ```
  89. from datetime import date
  90. from sqlalchemy import (MetaData, Table, Column, Integer,
  91.                         Date, select, literal, and_, exists)
  93. metadata = MetaData()
  95. visitors = Table('visitors', metadata,
  96.     Column('product_id', Integer, primary_key=True),
  97.     Column('date', Date, primary_key=True),
  98.     Column('count', Integer),
  99. )
  101. # add 5 visitors for the product_id == 1
  102. product_id = 1
  103. day = date.today()
  104. count = 5
  106. update_cte = (
  107.     visitors.update()
  108.     .where(and_(visitors.c.product_id == product_id,
  109.                 visitors.c.date == day))
  110.     .values(count=visitors.c.count + count)
  111.     .returning(literal(1))
  112.     .cte('update_cte')
  113. )
  115. upsert = visitors.insert().from_select(
  116.     [visitors.c.product_id, visitors.c.date, visitors.c.count],
  117.     select(literal(product_id), literal(day), literal(count))
  118.         .where(~exists(update_cte.select()))
  119. )
  121. connection.execute(upsert)
  122. ```
  124. Example 4, Nesting CTE (SQLAlchemy 1.4.24 and above):
  126. ```
  127. value_a = select(
  128.     literal("root").label("n")
  129. ).cte("value_a")
  131. # A nested CTE with the same name as the root one
  132. value_a_nested = select(
  133.     literal("nesting").label("n")
  134. ).cte("value_a", nesting=True)
  136. # Nesting CTEs takes ascendency locally
  137. # over the CTEs at a higher level
  138. value_b = select(value_a_nested.c.n).cte("value_b")
  140. value_ab = select(value_a.c.n.label("a"), value_b.c.n.label("b"))
  141. ```
  143. The above query will render the second CTE nested inside the first, shown with inline parameters below as:
  145. ```
  146. WITH
  147.     value_a AS
  148.         (SELECT 'root' AS n),
  149.     value_b AS
  150.         (WITH value_a AS
  151.             (SELECT 'nesting' AS n)
  152.         SELECT value_a.n AS n FROM value_a)
  153. SELECT value_a.n AS a, value_b.n AS b
  154. FROM value_a, value_b
  155. ```
  157. The same CTE can be set up using the [HasCTE.add\_cte()](#sqlalchemy.sql.expression.HasCTE.add_cte "sqlalchemy.sql.expression.HasCTE.add_cte") method as follows (SQLAlchemy 2.0 and above):
  159. ```
  160. value_a = select(
  161.     literal("root").label("n")
  162. ).cte("value_a")
  164. # A nested CTE with the same name as the root one
  165. value_a_nested = select(
  166.     literal("nesting").label("n")
  167. ).cte("value_a")
  169. # Nesting CTEs takes ascendency locally
  170. # over the CTEs at a higher level
  171. value_b = (
  172.     select(value_a_nested.c.n).
  173.     add_cte(value_a_nested, nest_here=True).
  174.     cte("value_b")
  175. )
  177. value_ab = select(value_a.c.n.label("a"), value_b.c.n.label("b"))
  178. ```
  180. Example 5, Non-Linear CTE (SQLAlchemy 1.4.28 and above):
  182. ```
  183. edge = Table(
  184.     "edge",
  185.     metadata,
  186.     Column("id", Integer, primary_key=True),
  187.     Column("left", Integer),
  188.     Column("right", Integer),
  189. )
  191. root_node = select(literal(1).label("node")).cte(
  192.     "nodes", recursive=True
  193. )
  195. left_edge = select(edge.c.left).join(
  196.     root_node, edge.c.right == root_node.c.node
  197. )
  198. right_edge = select(edge.c.right).join(
  199.     root_node, edge.c.left == root_node.c.node
  200. )
  202. subgraph_cte = root_node.union(left_edge, right_edge)
  204. subgraph = select(subgraph_cte)
  205. ```
  207. The above query will render 2 UNIONs inside the recursive CTE:
  209. ```
  210. WITH RECURSIVE nodes(node) AS (
  211.         SELECT 1 AS node
  212.     UNION
  213.         SELECT edge."left" AS "left"
  214.         FROM edge JOIN nodes ON edge."right" = nodes.node
  215.     UNION
  216.         SELECT edge."right" AS "right"
  217.         FROM edge JOIN nodes ON edge."left" = nodes.node
  218. )
  219. SELECT nodes.node FROM nodes
  220. ```
  222. See also
  224. [Query.cte()]($3d0cc000ec6c7150.md#sqlalchemy.orm.Query.cte "sqlalchemy.orm.Query.cte") - ORM version of [HasCTE.cte()](#sqlalchemy.sql.expression.HasCTE.cte "sqlalchemy.sql.expression.HasCTE.cte").

class sqlalchemy.sql.expression.Subquery

Represent a subquery of a SELECT.

A Subquery is created by invoking the SelectBase.subquery() method, or for convenience the SelectBase.alias() method, on any SelectBase subclass which includes Select, CompoundSelect, and TextualSelect. As rendered in a FROM clause, it represents the body of the SELECT statement inside of parenthesis, followed by the usual “AS <somename>” that defines all “alias” objects.

The Subquery object is very similar to the Alias object and can be used in an equivalent way. The difference between Alias and Subquery is that Alias always contains a FromClause object whereas Subquery always contains a SelectBase object.

New in version 1.4: The Subquery class was added which now serves the purpose of providing an aliased version of a SELECT statement.

Members

as_scalar(), inherit_cache

Class signature

class sqlalchemy.sql.expression.Subquery (sqlalchemy.sql.expression.AliasedReturnsRows)

  • method sqlalchemy.sql.expression.Subquery.as_scalar() → ScalarSelect[Any]

    Deprecated since version 1.4: The Subquery.as_scalar() method, which was previously Alias.as_scalar() prior to version 1.4, is deprecated and will be removed in a future release; Please use the Select.scalar_subquery() method of the select() construct before constructing a subquery object, or with the ORM use the Query.scalar_subquery() method.

  • attribute sqlalchemy.sql.expression.Subquery.inherit_cache: Optional[bool] = True

    Indicate if this HasCacheKey instance should make use of the cache key generation scheme used by its immediate superclass.

    The attribute defaults to None, which indicates that a construct has not yet taken into account whether or not its appropriate for it to participate in caching; this is functionally equivalent to setting the value to False, except that a warning is also emitted.

    This flag can be set to True on a particular class, if the SQL that corresponds to the object does not change based on attributes which are local to this class, and not its superclass.

    See also

    Enabling Caching Support for Custom Constructs - General guideslines for setting the HasCacheKey.inherit_cache attribute for third-party or user defined SQL constructs.

class sqlalchemy.sql.expression.TableClause

Represents a minimal “table” construct.

This is a lightweight table object that has only a name, a collection of columns, which are typically produced by the column() function, and a schema:

  1. from sqlalchemy import table, column
  3. user = table("user",
  4.         column("id"),
  5.         column("name"),
  6.         column("description"),
  7. )

The TableClause construct serves as the base for the more commonly used Table object, providing the usual set of FromClause services including the .c. collection and statement generation methods.

It does not provide all the additional schema-level services of Table, including constraints, references to other tables, or support for MetaData-level services. It’s useful on its own as an ad-hoc construct used to generate quick SQL statements when a more fully fledged Table is not on hand.

Members

alias(), c, columns, compare(), compile(), corresponding_column(), delete(), description, entity_namespace, exported_columns, foreign_keys, get_children(), implicit_returning, inherit_cache, insert(), is_derived_from(), join(), lateral(), outerjoin(), params(), primary_key, replace_selectable(), schema, select(), self_group(), table_valued(), tablesample(), unique_params(), update()

Class signature

class sqlalchemy.sql.expression.TableClause (sqlalchemy.sql.roles.DMLTableRole, sqlalchemy.sql.expression.Immutable, sqlalchemy.sql.expression.NamedFromClause)

  • method sqlalchemy.sql.expression.TableClause.alias(name: Optional[str] = None, flat: bool = False) → NamedFromClause

    inherited from the FromClause.alias() method of FromClause

    Return an alias of this FromClause.

    E.g.:

    1. a2 = some_table.alias('a2')

    The above code creates an Alias object which can be used as a FROM clause in any SELECT statement.

    See also

    Using Aliases

    alias()

  • attribute sqlalchemy.sql.expression.TableClause.c

    inherited from the FromClause.c attribute of FromClause

    A synonym for FromClause.columns

  • attribute sqlalchemy.sql.expression.TableClause.columns

    inherited from the FromClause.columns attribute of FromClause

    A named-based collection of ColumnElement objects maintained by this FromClause.

    The columns, or c collection, is the gateway to the construction of SQL expressions using table-bound or other selectable-bound columns:

    1. select(mytable).where(mytable.c.somecolumn == 5)

  • method sqlalchemy.sql.expression.TableClause.compare(other: ClauseElement, **kw: Any) → bool

    inherited from the ClauseElement.compare() method of ClauseElement

    Compare this ClauseElement to the given ClauseElement.

    Subclasses should override the default behavior, which is a straight identity comparison.

    **kw are arguments consumed by subclass compare() methods and may be used to modify the criteria for comparison (see ColumnElement).

  • method sqlalchemy.sql.expression.TableClause.compile(bind: Optional[Union[Engine, Connection]] = None, dialect: Optional[Dialect] = None, **kw: Any) → Compiled

    inherited from the CompilerElement.compile() method of CompilerElement

    Compile this SQL expression.

    The return value is a Compiled object. Calling str() or unicode() on the returned value will yield a string representation of the result. The Compiled object also can return a dictionary of bind parameter names and values using the params accessor.

    • Parameters:

      • bind – An Connection or Engine which can provide a Dialect in order to generate a Compiled object. If the bind and dialect parameters are both omitted, a default SQL compiler is used.

      • column_keys – Used for INSERT and UPDATE statements, a list of column names which should be present in the VALUES clause of the compiled statement. If None, all columns from the target table object are rendered.

      • dialect – A Dialect instance which can generate a Compiled object. This argument takes precedence over the bind argument.

      • compile_kwargs –

        optional dictionary of additional parameters that will be passed through to the compiler within all “visit” methods. This allows any custom flag to be passed through to a custom compilation construct, for example. It is also used for the case of passing the literal_binds flag through:

        1. from sqlalchemy.sql import table, column, select
        3. t = table('t', column('x'))
        5. s = select(t).where(t.c.x == 5)
        7. print(s.compile(compile_kwargs={"literal_binds": True}))

        New in version 0.9.0.

  1. See also
  3. [How do I render SQL expressions as strings, possibly with bound parameters inlined?]($e9fd44a49fe37bbb.md#faq-sql-expression-string)
  1. See also
  3. [Selectable.exported\_columns](#sqlalchemy.sql.expression.Selectable.exported_columns "sqlalchemy.sql.expression.Selectable.exported_columns") - the [ColumnCollection]($aafca12b71ff5dd3.md#sqlalchemy.sql.expression.ColumnCollection "sqlalchemy.sql.expression.ColumnCollection") that is used for the operation.
  5. [ColumnCollection.corresponding\_column()]($aafca12b71ff5dd3.md#sqlalchemy.sql.expression.ColumnCollection.corresponding_column "sqlalchemy.sql.expression.ColumnCollection.corresponding_column") - implementation method.
  1. See also
  3. [join()](#sqlalchemy.sql.expression.join "sqlalchemy.sql.expression.join") - standalone function
  5. [Join](#sqlalchemy.sql.expression.Join "sqlalchemy.sql.expression.Join") - the type of object produced

  • method sqlalchemy.sql.expression.TableClause.lateral(name: Optional[str] = None) → LateralFromClause

    inherited from the Selectable.lateral() method of Selectable

    Return a LATERAL alias of this Selectable.

    The return value is the Lateral construct also provided by the top-level lateral() function.

    New in version 1.1.

    See also

    LATERAL correlation - overview of usage.

  • method sqlalchemy.sql.expression.TableClause.outerjoin(right: _FromClauseArgument, onclause: Optional[_ColumnExpressionArgument[bool]] = None, full: bool = False) → Join

    inherited from the FromClause.outerjoin() method of FromClause

    Return a Join from this FromClause to another FromClause, with the “isouter” flag set to True.

    E.g.:

    1. from sqlalchemy import outerjoin
    3. j = user_table.outerjoin(address_table,
    4.                 user_table.c.id == address_table.c.user_id)

    The above is equivalent to:

    1. j = user_table.join(
    2.     address_table,
    3.     user_table.c.id == address_table.c.user_id,
    4.     isouter=True)

    • Parameters:

      • right – the right side of the join; this is any FromClause object such as a Table object, and may also be a selectable-compatible object such as an ORM-mapped class.

      • onclause – a SQL expression representing the ON clause of the join. If left at None, FromClause.join() will attempt to join the two tables based on a foreign key relationship.

      • full –

        if True, render a FULL OUTER JOIN, instead of LEFT OUTER JOIN.

        New in version 1.1.

  1. See also
  3. [FromClause.join()](#sqlalchemy.sql.expression.FromClause.join "sqlalchemy.sql.expression.FromClause.join")
  5. [Join](#sqlalchemy.sql.expression.Join "sqlalchemy.sql.expression.Join")

class sqlalchemy.sql.expression.TableSample

Represent a TABLESAMPLE clause.

This object is constructed from the tablesample() module level function as well as the FromClause.tablesample() method available on all FromClause subclasses.

New in version 1.1.

See also

tablesample()

Class signature

class sqlalchemy.sql.expression.TableSample (sqlalchemy.sql.expression.FromClauseAlias)

class sqlalchemy.sql.expression.TableValuedAlias

An alias against a “table valued” SQL function.

This construct provides for a SQL function that returns columns to be used in the FROM clause of a SELECT statement. The object is generated using the FunctionElement.table_valued() method, e.g.:

  1. >>> from sqlalchemy import select, func
  2. >>> fn = func.json_array_elements_text('["one", "two", "three"]').table_valued("value")
  3. >>> print(select(fn.c.value))
  4. SELECT anon_1.value
  5. FROM json_array_elements_text(:json_array_elements_text_1) AS anon_1

New in version 1.4.0b2.

See also

Table-Valued Functions - in the SQLAlchemy Unified Tutorial

Members

alias(), column, lateral(), render_derived()

Class signature

class sqlalchemy.sql.expression.TableValuedAlias (sqlalchemy.sql.expression.LateralFromClause, sqlalchemy.sql.expression.Alias)

  • method sqlalchemy.sql.expression.TableValuedAlias.alias(name: Optional[str] = None, flat: bool = False) → TableValuedAlias

    Return a new alias of this TableValuedAlias.

    This creates a distinct FROM object that will be distinguished from the original one when used in a SQL statement.

  • attribute sqlalchemy.sql.expression.TableValuedAlias.column

    Return a column expression representing this TableValuedAlias.

    This accessor is used to implement the FunctionElement.column_valued() method. See that method for further details.

    E.g.:

    1. >>> print(select(func.some_func().table_valued("value").column))
    2. SELECT anon_1 FROM some_func() AS anon_1

    See also

    FunctionElement.column_valued()

  • method sqlalchemy.sql.expression.TableValuedAlias.lateral(name: Optional[str] = None) → LateralFromClause

    Return a new TableValuedAlias with the lateral flag set, so that it renders as LATERAL.

    See also

    lateral()

  • method sqlalchemy.sql.expression.TableValuedAlias.render_derived(name: Optional[str] = None, with_types: bool = False) → TableValuedAlias

    Apply “render derived” to this TableValuedAlias.

    This has the effect of the individual column names listed out after the alias name in the “AS” sequence, e.g.:

    1. >>> print(
    2. ...     select(
    3. ...         func.unnest(array(["one", "two", "three"])).
    4.             table_valued("x", with_ordinality="o").render_derived()
    5. ...     )
    6. ... )
    7. SELECT anon_1.x, anon_1.o
    8. FROM unnest(ARRAY[%(param_1)s, %(param_2)s, %(param_3)s]) WITH ORDINALITY AS anon_1(x, o)

    The with_types keyword will render column types inline within the alias expression (this syntax currently applies to the PostgreSQL database):

    1. >>> print(
    2. ...     select(
    3. ...         func.json_to_recordset(
    4. ...             '[{"a":1,"b":"foo"},{"a":"2","c":"bar"}]'
    5. ...         )
    6. ...         .table_valued(column("a", Integer), column("b", String))
    7. ...         .render_derived(with_types=True)
    8. ...     )
    9. ... )
    10. SELECT anon_1.a, anon_1.b FROM json_to_recordset(:json_to_recordset_1)
    11. AS anon_1(a INTEGER, b VARCHAR)

    • Parameters:

      • name – optional string name that will be applied to the alias generated. If left as None, a unique anonymizing name will be used.

      • with_types – if True, the derived columns will include the datatype specification with each column. This is a special syntax currently known to be required by PostgreSQL for some SQL functions.

class sqlalchemy.sql.expression.TextualSelect

Wrap a TextClause construct within a SelectBase interface.

This allows the TextClause object to gain a .c collection and other FROM-like capabilities such as FromClause.alias(), SelectBase.cte(), etc.

The TextualSelect construct is produced via the TextClause.columns() method - see that method for details.

Changed in version 1.4: the TextualSelect class was renamed from TextAsFrom, to more correctly suit its role as a SELECT-oriented object and not a FROM clause.

See also

text()

TextClause.columns() - primary creation interface.

Members

add_cte(), alias(), as_scalar(), c, compare(), compile(), corresponding_column(), cte(), execution_options(), exists(), exported_columns, get_children(), get_execution_options(), get_label_style(), inherit_cache, is_derived_from(), label(), lateral(), options(), params(), replace_selectable(), scalar_subquery(), select(), selected_columns, self_group(), set_label_style(), subquery(), unique_params()

Class signature

class sqlalchemy.sql.expression.TextualSelect (sqlalchemy.sql.expression.SelectBase, sqlalchemy.sql.expression.Executable, sqlalchemy.sql.expression.Generative)

  • method sqlalchemy.sql.expression.TextualSelect.add_cte(*ctes: CTE, nest_here: bool = False) → SelfHasCTE

    inherited from the HasCTE.add_cte() method of HasCTE

    Add one or more CTE constructs to this statement.

    This method will associate the given CTE constructs with the parent statement such that they will each be unconditionally rendered in the WITH clause of the final statement, even if not referenced elsewhere within the statement or any sub-selects.

    The optional HasCTE.add_cte.nest_here parameter when set to True will have the effect that each given CTE will render in a WITH clause rendered directly along with this statement, rather than being moved to the top of the ultimate rendered statement, even if this statement is rendered as a subquery within a larger statement.

    This method has two general uses. One is to embed CTE statements that serve some purpose without being referenced explicitly, such as the use case of embedding a DML statement such as an INSERT or UPDATE as a CTE inline with a primary statement that may draw from its results indirectly. The other is to provide control over the exact placement of a particular series of CTE constructs that should remain rendered directly in terms of a particular statement that may be nested in a larger statement.

    E.g.:

    1. from sqlalchemy import table, column, select
    2. t = table('t', column('c1'), column('c2'))
    4. ins = t.insert().values({"c1": "x", "c2": "y"}).cte()
    6. stmt = select(t).add_cte(ins)

    Would render:

    1. WITH anon_1 AS
    2. (INSERT INTO t (c1, c2) VALUES (:param_1, :param_2))
    3. SELECT t.c1, t.c2
    4. FROM t

    Above, the “anon_1” CTE is not referred towards in the SELECT statement, however still accomplishes the task of running an INSERT statement.

    Similarly in a DML-related context, using the PostgreSQL Insert construct to generate an “upsert”:

    1. from sqlalchemy import table, column
    2. from sqlalchemy.dialects.postgresql import insert
    4. t = table("t", column("c1"), column("c2"))
    6. delete_statement_cte = (
    7.     t.delete().where(t.c.c1 < 1).cte("deletions")
    8. )
    10. insert_stmt = insert(t).values({"c1": 1, "c2": 2})
    11. update_statement = insert_stmt.on_conflict_do_update(
    12.     index_elements=[t.c.c1],
    13.     set_={
    14.         "c1": insert_stmt.excluded.c1,
    15.         "c2": insert_stmt.excluded.c2,
    16.     },
    17. ).add_cte(delete_statement_cte)
    19. print(update_statement)

    The above statement renders as:

    1. WITH deletions AS
    2. (DELETE FROM t WHERE t.c1 < %(c1_1)s)
    3. INSERT INTO t (c1, c2) VALUES (%(c1)s, %(c2)s)
    4. ON CONFLICT (c1) DO UPDATE SET c1 = excluded.c1, c2 = excluded.c2

    New in version 1.4.21.

    • Parameters:

      • *ctes –

        zero or more CTE constructs.

        Changed in version 2.0: Multiple CTE instances are accepted

      • nest_here –

        if True, the given CTE or CTEs will be rendered as though they specified the HasCTE.cte.nesting flag to True when they were added to this HasCTE. Assuming the given CTEs are not referenced in an outer-enclosing statement as well, the CTEs given should render at the level of this statement when this flag is given.

        New in version 2.0.

        See also

        HasCTE.cte.nesting

  1. See also
  3. [How do I render SQL expressions as strings, possibly with bound parameters inlined?]($e9fd44a49fe37bbb.md#faq-sql-expression-string)
  1. See also
  3. [Selectable.exported\_columns](#sqlalchemy.sql.expression.Selectable.exported_columns "sqlalchemy.sql.expression.Selectable.exported_columns") - the [ColumnCollection]($aafca12b71ff5dd3.md#sqlalchemy.sql.expression.ColumnCollection "sqlalchemy.sql.expression.ColumnCollection") that is used for the operation.
  5. [ColumnCollection.corresponding\_column()]($aafca12b71ff5dd3.md#sqlalchemy.sql.expression.ColumnCollection.corresponding_column "sqlalchemy.sql.expression.ColumnCollection.corresponding_column") - implementation method.

  • method sqlalchemy.sql.expression.TextualSelect.cte(name: Optional[str] = None, recursive: bool = False, nesting: bool = False) → CTE

    inherited from the HasCTE.cte() method of HasCTE

    Return a new CTE, or Common Table Expression instance.

    Common table expressions are a SQL standard whereby SELECT statements can draw upon secondary statements specified along with the primary statement, using a clause called “WITH”. Special semantics regarding UNION can also be employed to allow “recursive” queries, where a SELECT statement can draw upon the set of rows that have previously been selected.

    CTEs can also be applied to DML constructs UPDATE, INSERT and DELETE on some databases, both as a source of CTE rows when combined with RETURNING, as well as a consumer of CTE rows.

    Changed in version 1.1: Added support for UPDATE/INSERT/DELETE as CTE, CTEs added to UPDATE/INSERT/DELETE.

    SQLAlchemy detects CTE objects, which are treated similarly to Alias objects, as special elements to be delivered to the FROM clause of the statement as well as to a WITH clause at the top of the statement.

    For special prefixes such as PostgreSQL “MATERIALIZED” and “NOT MATERIALIZED”, the CTE.prefix_with() method may be used to establish these.

    Changed in version 1.3.13: Added support for prefixes. In particular - MATERIALIZED and NOT MATERIALIZED.

    • Parameters:

      • name – name given to the common table expression. Like FromClause.alias(), the name can be left as None in which case an anonymous symbol will be used at query compile time.

      • recursive – if True, will render WITH RECURSIVE. A recursive common table expression is intended to be used in conjunction with UNION ALL in order to derive rows from those already selected.

      • nesting –

        if True, will render the CTE locally to the statement in which it is referenced. For more complex scenarios, the HasCTE.add_cte() method using the HasCTE.add_cte.nest_here parameter may also be used to more carefully control the exact placement of a particular CTE.

        New in version 1.4.24.

        See also

        HasCTE.add_cte()

  1. The following examples include two from PostgreSQL’s documentation at [https://www.postgresql.org/docs/current/static/queries-with.html](https://www.postgresql.org/docs/current/static/queries-with.html), as well as additional examples.
  3. Example 1, non recursive:
  5. ```
  6. from sqlalchemy import (Table, Column, String, Integer,
  7.                         MetaData, select, func)
  9. metadata = MetaData()
  11. orders = Table('orders', metadata,
  12.     Column('region', String),
  13.     Column('amount', Integer),
  14.     Column('product', String),
  15.     Column('quantity', Integer)
  16. )
  18. regional_sales = select(
  19.                     orders.c.region,
  20.                     func.sum(orders.c.amount).label('total_sales')
  21.                 ).group_by(orders.c.region).cte("regional_sales")
  24. top_regions = select(regional_sales.c.region).\
  25.         where(
  26.             regional_sales.c.total_sales >
  27.             select(
  28.                 func.sum(regional_sales.c.total_sales) / 10
  29.             )
  30.         ).cte("top_regions")
  32. statement = select(
  33.             orders.c.region,
  34.             orders.c.product,
  35.             func.sum(orders.c.quantity).label("product_units"),
  36.             func.sum(orders.c.amount).label("product_sales")
  37.     ).where(orders.c.region.in_(
  38.         select(top_regions.c.region)
  39.     )).group_by(orders.c.region, orders.c.product)
  41. result = conn.execute(statement).fetchall()
  42. ```
  44. Example 2, WITH RECURSIVE:
  46. ```
  47. from sqlalchemy import (Table, Column, String, Integer,
  48.                         MetaData, select, func)
  50. metadata = MetaData()
  52. parts = Table('parts', metadata,
  53.     Column('part', String),
  54.     Column('sub_part', String),
  55.     Column('quantity', Integer),
  56. )
  58. included_parts = select(\
  59.     parts.c.sub_part, parts.c.part, parts.c.quantity\
  60.     ).\
  61.     where(parts.c.part=='our part').\
  62.     cte(recursive=True)
  65. incl_alias = included_parts.alias()
  66. parts_alias = parts.alias()
  67. included_parts = included_parts.union_all(
  68.     select(
  69.         parts_alias.c.sub_part,
  70.         parts_alias.c.part,
  71.         parts_alias.c.quantity
  72.     ).\
  73.     where(parts_alias.c.part==incl_alias.c.sub_part)
  74. )
  76. statement = select(
  77.             included_parts.c.sub_part,
  78.             func.sum(included_parts.c.quantity).
  79.               label('total_quantity')
  80.         ).\
  81.         group_by(included_parts.c.sub_part)
  83. result = conn.execute(statement).fetchall()
  84. ```
  86. Example 3, an upsert using UPDATE and INSERT with CTEs:
  88. ```
  89. from datetime import date
  90. from sqlalchemy import (MetaData, Table, Column, Integer,
  91.                         Date, select, literal, and_, exists)
  93. metadata = MetaData()
  95. visitors = Table('visitors', metadata,
  96.     Column('product_id', Integer, primary_key=True),
  97.     Column('date', Date, primary_key=True),
  98.     Column('count', Integer),
  99. )
  101. # add 5 visitors for the product_id == 1
  102. product_id = 1
  103. day = date.today()
  104. count = 5
  106. update_cte = (
  107.     visitors.update()
  108.     .where(and_(visitors.c.product_id == product_id,
  109.                 visitors.c.date == day))
  110.     .values(count=visitors.c.count + count)
  111.     .returning(literal(1))
  112.     .cte('update_cte')
  113. )
  115. upsert = visitors.insert().from_select(
  116.     [visitors.c.product_id, visitors.c.date, visitors.c.count],
  117.     select(literal(product_id), literal(day), literal(count))
  118.         .where(~exists(update_cte.select()))
  119. )
  121. connection.execute(upsert)
  122. ```
  124. Example 4, Nesting CTE (SQLAlchemy 1.4.24 and above):
  126. ```
  127. value_a = select(
  128.     literal("root").label("n")
  129. ).cte("value_a")
  131. # A nested CTE with the same name as the root one
  132. value_a_nested = select(
  133.     literal("nesting").label("n")
  134. ).cte("value_a", nesting=True)
  136. # Nesting CTEs takes ascendency locally
  137. # over the CTEs at a higher level
  138. value_b = select(value_a_nested.c.n).cte("value_b")
  140. value_ab = select(value_a.c.n.label("a"), value_b.c.n.label("b"))
  141. ```
  143. The above query will render the second CTE nested inside the first, shown with inline parameters below as:
  145. ```
  146. WITH
  147.     value_a AS
  148.         (SELECT 'root' AS n),
  149.     value_b AS
  150.         (WITH value_a AS
  151.             (SELECT 'nesting' AS n)
  152.         SELECT value_a.n AS n FROM value_a)
  153. SELECT value_a.n AS a, value_b.n AS b
  154. FROM value_a, value_b
  155. ```
  157. The same CTE can be set up using the [HasCTE.add\_cte()](#sqlalchemy.sql.expression.HasCTE.add_cte "sqlalchemy.sql.expression.HasCTE.add_cte") method as follows (SQLAlchemy 2.0 and above):
  159. ```
  160. value_a = select(
  161.     literal("root").label("n")
  162. ).cte("value_a")
  164. # A nested CTE with the same name as the root one
  165. value_a_nested = select(
  166.     literal("nesting").label("n")
  167. ).cte("value_a")
  169. # Nesting CTEs takes ascendency locally
  170. # over the CTEs at a higher level
  171. value_b = (
  172.     select(value_a_nested.c.n).
  173.     add_cte(value_a_nested, nest_here=True).
  174.     cte("value_b")
  175. )
  177. value_ab = select(value_a.c.n.label("a"), value_b.c.n.label("b"))
  178. ```
  180. Example 5, Non-Linear CTE (SQLAlchemy 1.4.28 and above):
  182. ```
  183. edge = Table(
  184.     "edge",
  185.     metadata,
  186.     Column("id", Integer, primary_key=True),
  187.     Column("left", Integer),
  188.     Column("right", Integer),
  189. )
  191. root_node = select(literal(1).label("node")).cte(
  192.     "nodes", recursive=True
  193. )
  195. left_edge = select(edge.c.left).join(
  196.     root_node, edge.c.right == root_node.c.node
  197. )
  198. right_edge = select(edge.c.right).join(
  199.     root_node, edge.c.left == root_node.c.node
  200. )
  202. subgraph_cte = root_node.union(left_edge, right_edge)
  204. subgraph = select(subgraph_cte)
  205. ```
  207. The above query will render 2 UNIONs inside the recursive CTE:
  209. ```
  210. WITH RECURSIVE nodes(node) AS (
  211.         SELECT 1 AS node
  212.     UNION
  213.         SELECT edge."left" AS "left"
  214.         FROM edge JOIN nodes ON edge."right" = nodes.node
  215.     UNION
  216.         SELECT edge."right" AS "right"
  217.         FROM edge JOIN nodes ON edge."left" = nodes.node
  218. )
  219. SELECT nodes.node FROM nodes
  220. ```
  222. See also
  224. [Query.cte()]($3d0cc000ec6c7150.md#sqlalchemy.orm.Query.cte "sqlalchemy.orm.Query.cte") - ORM version of [HasCTE.cte()](#sqlalchemy.sql.expression.HasCTE.cte "sqlalchemy.sql.expression.HasCTE.cte").

class sqlalchemy.sql.expression.Values

Represent a VALUES construct that can be used as a FROM element in a statement.

The Values object is created from the values() function.

New in version 1.4.

Members

alias(), data(), lateral(), scalar_values()

Class signature

class sqlalchemy.sql.expression.Values (sqlalchemy.sql.roles.InElementRole, sqlalchemy.sql.expression.Generative, sqlalchemy.sql.expression.LateralFromClause)

class sqlalchemy.sql.expression.ScalarValues

Represent a scalar VALUES construct that can be used as a COLUMN element in a statement.

The ScalarValues object is created from the Values.scalar_values() method. It’s also automatically generated when a Values is used in an IN or NOT IN condition.

New in version 2.0.0b4.

Class signature

class sqlalchemy.sql.expression.ScalarValues (sqlalchemy.sql.roles.InElementRole, sqlalchemy.sql.expression.GroupedElement, sqlalchemy.sql.expression.ColumnElement)

Label Style Constants

Constants used with the GenerativeSelect.set_label_style() method.

Object NameDescription

SelectLabelStyle

Label style constants that may be passed to Select.set_label_style().

class sqlalchemy.sql.expression.SelectLabelStyle

Label style constants that may be passed to Select.set_label_style().

Members

LABEL_STYLE_DEFAULT, LABEL_STYLE_DISAMBIGUATE_ONLY, LABEL_STYLE_NONE, LABEL_STYLE_TABLENAME_PLUS_COL

Class signature

class sqlalchemy.sql.expression.SelectLabelStyle (enum.Enum)

  • attribute sqlalchemy.sql.expression.SelectLabelStyle.LABEL_STYLE_DEFAULT = 2

    The default label style, refers to LABEL_STYLE_DISAMBIGUATE_ONLY.

    New in version 1.4.

  • attribute sqlalchemy.sql.expression.SelectLabelStyle.LABEL_STYLE_DISAMBIGUATE_ONLY = 2

    Label style indicating that columns with a name that conflicts with an existing name should be labeled with a semi-anonymizing label when generating the columns clause of a SELECT statement.

    Below, most column names are left unaffected, except for the second occurrence of the name columna, which is labeled using the label columna_1 to disambiguate it from that of tablea.columna:

    1. >>> from sqlalchemy import table, column, select, true, LABEL_STYLE_DISAMBIGUATE_ONLY
    2. >>> table1 = table("table1", column("columna"), column("columnb"))
    3. >>> table2 = table("table2", column("columna"), column("columnc"))
    4. >>> print(select(table1, table2).join(table2, true()).set_label_style(LABEL_STYLE_DISAMBIGUATE_ONLY))
    5. SELECT table1.columna, table1.columnb, table2.columna AS columna_1, table2.columnc
    6. FROM table1 JOIN table2 ON true

    Used with the GenerativeSelect.set_label_style() method, LABEL_STYLE_DISAMBIGUATE_ONLY is the default labeling style for all SELECT statements outside of 1.x style ORM queries.

    New in version 1.4.

  • attribute sqlalchemy.sql.expression.SelectLabelStyle.LABEL_STYLE_NONE = 0

    Label style indicating no automatic labeling should be applied to the columns clause of a SELECT statement.

    Below, the columns named columna are both rendered as is, meaning that the name columna can only refer to the first occurrence of this name within a result set, as well as if the statement were used as a subquery:

    1. >>> from sqlalchemy import table, column, select, true, LABEL_STYLE_NONE
    2. >>> table1 = table("table1", column("columna"), column("columnb"))
    3. >>> table2 = table("table2", column("columna"), column("columnc"))
    4. >>> print(select(table1, table2).join(table2, true()).set_label_style(LABEL_STYLE_NONE))
    5. SELECT table1.columna, table1.columnb, table2.columna, table2.columnc
    6. FROM table1 JOIN table2 ON true

    Used with the Select.set_label_style() method.

    New in version 1.4.

  • attribute sqlalchemy.sql.expression.SelectLabelStyle.LABEL_STYLE_TABLENAME_PLUS_COL = 1

    Label style indicating all columns should be labeled as <tablename>_<columnname> when generating the columns clause of a SELECT statement, to disambiguate same-named columns referenced from different tables, aliases, or subqueries.

    Below, all column names are given a label so that the two same-named columns columna are disambiguated as table1_columna and table2_columna:

    1. >>> from sqlalchemy import table, column, select, true, LABEL_STYLE_TABLENAME_PLUS_COL
    2. >>> table1 = table("table1", column("columna"), column("columnb"))
    3. >>> table2 = table("table2", column("columna"), column("columnc"))
    4. >>> print(select(table1, table2).join(table2, true()).set_label_style(LABEL_STYLE_TABLENAME_PLUS_COL))
    5. SELECT table1.columna AS table1_columna, table1.columnb AS table1_columnb, table2.columna AS table2_columna, table2.columnc AS table2_columnc
    6. FROM table1 JOIN table2 ON true

    Used with the GenerativeSelect.set_label_style() method. Equivalent to the legacy method Select.apply_labels(); LABEL_STYLE_TABLENAME_PLUS_COL is SQLAlchemy’s legacy auto-labeling style. LABEL_STYLE_DISAMBIGUATE_ONLY provides a less intrusive approach to disambiguation of same-named column expressions.

    New in version 1.4.

See also

Select.set_label_style()

Select.get_label_style()