SERVER-37188 Rename "All Paths" index to "Wildcard" index

1 files changed, 101 insertions, 0 deletions

diff --git a/src/mongo/db/query/planner_wildcard_helpers.h b/src/mongo/db/query/planner_wildcard_helpers.h
new file mode 100644
index 00000000000..7e56adb1ba0
--- /dev/null
+++ b/src/mongo/db/query/planner_wildcard_helpers.h
@@ -0,0 +1,101 @@
+/**
+ * Copyright (C) 2018 MongoDB Inc.
+ *
+ * This program is free software: you can redistribute it and/or  modify
+ * it under the terms of the GNU Affero General Public License, version 3,
+ * as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU Affero General Public License for more details.
+ *
+ * You should have received a copy of the GNU Affero General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ * As a special exception, the copyright holders give permission to link the
+ * code of portions of this program with the OpenSSL library under certain
+ * conditions as described in each individual source file and distribute
+ * linked combinations including the program with the OpenSSL library. You
+ * must comply with the GNU Affero General Public License in all respects
+ * for all of the code used other than as permitted herein. If you modify
+ * file(s) with this exception, you may extend this exception to your
+ * version of the file(s), but you are not obligated to do so. If you do not
+ * wish to do so, delete this exception statement from your version. If you
+ * delete this exception statement from all source files in the program,
+ * then also delete it in the license file.
+ */
+
+#pragma once
+
+#include <set>
+
+#include "mongo/db/field_ref.h"
+#include "mongo/db/index/multikey_paths.h"
+#include "mongo/db/query/index_bounds_builder.h"
+#include "mongo/db/query/query_solution.h"
+
+namespace mongo {
+namespace wildcard_planning {
+
+using BoundsTightness = IndexBoundsBuilder::BoundsTightness;
+
+/**
+ * Specifies the maximum depth of nested array indices through which a query may traverse before a
+ * $** declines to answer it, due to the exponential complexity of the bounds required.
+ */
+static constexpr size_t kWildcardMaxArrayIndexTraversalDepth = 8u;
+
+/**
+ * Returns a MultikeyPaths which indicates which components of 'indexedPath' are multikey, by
+ * looking up multikeyness in 'multikeyPathSet'.
+ */
+MultikeyPaths buildMultiKeyPathsForExpandedWildcardIndexEntry(
+    const FieldRef& indexedPath, const std::set<FieldRef>& multikeyPathSet);
+
+/**
+ * Given a query path and the set of known multikey path components, this function outputs the power
+ * set of paths generated by considering each relevant numerical path component first as a literal
+ * fieldname, and then as an array index. In other words, for each numerical path component that
+ * immediately follows a multikey field, we will output one path with the numerical component and
+ * one path without. If both 'a' and 'b' are multikey, then the queryPath 'a.0.b.1.c' will produce
+ * ['a.0.b.1.c', 'a.0.b.c', 'a.b.1.c', 'a.b.c'].
+ */
+std::set<FieldRef> generateFieldNameOrArrayIndexPathSet(const std::set<std::size_t>& multikeyPaths,
+                                                        const FieldRef& queryPath);
+
+/**
+ * In certain circumstances, it is necessary to adjust the tightness of the bounds generated by the
+ * planner for $** indexes. For instance, if the query traverses through one or more arrays via
+ * specific indices, then we must enforce INEXACT_FETCH to ensure correctness, regardless of the
+ * predicate. Given an IndexEntry representing an expanded $** index, we apply any necessary changes
+ * to the tightness here.
+ */
+BoundsTightness applyWildcardIndexScanBoundsTightness(const IndexEntry& index,
+                                                      BoundsTightness tightnessIn);
+
+/**
+ * Returns false if 'queryPath' includes any numerical path components which render it unanswerable
+ * by the $** index, true otherwise. Specifically, the $** index cannot answer the query if either
+ * of the following scenarios occur:
+ *
+ * - The query path traverses through more than 'kWildcardMaxArrayIndexTraversalDepth' nested arrays
+ * via explicit array indices.
+ * - The query path lies along a $** projection through an array index.
+ *
+ * For an example of the latter case, say that our query path is 'a.0.b', our projection includes
+ * {'a.0': 1}, and 'a' is multikey. The query semantics will match 'a.0.b' by either field name or
+ * array index against the documents, but because the $** index always projects numeric path
+ * components strictly as field names, the projection {'a.0': 1} cannot correctly support this
+ * query.
+ *
+ * To see why, consider the document {a: [1, 2, 3]}. Query {'a.0': 1} will match this document, but
+ * the projection {'a.0': 1} will produce output document {a: []}, and so we will not index any of
+ * the values [1, 2, 3] for 'a'.
+ */
+bool validateNumericPathComponents(const MultikeyPaths& multikeyPaths,
+                                   const std::set<FieldRef>& includedPaths,
+                                   const FieldRef& queryPath);
+
+}  // namespace wildcard_planning
+}  // namespace mongo