forked from mhgrove/Empire
-
Notifications
You must be signed in to change notification settings - Fork 0
/
CHANGES.txt
76 lines (73 loc) · 6.68 KB
/
CHANGES.txt
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
Change Log
==========
Empire 0.7 (TBD)
=========================
* MODIFIED: Repackaged all DataSource specific classes into the com.clarkparsia.empire.ds package. (issue #25)
* MODIFIED: Moved some DataSource specific functions from EmpireUtil to DataSourceUtil
* MODIFIED: Query results can returned proxied objects, allowing users to defer the loading of the results until they
iterate over them. Disable/Enabling this feature is controlled via EmpireOptions.
* MODIFIED: All contains to contains, and persistence operations (merge/persist/remove) in EntityManager have been
compacted; the amount of calls to the DataSource has been greatly reduced resulting in a large performance
increase when cascading persistence operations, and a modest increase overall.
* MODIFIED: Global config passed through to EntityManagers in the unit config so parameters do not have to be duplicated.
* MODIFIED: Properties mapped as collections, with no values, return empty collections instead of null. (issue #40)
* MODIFIED: Laundry list of names checked for configuration files, settling on 'empire.configuration' with the Properties format as the default
* MODIFIED: Upgrade to TDB 0.8.7, Jena 2.6.4, CP OpenRdfUtil 0.3
* ADDED: Support for loading the configuration from the jar (issue #58)
* ADDED: Support for using Empire as the Java Agent and an AnnotationProvider implementation based on this (issue #46)
* ADDED: TripleSource interface for allowing triple-level access to a DataSource
* ADDED: Ask & Describe query methods introduced to DataSource interface. Best effort implementations provided for DataSources which do not implement these natively.
* ADDED: Query type specific methods to RdfQuery to allow direct access to the query results from the DataSource
* ADDED: Default logging configuration
* ADDED: Flag to RdfProperty annotation to control how java.net.URI objects are marshalled into RDF.
* ADDED: Flag in EmpireOptions to disable the enforcement of @Entity annotations on classes
* ADDED: TDB support
* ADDED: Added support to automatically "infer" RdfProperty annotations when they are not present. When flag in EmpireOptions is enabled, RdfGenerator
will try to map all fields of a bean, regardless of whether or not RdfProperty annotation is present or not. Transient annotations can
'hide' these fields from being mapped. This is more inline with how normal JPA implementations work. (issue #19)
* ADDED: Simple cache to support reflection which increases performance of the serialization/deserialization into and out of Java. (issue #44)
* ADDED: Extensive improvements and additions to TDB & SDB support via contributions from Ultan O'Carroll -- includes support for JDBC data sources via DBCP & C3P0 DataSource wrappers. (issue #55)
* FIXED: Several issues with the SPARQL endpoint data source implementation
* FIXED: Bug in InstanceGenerator where the class/interface hierarchy was not being used to generate the bytecode resulting in AbstractMethodError's (issue #38)
* FIXED: Upgraded to Javassist 3.14.0 and altered how proxies are created when objects are constructed. This solves both the PermGen OutOfMemory error (issue #52)
regarding a bug in Javassist and resolves a significant memory leak within RdfGenerator.
* FIXED: Bug in BNode detection for building and using them as primary keys for instances
* FIXED: Bug in InstanceGenerator to allow creating instances of abstract classes (issue #38)
* FIXED: Ant/JUnit integration, annotation configuration was not correctly configured and the xml for the JUnit reports was missing (from patch from ManuelB) (issue #49-51)
* FIXED: Infinite loops in bi-directional cascades (issue #57)
* FIXED: Duplicated property assertions when a method is annotated with @RdfProperty in a concrete class and in an interface/abstract class (issue #56)
* FIXED: Calls to CtClass.getMethod -- catching NotFoundException and handling properly, was incorrectly expecting a null to be returned (issue #60)
* REMOVED: Global flag in EmpireOptions for handling proxying, this is now completely controlled via JPA annotations (issue #32)
Empire 0.6.6 (2010-07-08)
=========================
* MODIFIED: Updates in 4Store & OpenRDF utility libraries to resolve issue regarding escaping quotes in queries.
* MODIFIED: Better sniffing for XML formatted configuration files in DefaultEmpireModule
* ADDED: Support for MappedSuperclass
* ADDED: Build flag for enabling debug in jar files
* ADDED: Targets for producing separate builds for each platform (Jena, Sesame, 4Store, etc.)
* REMOVED: Global option to inspect bean hierarchy, this is now controlled by @MappedSuperclass
* FIXED: Bug in query parameter handling introduced in 0.6.5, optional parameters could produce invalid queries in certain cases.
* FIXED: Bug in Empire initialization. Static code in RdfGenerator was not getting executed to init its state in some cases.
* MODIFIED: Updates in 4Store & OpenRDF utility libraries to resolve issue regarding escaping quotes in queries.
Empire 0.6.5 (2010-06-04)
=========================
* MODIFIED: Behavior of a missing annotation index changed from error to warning.
* MODIFIED: Refactor of persistence API to include a method for creating a DataSource independent of an EntityManager
* MODIFIED: Query API clean-ups
* Limit & Offset are extracted from static query strings and used on the Query object
* Fix to detection of the type of query being executed
* Relaxing constraint that all variables in the parameterized query must be bound, any vars not bound are left as normal query variables.
* Query factory now public on DataSources
* Support for ARQ-style bnode encoding in SPARQL queries.
* ADDED: Implementation of DataSource interface for SPARQL endpoints.
* ADDED: Utility methods for getting values of a property for a specific subject
* ADDED: Utility method to load a class from any of the system, thread and current context class loaders
* ADDED: New example code and configuration for the Empire article published at SemanticUniverse
* ADDED: Annotation checker for scanning classes to make sure they have the required Empire & JPA annotations and that they are used correctly.
* ADDED: Support for @NamedQueries and @NamedNativeQueries
* ADDED: Parameter to Sesame DataSource factory which toggles the query dialect used, SeRQL or SPARQL.
* ADDED: support for language attribute awareness in handling literal values in Java/RDF conversion
* FIXED: InstanceGenerator to correct generate implementations for abstract classes as well as interfaces
* FIXED: Bug in Empire.init where the default module would not get loaded when no other modules were loaded
* FIXED: Bug in primary key creation
* FIXED: Bug in regex used for BNode detection