DEEP CLONING OF OBJECTS USING BINARY FORMAT

• US 20130124573A1
• Filed: 11/10/2011
• Published: 05/16/2013
• Est. Priority Date: 11/10/2011
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Seth, Gaurav, Kizer, Geoffrey M., Xu, Jianchun, Veluri, Koundinya
• Granted Patent: US 8,954,475 B2
• Time in Patent Office: Days
• US Class Current: 707/798
• CPC Class Codes: G06F 16/00 Information retrieval; Database structures therefor; File system structures therefor
• CPC Class Codes: G06F 16/2291 User-Defined Types; Storage management thereof
• CPC Class Codes: G06F 9/4493 Object persistence

Abstract

Techniques are described herein that are capable of deep cloning (a.k.a. deep copying) objects using a binary format. A deep clone of a designated object includes references to other objects (and potentially copies of the other objects) included in the designated object. A binary representation of each object in an object graph is generated to provide a binary clone of the object graph. Objects created by a dynamic language engine are called dynamic language objects. Objects created by a host (e.g., that hosts the dynamic language engine) are called host objects. Each host object is associated with an intermediate representation thereof via a property bag that describes properties of that host object. Each intermediate representation is understandable by the dynamic language engine. A binary representation of each dynamic language object and each host object may be generated in accordance with the binary format to provide the binary clone.

Claims

1. A method of deep cloning one or more objects in an object graph using a binary format, the method comprising:

determining whether each of the one or more objects in the object graph is a host object that is created by a host or a dynamic language object that is created by a dynamic language engine hosted by the host;

retrieving each dynamic language object that is included in the one or more objects from the dynamic language engine;

retrieving each host object that is included in the one or more objects from the host via an intermediate representation of that host object contained in a property bag; and

generating a binary representation of each dynamic language object and each host object in accordance with the binary format to provide a binary clone of the object graph.

2. The method of claim 1, further comprising:

instantiating serialization of the object graph via an application programming interface that is exposed by a serialization engine that is included in the dynamic language engine.

3. The method of claim 2, further comprising:

generating an object identifier by the serialization engine for each of the one or more objects, at least one of the object identifiers indicating a relationship of the corresponding object with at least one other object that is included in the object graph.

4. The method of claim 2, further comprising:

receiving a byte stream indicator from the host at the serialization engine, the byte stream indicator specifying a designated byte stream into which the binary clone of the object graph is to be written; and

writing the binary clone of the object graph into the designated byte stream by the serialization engine based on the byte stream indicator.

5. The method of claim 2, wherein retrieving each host object that is included in the one or more objects from the host comprises:

creating the respective property bag by the serialization engine;

calling a second application programming interface by the serialization engine, the second application programming interface being exposed by the host;

sharing the respective property bag with the host by the serialization engine via the second application programming interface;

generating the intermediate representation of each host object into the respective property bag by the host, the intermediate representation of each host object being understandable by the dynamic language engine; and

retrieving the intermediate representation for each host object that is included in the one or more objects from the respective property bag by the serialization engine.

6. The method of claim 1, further comprising:

reconstructing the object graph based on the binary clone to provide a reconstructed object graph, the reconstructing comprising;

for the binary representation of each host object in the binary clone,

creating a placeholder for the host object,

extracting the binary representation of the host object from the binary clone into a respective property bag object such that the respective property bag object includes properties of the host object,

extracting the properties from the property bag object, and

initializing the placeholder with the properties that are extracted from the property bag object to provide a respective initialized host object in the reconstructed object graph; and

recreating each dynamic language object in the reconstructed object graph based on the binary representation of that dynamic language object.

7. The method of claim 6, further comprising:

instantiating serialization of the object graph via an application programming interface that is exposed by a serialization engine that is included in the dynamic language engine;

wherein, for the binary representation of each host object in the binary clone, creating the placeholder for the host object comprises;

creating the placeholder for the host object by a second host that is different from the first host from which the property bag for each host object that is included in the one or more objects is retrieved;

wherein, for the binary representation of each host object in the binary clone, extracting the binary representation of the host object comprises;

extracting the binary representation of the host object from the binary clone into the respective property bag object by a de-serialization engine that is included in a second dynamic language engine that is hosted by the second host;

wherein extracting the properties from the property bag object comprises;

extracting the properties from the property bag object by the second host; and

wherein initializing the placeholder comprises;

initializing the placeholder by the second host.

8. The method of claim 7, wherein retrieving each dynamic language object that is included in the one or more objects comprises:

retrieving each dynamic language object that is included in the one or more objects from the first dynamic language engine configured to execute commands written in a first dynamic language; and

wherein recreating each dynamic language object in the reconstructed object graph comprises;

recreating each dynamic language object in the reconstructed object graph using the second dynamic language engine configured to execute commands written in a second dynamic language.

9. The method of claim 6, further comprising:

instantiating serialization of the object graph via an application programming interface that is exposed by a serialization engine that is included in the dynamic language engine;

wherein, for the binary representation of each host object in the binary clone, creating the placeholder for the host object comprises;

creating the placeholder for the host object by the host;

wherein, for the binary representation of each host object in the binary clone, extracting the binary representation of the host object comprises;

extracting the binary representation of the host object from the binary clone into the respective property bag object by a de-serialization engine that is included in a second dynamic language engine that is hosted by the host and that is different from the first dynamic language engine, which includes the serialization engine;

wherein extracting the properties from the property bag object comprises;

extracting the properties from the property bag object by the host; and

wherein initializing the placeholder comprises;

initializing the placeholder by the host.

10. The method of claim 9, wherein retrieving each dynamic language object that is included in the one or more objects comprises:

retrieving each dynamic language object that is included in the one or more objects from the first dynamic language engine configured to execute commands written in a first dynamic language; and

wherein recreating each dynamic language object in the reconstructed object graph comprises;

recreating each dynamic language object in the reconstructed object graph using the second dynamic language engine configured to execute commands written in a second dynamic language.

11. A method of reconstructing an object graph based on a binary clone of the object graph, the binary clone including one or more binary representations of one or more respective host objects, the method comprising:

for each of the one or more binary representations of the one or more respective host objects,

creating a placeholder for the respective host object,

extracting the binary representation of the host object from the binary clone into a respective property bag object such that the respective property bag object includes properties of the host object,

extracting the properties from the property bag object, and

initializing the placeholder with the properties that are extracted from the property bag object to provide a respective initialized host object in the reconstructed object graph.

12. The method of claim 11, wherein the binary clone further includes one or more binary representations of one or more respective dynamic language objects;

and

wherein the method further comprises;

recreating each of the one or more dynamic language objects in the object graph based on the binary representation of that dynamic language object in the binary clone.

13. The method of claim 11, further comprising:

instantiating de-serialization of the binary clone via an application programming interface that is exposed by a de-serialization engine that is included in a dynamic language engine;

wherein, for each of the one or more binary representations of the one or more respective host objects, creating the placeholder for the host object comprises;

creating the placeholder for the host object by a host that hosts the dynamic language engine;

wherein, for the binary representation of each host object in the binary clone, extracting the binary representation of the host object comprises;

extracting the binary representation of the host object from the binary clone into the respective property bag object by a de-serialization engine that is included in the dynamic language engine;

wherein extracting the properties from the property bag object comprises;

extracting the properties from the property bag object by the host; and

wherein initializing the placeholder comprises;

initializing the placeholder by the host.

14. A system to deep clone one or more objects in an object graph using a binary format, the system comprising:

a host configured to instantiate serialization of the object graph via an application programming interface; and

a serialization engine included in a dynamic language engine, the dynamic language engine being hosted by the host, the serialization engine comprising;

an API module configured to expose the application programming interface to the host;

an object module configured to determine whether each of the one or more objects in the object graph is a host object that is created by the host or a dynamic language object that is created by the dynamic language engine;

a retrieval module configured to retrieve each dynamic language object that is included in the one or more objects from the dynamic language engine, the retrieval module further configured to create one or more property bags to retrieve one or more respective host objects included in the one or more objects from the host, each property bag containing an intermediate representation of the corresponding host object; and

a binary module configured to generate a binary representation of each dynamic language object and each host object in accordance with the binary format to provide a binary clone of the object graph.

15. The system of claim 14, wherein the serialization engine further comprises:

an identifier module configured to generate an object identifier for each of the one or more objects, at least one of the object identifiers indicating a relationship of the corresponding object with at least one other object that is included in the object graph.

16. The system of claim 14, wherein the serialization engine further comprises:

a provision module configured to write the binary clone of the object graph into a designated byte stream based on a byte stream indicator and contextual information that is received from the host,

the byte stream indicator specifying that the binary clone is to be written into the designated byte stream, and

the contextual information specifying whether serialization of the object graph and de-serialization of a binary representation of the object graph are to be performed within at least one of a same dynamic language engine, a same host, or a same device, and whether the deep clone is being performed to persist each of the one or more objects.

17. The system of claim 14, wherein the retrieval module comprises:

a call module configured to call a second application programming interface that is exposed to the host;

a property bag module configured to create an empty property bag for each host object that is included in the one or more objects from the host via the second application programming interface; and

a receipt module configured to receive the property bag, which is filled by the host, for each host object that is included in the one or more objects from the host via the second application programming interface.

18. The system of claim 14, further comprising:

a de-serialization engine configured to cause reconstruction of the object graph based on the binary clone to provide a reconstructed object graph, the de-serialization engine comprising;

a first creation module configured to provide a first instruction to the host to cause the host to create an uninitialized host object for the intermediate representation of the host object contained in each property bag;

an intermediate module configured to provide a second instruction to the host to cause the host to initialize each uninitialized host object based on the intermediate representation of the corresponding host object to provide a respective initialized host object in the reconstructed object graph;

a second creation module configured to recreate each dynamic language object in the reconstructed object graph based on the binary representation of that dynamic language object; and

a reconstruction module configured to combine the host objects recreated by the first creation module and the dynamic language objects recreated by the second creation module in accordance with contextual information to provide the reconstructed object graph;

wherein the contextual information specifies whether the deep clone and the reconstruction are to be performed by at least one of a common dynamic language engine, a common host, or a common device, and whether the deep clone is being performed to persist each of the one or more objects.

19. The method of claim 18, wherein the host is further configured to host a second dynamic language engine that is different from the first dynamic language engine;

and

wherein the second dynamic language engine includes the de-serialization engine.

20. The system of claim 14, further comprising:

a second host configured to host a second dynamic language engine, the second host being different from the first host from which the property bag for each host object that is included in the one or more objects is retrieved; and

a de-serialization engine included in the second dynamic language engine, the de-serialization engine configured to cause reconstruction of the object graph based on the binary clone to provide a reconstructed object graph, the de-serialization engine comprising;

a first creation module configured to provide a first instruction to the second host to cause the second host to create an uninitialized host object for the binary representation of the host object contained in each property bag;

an intermediate module configured to provide a second instruction to the second host to cause the second host to initialize each uninitialized host object based on the intermediate representation of the corresponding host object to provide a respective initialized host object in the reconstructed object graph;

a second creation module configured to recreate each dynamic language object in the reconstructed object graph based on the binary representation of that dynamic language object; and

a reconstruction module configured to combine the host objects recreated by the first creation module and the dynamic language objects recreated by the second creation module in accordance with contextual information to provide the reconstructed object graph;

wherein the contextual information specifies whether the deep clone and reconstruction are to be performed by at least one of a common dynamic language engine, a common host, or a common device, and whether the deep clone is being performed to persist each of the one or more objects.