django1/django/db/backends/postgresql/schema.py

239 lines
11 KiB
Python

import psycopg2
from django.db.backends.base.schema import BaseDatabaseSchemaEditor
from django.db.backends.ddl_references import IndexColumns
from django.db.backends.utils import strip_quotes
class DatabaseSchemaEditor(BaseDatabaseSchemaEditor):
sql_create_sequence = "CREATE SEQUENCE %(sequence)s"
sql_delete_sequence = "DROP SEQUENCE IF EXISTS %(sequence)s CASCADE"
sql_set_sequence_max = "SELECT setval('%(sequence)s', MAX(%(column)s)) FROM %(table)s"
sql_set_sequence_owner = 'ALTER SEQUENCE %(sequence)s OWNED BY %(table)s.%(column)s'
sql_create_index = (
'CREATE INDEX %(name)s ON %(table)s%(using)s '
'(%(columns)s)%(include)s%(extra)s%(condition)s'
)
sql_create_index_concurrently = (
'CREATE INDEX CONCURRENTLY %(name)s ON %(table)s%(using)s '
'(%(columns)s)%(include)s%(extra)s%(condition)s'
)
sql_delete_index = "DROP INDEX IF EXISTS %(name)s"
sql_delete_index_concurrently = "DROP INDEX CONCURRENTLY IF EXISTS %(name)s"
# Setting the constraint to IMMEDIATE to allow changing data in the same
# transaction.
sql_create_column_inline_fk = (
'CONSTRAINT %(name)s REFERENCES %(to_table)s(%(to_column)s)%(deferrable)s'
'; SET CONSTRAINTS %(namespace)s%(name)s IMMEDIATE'
)
# Setting the constraint to IMMEDIATE runs any deferred checks to allow
# dropping it in the same transaction.
sql_delete_fk = "SET CONSTRAINTS %(name)s IMMEDIATE; ALTER TABLE %(table)s DROP CONSTRAINT %(name)s"
sql_delete_procedure = 'DROP FUNCTION %(procedure)s(%(param_types)s)'
def quote_value(self, value):
if isinstance(value, str):
value = value.replace('%', '%%')
adapted = psycopg2.extensions.adapt(value)
if hasattr(adapted, 'encoding'):
adapted.encoding = 'utf8'
# getquoted() returns a quoted bytestring of the adapted value.
return adapted.getquoted().decode()
def _field_indexes_sql(self, model, field):
output = super()._field_indexes_sql(model, field)
like_index_statement = self._create_like_index_sql(model, field)
if like_index_statement is not None:
output.append(like_index_statement)
return output
def _field_data_type(self, field):
if field.is_relation:
return field.rel_db_type(self.connection)
return self.connection.data_types.get(
field.get_internal_type(),
field.db_type(self.connection),
)
def _field_base_data_types(self, field):
# Yield base data types for array fields.
if field.base_field.get_internal_type() == 'ArrayField':
yield from self._field_base_data_types(field.base_field)
else:
yield self._field_data_type(field.base_field)
def _create_like_index_sql(self, model, field):
"""
Return the statement to create an index with varchar operator pattern
when the column type is 'varchar' or 'text', otherwise return None.
"""
db_type = field.db_type(connection=self.connection)
if db_type is not None and (field.db_index or field.unique):
# Fields with database column types of `varchar` and `text` need
# a second index that specifies their operator class, which is
# needed when performing correct LIKE queries outside the
# C locale. See #12234.
#
# The same doesn't apply to array fields such as varchar[size]
# and text[size], so skip them.
if '[' in db_type:
return None
if db_type.startswith('varchar'):
return self._create_index_sql(
model,
fields=[field],
suffix='_like',
opclasses=['varchar_pattern_ops'],
)
elif db_type.startswith('text'):
return self._create_index_sql(
model,
fields=[field],
suffix='_like',
opclasses=['text_pattern_ops'],
)
return None
def _alter_column_type_sql(self, model, old_field, new_field, new_type):
self.sql_alter_column_type = 'ALTER COLUMN %(column)s TYPE %(type)s'
# Cast when data type changed.
using_sql = ' USING %(column)s::%(type)s'
new_internal_type = new_field.get_internal_type()
old_internal_type = old_field.get_internal_type()
if new_internal_type == 'ArrayField' and new_internal_type == old_internal_type:
# Compare base data types for array fields.
if list(self._field_base_data_types(old_field)) != list(self._field_base_data_types(new_field)):
self.sql_alter_column_type += using_sql
elif self._field_data_type(old_field) != self._field_data_type(new_field):
self.sql_alter_column_type += using_sql
# Make ALTER TYPE with SERIAL make sense.
table = strip_quotes(model._meta.db_table)
serial_fields_map = {'bigserial': 'bigint', 'serial': 'integer', 'smallserial': 'smallint'}
if new_type.lower() in serial_fields_map:
column = strip_quotes(new_field.column)
sequence_name = "%s_%s_seq" % (table, column)
return (
(
self.sql_alter_column_type % {
"column": self.quote_name(column),
"type": serial_fields_map[new_type.lower()],
},
[],
),
[
(
self.sql_delete_sequence % {
"sequence": self.quote_name(sequence_name),
},
[],
),
(
self.sql_create_sequence % {
"sequence": self.quote_name(sequence_name),
},
[],
),
(
self.sql_alter_column % {
"table": self.quote_name(table),
"changes": self.sql_alter_column_default % {
"column": self.quote_name(column),
"default": "nextval('%s')" % self.quote_name(sequence_name),
}
},
[],
),
(
self.sql_set_sequence_max % {
"table": self.quote_name(table),
"column": self.quote_name(column),
"sequence": self.quote_name(sequence_name),
},
[],
),
(
self.sql_set_sequence_owner % {
'table': self.quote_name(table),
'column': self.quote_name(column),
'sequence': self.quote_name(sequence_name),
},
[],
),
],
)
elif old_field.db_parameters(connection=self.connection)['type'] in serial_fields_map:
# Drop the sequence if migrating away from AutoField.
column = strip_quotes(new_field.column)
sequence_name = '%s_%s_seq' % (table, column)
fragment, _ = super()._alter_column_type_sql(model, old_field, new_field, new_type)
return fragment, [
(
self.sql_delete_sequence % {
'sequence': self.quote_name(sequence_name),
},
[],
),
]
else:
return super()._alter_column_type_sql(model, old_field, new_field, new_type)
def _alter_field(self, model, old_field, new_field, old_type, new_type,
old_db_params, new_db_params, strict=False):
# Drop indexes on varchar/text/citext columns that are changing to a
# different type.
if (old_field.db_index or old_field.unique) and (
(old_type.startswith('varchar') and not new_type.startswith('varchar')) or
(old_type.startswith('text') and not new_type.startswith('text')) or
(old_type.startswith('citext') and not new_type.startswith('citext'))
):
index_name = self._create_index_name(model._meta.db_table, [old_field.column], suffix='_like')
self.execute(self._delete_index_sql(model, index_name))
super()._alter_field(
model, old_field, new_field, old_type, new_type, old_db_params,
new_db_params, strict,
)
# Added an index? Create any PostgreSQL-specific indexes.
if ((not (old_field.db_index or old_field.unique) and new_field.db_index) or
(not old_field.unique and new_field.unique)):
like_index_statement = self._create_like_index_sql(model, new_field)
if like_index_statement is not None:
self.execute(like_index_statement)
# Removed an index? Drop any PostgreSQL-specific indexes.
if old_field.unique and not (new_field.db_index or new_field.unique):
index_to_remove = self._create_index_name(model._meta.db_table, [old_field.column], suffix='_like')
self.execute(self._delete_index_sql(model, index_to_remove))
def _index_columns(self, table, columns, col_suffixes, opclasses):
if opclasses:
return IndexColumns(table, columns, self.quote_name, col_suffixes=col_suffixes, opclasses=opclasses)
return super()._index_columns(table, columns, col_suffixes, opclasses)
def add_index(self, model, index, concurrently=False):
self.execute(index.create_sql(model, self, concurrently=concurrently), params=None)
def remove_index(self, model, index, concurrently=False):
self.execute(index.remove_sql(model, self, concurrently=concurrently))
def _delete_index_sql(self, model, name, sql=None, concurrently=False):
sql = self.sql_delete_index_concurrently if concurrently else self.sql_delete_index
return super()._delete_index_sql(model, name, sql)
def _create_index_sql(
self, model, *, fields=None, name=None, suffix='', using='',
db_tablespace=None, col_suffixes=(), sql=None, opclasses=(),
condition=None, concurrently=False, include=None, expressions=None,
):
sql = self.sql_create_index if not concurrently else self.sql_create_index_concurrently
return super()._create_index_sql(
model, fields=fields, name=name, suffix=suffix, using=using,
db_tablespace=db_tablespace, col_suffixes=col_suffixes, sql=sql,
opclasses=opclasses, condition=condition, include=include,
expressions=expressions,
)