Merge remote-tracking branch 'goog/upstream-master' into androidx-platform-dev

* goog/upstream-master:
  Update Icing from upstream.
  Update Icing from upstream.

Descriptions:
========================================================================
Cache an instance of UBreakIterator to reduce unnecessary creations.
========================================================================
Cap number of individual IntegerIndexStorages that IntegerIndex creates.
========================================================================
Change error in trimRightMostNode from Unimplemented to InvalidArgument.
========================================================================
Add detection for new language features of List Filters Query Language.
========================================================================
Add option to control threshold to rebuild index during optimize by flag
========================================================================
Add option to control use of namespace id to build urimapper by flag.
========================================================================
Enforce schema validation for joinable config.
========================================================================
Adopt bucket splitting for IntegerIndexStorage.
========================================================================
Implement bucket splitting function.
========================================================================
Add Icing initialization unit tests for QualifiedIdTypeJoinableIndex.
========================================================================
Add Icing schema change unit tests for QualifiedIdTypeJoinableIndex.
========================================================================
Add Icing optimization unit tests for QualifiedIdTypeJoinableIndex.
========================================================================
Integrate QualifiedIdTypeJoinableIndex into IcingSearchEngine.
========================================================================
Implement QualifiedIdJoinablePropertyIndexingHandler.
========================================================================
Change QualifiedIdTypeJoinableIndex to store raw qualified id string.
========================================================================
Pass info about unnormalized query terms through lexer/parser/visitor.
========================================================================
Integrate Advanced Query w/ Suggest, make ADVANCED_QUERY default parser.
======================================================================

Bug: 208654892
Bug: 263890397
Bug: 259743562
Bug: 272145329
Bug: 227356108
Change-Id: I44de5853bb6c55b42800ae34d8071016be6c87cd

1 files changed, 42 insertions, 4 deletions

diff --git a/icing/schema/schema-util.h b/icing/schema/schema-util.h
index e5747bb..47bb76b 100644
--- a/icing/schema/schema-util.h
+++ b/icing/schema/schema-util.h
@@ -34,9 +34,19 @@ class SchemaUtil {
       std::unordered_map<std::string, const SchemaTypeConfigProto>;
 
   // If A -> B is indicated in the map, then type A must be built before
-  // building type B, i.e. B depends on A.
-  using DependentMap = std::unordered_map<std::string_view,
-                                          std::unordered_set<std::string_view>>;
+  // building type B, i.e. B has a property of type A. Also include all
+  // PropertyConfigProto (with DOCUMENT data_type) pointers which directly
+  // connects type A and B. IOW, this vector of PropertyConfigProto* are "direct
+  // edges" connecting A and B directly. It will be an empty vector if A and B
+  // are not "directly" connected, but instead via another intermediate level of
+  // schema type. For example, the actual dependency is A -> C -> B, so there
+  // will be A -> C and C -> B with valid PropertyConfigProto* respectively in
+  // this map, but we will also expand transitive dependents: add A -> B into
+  // dependent map with empty vector of "edges".
+  using DependentMap = std::unordered_map<
+      std::string_view,
+      std::unordered_map<std::string_view,
+                         std::vector<const PropertyConfigProto*>>>;
 
   struct SchemaDelta {
     // Which schema types were present in the old schema, but were deleted from
@@ -113,6 +123,11 @@ class SchemaUtil {
   //      itself, thus creating an infinite loop.
   //  13. Two SchemaTypeConfigProtos cannot have properties that reference each
   //      other's schema_type, thus creating an infinite loop.
+  //  14. PropertyConfigProtos.joinable_config must be valid. See
+  //      ValidateJoinableConfig for more details.
+  //  15. Any PropertyConfigProtos with nested DOCUMENT data type must not have
+  //      REPEATED cardinality if they reference a schema type containing
+  //      joinable property.
   //
   //  TODO(b/171996137): Clarify 12 and 13 are only for indexed properties, once
   //  document properties can be opted out of indexing.
@@ -121,7 +136,7 @@ class SchemaUtil {
   //   On success, a dependent map from each types to their dependent types
   //   that depend on it directly or indirectly.
   //   ALREADY_EXISTS for case 1 and 2
-  //   INVALID_ARGUMENT for 3-13
+  //   INVALID_ARGUMENT for 3-15
   static libtextclassifier3::StatusOr<DependentMap> Validate(
       const SchemaProto& schema);
 
@@ -145,6 +160,8 @@ class SchemaUtil {
   //      `SchemaDelta.schema_types_deleted`
   //   3. A schema type's new definition would mean any existing data of the old
   //      definition is now incompatible.
+  //   4. The derived join index would be incompatible. This is held in
+  //      `SchemaDelta.join_incompatible`.
   //
   // For case 1, the two schemas would result in an incompatible index if:
   //   1.1. The new SchemaProto has a different set of indexed properties than
@@ -167,6 +184,11 @@ class SchemaUtil {
   //        scale defined as:
   //          LEAST <REPEATED - OPTIONAL - REQUIRED> MOST
   //
+  // For case 4, the two schemas would result in an incompatible join if:
+  //   4.1. A SchematypeConfig exists in the new SchemaProto that has a
+  //        different set of joinable properties than it did in the old
+  //        SchemaProto.
+  //
   // A property is defined by the combination of the
   // SchemaTypeConfig.schema_type and the PropertyConfigProto.property_name.
   //
@@ -227,6 +249,22 @@ class SchemaUtil {
       const StringIndexingConfig& config,
       PropertyConfigProto::DataType::Code data_type,
       std::string_view schema_type, std::string_view property_name);
+
+  // Checks that the 'joinable_config' satisfies the following rules:
+  //   1. If the data type matches joinable value type
+  //      a. Only STRING data types can use QUALIFIED_ID joinable value type
+  //   2. Only QUALIFIED_ID joinable value type can have delete propagation
+  //      enabled
+  //   3. Any joinable property should have non-REPEATED cardinality
+  //
+  // Returns:
+  //   INVALID_ARGUMENT if any of the rules are not followed
+  //   OK on success
+  static libtextclassifier3::Status ValidateJoinableConfig(
+      const JoinableConfig& config,
+      PropertyConfigProto::DataType::Code data_type,
+      PropertyConfigProto::Cardinality::Code cardinality,
+      std::string_view schema_type, std::string_view property_name);
 };
 
 }  // namespace lib