The core auditing implementation now skips the invocation of auditing in case the candidate aggregate doesn't need any auditing in the first place. We needed to adapt the sample class we use to actually carry the necessary auditing annotations.
Related ticket: DATACMNS-957.
We now use more type information to create a better empty collection in the first place. The previous algorithm always used an empty HashSet plus a subsequent conversion using the raw collection type. Especially the latter caused problems for EnumSets as the conversion into one requires the presence of component type information.
We now use Spring's collection factory and more available type information to create a proper collection in the first place and only rely on a subsequent conversion for arrays.
The methods in MongoTemplate inserting a batch of documents previously only returned database generated identifiers, more especially ObjectId ones. This caused non-ObjectId identifiers potentially generated by other parties — i.e. an event listener reacting to a BeforeSaveEvent — not being considered for source object identifier population.
This commit adds a workaround augmenting the list of database generated identifiers with the ones actually present in the documents to be inserted. A follow-up ticket DATAMONGO-1519 was created to track the removal of the workaround in favor of a proper fix unfortunately requiring a change in public API (so a 2.0 candidate only).
Related tickets: DATAMONGO-1519.
SpringDataMongodbQuery is exposed publicly in QuerydslRepositorySupport, that's we've got to make it public to make sure class to the exposed methods from outside the package actually compile.
Original pull request: #401.
We now use DbObjectAccessor also for preliminary inspections of the source DBObject (e.g. whether a value is present at all). Previously we operated on the DBObject directly which caused issues with properties mapped to nested fields as the keys weren't exploded correctly and thus the check always failed.
We now removed all eager query parameter conversions during query creation to be sure that the query mapping later on sees all to property names and is able to derive proper conversions.
By turning `AggregationExpression` public we allow adding custom expressions without workarounds. It is now possible to create eg. `ProjectionOperation` like:
ProjectionOperation agg = Aggregation.project()
.and(new AggregationExpression() {
@Override
public DBObject toDbObject(AggregationOperationContext context) {
DBObject filterExpression = new BasicDBObject();
filterExpression.put("input", "$x");
filterExpression.put("as", "y");
filterExpression.put("cond", new BasicDBObject("$eq", Arrays.<Object> asList("$$y.z", 2)));
return new BasicDBObject("$filter", filterExpression);
}
}).as("profile");
Original pull request: #392.
We now take potential registered converters for enums into account when serializing path expressions via SpringDataMongodbSerializer.
Original pull request: #388.
We now propagate the PersistentEntity when mapping nested keywords so that the criteria mapping chain for nested keywords and properties has now access to the PersistentEntity and can use configured field names.
Previously the plain property names have been used as field names and potential customizations via @Field have been ignored.
Original Pull Request: #384
Replace boolean flag in convertAndJoinScriptArgs with literal. Joined args are rendered to JavaScript and require always string quotation.
Original pull request: #383.
Setting the value for the identifier property is an explicit step in MappingMongoConverter and always executed if the type to be created has an identifier property. If the source document doesn't contain an _id field (e.g. because it has been excluded explicitly) that previously caused null to be set on the identifier. This caused an exception if the identifier property is a primitive type.
We now explicitly check whether the field backing the identifier property is actually present in the source document and only explicitly set the value if so.
This should result in minor to moderate performance improvement for iteration on Collections/Arrays during DBObject to object mapping.
Original pull request: #368.
Fixed broken highlighting using backticks followed by chars/single quotes. Convert single quote emphasis of id to backtick code fences. Add missing spaces between words and backticks.
Original Pull Request: #359
We now ignore property reference exceptions when resolving field values that have already been mapped. Eg. in case of an already mapped update extracted from an actual domain type instance.
Original pull request: #351.
We changed BasicQuery to consider its parent getFieldsObject() when not given an explicit fields DBObject.
Original Pull Request: #345
CLA: 165520160303021604 (John Willemin)
