Coverage for python/lsst/daf/butler/registry/databases/postgresql.py: 33%
222 statements
« prev ^ index » next coverage.py v7.4.4, created at 2024-03-30 02:51 -0700
« prev ^ index » next coverage.py v7.4.4, created at 2024-03-30 02:51 -0700
1# This file is part of daf_butler.
2#
3# Developed for the LSST Data Management System.
4# This product includes software developed by the LSST Project
5# (http://www.lsst.org).
6# See the COPYRIGHT file at the top-level directory of this distribution
7# for details of code ownership.
8#
9# This software is dual licensed under the GNU General Public License and also
10# under a 3-clause BSD license. Recipients may choose which of these licenses
11# to use; please see the files gpl-3.0.txt and/or bsd_license.txt,
12# respectively. If you choose the GPL option then the following text applies
13# (but note that there is still no warranty even if you opt for BSD instead):
14#
15# This program is free software: you can redistribute it and/or modify
16# it under the terms of the GNU General Public License as published by
17# the Free Software Foundation, either version 3 of the License, or
18# (at your option) any later version.
19#
20# This program is distributed in the hope that it will be useful,
21# but WITHOUT ANY WARRANTY; without even the implied warranty of
22# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
23# GNU General Public License for more details.
24#
25# You should have received a copy of the GNU General Public License
26# along with this program. If not, see <http://www.gnu.org/licenses/>.
27from __future__ import annotations
29from sqlalchemy.sql.expression import ColumnElement as ColumnElement
31from ... import ddl, time_utils
33__all__ = ["PostgresqlDatabase"]
35import re
36from collections.abc import Callable, Iterable, Iterator, Mapping
37from contextlib import closing, contextmanager
38from typing import Any
40import psycopg2
41import sqlalchemy
42import sqlalchemy.dialects.postgresql
43from sqlalchemy import sql
45from ..._named import NamedValueAbstractSet
46from ..._timespan import Timespan
47from ...timespan_database_representation import TimespanDatabaseRepresentation
48from ..interfaces import Database
49from ..nameShrinker import NameShrinker
51_SERVER_VERSION_REGEX = re.compile(r"(?P<major>\d+)\.(?P<minor>\d+)")
54class PostgresqlDatabase(Database):
55 """An implementation of the `Database` interface for PostgreSQL.
57 Parameters
58 ----------
59 engine : `sqlalchemy.engine.Engine`
60 Engine to use for this connection.
61 origin : `int`
62 An integer ID that should be used as the default for any datasets,
63 quanta, or other entities that use a (autoincrement, origin) compound
64 primary key.
65 namespace : `str`, optional
66 The namespace (schema) this database is associated with. If `None`,
67 the default schema for the connection is used (which may be `None`).
68 writeable : `bool`, optional
69 If `True`, allow write operations on the database, including
70 ``CREATE TABLE``.
72 Notes
73 -----
74 This currently requires the psycopg2 driver to be used as the backend for
75 SQLAlchemy. Running the tests for this class requires the
76 ``testing.postgresql`` be installed, which we assume indicates that a
77 PostgreSQL server is installed and can be run locally in userspace.
79 Some functionality provided by this class (and used by `Registry`) requires
80 the ``btree_gist`` PostgreSQL server extension to be installed an enabled
81 on the database being connected to; this is checked at connection time.
82 """
84 def __init__(
85 self,
86 *,
87 engine: sqlalchemy.engine.Engine,
88 origin: int,
89 namespace: str | None = None,
90 writeable: bool = True,
91 ):
92 with engine.connect() as connection:
93 # `Any` to make mypy ignore the line below, can't use type: ignore
94 dbapi: Any = connection.connection
95 try:
96 dsn = dbapi.get_dsn_parameters()
97 except (AttributeError, KeyError) as err:
98 raise RuntimeError("Only the psycopg2 driver for PostgreSQL is supported.") from err
99 if namespace is None:
100 query = sql.select(sql.func.current_schema())
101 namespace = connection.execute(query).scalar()
102 query_text = "SELECT COUNT(*) FROM pg_extension WHERE extname='btree_gist';"
103 if not connection.execute(sqlalchemy.text(query_text)).scalar():
104 raise RuntimeError(
105 "The Butler PostgreSQL backend requires the btree_gist extension. "
106 "As extensions are enabled per-database, this may require an administrator to run "
107 "`CREATE EXTENSION btree_gist;` in a database before a butler client for it is "
108 " initialized."
109 )
110 raw_pg_version = connection.execute(sqlalchemy.text("SHOW server_version")).scalar()
111 if raw_pg_version is not None and (m := _SERVER_VERSION_REGEX.search(raw_pg_version)):
112 pg_version = (int(m.group("major")), int(m.group("minor")))
113 else:
114 raise RuntimeError("Failed to get PostgreSQL server version.")
115 self._init(
116 engine=engine,
117 origin=origin,
118 namespace=namespace,
119 writeable=writeable,
120 dbname=dsn.get("dbname"),
121 metadata=None,
122 pg_version=pg_version,
123 )
125 def _init(
126 self,
127 *,
128 engine: sqlalchemy.engine.Engine,
129 origin: int,
130 namespace: str | None = None,
131 writeable: bool = True,
132 dbname: str,
133 metadata: sqlalchemy.schema.MetaData | None,
134 pg_version: tuple[int, int],
135 ) -> None:
136 # Initialization logic shared between ``__init__`` and ``clone``.
137 super().__init__(origin=origin, engine=engine, namespace=namespace, metadata=metadata)
138 self._writeable = writeable
139 self.dbname = dbname
140 self._pg_version = pg_version
141 self._shrinker = NameShrinker(self.dialect.max_identifier_length)
143 def clone(self) -> PostgresqlDatabase:
144 clone = self.__new__(type(self))
145 clone._init(
146 origin=self.origin,
147 engine=self._engine,
148 namespace=self.namespace,
149 writeable=self._writeable,
150 dbname=self.dbname,
151 metadata=self._metadata,
152 pg_version=self._pg_version,
153 )
154 return clone
156 @classmethod
157 def makeEngine(
158 cls, uri: str | sqlalchemy.engine.URL, *, writeable: bool = True
159 ) -> sqlalchemy.engine.Engine:
160 return sqlalchemy.engine.create_engine(uri, pool_size=1)
162 @classmethod
163 def fromEngine(
164 cls,
165 engine: sqlalchemy.engine.Engine,
166 *,
167 origin: int,
168 namespace: str | None = None,
169 writeable: bool = True,
170 ) -> Database:
171 return cls(engine=engine, origin=origin, namespace=namespace, writeable=writeable)
173 @contextmanager
174 def _transaction(
175 self,
176 *,
177 interrupting: bool = False,
178 savepoint: bool = False,
179 lock: Iterable[sqlalchemy.schema.Table] = (),
180 for_temp_tables: bool = False,
181 ) -> Iterator[tuple[bool, sqlalchemy.engine.Connection]]:
182 with super()._transaction(interrupting=interrupting, savepoint=savepoint, lock=lock) as (
183 is_new,
184 connection,
185 ):
186 if is_new:
187 # pgbouncer with transaction-level pooling (which we aim to
188 # support) says that SET cannot be used, except for a list of
189 # "Startup parameters" that includes "timezone" (see
190 # https://www.pgbouncer.org/features.html#fnref:0). But I
191 # don't see "timezone" in PostgreSQL's list of parameters
192 # passed when creating a new connection
193 # (https://www.postgresql.org/docs/current/libpq-connect.html#LIBPQ-PARAMKEYWORDS).
194 # Given that the pgbouncer docs say, "PgBouncer detects their
195 # changes and so it can guarantee they remain consistent for
196 # the client", I assume we can use "SET TIMESPAN" and pgbouncer
197 # will take care of clients that share connections being set
198 # consistently. And if that assumption is wrong, we should
199 # still probably be okay, since all clients should be Butler
200 # clients, and they'll all be setting the same thing.
201 #
202 # The "SET TRANSACTION READ ONLY" should also be safe, because
203 # it only ever acts on the current transaction; I think it's
204 # not included in pgbouncer's declaration that SET is
205 # incompatible with transaction-level pooling because
206 # PostgreSQL actually considers SET TRANSACTION to be a
207 # fundamentally different statement from SET (they have their
208 # own distinct doc pages, at least).
209 if not (self.isWriteable() or for_temp_tables):
210 # PostgreSQL permits writing to temporary tables inside
211 # read-only transactions, but it doesn't permit creating
212 # them.
213 with closing(connection.connection.cursor()) as cursor:
214 cursor.execute("SET TRANSACTION READ ONLY")
215 cursor.execute("SET TIME ZONE 0")
216 else:
217 with closing(connection.connection.cursor()) as cursor:
218 # Make timestamps UTC, because we didn't use TIMESTAMPZ
219 # for the column type. When we can tolerate a schema
220 # change, we should change that type and remove this
221 # line.
222 cursor.execute("SET TIME ZONE 0")
223 yield is_new, connection
225 @contextmanager
226 def temporary_table(
227 self, spec: ddl.TableSpec, name: str | None = None
228 ) -> Iterator[sqlalchemy.schema.Table]:
229 # Docstring inherited.
230 with self.transaction(for_temp_tables=True), super().temporary_table(spec, name) as table:
231 yield table
233 def _lockTables(
234 self, connection: sqlalchemy.engine.Connection, tables: Iterable[sqlalchemy.schema.Table] = ()
235 ) -> None:
236 # Docstring inherited.
237 for table in tables:
238 connection.execute(sqlalchemy.text(f"LOCK TABLE {table.key} IN EXCLUSIVE MODE"))
240 def isWriteable(self) -> bool:
241 return self._writeable
243 def __str__(self) -> str:
244 return f"PostgreSQL@{self.dbname}:{self.namespace}"
246 def shrinkDatabaseEntityName(self, original: str) -> str:
247 return self._shrinker.shrink(original)
249 def expandDatabaseEntityName(self, shrunk: str) -> str:
250 return self._shrinker.expand(shrunk)
252 def _convertExclusionConstraintSpec(
253 self,
254 table: str,
255 spec: tuple[str | type[TimespanDatabaseRepresentation], ...],
256 metadata: sqlalchemy.MetaData,
257 ) -> sqlalchemy.schema.Constraint:
258 # Docstring inherited.
259 args: list[tuple[sqlalchemy.schema.Column, str]] = []
260 names = ["excl"]
261 for item in spec:
262 if isinstance(item, str):
263 args.append((sqlalchemy.schema.Column(item), "="))
264 names.append(item)
265 elif issubclass(item, TimespanDatabaseRepresentation):
266 assert item is self.getTimespanRepresentation()
267 args.append((sqlalchemy.schema.Column(TimespanDatabaseRepresentation.NAME), "&&"))
268 names.append(TimespanDatabaseRepresentation.NAME)
269 return sqlalchemy.dialects.postgresql.ExcludeConstraint(
270 *args,
271 name=self.shrinkDatabaseEntityName("_".join(names)),
272 )
274 def _make_temporary_table(
275 self,
276 connection: sqlalchemy.engine.Connection,
277 spec: ddl.TableSpec,
278 name: str | None = None,
279 **kwargs: Any,
280 ) -> sqlalchemy.schema.Table:
281 # Docstring inherited
282 # Adding ON COMMIT DROP here is really quite defensive: we already
283 # manually drop the table at the end of the temporary_table context
284 # manager, and that will usually happen first. But this will guarantee
285 # that we drop the table at the end of the transaction even if the
286 # connection lasts longer, and that's good citizenship when connections
287 # may be multiplexed by e.g. pgbouncer.
288 return super()._make_temporary_table(connection, spec, name, postgresql_on_commit="DROP", **kwargs)
290 @classmethod
291 def getTimespanRepresentation(cls) -> type[TimespanDatabaseRepresentation]:
292 # Docstring inherited.
293 return _RangeTimespanRepresentation
295 def replace(self, table: sqlalchemy.schema.Table, *rows: dict) -> None:
296 self.assertTableWriteable(table, f"Cannot replace into read-only table {table}.")
297 if not rows:
298 return
299 # This uses special support for UPSERT in PostgreSQL backend:
300 # https://docs.sqlalchemy.org/en/13/dialects/postgresql.html#insert-on-conflict-upsert
301 query = sqlalchemy.dialects.postgresql.dml.insert(table)
302 # In the SET clause assign all columns using special `excluded`
303 # pseudo-table. If some column in the table does not appear in the
304 # INSERT list this will set it to NULL.
305 excluded = query.excluded
306 data = {
307 column.name: getattr(excluded, column.name)
308 for column in table.columns
309 if column.name not in table.primary_key
310 }
311 query = query.on_conflict_do_update(constraint=table.primary_key, set_=data)
312 with self._transaction() as (_, connection):
313 connection.execute(query, rows)
315 def ensure(self, table: sqlalchemy.schema.Table, *rows: dict, primary_key_only: bool = False) -> int:
316 # Docstring inherited.
317 self.assertTableWriteable(table, f"Cannot ensure into read-only table {table}.")
318 if not rows:
319 return 0
320 # Like `replace`, this uses UPSERT.
321 base_insert = sqlalchemy.dialects.postgresql.dml.insert(table)
322 if primary_key_only:
323 query = base_insert.on_conflict_do_nothing(constraint=table.primary_key)
324 else:
325 query = base_insert.on_conflict_do_nothing()
326 with self._transaction() as (_, connection):
327 return connection.execute(query, rows).rowcount
329 def constant_rows(
330 self,
331 fields: NamedValueAbstractSet[ddl.FieldSpec],
332 *rows: dict,
333 name: str | None = None,
334 ) -> sqlalchemy.sql.FromClause:
335 # Docstring inherited.
336 return super().constant_rows(fields, *rows, name=name)
338 @property
339 def has_distinct_on(self) -> bool:
340 # Docstring inherited.
341 return True
343 @property
344 def has_any_aggregate(self) -> bool:
345 # Docstring inherited.
346 return self._pg_version >= (16, 0)
348 def apply_any_aggregate(self, column: sqlalchemy.ColumnElement[Any]) -> sqlalchemy.ColumnElement[Any]:
349 # Docstring inherited.x
350 return sqlalchemy.func.any_value(column)
353class _RangeTimespanType(sqlalchemy.TypeDecorator):
354 """A single-column `Timespan` representation usable only with
355 PostgreSQL.
357 This type should be able to take advantage of PostgreSQL's built-in
358 range operators, and the indexing and EXCLUSION table constraints built
359 off of them.
360 """
362 impl = sqlalchemy.dialects.postgresql.INT8RANGE
364 cache_ok = True
366 def process_bind_param(
367 self, value: Timespan | None, dialect: sqlalchemy.engine.Dialect
368 ) -> psycopg2.extras.NumericRange | None:
369 if value is None:
370 return None
371 if not isinstance(value, Timespan):
372 raise TypeError(f"Unsupported type: {type(value)}, expected Timespan.")
373 if value.isEmpty():
374 return psycopg2.extras.NumericRange(empty=True)
375 else:
376 converter = time_utils.TimeConverter()
377 assert value._nsec[0] >= converter.min_nsec, "Guaranteed by Timespan.__init__."
378 assert value._nsec[1] <= converter.max_nsec, "Guaranteed by Timespan.__init__."
379 lower = None if value._nsec[0] == converter.min_nsec else value._nsec[0]
380 upper = None if value._nsec[1] == converter.max_nsec else value._nsec[1]
381 return psycopg2.extras.NumericRange(lower=lower, upper=upper)
383 def process_result_value(
384 self, value: psycopg2.extras.NumericRange | None, dialect: sqlalchemy.engine.Dialect
385 ) -> Timespan | None:
386 if value is None:
387 return None
388 if value.isempty:
389 return Timespan.makeEmpty()
390 converter = time_utils.TimeConverter()
391 begin_nsec = converter.min_nsec if value.lower is None else value.lower
392 end_nsec = converter.max_nsec if value.upper is None else value.upper
393 return Timespan(begin=None, end=None, _nsec=(begin_nsec, end_nsec))
396class _RangeTimespanRepresentation(TimespanDatabaseRepresentation):
397 """An implementation of `TimespanDatabaseRepresentation` that uses
398 `_RangeTimespanType` to store a timespan in a single
399 PostgreSQL-specific field.
401 Parameters
402 ----------
403 column : `sqlalchemy.sql.ColumnElement`
404 SQLAlchemy object representing the column.
405 """
407 def __init__(self, column: sqlalchemy.sql.ColumnElement, name: str):
408 self.column = column
409 self._name = name
411 __slots__ = ("column", "_name")
413 @classmethod
414 def makeFieldSpecs(
415 cls, nullable: bool, name: str | None = None, **kwargs: Any
416 ) -> tuple[ddl.FieldSpec, ...]:
417 # Docstring inherited.
418 if name is None:
419 name = cls.NAME
420 return (
421 ddl.FieldSpec(
422 name,
423 dtype=_RangeTimespanType,
424 nullable=nullable,
425 default=(None if nullable else sqlalchemy.sql.text("'(,)'::int8range")),
426 **kwargs,
427 ),
428 )
430 @classmethod
431 def getFieldNames(cls, name: str | None = None) -> tuple[str, ...]:
432 # Docstring inherited.
433 if name is None:
434 name = cls.NAME
435 return (name,)
437 @classmethod
438 def update(
439 cls, extent: Timespan | None, name: str | None = None, result: dict[str, Any] | None = None
440 ) -> dict[str, Any]:
441 # Docstring inherited.
442 if name is None:
443 name = cls.NAME
444 if result is None:
445 result = {}
446 result[name] = extent
447 return result
449 @classmethod
450 def extract(cls, mapping: Mapping[str, Any], name: str | None = None) -> Timespan | None:
451 # Docstring inherited.
452 if name is None:
453 name = cls.NAME
454 return mapping[name]
456 @classmethod
457 def fromLiteral(cls, timespan: Timespan | None) -> _RangeTimespanRepresentation:
458 # Docstring inherited.
459 if timespan is None:
460 return cls(column=sqlalchemy.sql.null(), name=cls.NAME)
461 return cls(
462 column=sqlalchemy.sql.cast(
463 sqlalchemy.sql.literal(timespan, type_=_RangeTimespanType), type_=_RangeTimespanType
464 ),
465 name=cls.NAME,
466 )
468 @classmethod
469 def from_columns(
470 cls, columns: sqlalchemy.sql.ColumnCollection, name: str | None = None
471 ) -> _RangeTimespanRepresentation:
472 # Docstring inherited.
473 if name is None:
474 name = cls.NAME
475 return cls(columns[name], name)
477 @property
478 def name(self) -> str:
479 # Docstring inherited.
480 return self._name
482 def isNull(self) -> sqlalchemy.sql.ColumnElement:
483 # Docstring inherited.
484 return self.column.is_(None)
486 def isEmpty(self) -> sqlalchemy.sql.ColumnElement:
487 # Docstring inherited
488 return sqlalchemy.sql.func.isempty(self.column)
490 def __lt__(
491 self, other: _RangeTimespanRepresentation | sqlalchemy.sql.ColumnElement
492 ) -> sqlalchemy.sql.ColumnElement:
493 # Docstring inherited.
494 if isinstance(other, sqlalchemy.sql.ColumnElement):
495 return sqlalchemy.sql.and_(
496 sqlalchemy.sql.not_(sqlalchemy.sql.func.upper_inf(self.column)),
497 sqlalchemy.sql.not_(sqlalchemy.sql.func.isempty(self.column)),
498 sqlalchemy.sql.func.upper(self.column) <= other,
499 )
500 else:
501 return self.column << other.column
503 def __gt__(
504 self, other: _RangeTimespanRepresentation | sqlalchemy.sql.ColumnElement
505 ) -> sqlalchemy.sql.ColumnElement:
506 # Docstring inherited.
507 if isinstance(other, sqlalchemy.sql.ColumnElement):
508 return sqlalchemy.sql.and_(
509 sqlalchemy.sql.not_(sqlalchemy.sql.func.lower_inf(self.column)),
510 sqlalchemy.sql.not_(sqlalchemy.sql.func.isempty(self.column)),
511 sqlalchemy.sql.func.lower(self.column) > other,
512 )
513 else:
514 return self.column >> other.column
516 def overlaps(
517 self, other: _RangeTimespanRepresentation | sqlalchemy.sql.ColumnElement
518 ) -> sqlalchemy.sql.ColumnElement:
519 # Docstring inherited.
520 if not isinstance(other, _RangeTimespanRepresentation):
521 return self.contains(other)
522 return self.column.overlaps(other.column)
524 def contains(
525 self, other: _RangeTimespanRepresentation | sqlalchemy.sql.ColumnElement
526 ) -> sqlalchemy.sql.ColumnElement:
527 # Docstring inherited
528 if isinstance(other, _RangeTimespanRepresentation):
529 return self.column.contains(other.column)
530 else:
531 return self.column.contains(other)
533 def lower(self) -> sqlalchemy.sql.ColumnElement:
534 # Docstring inherited.
535 return sqlalchemy.sql.functions.coalesce(
536 sqlalchemy.sql.func.lower(self.column), sqlalchemy.sql.literal(0)
537 )
539 def upper(self) -> sqlalchemy.sql.ColumnElement:
540 # Docstring inherited.
541 return sqlalchemy.sql.functions.coalesce(
542 sqlalchemy.sql.func.upper(self.column), sqlalchemy.sql.literal(0)
543 )
545 def flatten(self, name: str | None = None) -> tuple[sqlalchemy.sql.ColumnElement]:
546 # Docstring inherited.
547 if name is None:
548 return (self.column,)
549 else:
550 return (self.column.label(name),)
552 def apply_any_aggregate(
553 self, func: Callable[[ColumnElement[Any]], ColumnElement[Any]]
554 ) -> TimespanDatabaseRepresentation:
555 # Docstring inherited.
556 return _RangeTimespanRepresentation(func(self.column), self.name)