Split out manager

git-svn-id: http://code.djangoproject.com/svn/django/branches/magic-removal@1695 bcc190cf-cafb-0310-a4f2-bffc1f526a37
This commit is contained in:
Robert Wittams 2005-12-16 13:00:13 +00:00
parent e5b3fd474b
commit 350e8a3523
3 changed files with 250 additions and 239 deletions

View File

@ -9,14 +9,12 @@ import copy, datetime, os, re, sys, types
from django.db.models.loading import get_installed_models, get_installed_model_modules from django.db.models.loading import get_installed_models, get_installed_model_modules
from django.db.models.manipulators import ManipulatorDescriptor, ModelAddManipulator, ModelChangeManipulator from django.db.models.manipulators import ManipulatorDescriptor, ModelAddManipulator, ModelChangeManipulator
from django.db.models.query import Q, parse_lookup, get_where_clause, get_cached_row, fill_table_cache, throw_bad_kwarg_error from django.db.models.query import Q, orderlist2sql
from django.db.models.manager import Manager
# Admin stages. # Admin stages.
ADD, CHANGE, BOTH = 1, 2, 3 ADD, CHANGE, BOTH = 1, 2, 3
# Size of each "chunk" for get_iterator calls.
# Larger values are slightly faster at the expense of more storage space.
GET_ITERATOR_CHUNK_SIZE = 100
# Prefix (in Python path style) to location of models. # Prefix (in Python path style) to location of models.
MODEL_PREFIX = 'django.models' MODEL_PREFIX = 'django.models'
@ -31,47 +29,7 @@ MANIPULATOR_FUNCTIONS_PREFIX = '_manipulator_'
####################
# HELPER FUNCTIONS #
####################
# Django currently supports two forms of ordering.
# Form 1 (deprecated) example:
# order_by=(('pub_date', 'DESC'), ('headline', 'ASC'), (None, 'RANDOM'))
# Form 2 (new-style) example:
# order_by=('-pub_date', 'headline', '?')
# Form 1 is deprecated and will no longer be supported for Django's first
# official release. The following code converts from Form 1 to Form 2.
LEGACY_ORDERING_MAPPING = {'ASC': '_', 'DESC': '-_', 'RANDOM': '?'}
def handle_legacy_orderlist(order_list):
if not order_list or isinstance(order_list[0], basestring):
return order_list
else:
import warnings
new_order_list = [LEGACY_ORDERING_MAPPING[j.upper()].replace('_', str(i)) for i, j in order_list]
warnings.warn("%r ordering syntax is deprecated. Use %r instead." % (order_list, new_order_list), DeprecationWarning)
return new_order_list
def orderfield2column(f, opts):
try:
return opts.get_field(f, False).column
except FieldDoesNotExist:
return f
def orderlist2sql(order_list, opts, prefix=''):
if prefix.endswith('.'):
prefix = backend.quote_name(prefix[:-1]) + '.'
output = []
for f in handle_legacy_orderlist(order_list):
if f.startswith('-'):
output.append('%s%s DESC' % (prefix, backend.quote_name(orderfield2column(f[1:], opts))))
elif f == '?':
output.append(backend.get_random_function_sql())
else:
output.append('%s%s ASC' % (prefix, backend.quote_name(orderfield2column(f, opts))))
return ', '.join(output)
#def get_module(app_label, module_name): #def get_module(app_label, module_name):
# return __import__('%s.%s.%s' % (MODEL_PREFIX, app_label, module_name), '', '', ['']) # return __import__('%s.%s.%s' % (MODEL_PREFIX, app_label, module_name), '', '', [''])
@ -440,201 +398,6 @@ get_module_name = lambda class_name: class_name.lower() + 's'
# Calculate the verbose_name by converting from InitialCaps to "lowercase with spaces". # Calculate the verbose_name by converting from InitialCaps to "lowercase with spaces".
get_verbose_name = lambda class_name: re.sub('([A-Z])', ' \\1', class_name).lower().strip() get_verbose_name = lambda class_name: re.sub('([A-Z])', ' \\1', class_name).lower().strip()
class Manager(object):
# Tracks each time a Field instance is created. Used to retain order.
creation_counter = 0
def __init__(self):
# Increase the creation counter, and save our local copy.
self.creation_counter = Manager.creation_counter
Manager.creation_counter += 1
def __get__(self, instance, type=None):
if instance != None:
raise AttributeError, "Manager isn't accessible via %s instances" % self.klass.__name__
return self
def _prepare(self, klass):
# Creates some methods once self.klass._meta has been populated.
self.klass = klass
if self.klass._meta.get_latest_by:
self.get_latest = self.__get_latest
for f in self.klass._meta.fields:
if isinstance(f, DateField):
setattr(self, 'get_%s_list' % f.name, curry(self.__get_date_list, f))
def _get_sql_clause(self, **kwargs):
def quote_only_if_word(word):
if ' ' in word:
return word
else:
return backend.quote_name(word)
opts = self.klass._meta
# Construct the fundamental parts of the query: SELECT X FROM Y WHERE Z.
select = ["%s.%s" % (backend.quote_name(opts.db_table), backend.quote_name(f.column)) for f in opts.fields]
tables = [opts.db_table] + (kwargs.get('tables') and kwargs['tables'][:] or [])
tables = [quote_only_if_word(t) for t in tables]
where = kwargs.get('where') and kwargs['where'][:] or []
params = kwargs.get('params') and kwargs['params'][:] or []
# Convert the kwargs into SQL.
tables2, join_where2, where2, params2, _ = parse_lookup(kwargs.items(), opts)
tables.extend(tables2)
where.extend(join_where2 + where2)
params.extend(params2)
# Add any additional constraints from the "where_constraints" parameter.
where.extend(opts.where_constraints)
# Add additional tables and WHERE clauses based on select_related.
if kwargs.get('select_related') is True:
fill_table_cache(opts, select, tables, where, opts.db_table, [opts.db_table])
# Add any additional SELECTs passed in via kwargs.
if kwargs.get('select'):
select.extend(['(%s) AS %s' % (quote_only_if_word(s[1]), backend.quote_name(s[0])) for s in kwargs['select']])
# ORDER BY clause
order_by = []
for f in handle_legacy_orderlist(kwargs.get('order_by', opts.ordering)):
if f == '?': # Special case.
order_by.append(backend.get_random_function_sql())
else:
if f.startswith('-'):
col_name = f[1:]
order = "DESC"
else:
col_name = f
order = "ASC"
if "." in col_name:
table_prefix, col_name = col_name.split('.', 1)
table_prefix = backend.quote_name(table_prefix) + '.'
else:
# Use the database table as a column prefix if it wasn't given,
# and if the requested column isn't a custom SELECT.
if "." not in col_name and col_name not in [k[0] for k in kwargs.get('select', [])]:
table_prefix = backend.quote_name(opts.db_table) + '.'
else:
table_prefix = ''
order_by.append('%s%s %s' % (table_prefix, backend.quote_name(orderfield2column(col_name, opts)), order))
order_by = ", ".join(order_by)
# LIMIT and OFFSET clauses
if kwargs.get('limit') is not None:
limit_sql = " %s " % backend.get_limit_offset_sql(kwargs['limit'], kwargs.get('offset'))
else:
assert kwargs.get('offset') is None, "'offset' is not allowed without 'limit'"
limit_sql = ""
return select, " FROM " + ",".join(tables) + (where and " WHERE " + " AND ".join(where) or "") + (order_by and " ORDER BY " + order_by or "") + limit_sql, params
def get_iterator(self, **kwargs):
# kwargs['select'] is a dictionary, and dictionaries' key order is
# undefined, so we convert it to a list of tuples internally.
kwargs['select'] = kwargs.get('select', {}).items()
cursor = connection.cursor()
select, sql, params = self._get_sql_clause(**kwargs)
cursor.execute("SELECT " + (kwargs.get('distinct') and "DISTINCT " or "") + ",".join(select) + sql, params)
fill_cache = kwargs.get('select_related')
index_end = len(self.klass._meta.fields)
while 1:
rows = cursor.fetchmany(GET_ITERATOR_CHUNK_SIZE)
if not rows:
raise StopIteration
for row in rows:
if fill_cache:
obj, index_end = get_cached_row(self.klass, row, 0)
else:
obj = self.klass(*row[:index_end])
for i, k in enumerate(kwargs['select']):
setattr(obj, k[0], row[index_end+i])
yield obj
def get_list(self, **kwargs):
return list(self.get_iterator(**kwargs))
def get_count(self, **kwargs):
kwargs['order_by'] = []
kwargs['offset'] = None
kwargs['limit'] = None
kwargs['select_related'] = False
_, sql, params = self._get_sql_clause(**kwargs)
cursor = connection.cursor()
cursor.execute("SELECT COUNT(*)" + sql, params)
return cursor.fetchone()[0]
def get_object(self, **kwargs):
obj_list = self.get_list(**kwargs)
if len(obj_list) < 1:
raise self.klass.DoesNotExist, "%s does not exist for %s" % (self.klass._meta.object_name, kwargs)
assert len(obj_list) == 1, "get_object() returned more than one %s -- it returned %s! Lookup parameters were %s" % (self.klass._meta.object_name, len(obj_list), kwargs)
return obj_list[0]
def get_in_bulk(self, *args, **kwargs):
id_list = args and args[0] or kwargs['id_list']
assert id_list != [], "get_in_bulk() cannot be passed an empty list."
kwargs['where'] = ["%s.%s IN (%s)" % (backend.quote_name(self.klass._meta.db_table), backend.quote_name(self.klass._meta.pk.column), ",".join(['%s'] * len(id_list)))]
kwargs['params'] = id_list
obj_list = self.get_list(**kwargs)
return dict([(getattr(o, self.klass._meta.pk.attname), o) for o in obj_list])
def get_values_iterator(self, **kwargs):
# select_related and select aren't supported in get_values().
kwargs['select_related'] = False
kwargs['select'] = {}
# 'fields' is a list of field names to fetch.
try:
fields = [self.klass._meta.get_field(f).column for f in kwargs.pop('fields')]
except KeyError: # Default to all fields.
fields = [f.column for f in self.klass._meta.fields]
cursor = connection.cursor()
_, sql, params = self._get_sql_clause(**kwargs)
select = ['%s.%s' % (backend.quote_name(self.klass._meta.db_table), backend.quote_name(f)) for f in fields]
cursor.execute("SELECT " + (kwargs.get('distinct') and "DISTINCT " or "") + ",".join(select) + sql, params)
while 1:
rows = cursor.fetchmany(GET_ITERATOR_CHUNK_SIZE)
if not rows:
raise StopIteration
for row in rows:
yield dict(zip(fields, row))
def get_values(self, **kwargs):
return list(self.get_values_iterator(**kwargs))
def __get_latest(self, **kwargs):
kwargs['order_by'] = ('-' + self.klass._meta.get_latest_by,)
kwargs['limit'] = 1
return self.get_object(**kwargs)
def __get_date_list(self, field, *args, **kwargs):
from django.db.backends.util import typecast_timestamp
kind = args and args[0] or kwargs['kind']
assert kind in ("month", "year", "day"), "'kind' must be one of 'year', 'month' or 'day'."
order = 'ASC'
if kwargs.has_key('order'):
order = kwargs['order']
del kwargs['order']
assert order in ('ASC', 'DESC'), "'order' must be either 'ASC' or 'DESC'"
kwargs['order_by'] = () # Clear this because it'll mess things up otherwise.
if field.null:
kwargs.setdefault('where', []).append('%s.%s IS NOT NULL' % \
(backend.quote_name(self.klass._meta.db_table), backend.quote_name(field.column)))
select, sql, params = self._get_sql_clause(**kwargs)
sql = 'SELECT %s %s GROUP BY 1 ORDER BY 1 %s' % \
(backend.get_date_trunc_sql(kind, '%s.%s' % (backend.quote_name(self.klass._meta.db_table),
backend.quote_name(field.column))), sql, order)
cursor = connection.cursor()
cursor.execute(sql, params)
# We have to manually run typecast_timestamp(str()) on the results, because
# MySQL doesn't automatically cast the result of date functions as datetime
# objects -- MySQL returns the values as strings, instead.
return [typecast_timestamp(str(row[0])) for row in cursor.fetchall()]

207
django/db/models/manager.py Normal file
View File

@ -0,0 +1,207 @@
from django.db.models.fields import DateField
from django.utils.functional import curry
from django.db import backend, connection
from django.db.models.query import Q, parse_lookup, fill_table_cache, get_cached_row
from django.db.models.query import handle_legacy_orderlist, orderlist2sql, orderfield2column
# Size of each "chunk" for get_iterator calls.
# Larger values are slightly faster at the expense of more storage space.
GET_ITERATOR_CHUNK_SIZE = 100
class Manager(object):
# Tracks each time a Field instance is created. Used to retain order.
creation_counter = 0
def __init__(self):
# Increase the creation counter, and save our local copy.
self.creation_counter = Manager.creation_counter
Manager.creation_counter += 1
def __get__(self, instance, type=None):
if instance != None:
raise AttributeError, "Manager isn't accessible via %s instances" % self.klass.__name__
return self
def _prepare(self, klass):
# Creates some methods once self.klass._meta has been populated.
self.klass = klass
if self.klass._meta.get_latest_by:
self.get_latest = self.__get_latest
for f in self.klass._meta.fields:
if isinstance(f, DateField):
setattr(self, 'get_%s_list' % f.name, curry(self.__get_date_list, f))
def _get_sql_clause(self, **kwargs):
def quote_only_if_word(word):
if ' ' in word:
return word
else:
return backend.quote_name(word)
opts = self.klass._meta
# Construct the fundamental parts of the query: SELECT X FROM Y WHERE Z.
select = ["%s.%s" % (backend.quote_name(opts.db_table), backend.quote_name(f.column)) for f in opts.fields]
tables = [opts.db_table] + (kwargs.get('tables') and kwargs['tables'][:] or [])
tables = [quote_only_if_word(t) for t in tables]
where = kwargs.get('where') and kwargs['where'][:] or []
params = kwargs.get('params') and kwargs['params'][:] or []
# Convert the kwargs into SQL.
tables2, join_where2, where2, params2, _ = parse_lookup(kwargs.items(), opts)
tables.extend(tables2)
where.extend(join_where2 + where2)
params.extend(params2)
# Add any additional constraints from the "where_constraints" parameter.
where.extend(opts.where_constraints)
# Add additional tables and WHERE clauses based on select_related.
if kwargs.get('select_related') is True:
fill_table_cache(opts, select, tables, where, opts.db_table, [opts.db_table])
# Add any additional SELECTs passed in via kwargs.
if kwargs.get('select'):
select.extend(['(%s) AS %s' % (quote_only_if_word(s[1]), backend.quote_name(s[0])) for s in kwargs['select']])
# ORDER BY clause
order_by = []
for f in handle_legacy_orderlist(kwargs.get('order_by', opts.ordering)):
if f == '?': # Special case.
order_by.append(backend.get_random_function_sql())
else:
if f.startswith('-'):
col_name = f[1:]
order = "DESC"
else:
col_name = f
order = "ASC"
if "." in col_name:
table_prefix, col_name = col_name.split('.', 1)
table_prefix = backend.quote_name(table_prefix) + '.'
else:
# Use the database table as a column prefix if it wasn't given,
# and if the requested column isn't a custom SELECT.
if "." not in col_name and col_name not in [k[0] for k in kwargs.get('select', [])]:
table_prefix = backend.quote_name(opts.db_table) + '.'
else:
table_prefix = ''
order_by.append('%s%s %s' % (table_prefix, backend.quote_name(orderfield2column(col_name, opts)), order))
order_by = ", ".join(order_by)
# LIMIT and OFFSET clauses
if kwargs.get('limit') is not None:
limit_sql = " %s " % backend.get_limit_offset_sql(kwargs['limit'], kwargs.get('offset'))
else:
assert kwargs.get('offset') is None, "'offset' is not allowed without 'limit'"
limit_sql = ""
return select, " FROM " + ",".join(tables) + (where and " WHERE " + " AND ".join(where) or "") + (order_by and " ORDER BY " + order_by or "") + limit_sql, params
def get_iterator(self, **kwargs):
# kwargs['select'] is a dictionary, and dictionaries' key order is
# undefined, so we convert it to a list of tuples internally.
kwargs['select'] = kwargs.get('select', {}).items()
cursor = connection.cursor()
select, sql, params = self._get_sql_clause(**kwargs)
cursor.execute("SELECT " + (kwargs.get('distinct') and "DISTINCT " or "") + ",".join(select) + sql, params)
fill_cache = kwargs.get('select_related')
index_end = len(self.klass._meta.fields)
while 1:
rows = cursor.fetchmany(GET_ITERATOR_CHUNK_SIZE)
if not rows:
raise StopIteration
for row in rows:
if fill_cache:
obj, index_end = get_cached_row(self.klass, row, 0)
else:
obj = self.klass(*row[:index_end])
for i, k in enumerate(kwargs['select']):
setattr(obj, k[0], row[index_end+i])
yield obj
def get_list(self, **kwargs):
return list(self.get_iterator(**kwargs))
def get_count(self, **kwargs):
kwargs['order_by'] = []
kwargs['offset'] = None
kwargs['limit'] = None
kwargs['select_related'] = False
_, sql, params = self._get_sql_clause(**kwargs)
cursor = connection.cursor()
cursor.execute("SELECT COUNT(*)" + sql, params)
return cursor.fetchone()[0]
def get_object(self, **kwargs):
obj_list = self.get_list(**kwargs)
if len(obj_list) < 1:
raise self.klass.DoesNotExist, "%s does not exist for %s" % (self.klass._meta.object_name, kwargs)
assert len(obj_list) == 1, "get_object() returned more than one %s -- it returned %s! Lookup parameters were %s" % (self.klass._meta.object_name, len(obj_list), kwargs)
return obj_list[0]
def get_in_bulk(self, *args, **kwargs):
id_list = args and args[0] or kwargs['id_list']
assert id_list != [], "get_in_bulk() cannot be passed an empty list."
kwargs['where'] = ["%s.%s IN (%s)" % (backend.quote_name(self.klass._meta.db_table), backend.quote_name(self.klass._meta.pk.column), ",".join(['%s'] * len(id_list)))]
kwargs['params'] = id_list
obj_list = self.get_list(**kwargs)
return dict([(getattr(o, self.klass._meta.pk.attname), o) for o in obj_list])
def get_values_iterator(self, **kwargs):
# select_related and select aren't supported in get_values().
kwargs['select_related'] = False
kwargs['select'] = {}
# 'fields' is a list of field names to fetch.
try:
fields = [self.klass._meta.get_field(f).column for f in kwargs.pop('fields')]
except KeyError: # Default to all fields.
fields = [f.column for f in self.klass._meta.fields]
cursor = connection.cursor()
_, sql, params = self._get_sql_clause(**kwargs)
select = ['%s.%s' % (backend.quote_name(self.klass._meta.db_table), backend.quote_name(f)) for f in fields]
cursor.execute("SELECT " + (kwargs.get('distinct') and "DISTINCT " or "") + ",".join(select) + sql, params)
while 1:
rows = cursor.fetchmany(GET_ITERATOR_CHUNK_SIZE)
if not rows:
raise StopIteration
for row in rows:
yield dict(zip(fields, row))
def get_values(self, **kwargs):
return list(self.get_values_iterator(**kwargs))
def __get_latest(self, **kwargs):
kwargs['order_by'] = ('-' + self.klass._meta.get_latest_by,)
kwargs['limit'] = 1
return self.get_object(**kwargs)
def __get_date_list(self, field, *args, **kwargs):
from django.db.backends.util import typecast_timestamp
kind = args and args[0] or kwargs['kind']
assert kind in ("month", "year", "day"), "'kind' must be one of 'year', 'month' or 'day'."
order = 'ASC'
if kwargs.has_key('order'):
order = kwargs['order']
del kwargs['order']
assert order in ('ASC', 'DESC'), "'order' must be either 'ASC' or 'DESC'"
kwargs['order_by'] = () # Clear this because it'll mess things up otherwise.
if field.null:
kwargs.setdefault('where', []).append('%s.%s IS NOT NULL' % \
(backend.quote_name(self.klass._meta.db_table), backend.quote_name(field.column)))
select, sql, params = self._get_sql_clause(**kwargs)
sql = 'SELECT %s %s GROUP BY 1 ORDER BY 1 %s' % \
(backend.get_date_trunc_sql(kind, '%s.%s' % (backend.quote_name(self.klass._meta.db_table),
backend.quote_name(field.column))), sql, order)
cursor = connection.cursor()
cursor.execute(sql, params)
# We have to manually run typecast_timestamp(str()) on the results, because
# MySQL doesn't automatically cast the result of date functions as datetime
# objects -- MySQL returns the values as strings, instead.
return [typecast_timestamp(str(row[0])) for row in cursor.fetchall()]

View File

@ -1,6 +1,47 @@
from django.db import backend, connection from django.db import backend, connection
LOOKUP_SEPARATOR = '__' LOOKUP_SEPARATOR = '__'
####################
# HELPER FUNCTIONS #
####################
# Django currently supports two forms of ordering.
# Form 1 (deprecated) example:
# order_by=(('pub_date', 'DESC'), ('headline', 'ASC'), (None, 'RANDOM'))
# Form 2 (new-style) example:
# order_by=('-pub_date', 'headline', '?')
# Form 1 is deprecated and will no longer be supported for Django's first
# official release. The following code converts from Form 1 to Form 2.
LEGACY_ORDERING_MAPPING = {'ASC': '_', 'DESC': '-_', 'RANDOM': '?'}
def handle_legacy_orderlist(order_list):
if not order_list or isinstance(order_list[0], basestring):
return order_list
else:
import warnings
new_order_list = [LEGACY_ORDERING_MAPPING[j.upper()].replace('_', str(i)) for i, j in order_list]
warnings.warn("%r ordering syntax is deprecated. Use %r instead." % (order_list, new_order_list), DeprecationWarning)
return new_order_list
def orderfield2column(f, opts):
try:
return opts.get_field(f, False).column
except FieldDoesNotExist:
return f
def orderlist2sql(order_list, opts, prefix=''):
if prefix.endswith('.'):
prefix = backend.quote_name(prefix[:-1]) + '.'
output = []
for f in handle_legacy_orderlist(order_list):
if f.startswith('-'):
output.append('%s%s DESC' % (prefix, backend.quote_name(orderfield2column(f[1:], opts))))
elif f == '?':
output.append(backend.get_random_function_sql())
else:
output.append('%s%s ASC' % (prefix, backend.quote_name(orderfield2column(f, opts))))
return ', '.join(output)
class QBase: class QBase:
"Base class for QAnd and QOr" "Base class for QAnd and QOr"