1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175
|
import datetime
from django.conf import settings
from django.core.exceptions import FieldError
from django.db.backends.utils import truncate_name
from django.db.models.constants import LOOKUP_SEP
from django.db.models.expressions import ExpressionNode
from django.db.models.query_utils import select_related_descend, QueryWrapper
from django.db.models.sql.constants import (CURSOR, SINGLE, MULTI, NO_RESULTS,
ORDER_DIR, GET_ITERATOR_CHUNK_SIZE, SelectInfo)
from django.db.models.sql.datastructures import EmptyResultSet
from django.db.models.sql.expressions import SQLEvaluator
from django.db.models.sql.query import get_order_dir, Query
from django.db.transaction import TransactionManagementError
from django.db.utils import DatabaseError
from django.utils import six
from django.utils.six.moves import zip
from django.utils import timezone
class SQLCompiler(object):
def __init__(self, query, connection, using):
self.query = query
self.connection = connection
self.using = using
self.quote_cache = {'*': '*'}
# When ordering a queryset with distinct on a column not part of the
# select set, the ordering column needs to be added to the select
# clause. This information is needed both in SQL construction and
# masking away the ordering selects from the returned row.
self.ordering_aliases = []
self.ordering_params = []
def pre_sql_setup(self):
"""
Does any necessary class setup immediately prior to producing SQL. This
is for things that can't necessarily be done in __init__ because we
might not have all the pieces in place at that time.
# TODO: after the query has been executed, the altered state should be
# cleaned. We are not using a clone() of the query here.
"""
if not self.query.tables:
self.query.join((None, self.query.get_meta().db_table, None))
if (not self.query.select and self.query.default_cols and not
self.query.included_inherited_models):
self.query.setup_inherited_models()
if self.query.select_related and not self.query.related_select_cols:
self.fill_related_selections()
def __call__(self, name):
"""
A wrapper around connection.ops.quote_name that doesn't quote aliases
for table names. This avoids problems with some SQL dialects that treat
quoted strings specially (e.g. PostgreSQL).
"""
if name in self.quote_cache:
return self.quote_cache[name]
if ((name in self.query.alias_map and name not in self.query.table_map) or
name in self.query.extra_select or (
name in self.query.external_aliases and name not in self.query.table_map)):
self.quote_cache[name] = name
return name
r = self.connection.ops.quote_name(name)
self.quote_cache[name] = r
return r
def quote_name_unless_alias(self, name):
"""
A wrapper around connection.ops.quote_name that doesn't quote aliases
for table names. This avoids problems with some SQL dialects that treat
quoted strings specially (e.g. PostgreSQL).
"""
return self(name)
def compile(self, node):
vendor_impl = getattr(
node, 'as_' + self.connection.vendor, None)
if vendor_impl:
return vendor_impl(self, self.connection)
else:
return node.as_sql(self, self.connection)
def as_sql(self, with_limits=True, with_col_aliases=False):
"""
Creates the SQL for this query. Returns the SQL string and list of
parameters.
If 'with_limits' is False, any limit/offset information is not included
in the query.
"""
if with_limits and self.query.low_mark == self.query.high_mark:
return '', ()
self.pre_sql_setup()
# After executing the query, we must get rid of any joins the query
# setup created. So, take note of alias counts before the query ran.
# However we do not want to get rid of stuff done in pre_sql_setup(),
# as the pre_sql_setup will modify query state in a way that forbids
# another run of it.
self.refcounts_before = self.query.alias_refcount.copy()
out_cols, s_params = self.get_columns(with_col_aliases)
ordering, o_params, ordering_group_by = self.get_ordering()
distinct_fields = self.get_distinct()
# This must come after 'select', 'ordering' and 'distinct' -- see
# docstring of get_from_clause() for details.
from_, f_params = self.get_from_clause()
where, w_params = self.compile(self.query.where)
having, h_params = self.compile(self.query.having)
having_group_by = self.query.having.get_group_by_cols()
params = []
for val in six.itervalues(self.query.extra_select):
params.extend(val[1])
result = ['SELECT']
if self.query.distinct:
result.append(self.connection.ops.distinct_sql(distinct_fields))
result.append(', '.join(out_cols + self.ordering_aliases))
params.extend(s_params)
params.extend(self.ordering_params)
result.append('FROM')
result.extend(from_)
params.extend(f_params)
if where:
result.append('WHERE %s' % where)
params.extend(w_params)
grouping, gb_params = self.get_grouping(having_group_by, ordering_group_by)
if grouping:
if distinct_fields:
raise NotImplementedError(
"annotate() + distinct(fields) not implemented.")
if not ordering:
ordering = self.connection.ops.force_no_ordering()
result.append('GROUP BY %s' % ', '.join(grouping))
params.extend(gb_params)
if having:
result.append('HAVING %s' % having)
params.extend(h_params)
if ordering:
result.append('ORDER BY %s' % ', '.join(ordering))
params.extend(o_params)
if with_limits:
if self.query.high_mark is not None:
result.append('LIMIT %d' % (self.query.high_mark - self.query.low_mark))
if self.query.low_mark:
if self.query.high_mark is None:
val = self.connection.ops.no_limit_value()
if val:
result.append('LIMIT %d' % val)
result.append('OFFSET %d' % self.query.low_mark)
if self.query.select_for_update and self.connection.features.has_select_for_update:
if self.connection.get_autocommit():
raise TransactionManagementError("select_for_update cannot be used outside of a transaction.")
# If we've been asked for a NOWAIT query but the backend does not support it,
# raise a DatabaseError otherwise we could get an unexpected deadlock.
nowait = self.query.select_for_update_nowait
if nowait and not self.connection.features.has_select_for_update_nowait:
raise DatabaseError('NOWAIT is not supported on this database backend.')
result.append(self.connection.ops.for_update_sql(nowait=nowait))
# Finally do cleanup - get rid of the joins we created above.
self.query.reset_refcounts(self.refcounts_before)
return ' '.join(result), tuple(params)
def as_nested_sql(self):
"""
Perform the same functionality as the as_sql() method, returning an
SQL string and parameters. However, the alias prefixes are bumped
beforehand (in a copy -- the current query isn't changed), and any
ordering is removed if the query is unsliced.
Used when nesting this query inside another.
"""
obj = self.query.clone()
if obj.low_mark == 0 and obj.high_mark is None and not self.query.distinct_fields:
# If there is no slicing in use, then we can safely drop all ordering
obj.clear_ordering(True)
return obj.get_compiler(connection=self.connection).as_sql()
def get_columns(self, with_aliases=False):
"""
Returns the list of columns to use in the select statement, as well as
a list any extra parameters that need to be included. If no columns
have been specified, returns all columns relating to fields in the
model.
If 'with_aliases' is true, any column names that are duplicated
(without the table names) are given unique aliases. This is needed in
some cases to avoid ambiguity with nested queries.
"""
qn = self
qn2 = self.connection.ops.quote_name
result = ['(%s) AS %s' % (col[0], qn2(alias)) for alias, col in six.iteritems(self.query.extra_select)]
params = []
aliases = set(self.query.extra_select.keys())
if with_aliases:
col_aliases = aliases.copy()
else:
col_aliases = set()
if self.query.select:
only_load = self.deferred_to_columns()
for col, _ in self.query.select:
if isinstance(col, (list, tuple)):
alias, column = col
table = self.query.alias_map[alias].table_name
if table in only_load and column not in only_load[table]:
continue
r = '%s.%s' % (qn(alias), qn(column))
if with_aliases:
if col[1] in col_aliases:
c_alias = 'Col%d' % len(col_aliases)
result.append('%s AS %s' % (r, c_alias))
aliases.add(c_alias)
col_aliases.add(c_alias)
else:
result.append('%s AS %s' % (r, qn2(col[1])))
aliases.add(r)
col_aliases.add(col[1])
else:
result.append(r)
aliases.add(r)
col_aliases.add(col[1])
else:
col_sql, col_params = self.compile(col)
result.append(col_sql)
params.extend(col_params)
if hasattr(col, 'alias'):
aliases.add(col.alias)
col_aliases.add(col.alias)
elif self.query.default_cols:
cols, new_aliases = self.get_default_columns(with_aliases,
col_aliases)
result.extend(cols)
aliases.update(new_aliases)
max_name_length = self.connection.ops.max_name_length()
for alias, aggregate in self.query.aggregate_select.items():
agg_sql, agg_params = self.compile(aggregate)
if alias is None:
result.append(agg_sql)
else:
result.append('%s AS %s' % (agg_sql, qn(truncate_name(alias, max_name_length))))
params.extend(agg_params)
for (table, col), _ in self.query.related_select_cols:
r = '%s.%s' % (qn(table), qn(col))
if with_aliases and col in col_aliases:
c_alias = 'Col%d' % len(col_aliases)
result.append('%s AS %s' % (r, c_alias))
aliases.add(c_alias)
col_aliases.add(c_alias)
else:
result.append(r)
aliases.add(r)
col_aliases.add(col)
self._select_aliases = aliases
return result, params
def get_default_columns(self, with_aliases=False, col_aliases=None,
start_alias=None, opts=None, as_pairs=False, from_parent=None):
"""
Computes the default columns for selecting every field in the base
model. Will sometimes be called to pull in related models (e.g. via
select_related), in which case "opts" and "start_alias" will be given
to provide a starting point for the traversal.
Returns a list of strings, quoted appropriately for use in SQL
directly, as well as a set of aliases used in the select statement (if
'as_pairs' is True, returns a list of (alias, col_name) pairs instead
of strings as the first component and None as the second component).
"""
result = []
if opts is None:
opts = self.query.get_meta()
qn = self
qn2 = self.connection.ops.quote_name
aliases = set()
only_load = self.deferred_to_columns()
if not start_alias:
start_alias = self.query.get_initial_alias()
# The 'seen_models' is used to optimize checking the needed parent
# alias for a given field. This also includes None -> start_alias to
# be used by local fields.
seen_models = {None: start_alias}
for field, model in opts.get_concrete_fields_with_model():
if from_parent and model is not None and issubclass(from_parent, model):
# Avoid loading data for already loaded parents.
continue
alias = self.query.join_parent_model(opts, model, start_alias,
seen_models)
column = field.column
for seen_model, seen_alias in seen_models.items():
if seen_model and seen_alias == alias:
ancestor_link = seen_model._meta.get_ancestor_link(model)
if ancestor_link:
column = ancestor_link.column
break
table = self.query.alias_map[alias].table_name
if table in only_load and column not in only_load[table]:
continue
if as_pairs:
result.append((alias, field))
aliases.add(alias)
continue
if with_aliases and column in col_aliases:
c_alias = 'Col%d' % len(col_aliases)
result.append('%s.%s AS %s' % (qn(alias),
qn2(column), c_alias))
col_aliases.add(c_alias)
aliases.add(c_alias)
else:
r = '%s.%s' % (qn(alias), qn2(column))
result.append(r)
aliases.add(r)
if with_aliases:
col_aliases.add(column)
return result, aliases
def get_distinct(self):
"""
Returns a quoted list of fields to use in DISTINCT ON part of the query.
Note that this method can alter the tables in the query, and thus it
must be called before get_from_clause().
"""
qn = self
qn2 = self.connection.ops.quote_name
result = []
opts = self.query.get_meta()
for name in self.query.distinct_fields:
parts = name.split(LOOKUP_SEP)
_, targets, alias, joins, path, _ = self._setup_joins(parts, opts, None)
targets, alias, _ = self.query.trim_joins(targets, joins, path)
for target in targets:
result.append("%s.%s" % (qn(alias), qn2(target.column)))
return result
def get_ordering(self):
"""
Returns a tuple containing a list representing the SQL elements in the
"order by" clause, and the list of SQL elements that need to be added
to the GROUP BY clause as a result of the ordering.
Also sets the ordering_aliases attribute on this instance to a list of
extra aliases needed in the select.
Determining the ordering SQL can change the tables we need to include,
so this should be run *before* get_from_clause().
"""
if self.query.extra_order_by:
ordering = self.query.extra_order_by
elif not self.query.default_ordering:
ordering = self.query.order_by
else:
ordering = (self.query.order_by
or self.query.get_meta().ordering
or [])
qn = self
qn2 = self.connection.ops.quote_name
distinct = self.query.distinct
select_aliases = self._select_aliases
result = []
group_by = []
ordering_aliases = []
if self.query.standard_ordering:
asc, desc = ORDER_DIR['ASC']
else:
asc, desc = ORDER_DIR['DESC']
# It's possible, due to model inheritance, that normal usage might try
# to include the same field more than once in the ordering. We track
# the table/column pairs we use and discard any after the first use.
processed_pairs = set()
params = []
ordering_params = []
# For plain DISTINCT queries any ORDER BY clause must appear
# in SELECT clause.
# http://www.postgresql.org/message-id/27009.1171559417@sss.pgh.pa.us
must_append_to_select = distinct and not self.query.distinct_fields
for pos, field in enumerate(ordering):
if field == '?':
result.append(self.connection.ops.random_function_sql())
continue
if isinstance(field, int):
if field < 0:
order = desc
field = -field
else:
order = asc
result.append('%s %s' % (field, order))
group_by.append((str(field), []))
continue
col, order = get_order_dir(field, asc)
if col in self.query.aggregate_select:
result.append('%s %s' % (qn(col), order))
continue
if '.' in field:
# This came in through an extra(order_by=...) addition. Pass it
# on verbatim.
table, col = col.split('.', 1)
if (table, col) not in processed_pairs:
elt = '%s.%s' % (qn(table), col)
processed_pairs.add((table, col))
if not must_append_to_select or elt in select_aliases:
result.append('%s %s' % (elt, order))
group_by.append((elt, []))
elif not self.query._extra or get_order_dir(field)[0] not in self.query._extra:
# 'col' is of the form 'field' or 'field1__field2' or
# '-field1__field2__field', etc.
for table, cols, order in self.find_ordering_name(field,
self.query.get_meta(), default_order=asc):
for col in cols:
if (table, col) not in processed_pairs:
elt = '%s.%s' % (qn(table), qn2(col))
processed_pairs.add((table, col))
if must_append_to_select and elt not in select_aliases:
ordering_aliases.append(elt)
result.append('%s %s' % (elt, order))
group_by.append((elt, []))
else:
elt = qn2(col)
if col not in self.query.extra_select:
if must_append_to_select:
sql = "(%s) AS %s" % (self.query.extra[col][0], elt)
ordering_aliases.append(sql)
ordering_params.extend(self.query.extra[col][1])
result.append('%s %s' % (elt, order))
else:
result.append("(%s) %s" % (self.query.extra[col][0], order))
params.extend(self.query.extra[col][1])
else:
result.append('%s %s' % (elt, order))
group_by.append(self.query.extra[col])
self.ordering_aliases = ordering_aliases
self.ordering_params = ordering_params
return result, params, group_by
def find_ordering_name(self, name, opts, alias=None, default_order='ASC',
already_seen=None):
"""
Returns the table alias (the name might be ambiguous, the alias will
not be) and column name for ordering by the given 'name' parameter.
The 'name' is of the form 'field1__field2__...__fieldN'.
"""
name, order = get_order_dir(name, default_order)
pieces = name.split(LOOKUP_SEP)
field, targets, alias, joins, path, opts = self._setup_joins(pieces, opts, alias)
# If we get to this point and the field is a relation to another model,
# append the default ordering for that model unless the attribute name
# of the field is specified.
if field.rel and path and opts.ordering and name != field.attname:
# Firstly, avoid infinite loops.
if not already_seen:
already_seen = set()
join_tuple = tuple(self.query.alias_map[j].table_name for j in joins)
if join_tuple in already_seen:
raise FieldError('Infinite loop caused by ordering.')
already_seen.add(join_tuple)
results = []
for item in opts.ordering:
results.extend(self.find_ordering_name(item, opts, alias,
order, already_seen))
return results
targets, alias, _ = self.query.trim_joins(targets, joins, path)
return [(alias, [t.column for t in targets], order)]
def _setup_joins(self, pieces, opts, alias):
"""
A helper method for get_ordering and get_distinct.
Note that get_ordering and get_distinct must produce same target
columns on same input, as the prefixes of get_ordering and get_distinct
must match. Executing SQL where this is not true is an error.
"""
if not alias:
alias = self.query.get_initial_alias()
field, targets, opts, joins, path = self.query.setup_joins(
pieces, opts, alias)
alias = joins[-1]
return field, targets, alias, joins, path, opts
def get_from_clause(self):
"""
Returns a list of strings that are joined together to go after the
"FROM" part of the query, as well as a list any extra parameters that
need to be included. Sub-classes, can override this to create a
from-clause via a "select".
This should only be called after any SQL construction methods that
might change the tables we need. This means the select columns,
ordering and distinct must be done first.
"""
result = []
qn = self
qn2 = self.connection.ops.quote_name
first = True
from_params = []
for alias in self.query.tables:
if not self.query.alias_refcount[alias]:
continue
try:
name, alias, join_type, lhs, join_cols, _, join_field = self.query.alias_map[alias]
except KeyError:
# Extra tables can end up in self.tables, but not in the
# alias_map if they aren't in a join. That's OK. We skip them.
continue
alias_str = '' if alias == name else (' %s' % alias)
if join_type and not first:
extra_cond = join_field.get_extra_restriction(
self.query.where_class, alias, lhs)
if extra_cond:
extra_sql, extra_params = self.compile(extra_cond)
extra_sql = 'AND (%s)' % extra_sql
from_params.extend(extra_params)
else:
extra_sql = ""
result.append('%s %s%s ON ('
% (join_type, qn(name), alias_str))
for index, (lhs_col, rhs_col) in enumerate(join_cols):
if index != 0:
result.append(' AND ')
result.append('%s.%s = %s.%s' %
(qn(lhs), qn2(lhs_col), qn(alias), qn2(rhs_col)))
result.append('%s)' % extra_sql)
else:
connector = '' if first else ', '
result.append('%s%s%s' % (connector, qn(name), alias_str))
first = False
for t in self.query.extra_tables:
alias, unused = self.query.table_alias(t)
# Only add the alias if it's not already present (the table_alias()
# calls increments the refcount, so an alias refcount of one means
# this is the only reference.
if alias not in self.query.alias_map or self.query.alias_refcount[alias] == 1:
connector = '' if first else ', '
result.append('%s%s' % (connector, qn(alias)))
first = False
return result, from_params
def get_grouping(self, having_group_by, ordering_group_by):
"""
Returns a tuple representing the SQL elements in the "group by" clause.
"""
qn = self
result, params = [], []
if self.query.group_by is not None:
select_cols = self.query.select + self.query.related_select_cols
# Just the column, not the fields.
select_cols = [s[0] for s in select_cols]
if (len(self.query.get_meta().concrete_fields) == len(self.query.select)
and self.connection.features.allows_group_by_pk):
self.query.group_by = [
(self.query.get_meta().db_table, self.query.get_meta().pk.column)
]
select_cols = []
seen = set()
cols = self.query.group_by + having_group_by + select_cols
for col in cols:
col_params = ()
if isinstance(col, (list, tuple)):
sql = '%s.%s' % (qn(col[0]), qn(col[1]))
elif hasattr(col, 'as_sql'):
self.compile(col)
else:
sql = '(%s)' % str(col)
if sql not in seen:
result.append(sql)
params.extend(col_params)
seen.add(sql)
# Still, we need to add all stuff in ordering (except if the backend can
# group by just by PK).
if ordering_group_by and not self.connection.features.allows_group_by_pk:
for order, order_params in ordering_group_by:
# Even if we have seen the same SQL string, it might have
# different params, so, we add same SQL in "has params" case.
if order not in seen or order_params:
result.append(order)
params.extend(order_params)
seen.add(order)
# Unconditionally add the extra_select items.
for extra_select, extra_params in self.query.extra_select.values():
sql = '(%s)' % str(extra_select)
result.append(sql)
params.extend(extra_params)
return result, params
def fill_related_selections(self, opts=None, root_alias=None, cur_depth=1,
requested=None, restricted=None):
"""
Fill in the information needed for a select_related query. The current
depth is measured as the number of connections away from the root model
(for example, cur_depth=1 means we are looking at models with direct
connections to the root model).
"""
if not restricted and self.query.max_depth and cur_depth > self.query.max_depth:
# We've recursed far enough; bail out.
return
if not opts:
opts = self.query.get_meta()
root_alias = self.query.get_initial_alias()
self.query.related_select_cols = []
only_load = self.query.get_loaded_field_names()
# Setup for the case when only particular related fields should be
# included in the related selection.
if requested is None:
if isinstance(self.query.select_related, dict):
requested = self.query.select_related
restricted = True
else:
restricted = False
for f, model in opts.get_fields_with_model():
# The get_fields_with_model() returns None for fields that live
# in the field's local model. So, for those fields we want to use
# the f.model - that is the field's local model.
field_model = model or f.model
if not select_related_descend(f, restricted, requested,
only_load.get(field_model)):
continue
_, _, _, joins, _ = self.query.setup_joins(
[f.name], opts, root_alias)
alias = joins[-1]
columns, _ = self.get_default_columns(start_alias=alias,
opts=f.rel.to._meta, as_pairs=True)
self.query.related_select_cols.extend(
SelectInfo((col[0], col[1].column), col[1]) for col in columns)
if restricted:
next = requested.get(f.name, {})
else:
next = False
self.fill_related_selections(f.rel.to._meta, alias, cur_depth + 1,
next, restricted)
if restricted:
related_fields = [
(o.field, o.model)
for o in opts.get_all_related_objects()
if o.field.unique
]
for f, model in related_fields:
if not select_related_descend(f, restricted, requested,
only_load.get(model), reverse=True):
continue
_, _, _, joins, _ = self.query.setup_joins(
[f.related_query_name()], opts, root_alias)
alias = joins[-1]
from_parent = (opts.model if issubclass(model, opts.model)
else None)
columns, _ = self.get_default_columns(start_alias=alias,
opts=model._meta, as_pairs=True, from_parent=from_parent)
self.query.related_select_cols.extend(
SelectInfo((col[0], col[1].column), col[1]) for col in columns)
next = requested.get(f.related_query_name(), {})
self.fill_related_selections(model._meta, alias, cur_depth + 1,
next, restricted)
def deferred_to_columns(self):
"""
Converts the self.deferred_loading data structure to mapping of table
names to sets of column names which are to be loaded. Returns the
dictionary.
"""
columns = {}
self.query.deferred_to_data(columns, self.query.deferred_to_columns_cb)
return columns
def results_iter(self):
"""
Returns an iterator over the results from executing this query.
"""
resolve_columns = hasattr(self, 'resolve_columns')
fields = None
has_aggregate_select = bool(self.query.aggregate_select)
for rows in self.execute_sql(MULTI):
for row in rows:
if has_aggregate_select:
loaded_fields = self.query.get_loaded_field_names().get(self.query.model, set()) or self.query.select
aggregate_start = len(self.query.extra_select) + len(loaded_fields)
aggregate_end = aggregate_start + len(self.query.aggregate_select)
if resolve_columns:
if fields is None:
# We only set this up here because
# related_select_cols isn't populated until
# execute_sql() has been called.
# We also include types of fields of related models that
# will be included via select_related() for the benefit
# of MySQL/MySQLdb when boolean fields are involved
# (#15040).
# This code duplicates the logic for the order of fields
# found in get_columns(). It would be nice to clean this up.
if self.query.select:
fields = [f.field for f in self.query.select]
elif self.query.default_cols:
fields = self.query.get_meta().concrete_fields
else:
fields = []
fields = fields + [f.field for f in self.query.related_select_cols]
# If the field was deferred, exclude it from being passed
# into `resolve_columns` because it wasn't selected.
only_load = self.deferred_to_columns()
if only_load:
fields = [f for f in fields if f.model._meta.db_table not in only_load or
f.column in only_load[f.model._meta.db_table]]
if has_aggregate_select:
# pad None in to fields for aggregates
fields = fields[:aggregate_start] + [
None for x in range(0, aggregate_end - aggregate_start)
] + fields[aggregate_start:]
row = self.resolve_columns(row, fields)
if has_aggregate_select:
row = tuple(row[:aggregate_start]) + tuple(
self.query.resolve_aggregate(value, aggregate, self.connection)
for (alias, aggregate), value
in zip(self.query.aggregate_select.items(), row[aggregate_start:aggregate_end])
) + tuple(row[aggregate_end:])
yield row
def has_results(self):
"""
Backends (e.g. NoSQL) can override this in order to use optimized
versions of "query has any results."
"""
# This is always executed on a query clone, so we can modify self.query
self.query.add_extra({'a': 1}, None, None, None, None, None)
self.query.set_extra_mask(['a'])
return bool(self.execute_sql(SINGLE))
def execute_sql(self, result_type=MULTI):
"""
Run the query against the database and returns the result(s). The
return value is a single data item if result_type is SINGLE, or an
iterator over the results if the result_type is MULTI.
result_type is either MULTI (use fetchmany() to retrieve all rows),
SINGLE (only retrieve a single row), or None. In this last case, the
cursor is returned if any query is executed, since it's used by
subclasses such as InsertQuery). It's possible, however, that no query
is needed, as the filters describe an empty set. In that case, None is
returned, to avoid any unnecessary database interaction.
"""
if not result_type:
result_type = NO_RESULTS
try:
sql, params = self.as_sql()
if not sql:
raise EmptyResultSet
except EmptyResultSet:
if result_type == MULTI:
return iter([])
else:
return
cursor = self.connection.cursor()
try:
cursor.execute(sql, params)
except Exception:
cursor.close()
raise
if result_type == CURSOR:
# Caller didn't specify a result_type, so just give them back the
# cursor to process (and close).
return cursor
if result_type == SINGLE:
try:
if self.ordering_aliases:
return cursor.fetchone()[:-len(self.ordering_aliases)]
return cursor.fetchone()
finally:
# done with the cursor
cursor.close()
if result_type == NO_RESULTS:
cursor.close()
return
# The MULTI case.
if self.ordering_aliases:
result = order_modified_iter(cursor, len(self.ordering_aliases),
self.connection.features.empty_fetchmany_value)
else:
result = cursor_iter(cursor,
self.connection.features.empty_fetchmany_value)
if not self.connection.features.can_use_chunked_reads:
try:
# If we are using non-chunked reads, we return the same data
# structure as normally, but ensure it is all read into memory
# before going any further.
return list(result)
finally:
# done with the cursor
cursor.close()
return result
def as_subquery_condition(self, alias, columns, qn):
inner_qn = self
qn2 = self.connection.ops.quote_name
if len(columns) == 1:
sql, params = self.as_sql()
return '%s.%s IN (%s)' % (qn(alias), qn2(columns[0]), sql), params
for index, select_col in enumerate(self.query.select):
lhs = '%s.%s' % (inner_qn(select_col.col[0]), qn2(select_col.col[1]))
rhs = '%s.%s' % (qn(alias), qn2(columns[index]))
self.query.where.add(
QueryWrapper('%s = %s' % (lhs, rhs), []), 'AND')
sql, params = self.as_sql()
return 'EXISTS (%s)' % sql, params
class SQLInsertCompiler(SQLCompiler):
def __init__(self, *args, **kwargs):
self.return_id = False
super(SQLInsertCompiler, self).__init__(*args, **kwargs)
def placeholder(self, field, val):
if field is None:
# A field value of None means the value is raw.
return val
elif hasattr(field, 'get_placeholder'):
# Some fields (e.g. geo fields) need special munging before
# they can be inserted.
return field.get_placeholder(val, self.connection)
else:
# Return the common case for the placeholder
return '%s'
def as_sql(self):
# We don't need quote_name_unless_alias() here, since these are all
# going to be column names (so we can avoid the extra overhead).
qn = self.connection.ops.quote_name
opts = self.query.get_meta()
result = ['INSERT INTO %s' % qn(opts.db_table)]
has_fields = bool(self.query.fields)
fields = self.query.fields if has_fields else [opts.pk]
result.append('(%s)' % ', '.join(qn(f.column) for f in fields))
if has_fields:
params = values = [
[
f.get_db_prep_save(getattr(obj, f.attname) if self.query.raw else f.pre_save(obj, True), connection=self.connection)
for f in fields
]
for obj in self.query.objs
]
else:
values = [[self.connection.ops.pk_default_value()] for obj in self.query.objs]
params = [[]]
fields = [None]
can_bulk = (not any(hasattr(field, "get_placeholder") for field in fields) and
not self.return_id and self.connection.features.has_bulk_insert)
if can_bulk:
placeholders = [["%s"] * len(fields)]
else:
placeholders = [
[self.placeholder(field, v) for field, v in zip(fields, val)]
for val in values
]
# Oracle Spatial needs to remove some values due to #10888
params = self.connection.ops.modify_insert_params(placeholders, params)
if self.return_id and self.connection.features.can_return_id_from_insert:
params = params[0]
col = "%s.%s" % (qn(opts.db_table), qn(opts.pk.column))
result.append("VALUES (%s)" % ", ".join(placeholders[0]))
r_fmt, r_params = self.connection.ops.return_insert_id()
# Skip empty r_fmt to allow subclasses to customize behavior for
# 3rd party backends. Refs #19096.
if r_fmt:
result.append(r_fmt % col)
params += r_params
return [(" ".join(result), tuple(params))]
if can_bulk:
result.append(self.connection.ops.bulk_insert_sql(fields, len(values)))
return [(" ".join(result), tuple(v for val in values for v in val))]
else:
return [
(" ".join(result + ["VALUES (%s)" % ", ".join(p)]), vals)
for p, vals in zip(placeholders, params)
]
def execute_sql(self, return_id=False):
assert not (return_id and len(self.query.objs) != 1)
self.return_id = return_id
with self.connection.cursor() as cursor:
for sql, params in self.as_sql():
cursor.execute(sql, params)
if not (return_id and cursor):
return
if self.connection.features.can_return_id_from_insert:
return self.connection.ops.fetch_returned_insert_id(cursor)
return self.connection.ops.last_insert_id(cursor,
self.query.get_meta().db_table, self.query.get_meta().pk.column)
class SQLDeleteCompiler(SQLCompiler):
def as_sql(self):
"""
Creates the SQL for this query. Returns the SQL string and list of
parameters.
"""
assert len(self.query.tables) == 1, \
"Can only delete from one table at a time."
qn = self
result = ['DELETE FROM %s' % qn(self.query.tables[0])]
where, params = self.compile(self.query.where)
if where:
result.append('WHERE %s' % where)
return ' '.join(result), tuple(params)
class SQLUpdateCompiler(SQLCompiler):
def as_sql(self):
"""
Creates the SQL for this query. Returns the SQL string and list of
parameters.
"""
self.pre_sql_setup()
if not self.query.values:
return '', ()
table = self.query.tables[0]
qn = self
result = ['UPDATE %s' % qn(table)]
result.append('SET')
values, update_params = [], []
for field, model, val in self.query.values:
if hasattr(val, 'prepare_database_save'):
if field.rel or isinstance(val, ExpressionNode):
val = val.prepare_database_save(field)
else:
raise TypeError("Database is trying to update a relational field "
"of type %s with a value of type %s. Make sure "
"you are setting the correct relations" %
(field.__class__.__name__, val.__class__.__name__))
else:
val = field.get_db_prep_save(val, connection=self.connection)
# Getting the placeholder for the field.
if hasattr(field, 'get_placeholder'):
placeholder = field.get_placeholder(val, self.connection)
else:
placeholder = '%s'
if hasattr(val, 'evaluate'):
val = SQLEvaluator(val, self.query, allow_joins=False)
name = field.column
if hasattr(val, 'as_sql'):
sql, params = self.compile(val)
values.append('%s = %s' % (qn(name), sql))
update_params.extend(params)
elif val is not None:
values.append('%s = %s' % (qn(name), placeholder))
update_params.append(val)
else:
values.append('%s = NULL' % qn(name))
if not values:
return '', ()
result.append(', '.join(values))
where, params = self.compile(self.query.where)
if where:
result.append('WHERE %s' % where)
return ' '.join(result), tuple(update_params + params)
def execute_sql(self, result_type):
"""
Execute the specified update. Returns the number of rows affected by
the primary update query. The "primary update query" is the first
non-empty query that is executed. Row counts for any subsequent,
related queries are not available.
"""
cursor = super(SQLUpdateCompiler, self).execute_sql(result_type)
try:
rows = cursor.rowcount if cursor else 0
is_empty = cursor is None
finally:
if cursor:
cursor.close()
for query in self.query.get_related_updates():
aux_rows = query.get_compiler(self.using).execute_sql(result_type)
if is_empty and aux_rows:
rows = aux_rows
is_empty = False
return rows
def pre_sql_setup(self):
"""
If the update depends on results from other tables, we need to do some
munging of the "where" conditions to match the format required for
(portable) SQL updates. That is done here.
Further, if we are going to be running multiple updates, we pull out
the id values to update at this point so that they don't change as a
result of the progressive updates.
"""
self.query.select_related = False
self.query.clear_ordering(True)
super(SQLUpdateCompiler, self).pre_sql_setup()
count = self.query.count_active_tables()
if not self.query.related_updates and count == 1:
return
# We need to use a sub-select in the where clause to filter on things
# from other tables.
query = self.query.clone(klass=Query)
query._extra = {}
query.select = []
query.add_fields([query.get_meta().pk.name])
# Recheck the count - it is possible that fiddling with the select
# fields above removes tables from the query. Refs #18304.
count = query.count_active_tables()
if not self.query.related_updates and count == 1:
return
must_pre_select = count > 1 and not self.connection.features.update_can_self_select
# Now we adjust the current query: reset the where clause and get rid
# of all the tables we don't need (since they're in the sub-select).
self.query.where = self.query.where_class()
if self.query.related_updates or must_pre_select:
# Either we're using the idents in multiple update queries (so
# don't want them to change), or the db backend doesn't support
# selecting from the updating table (e.g. MySQL).
idents = []
for rows in query.get_compiler(self.using).execute_sql(MULTI):
idents.extend(r[0] for r in rows)
self.query.add_filter(('pk__in', idents))
self.query.related_ids = idents
else:
# The fast path. Filters and updates in one query.
self.query.add_filter(('pk__in', query))
for alias in self.query.tables[1:]:
self.query.alias_refcount[alias] = 0
class SQLAggregateCompiler(SQLCompiler):
def as_sql(self, qn=None):
"""
Creates the SQL for this query. Returns the SQL string and list of
parameters.
"""
if qn is None:
qn = self
sql, params = [], []
for aggregate in self.query.aggregate_select.values():
agg_sql, agg_params = self.compile(aggregate)
sql.append(agg_sql)
params.extend(agg_params)
sql = ', '.join(sql)
params = tuple(params)
sql = 'SELECT %s FROM (%s) subquery' % (sql, self.query.subquery)
params = params + self.query.sub_params
return sql, params
class SQLDateCompiler(SQLCompiler):
def results_iter(self):
"""
Returns an iterator over the results from executing this query.
"""
resolve_columns = hasattr(self, 'resolve_columns')
if resolve_columns:
from django.db.models.fields import DateField
fields = [DateField()]
else:
from django.db.backends.utils import typecast_date
needs_string_cast = self.connection.features.needs_datetime_string_cast
offset = len(self.query.extra_select)
for rows in self.execute_sql(MULTI):
for row in rows:
date = row[offset]
if resolve_columns:
date = self.resolve_columns(row, fields)[offset]
elif needs_string_cast:
date = typecast_date(str(date))
if isinstance(date, datetime.datetime):
date = date.date()
yield date
class SQLDateTimeCompiler(SQLCompiler):
def results_iter(self):
"""
Returns an iterator over the results from executing this query.
"""
resolve_columns = hasattr(self, 'resolve_columns')
if resolve_columns:
from django.db.models.fields import DateTimeField
fields = [DateTimeField()]
else:
from django.db.backends.utils import typecast_timestamp
needs_string_cast = self.connection.features.needs_datetime_string_cast
offset = len(self.query.extra_select)
for rows in self.execute_sql(MULTI):
for row in rows:
datetime = row[offset]
if resolve_columns:
datetime = self.resolve_columns(row, fields)[offset]
elif needs_string_cast:
datetime = typecast_timestamp(str(datetime))
# Datetimes are artificially returned in UTC on databases that
# don't support time zone. Restore the zone used in the query.
if settings.USE_TZ:
if datetime is None:
raise ValueError("Database returned an invalid value "
"in QuerySet.datetimes(). Are time zone "
"definitions for your database and pytz installed?")
datetime = datetime.replace(tzinfo=None)
datetime = timezone.make_aware(datetime, self.query.tzinfo)
yield datetime
def cursor_iter(cursor, sentinel):
"""
Yields blocks of rows from a cursor and ensures the cursor is closed when
done.
"""
try:
for rows in iter((lambda: cursor.fetchmany(GET_ITERATOR_CHUNK_SIZE)),
sentinel):
yield rows
finally:
cursor.close()
def order_modified_iter(cursor, trim, sentinel):
"""
Yields blocks of rows from a cursor. We use this iterator in the special
case when extra output columns have been added to support ordering
requirements. We must trim those extra columns before anything else can use
the results, since they're only needed to make the SQL valid.
"""
try:
for rows in iter((lambda: cursor.fetchmany(GET_ITERATOR_CHUNK_SIZE)),
sentinel):
yield [r[:-trim] for r in rows]
finally:
cursor.close()
|