d5f390f55c
Check explicitly for the cases where no "sql_connection" attribute was set when running _start(), so that the lack of this parameter is documented by the exception rather than failing into create_engine() with an unclear failure mode. If _start() fails as it will here, make sure _started stays False so that repeated calls to _start() continue to raise the same exception, rather than raising attribute errors. When accessing the "session" or "connection" attributes of the context when these attributes were not requested by the decorator or context manager, raise explicit exceptions for each, rather than returning None which leads to hard-to-debug NoneType errors. Change-Id: Iadfbf4707daed4140285a3a472009f6863b18275 Closes-bug: 1477080
3.5 KiB
Usage
To use oslo.db in a project:
Session Handling
Session handling is achieved using the oslo_db.sqlalchemy.enginefacade system. This module
presents a function decorator as well as a context manager approach to
delivering .Session
as well as .Connection objects to a function or block.
Both calling styles require the use of a context object. This object may be of any class, though when used with the decorator form, requires special instrumentation.
The context manager form is as follows:
from oslo_db.sqlalchemy import enginefacade
class MyContext(object):
"User-defined context class."
def some_reader_api_function(context):
with enginefacade.reader.using(context) as session:
return session.query(SomeClass).all()
def some_writer_api_function(context, x, y):
with enginefacade.writer.using(context) as session:
session.add(SomeClass(x, y))
def run_some_database_calls():
context = MyContext()
results = some_reader_api_function(context)
some_writer_api_function(context, 5, 10)The decorator form accesses attributes off the user-defined context
directly; the context must be decorated with the oslo_db.sqlalchemy.enginefacade.transaction_context_provider
decorator. Each function must receive the context as the first
positional argument:
from oslo_db.sqlalchemy import enginefacade
@enginefacade.transaction_context_provider
class MyContext(object):
"User-defined context class."
@enginefacade.reader
def some_reader_api_function(context):
return context.session.query(SomeClass).all()
@enginefacade.writer
def some_writer_api_function(context, x, y):
context.session.add(SomeClass(x, y))
def run_some_database_calls():
context = MyContext()
results = some_reader_api_function(context)
some_writer_api_function(context, 5, 10)Note
The context.session and context.connection
attributes must be accessed within the scope of an appropriate
writer/reader block (either the decorator or contextmanager approach).
An AttributeError is raised otherwise.
The scope of transaction and connectivity for both approaches is
managed transparently. The configuration for the connection comes from
the standard oslo_config.cfg.CONF collection. Additional
configurations can be established for the enginefacade using the oslo_db.sqlalchemy.enginefacade.configure function,
before any use of the database begins:
from oslo_db.sqlalchemy import enginefacade
enginefacade.configure(
sqlite_fk=True,
max_retries=5,
mysql_sql_mode='ANSI'
)Base class for models usage
from oslo.db import models
class ProjectSomething(models.TimestampMixin,
models.ModelBase):
id = Column(Integer, primary_key=True)
...DB API backend support
from oslo.config import cfg
from oslo.db import api as db_api
_BACKEND_MAPPING = {'sqlalchemy': 'project.db.sqlalchemy.api'}
IMPL = db_api.DBAPI.from_config(cfg.CONF, backend_mapping=_BACKEND_MAPPING)
def get_engine():
return IMPL.get_engine()
def get_session():
return IMPL.get_session()
# DB-API method
def do_something(somethind_id):
return IMPL.do_something(somethind_id)