simple but unusual system allows for a dramatic savings in Python
overhead for the construction and processing of orm :class:`.Query`
objects, from query construction up through rendering of a string
SQL statement.
fixes#3054
become an externally usable package but still remains within the main sqlalchemy parent package.
in this system, we use kind of an ugly hack to get the noseplugin imported outside of the
"sqlalchemy" package, while still making it available within sqlalchemy for usage by
third party libraries.
inheriting mappers to be GC'ed when the
class itself is deferenced. The mapper
must not have its own table (i.e.
single table inh only) without polymorphic
attributes in place.
This allows for the use case of
creating a temporary subclass of a declarative
mapped class, with no table or mapping
directives of its own, to be garbage collected
when dereferenced by a unit test.
[ticket:2526]
now accepts aliased() class constructs as well
as with_polymorphic constructs, and works with
query.join(), any(), has(), and also
eager loaders subqueryload(), joinedload(),
contains_eager()
[ticket:2438] [ticket:1106]
- a rewrite of the query path system to use an
object based approach for more succinct usage. the system
has been designed carefully to not add an excessive method overhead.
- [feature] select() features a correlate_except()
method, auto correlates all selectables except those
passed. Is needed here for the updated any()/has()
functionality.
- remove some old cruft from LoaderStrategy, init(),debug_callable()
- use a namedtuple for _extended_entity_info. This method should
become standard within the orm internals
- some tweaks to the memory profile tests, number of runs can
be customized to work around pysqlite's very annoying behavior
- try to simplify PropertyOption._get_paths(), rename to _process_paths(),
returns a single list now. overall works more completely as was needed
for of_type() functionality
all referncing tests to not use globals
- tests that deal with pickle specifically load the fixture classes
from test.lib.pickleable, which gets some more classes added
- removed weird sa05 pickling tests that don't matter
- add --with-cdecimal flag to tests, monkeypatches cdecimal in
- fix mssql/pyodbc.py to not use private '_int' accessor in decimal conversion
routines
- pyodbc version 2.1.8 is needed for cdecimal in any case as
previous versions also called '_int', 2.1.8 adds the same string
logic as our own dialect, so that logic is skipped for modern
pyodbc version
- make the imports for "Decimal" consistent across the whole lib. not sure
yet how we should be importing "Decimal" or what the best way forward
is that would allow a clean user-invoked swap of cdecimal; for now,
added docs suggesting a global monkeypatch - the two decimal libs
are not compatible with each other so any chance of mixing produces
serious issues. adding adapters to DBAPIs tedious and adds in-python
overhead. suggestions welcome on how we should be doing
Decimal/cdecimal.
- Added "map_to()" method to SqlSoup, which is a "master"
method which accepts explicit arguments for each aspect of
the selectable and mapping, including a base class per
mapping. [ticket:1975]
- Mapped selectables used with the map(), with_labels(),
join() methods no longer put the given argument into the
internal "cache" dictionary. Particularly since the
join() and select() objects are created in the method
itself this was pretty much a pure memory leaking behavior.
- with m2m we have to go back to the previous approach of having both sides of
the DP fire off, tracking each pair of objects. history may not be consistently present
in one side or the other
- this revealed a whole lot of issues with self-referential m2m, which are fixed
- one-to-many relationships now maintain a list of positive
parent-child associations within the flush, preventing
previous parents marked as deleted from cascading a
delete or NULL foreign key set on those child objects,
despite the end-user not removing the child from the old
association. [ticket:1764]
- re-established Preprocess as unique on their arguments,
as they were definitely duped in inheritance scenarios
- added a "memo" feature to UOWTransaction which represents the usual
pattern of using the .attributes collection
- added the test case from [ticket:1081] into perf/
loading available, the new names for eagerload() and
eagerload_all() are joinedload() and joinedload_all(). The
old names will remain as synonyms for the foreseeable future.
- The "lazy" flag on the relationship() function now accepts
a string argument for all kinds of loading: "select", "joined",
"subquery", "noload" and "dynamic", where the default is now
"select". The old values of True/
False/None still retain their usual meanings and will remain
as synonyms for the foreseeable future.
- Added documentation to tutorial,mapper doc, api docs
for subqueryload, subqueryload_all, and other options.
relationship(), to eliminate confusion over the relational
algebra term. relation() however will remain available
in equal capacity for the foreseeable future. [ticket:1740]
reimplementing the highest impact functions.
The actual speedups will depend heavily on your DBAPI and
the mix of datatypes used in your tables, and can vary from
a 50% improvement to more than 200%. It also provides a modest
(~20%) indirect improvement to ORM speed for large queries.
Note that it is *not* built/installed by default.
See README for installation instructions.
- The most common result processors conversion function were
moved to the new "processors" module. Dialect authors are
encouraged to use those functions whenever they correspond
to their needs instead of implementing custom ones.
the version called by column loaders on an incomplete row (i.e.
joined table inheritance). there are more dramatic changes
that can be made here but this one is conservative so far
as far as how much we're altering how InstanceState tracks
"expired" attributes.
sqlalchemy.orm.strategies, most logging calls during row
loading have been removed. These were never very helpful
and cluttered up the code.
- Some internal streamlining of object loading grants a
small speedup for large results, estimates are around
10-15%.
- Importing testenv has no side effects- explicit functions provide similar behavior to the old immediate behavior of testbase
- testing.db has the configured db
- Fixed up the perf/* scripts
Khairi Hafsham