Fix count operation inside transaction and avoid superfluous client session instantiation.
Default MongoDatabase emission in case of non active transaction, update documentation, move test to another package.
Delay reactive collection re/creation in test to cope with issues in server version 4.1.10.
Original Pull Request: #745
Support declarative reactive transaction via the Transactional annotation via a MongoDB specific ReactiveTransactionManager implementation.
@Bean
ReactiveMongoTransactionManager transactionManager(ReactiveDatabaseFactory factory) {
return new ReactiveMongoTransactionManager(factory);
}
@Component
public class StateService {
@Transactional
Mono<UpdateResult> someBusinessFunction(Step step) {
return template.insert(step)
.then(process(step))
.then(template.update(Step.class).apply(Update.set("state", …));
};
});
Original Pull Request: #745
Store result to getPersistentEntity() in local variable to reduce invocation count. Use numbered postfix instead of object instantiation to generate a unique distance field name.
Deprecate geoNear methods with hint to aggregations. Fix distance field usage in ReactiveMongoTemplate.
Tweak docs.
Original pull request: #744.
Deprecate and guard tests for removed commands:
- eval
- group
- maxScan
- geoNear
Transition from removed geoNear command to $geoNear aggregation pipeline stage.
This allows users to still use NearQuery along with the template that now transforms the query into a GeoNearOperation applying skip (previously in memory skipping) and limit (previous num parameter) as additional aggregation pipeline stages.
Tested against:
- 4.1.10
- 4.0.4
- 3.6.12
- 4.4.20
Original pull request: #744.
The collation can now also be configured on entity level and gets applied via MongoTemplate. However one can alway override a default collation by adding the collation explicitly to either the Query or one of the Options available for various operations.
When it comes to the repository level the hierarchy is method parameter over query annotation over entity metadata.
Remove collation annotation in favor of attributes of Query / Document.
Original pull request: #644.
We now allow to specify the collation via the @Query annotation.
public interface PersonRepository extends MongoRepository<Person, String> {
@Query(collation = "en_US")
List<Person> findByFirstname(String firstname);
@Query(collation = "{ 'locale' : 'en_US' }")
List<Person> findPersonByFirstname(String firstname);
@Query(collation = "?1")
List<Person> findByFirstname(String firstname, Object collation);
@Query(collation = "{ 'locale' : '?1' }")
List<Person> findByFirstname(String firstname, String collation);
List<Person> findByFirstname(String firstname, Collation collation);
}
We now make sure to include collation information derived from the Query method if the collation is a fixed value.
Original pull request: #644.
Align Indexed(expireAfter) default to empty string according to the documentation. Slightly reword Javadoc.
Import DurationStyle to reuse duration parsing until Spring Framework provides a similar utility. Document ISO-8601 duration style. Prevent null index name evaluation to render "null" as String.
Convert assertions from Hamcrest to AssertJ.
Original pull request: #647.
We now also evaluate SpEL expressions for the name of indices as well as the def attribute of the compound index definition.
@CompoundIndex(name = "#{'cmp' + 2 + 'name‘}“, def = "#{T(org.bson.Document).parse(\"{ 'foo': 1, 'bar': -1 }\")}")
class WithCompoundIndexFromExpression {
// …
}
An expression used for Indexed.expireAfter may now not only return a plain String value with the timeout but also a java.time.Duration.
Original pull request: #647.
We added expireAfter which accepts numeric values followed by the unit of measure (d(ays), h(ours), m(inutes), s(econds)) or a Spring template expression to the Indexed annotation.
@Indexed(expireAfter = "10s")
String expireAfterTenSeconds;
@Indexed(expireAfter = "1d")
String expireAfterOneDay;
@Indexed(expireAfter = "#{@mySpringBean.timeout}")
String expireAfterTimeoutObtainedFromSpringBean;
Original pull request: #647.
Fix generics usage in MappingMongoJsonSchemaCreator. Make fields final. Rename MappingMongoConverter.computeWriteTarget to getWriteTarget and expose it publicly for reuse in custom DefaultTypeMapper setups without the need to subclass MappingMongoConverter.
Remove Nullability functionality for required fields as nullability indicators should originate from PersistentProperty and PreferredConstructor. Update documentation.
Related ticket: DATACMNS-1513
Original pull request: #733.
MongoJsonSchemaCreator extracts the MongoJsonSchema for a given Class by applying the following mapping rules:
Required Properties:
- Properties of primitive type.
Ignored Properties:
- All properties annotated with Transient.
Property Type Mapping:
- java.lang.Object -> { type : 'object' }
- java.util.Arrays -> { type : 'array' }
- java.util.Collection -> { type : 'array'}
- java.util.Map -> { type : 'object'}
- java.lang.Enum -> { type : 'string', enum : [ ... ] }
- Simple Types -> { type : 'the corresponding bson type' }
- Domain Types -> { type : 'object', properties : { ... } }
_id properties using types that can be converted into
ObjectId like String will be mapped to { type : 'object' } unless there is more specific information available via the MongoId annotation.
By using Field#targetType it is now possible to pass down a type hint to the conversion subsystem. This allows specifying the desired target type for a property so that eg. a plain String can be stored as Code.
Original pull request: #733.
We now pass on the range defined by Query.skip and Query.limit to MongoDB. This allows to count documents within a certain range so that it is possible to find the number of matches within eg. the first 10,000 documents.
Original pull request: #650.
We now reuse the MongoConverter instance that is associated with a MongoTemplate when using withSession(…) instead of creating a new converter instance.
In consequence, we're reusing PersistentEntity instances, EntityInstantiators and generated accessor classes.
Original Pull Request: #734
We now make sure we to not convert Strings representing valid ObjectIds into the such for paths that cannot be resolved to a Property.
Original pull request: #732.