Mike Bayer [Sat, 13 Apr 2019 03:44:42 +0000 (23:44 -0400)]
Use pytest items in custom collection
We have a custom test collection hook that did not take
node of the actual list of functions in items. By looking
in this list we now support the class/function arguments
passed to the py.test command line.
Mike Bayer [Wed, 10 Apr 2019 15:51:27 +0000 (11:51 -0400)]
Don't use and_() inside of Query.filter_by
Adjusted the :meth:`.Query.filter_by` method to not call :func:`.and()`
internally against multiple criteria, instead passing it off to
:meth:`.Query.filter` as a series of criteria, instead of a single criteria.
This allows :meth:`.Query.filter_by` to defer to :meth:`.Query.filter`'s
treatment of variable numbers of clauses, including the case where the list
is empty. In this case, the :class:`.Query` object will not have a
``.whereclause``, which allows subsequent "no whereclause" methods like
:meth:`.Query.select_from` to behave consistently.
Mike Bayer [Tue, 9 Apr 2019 21:38:53 +0000 (17:38 -0400)]
Add __reduce_ex__ to MutableList; add compat for older pickles
Fixed bug where using ``copy.copy()`` or ``copy.deepcopy()`` on
:class:`.MutableList` would cause the items within the list to be
duplicated, due to an inconsistency in how Python pickle and copy both make
use of ``__getstate__()`` and ``__setstate__()`` regarding lists. In order
to resolve, a ``__reduce_ex__`` method had to be added to
:class:`.MutableList`. In order to maintain backwards compatibility with
existing pickles based on ``__getstate__()``, the ``__setstate__()`` method
remains as well; the test suite asserts that pickles made against the old
version of the class can still be deserialized by the pickle module.
Also modified sqlalchemy.testing.util.picklers to return picklers all the way through
pickle.HIGHEST_PROTOCOL.
Mike Bayer [Tue, 9 Apr 2019 13:16:16 +0000 (09:16 -0400)]
Propagate query-arg-only URL to psycopg2; don't send blank host
Fixed regression from release 1.3.2 caused by :ticket:`4562` where a URL
that contained only a query string and no hostname, such as for the
purposes of specifying a service file with connection information, would no
longer be propagated to psycopg2 properly. The change in :ticket:`4562`
has been adjusted to further suit psycopg2's exact requirements, which is
that if there are any connection parameters whatsoever, the "dsn" parameter
is no longer required, so in this case the query string parameters are
passed alone.
Mike Bayer [Fri, 5 Apr 2019 01:43:12 +0000 (21:43 -0400)]
Enhance documentation for string compilation use cases
- Add a web link for UnsupportedCompilationError
- Add new section to errors.rst
- add more detail and cross-linking to the FAQ
- include security caveats for parameter rendering
Matt Lewellyn [Wed, 3 Apr 2019 22:39:15 +0000 (18:39 -0400)]
MSSQL: only compile ORDER BY if it will be rendered
Fixed issue in SQL Server dialect where if a bound parameter were present in
an ORDER BY expression that would ultimately not be rendered in the SQL
Server version of the statement, the parameters would still be part of the
execution parameters, leading to DBAPI-level errors. Pull request courtesy
Matt Lewellyn.
Mike Bayer [Wed, 3 Apr 2019 14:03:17 +0000 (10:03 -0400)]
Restore use_threadlocal equivalent behavior to SingletonThreadPool
Fixed behavioral regression as a result of deprecating the "use_threadlocal"
flag for :class:`.Pool`, where the :class:`.SingletonThreadPool` no longer
makes use of this option which causes the "rollback on return" logic to take
place when the same :class:`.Engine` is used multiple times in the context
of a transaction to connect or implicitly execute, thereby cancelling the
transaction. While this is not the recommended way to work with engines
and connections, it is nonetheless a confusing behavioral change as when
using :class:`.SingletonThreadPool`, the transaction should stay open
regardless of what else is done with the same engine in the same thread.
The ``use_threadlocal`` flag remains deprecated however the
:class:`.SingletonThreadPool` now implements its own version of the same
logic.
Mike Bayer [Wed, 3 Apr 2019 02:12:09 +0000 (22:12 -0400)]
Fold entities into existing joins when resolving FROM ambiguity
Fixed 1.3 regression in new "ambiguous FROMs" query logic introduced in
:ref:`change_4365` where a :class:`.Query` that explicitly places an entity
in the FROM clause with :meth:`.Query.select_from` and also joins to it
using :meth:`.Query.join` would later cause an "ambiguous FROM" error if
that entity were used in additional joins, as the entity appears twice in
the "from" list of the :class:`.Query`. The fix resolves this ambiguity by
folding the standalone entity into the join that it's already a part of in
the same way that ultimately happens when the SELECT statement is rendered.
Mike Bayer [Fri, 29 Mar 2019 13:42:43 +0000 (09:42 -0400)]
Remove "subclass existing types" use case
Thanks to :ref:`change_3981`, we no longer need to rely on recipes that
subclass dialect-specific types directly, :class:`.TypeDecorator` can now
handle all cases. Additionally, the above change made it slightly less
likely that a direct subclass of a base SQLAlchemy type would work as
expected, which could be misleading. Documentation has been updated to use
:class:`.TypeDecorator` for these examples including the PostgreSQL
"ArrayOfEnum" example datatype and direct support for the "subclass a type
directly" has been removed.
Mike Bayer [Tue, 26 Mar 2019 16:58:42 +0000 (12:58 -0400)]
Refine ambiguous access for unknown attribute types
Restored instance-level support for plain Python descriptors, e.g.
``@property`` objects, in conjunction with association proxies, in that if
the proxied object is not within ORM scope at all, it gets classified as
"ambiguous" but is proxed directly. For class level access, a basic class
level``__get__()`` now returns the
:class:`.AmbiguousAssociationProxyInstance` directly, rather than raising
its exception, which is the closest approximation to the previous behavior
that returned the :class:`.AssociationProxy` itself that's possible. Also
improved the stringification of these objects to be more descriptive of
current state.
Mike Bayer [Tue, 26 Mar 2019 17:29:57 +0000 (13:29 -0400)]
Test double colons without escaping in text()
Apparently the BIND_PARAMS regex passes over double colons,
it just doesn't accommodate for a bound parameter in that case.
add this use case to current tests as people can be relying upon it.
Mike Bayer [Sun, 24 Mar 2019 02:05:22 +0000 (22:05 -0400)]
Fix boolean check in new path comparison logic
Fixed regression where a new error message that was supposed to raise when
attempting to link a relationship option to an AliasedClass without using
:meth:`.PropComparator.of_type` would instead raise an ``AttributeError``.
Note that in 1.3, it is no longer valid to create an option path from a
plain mapper relationship to an :class:`.AliasedClass` without using
:meth:`.PropComparator.of_type`.
Julian Mehnle [Thu, 21 Mar 2019 21:54:22 +0000 (17:54 -0400)]
Support DNS-less connections for psycopg2
Added support for parameter-less connection URLs for the psycopg2 dialect,
meaning, the URL can be passed to :func:`.create_engine` as
``"postgresql+psycopg2://"`` with no additional arguments to indicate an
empty DSN passed to libpq, which indicates to connect to "localhost" with
no username, password, or database given. Pull request courtesy Julian
Mehnle.
raylu [Wed, 20 Mar 2019 21:22:19 +0000 (17:22 -0400)]
Expand joins when calculating PostgreSQL "WITH FOR UPDATE OF"
Modified the :paramref:`.Select.with_for_update.of` parameter so that if a
join or other composed selectable is passed, the individual :class:`.Table`
objects will be filtered from it, allowing one to pass a join() object to
the parameter, as occurs normally when using joined table inheritance with
the ORM. Pull request courtesy Raymond Lu.
Mike Bayer [Mon, 11 Mar 2019 03:34:33 +0000 (23:34 -0400)]
Clone internals for Select._correlate_except collection as well as _correlate
Fixed bug where use of :func:`.with_polymorphic` or other aliased construct
would not properly adapt when the aliased target were used as the
:meth:`.Select.correlate_except` target of a subquery used inside of a
:func:`.column_property`. This required a fix to the clause adaption
mechanics to properly handle a selectable that shows up in the "correlate
except" list, in a similar manner as which occurs for selectables that show
up in the "correlate" list. This is ultimately a fairly fundamental bug
that has lasted for a long time but it is hard to come across it.
Mike Bayer [Fri, 8 Mar 2019 23:36:33 +0000 (18:36 -0500)]
Commit transaction after SNAPSHOT isolation change
A commit() is emitted after an isolation level change to SNAPSHOT, as both
pyodbc and pymssql open an implicit transaction which blocks subsequent SQL
from being emitted in the current transaction.
Mike Bayer [Fri, 8 Mar 2019 14:54:58 +0000 (09:54 -0500)]
Update the mysql-connector-python docs for now
mysql-connector-python still seems to have issues with binary datatypes
but the other issues here, with the possible exception of the deadlock
still being investigated, seem to have been resolved in the recent
8.x series of mysql-connector-python. The driver is still not
recommended due to these issues however we don't need a special
class of warnings at this point.
Fix large_resultsets performance example field name
Fixed bug in large_resultsets example case where a re-named "id" variable
due to code reformatting caused the test to fail. Pull request courtesy
Matt Schuchhardt.
Mike Bayer [Wed, 6 Mar 2019 14:05:23 +0000 (09:05 -0500)]
Ensure scale param not sent to float types
Fixed regression in SQL Server reflection due to :ticket:`4393` where the
removal of open-ended ``**kw`` from the :class:`.Float` datatype caused
reflection of this type to fail due to a "scale" argument being passed.
Sanjana [Thu, 28 Feb 2019 15:16:13 +0000 (10:16 -0500)]
Add port comparison in __eq__() and __ne__() method to URL
Comparing two objects of :class:`.URL` using ``__eq__()`` did not take port
number into consideration, two objects differing only by port number were
considered equal. Port comparison is now added in ``__eq__()`` method of
:class:`.URL`, objects differing by port number are now not equal.
Additionally, ``__ne__()`` was not implemented for :class:`.URL` which
caused unexpected result when ``!=`` was used in Python2, since there are no
implied relationships among the comparison operators in Python2.
Mike Bayer [Thu, 28 Feb 2019 15:59:41 +0000 (10:59 -0500)]
Add resolve_fks=False option for reflection
Added new parameters :paramref:`.Table.resolve_fks` and
:paramref:`.MetaData.reflect.resolve_fks` which when set to False will
disable the automatic reflection of related tables encountered in
:class:`.ForeignKey` objects, which can both reduce SQL overhead for omitted
tables as well as avoid tables that can't be reflected for database-specific
reasons. Two :class:`.Table` objects present in the same :class:`.MetaData`
collection can still refer to each other even if the reflection of the two
tables occurred separately.
Toni Karppi [Thu, 21 Feb 2019 19:22:48 +0000 (14:22 -0500)]
Add test to as_interface when no cls or methods
Fixes #4511
### Description
The `as_interface` function requires that an argument be provided for at least one of the parameters `cls` or `methods`. There was no test that tested the case when neither of these arguments were provided.
### Checklist
<!-- go over following points. check them with an `x` if they do apply, (they turn into clickable checkboxes once the PR is submitted, so no need to do everything at once)
-->
This pull request is:
- [ ] A documentation / typographical error fix
- Good to go, no issue or tests are needed
- [x] A short code fix
- please include the issue number, and create an issue if none exists, which
must include a complete example of the issue. one line code fixes without an
issue and demonstration will not be accepted.
- Please include: `Fixes: #<issue number>` in the commit message
- please include tests. one line code fixes without tests will not be accepted.
- [ ] A new feature implementation
- please include the issue number, and create an issue if none exists, which must
include a complete example of how the feature would look.
- Please include: `Fixes: #<issue number>` in the commit message
- please include tests.
Mike Bayer [Wed, 20 Feb 2019 20:00:09 +0000 (15:00 -0500)]
Prevent __init__ from being called for Alias, subclasses
The :class:`.Alias` class and related subclasses :class:`.CTE`,
:class:`.Lateral` and :class:`.TableSample` have been reworked so that it is
not possible for a user to construct the objects directly. These constructs
require that the standalone construction function or selectable-bound method
be used to instantiate new objects.
Nate Clark [Wed, 20 Feb 2019 17:58:18 +0000 (12:58 -0500)]
Include newlines in StatementError formatting
Revised the formatting for :class:`.StatementError` when stringified. Each
error detail is broken up over multiple newlines instead of spaced out on a
single line. Additionally, the SQL representation now stringifies the SQL
statement rather than using ``repr()``, so that newlines are rendered as is.
Pull request courtesy Nate Clark.
sanjana [Wed, 20 Feb 2019 04:07:12 +0000 (23:07 -0500)]
Add support for key-word based get()
The :meth:`.Query.get` method can now accept a dictionary of attribute keys
and values as a means of indicating the primary key value to load; is
particularly useful for composite primary keys. Pull request courtesy
Sanjana S.
Mike Bayer [Wed, 20 Feb 2019 00:46:17 +0000 (19:46 -0500)]
Ensure _simple_lazy_clause bind names are fixed before cloning
Fixed a regression in 1.2 due to the introduction of baked queries for
relationship lazy loaders, where a race condition is created during the
generation of the "lazy clause" which occurs within a memoized attribute. If
two threads initialize the memoized attribute concurrently, the baked query
could be generated with bind parameter keys that are then replaced with new
keys by the next run, leading to a lazy load query that specifies the
related criteria as ``None``. The fix establishes that the parameter names
are fixed before the new clause and parameter objects are generated, so that
the names are the same every time.
Mike Bayer [Tue, 19 Feb 2019 17:13:04 +0000 (12:13 -0500)]
Add NCHAR to Oracle dialect
Added support for reflection of the :class:`.NCHAR` datatype to the Oracle
dialect, and added :class:`.NCHAR` to the list of types exported by the
Oracle dialect.
sanjana [Wed, 13 Feb 2019 15:17:46 +0000 (10:17 -0500)]
Adding setter to should_evaluate_none property
Fixed issue where the :class:`.JSON` type had a read-only
:attr:`.JSON.should_evaluate_none` attribute, which would cause failures
when making use of the :meth:`.TypeEngine.evaluates_none` method in
conjunction with this type. Pull request courtesy Sanjana S.
Mike Bayer [Wed, 13 Feb 2019 16:26:54 +0000 (11:26 -0500)]
Set IDENTITY_INSERT for insert.values({column: expr})
Fixed bug where the SQL Server "IDENTITY_INSERT" logic that allows an INSERT
to proceed with an explicit value on an IDENTITY column was not detecting
the case where :meth:`.Insert.values` were used with a dictionary that
contained a :class:`.Column` as key and a SQL expression as a value.
Mike Bayer [Mon, 11 Feb 2019 22:00:47 +0000 (17:00 -0500)]
Allow SQL expression for ORM primary keys
A SQL expression can now be assigned to a primary key attribute for an ORM
flush in the same manner as ordinary attributes as described in
:ref:`flush_embedded_sql_expressions` where the expression will be evaulated
and then returned to the ORM using RETURNING, or in the case of pysqlite,
works using the cursor.lastrowid attribute.Requires either a database that
supports RETURNING (e.g. Postgresql, Oracle, SQL Server) or pysqlite.
Mike Bayer [Mon, 11 Feb 2019 14:07:03 +0000 (09:07 -0500)]
Add complete coverage and fix lower() for MySQL 88718 workaround
Fixed a second regression caused by :ticket:`4344` (the first was
:ticket:`4361`), which works around MySQL issue 88718, where the lower
casing function used was not correct for Python 2 with OSX/Windows casing
conventions, which would then raise ``TypeError``. Full coverage has been
added to this logic so that every codepath is exercised in a mock style for
all three casing conventions on all versions of Python. MySQL 8.0 has
meanwhile fixed issue 88718 so the workaround is only applies to a
particular span of MySQL 8.0 versions.
Mike Bayer [Sat, 9 Feb 2019 06:46:06 +0000 (01:46 -0500)]
Don't run pending_to_persistent for non-new objects
Fixed fairly simple but critical issue where the
:meth:`.SessionEvents.pending_to_persistent` event would be invoked for
objects not just when they move from pending to persistent, but when they
were also already persistent and just being updated, thus causing the event
to be invoked for all objects on every update.
Ivan Levkivskyi [Fri, 8 Feb 2019 19:10:19 +0000 (19:10 +0000)]
Add type stubs link to the docs
Moved by Mike to be a sidebar in the Declarative docs, as
the plugin appears to be specific to Declarative. These docs will
eventually be moved out of the "extension" section and merged
into the primary narrative documentation.
Co-authored-by: Mike Bayer <mike_mp@zzzcomputing.com>
Change-Id: Idba4c13ffc920b1140b5e51d206ab02f19015e8e
Mike Bayer [Mon, 4 Feb 2019 20:50:29 +0000 (15:50 -0500)]
Remove all remaining text() coercions and ensure identifiers are safe
Fully removed the behavior of strings passed directly as components of a
:func:`.select` or :class:`.Query` object being coerced to :func:`.text`
constructs automatically; the warning that has been emitted is now an
ArgumentError or in the case of order_by() / group_by() a CompileError.
This has emitted a warning since version 1.0 however its presence continues
to create concerns for the potential of mis-use of this behavior.
Note that public CVEs have been posted for order_by() / group_by() which
are resolved by this commit: CVE-2019-7164 CVE-2019-7548
Added "SQL phrase validation" to key DDL phrases that are accepted as plain
strings, including :paramref:`.ForeignKeyConstraint.on_delete`,
:paramref:`.ForeignKeyConstraint.on_update`,
:paramref:`.ExcludeConstraint.using`,
:paramref:`.ForeignKeyConstraint.initially`, for areas where a series of SQL
keywords only are expected.Any non-space characters that suggest the phrase
would need to be quoted will raise a :class:`.CompileError`. This change
is related to the series of changes committed as part of :ticket:`4481`.
Fixed issue where using an uppercase name for an index type (e.g. GIST,
BTREE, etc. ) or an EXCLUDE constraint would treat it as an identifier to
be quoted, rather than rendering it as is. The new behavior converts these
types to lowercase and ensures they contain only valid SQL characters.
Quoting is applied to :class:`.Function` names, those which are usually but
not necessarily generated from the :attr:`.sql.func` construct, at compile
time if they contain illegal characters, such as spaces or punctuation. The
names are as before treated as case insensitive however, meaning if the
names contain uppercase or mixed case characters, that alone does not
trigger quoting. The case insensitivity is currently maintained for
backwards compatibility.
projects / thirdparty / sqlalchemy / sqlalchemy.git / log
