import json from functools import lru_cache, partial from django.conf import settings from django.db.backends.base.operations import BaseDatabaseOperations from django.db.backends.postgresql.psycopg_any import ( Inet, Jsonb, errors, is_psycopg3, mogrify, ) from django.db.backends.utils import split_tzname_delta from django.db.models.constants import OnConflict from django.utils.regex_helper import _lazy_re_compile @lru_cache def get_json_dumps(encoder): if encoder is None: return json.dumps return partial(json.dumps, cls=encoder) class DatabaseOperations(BaseDatabaseOperations): cast_char_field_without_max_length = "varchar" explain_prefix = "EXPLAIN" explain_options = frozenset( [ "ANALYZE", "BUFFERS", "COSTS", "SETTINGS", "SUMMARY", "TIMING", "VERBOSE", "WAL", ] ) cast_data_types = { "AutoField": "integer", "BigAutoField": "bigint", "SmallAutoField": "smallint", } if is_psycopg3: from psycopg.types import numeric integerfield_type_map = { "SmallIntegerField": numeric.Int2, "IntegerField": numeric.Int4, "BigIntegerField": numeric.Int8, "PositiveSmallIntegerField": numeric.Int2, "PositiveIntegerField": numeric.Int4, "PositiveBigIntegerField": numeric.Int8, } def unification_cast_sql(self, output_field): internal_type = output_field.get_internal_type() if internal_type in ( "GenericIPAddressField", "IPAddressField", "TimeField", "UUIDField", ): # PostgreSQL will resolve a union as type 'text' if input types are # 'unknown'. # https://www.postgresql.org/docs/current/typeconv-union-case.html # These fields cannot be implicitly cast back in the default # PostgreSQL configuration so we need to explicitly cast them. # We must also remove components of the type within brackets: # varchar(255) -> varchar. return ( "CAST(%%s AS %s)" % output_field.db_type(self.connection).split("(")[0] ) return "%s" # EXTRACT format cannot be passed in parameters. _extract_format_re = _lazy_re_compile(r"[A-Z_]+") def date_extract_sql(self, lookup_type, sql, params): # https://www.postgresql.org/docs/current/functions-datetime.html#FUNCTIONS-DATETIME-EXTRACT if lookup_type == "week_day": # For consistency across backends, we return Sunday=1, Saturday=7. return f"EXTRACT(DOW FROM {sql}) + 1", params elif lookup_type == "iso_week_day": return f"EXTRACT(ISODOW FROM {sql})", params elif lookup_type == "iso_year": return f"EXTRACT(ISOYEAR FROM {sql})", params lookup_type = lookup_type.upper() if not self._extract_format_re.fullmatch(lookup_type): raise ValueError(f"Invalid lookup type: {lookup_type!r}") return f"EXTRACT({lookup_type} FROM {sql})", params def date_trunc_sql(self, lookup_type, sql, params, tzname=None): sql, params = self._convert_sql_to_tz(sql, params, tzname) # https://www.postgresql.org/docs/current/functions-datetime.html#FUNCTIONS-DATETIME-TRUNC return f"DATE_TRUNC(%s, {sql})", (lookup_type, *params) def _prepare_tzname_delta(self, tzname): tzname, sign, offset = split_tzname_delta(tzname) if offset: sign = "-" if sign == "+" else "+" return f"{tzname}{sign}{offset}" return tzname def _convert_sql_to_tz(self, sql, params, tzname): if tzname and settings.USE_TZ: tzname_param = self._prepare_tzname_delta(tzname) return f"{sql} AT TIME ZONE %s", (*params, tzname_param) return sql, params def datetime_cast_date_sql(self, sql, params, tzname): sql, params = self._convert_sql_to_tz(sql, params, tzname) return f"({sql})::date", params def datetime_cast_time_sql(self, sql, params, tzname): sql, params = self._convert_sql_to_tz(sql, params, tzname) return f"({sql})::time", params def datetime_extract_sql(self, lookup_type, sql, params, tzname): sql, params = self._convert_sql_to_tz(sql, params, tzname) if lookup_type == "second": # Truncate fractional seconds. return f"EXTRACT(SECOND FROM DATE_TRUNC(%s, {sql}))", ("second", *params) return self.date_extract_sql(lookup_type, sql, params) def datetime_trunc_sql(self, lookup_type, sql, params, tzname): sql, params = self._convert_sql_to_tz(sql, params, tzname) # https://www.postgresql.org/docs/current/functions-datetime.html#FUNCTIONS-DATETIME-TRUNC return f"DATE_TRUNC(%s, {sql})", (lookup_type, *params) def time_extract_sql(self, lookup_type, sql, params): if lookup_type == "second": # Truncate fractional seconds. return f"EXTRACT(SECOND FROM DATE_TRUNC(%s, {sql}))", ("second", *params) return self.date_extract_sql(lookup_type, sql, params) def time_trunc_sql(self, lookup_type, sql, params, tzname=None): sql, params = self._convert_sql_to_tz(sql, params, tzname) return f"DATE_TRUNC(%s, {sql})::time", (lookup_type, *params) def deferrable_sql(self): return " DEFERRABLE INITIALLY DEFERRED" def fetch_returned_insert_rows(self, cursor): """ Given a cursor object that has just performed an INSERT...RETURNING statement into a table, return the tuple of returned data. """ return cursor.fetchall() def lookup_cast(self, lookup_type, internal_type=None): lookup = "%s" if lookup_type == "isnull" and internal_type in ( "CharField", "EmailField", "TextField", "CICharField", "CIEmailField", "CITextField", ): return "%s::text" # Cast text lookups to text to allow things like filter(x__contains=4) if lookup_type in ( "iexact", "contains", "icontains", "startswith", "istartswith", "endswith", "iendswith", "regex", "iregex", ): if internal_type in ("IPAddressField", "GenericIPAddressField"): lookup = "HOST(%s)" # RemovedInDjango51Warning. elif internal_type in ("CICharField", "CIEmailField", "CITextField"): lookup = "%s::citext" else: lookup = "%s::text" # Use UPPER(x) for case-insensitive lookups; it's faster. if lookup_type in ("iexact", "icontains", "istartswith", "iendswith"): lookup = "UPPER(%s)" % lookup return lookup def no_limit_value(self): return None def prepare_sql_script(self, sql): return [sql] def quote_name(self, name): if name.startswith('"') and name.endswith('"'): return name # Quoting once is enough. return '"%s"' % name def compose_sql(self, sql, params): return mogrify(sql, params, self.connection) def set_time_zone_sql(self): return "SELECT set_config('TimeZone', %s, false)" def sql_flush(self, style, tables, *, reset_sequences=False, allow_cascade=False): if not tables: return [] # Perform a single SQL 'TRUNCATE x, y, z...;' statement. It allows us # to truncate tables referenced by a foreign key in any other table. sql_parts = [ style.SQL_KEYWORD("TRUNCATE"), ", ".join(style.SQL_FIELD(self.quote_name(table)) for table in tables), ] if reset_sequences: sql_parts.append(style.SQL_KEYWORD("RESTART IDENTITY")) if allow_cascade: sql_parts.append(style.SQL_KEYWORD("CASCADE")) return ["%s;" % " ".join(sql_parts)] def sequence_reset_by_name_sql(self, style, sequences): # 'ALTER SEQUENCE sequence_name RESTART WITH 1;'... style SQL statements # to reset sequence indices sql = [] for sequence_info in sequences: table_name = sequence_info["table"] # 'id' will be the case if it's an m2m using an autogenerated # intermediate table (see BaseDatabaseIntrospection.sequence_list). column_name = sequence_info["column"] or "id" sql.append( "%s setval(pg_get_serial_sequence('%s','%s'), 1, false);" % ( style.SQL_KEYWORD("SELECT"), style.SQL_TABLE(self.quote_name(table_name)), style.SQL_FIELD(column_name), ) ) return sql def tablespace_sql(self, tablespace, inline=False): if inline: return "USING INDEX TABLESPACE %s" % self.quote_name(tablespace) else: return "TABLESPACE %s" % self.quote_name(tablespace) def sequence_reset_sql(self, style, model_list): from django.db import models output = [] qn = self.quote_name for model in model_list: # Use `coalesce` to set the sequence for each model to the max pk # value if there are records, or 1 if there are none. Set the # `is_called` property (the third argument to `setval`) to true if # there are records (as the max pk value is already in use), # otherwise set it to false. Use pg_get_serial_sequence to get the # underlying sequence name from the table name and column name. for f in model._meta.local_fields: if isinstance(f, models.AutoField): output.append( "%s setval(pg_get_serial_sequence('%s','%s'), " "coalesce(max(%s), 1), max(%s) %s null) %s %s;" % ( style.SQL_KEYWORD("SELECT"), style.SQL_TABLE(qn(model._meta.db_table)), style.SQL_FIELD(f.column), style.SQL_FIELD(qn(f.column)), style.SQL_FIELD(qn(f.column)), style.SQL_KEYWORD("IS NOT"), style.SQL_KEYWORD("FROM"), style.SQL_TABLE(qn(model._meta.db_table)), ) ) # Only one AutoField is allowed per model, so don't bother # continuing. break return output def prep_for_iexact_query(self, x): return x def max_name_length(self): """ Return the maximum length of an identifier. The maximum length of an identifier is 63 by default, but can be changed by recompiling PostgreSQL after editing the NAMEDATALEN macro in src/include/pg_config_manual.h. This implementation returns 63, but can be overridden by a custom database backend that inherits most of its behavior from this one. """ return 63 def distinct_sql(self, fields, params): if fields: params = [param for param_list in params for param in param_list] return (["DISTINCT ON (%s)" % ", ".join(fields)], params) else: return ["DISTINCT"], [] if is_psycopg3: def last_executed_query(self, cursor, sql, params): try: return self.compose_sql(sql, params) except errors.DataError: return None else: def last_executed_query(self, cursor, sql, params): # https://www.psycopg.org/docs/cursor.html#cursor.query # The query attribute is a Psycopg extension to the DB API 2.0. if cursor.query is not None: return cursor.query.decode() return None def return_insert_columns(self, fields): if not fields: return "", () columns = [ "%s.%s" % ( self.quote_name(field.model._meta.db_table), self.quote_name(field.column), ) for field in fields ] return "RETURNING %s" % ", ".join(columns), () def bulk_insert_sql(self, fields, placeholder_rows): placeholder_rows_sql = (", ".join(row) for row in placeholder_rows) values_sql = ", ".join("(%s)" % sql for sql in placeholder_rows_sql) return "VALUES " + values_sql if is_psycopg3: def adapt_integerfield_value(self, value, internal_type): if value is None or hasattr(value, "resolve_expression"): return value return self.integerfield_type_map[internal_type](value) def adapt_datefield_value(self, value): return value def adapt_datetimefield_value(self, value): return value def adapt_timefield_value(self, value): return value def adapt_decimalfield_value(self, value, max_digits=None, decimal_places=None): return value def adapt_ipaddressfield_value(self, value): if value: return Inet(value) return None def adapt_json_value(self, value, encoder): return Jsonb(value, dumps=get_json_dumps(encoder)) def subtract_temporals(self, internal_type, lhs, rhs): if internal_type == "DateField": lhs_sql, lhs_params = lhs rhs_sql, rhs_params = rhs params = (*lhs_params, *rhs_params) return "(interval '1 day' * (%s - %s))" % (lhs_sql, rhs_sql), params return super().subtract_temporals(internal_type, lhs, rhs) def explain_query_prefix(self, format=None, **options): extra = {} # Normalize options. if options: options = { name.upper(): "true" if value else "false" for name, value in options.items() } for valid_option in self.explain_options: value = options.pop(valid_option, None) if value is not None: extra[valid_option] = value prefix = super().explain_query_prefix(format, **options) if format: extra["FORMAT"] = format if extra: prefix += " (%s)" % ", ".join("%s %s" % i for i in extra.items()) return prefix def on_conflict_suffix_sql(self, fields, on_conflict, update_fields, unique_fields): if on_conflict == OnConflict.IGNORE: return "ON CONFLICT DO NOTHING" if on_conflict == OnConflict.UPDATE: return "ON CONFLICT(%s) DO UPDATE SET %s" % ( ", ".join(map(self.quote_name, unique_fields)), ", ".join( [ f"{field} = EXCLUDED.{field}" for field in map(self.quote_name, update_fields) ] ), ) return super().on_conflict_suffix_sql( fields, on_conflict, update_fields, unique_fields, )