Method and apparatus for providing schema evolution without recompilation

Claims

What is claimed is:

1. A method for supplying data to an application, the method comprising the computer-implemented steps of:

receiving a request initiated by the application for stored data to be supplied to the application;

in response to the request to supply the stored data to the application,

performing the steps of

determining a first format, said first format being the format in which said stored data is stored, and

determining a second format, said second format being the format in which said application expects to receive said stored data;

creating a target instance of said stored data by converting said stored data from said first format to said second format without said application that requested the stored data or users of the application that requested the stored data knowing the format in which the stored data is stored; and

supplying said target instance to the application that requested the stored data.

2. The method of claim 1 wherein the step of determining the second format comprises the steps of:

determining a data type that corresponds to said stored data;

determining a first version identifier that identifies a version of said data type that said application uses;

searching for a first schema version record that corresponds to said data type;

comparing a second version identifier in said first schema version record with said first version identifier;

if said second version identifier matches said first version identifier, then reading data that describes said second format from format data contained in said first schema version record;

if said second version identifier does not match said first version identifier, then

finding a second schema version record that is associated with said data type and which contains said first version identifier; and

reading data that describes said second format from format data contained in said second schema version record.

3. The method of claim 2 wherein the step of searching for a first schema version record comprises the steps of:

determining a type identifier that uniquely identifies said data type; and performing a search on a plurality of schema version records based on said type identifier.

4. The method of claim 1 wherein the step of determining the second format comprises the steps of:

determining a data type that corresponds to said stored data;

determining a first version identifier that identifies a version of said data type that said application uses;

using said data type and first version identifier to locate an entry in a table; and

using information in said entry to locate a schema version record that specifies said second format.

5. The method of claim 2 where in the step of determining the first version identifier includes the steps of:

inspecting an entry associated with said data type in a type version table associated with said application; and

reading said first version identifier from said entry.

6. The method of claim 5 further comprising the step of causing said application to build said type version table upon initialization by:

causing an initialization routine with in the application to call registration routines for a plurality of libraries used by said application; and

causing said registration routines to add entries to said type version table to indicate which versions of data types are used by said plurality of libraries.

7. The method of claim 1 wherein said step of determining said first format includes the steps of:

determining a data type that corresponds to said stored data;

determining a first version identifier that identifies a version of said data type that was used to store said stored data;

searching for a first schema version record that corresponds to said data type;

comparing a second version identifier in said first schema version record with said first version identifier;

if said second version identifier matches said first version identifier, then reading data that describes said first format from format data contained in said first schema version record;

if said second version identifier does not match said first version identifier, then

finding a second schema version record that is associated with said data type and which contains said first version identifier; and

reading data that describes said first format from format data contained in said first schema version record.

8. The method of claim 7 wherein said step of determining said first version identifier includes the step of reading schema version information stored with said stored data.

9. The method of claim 1 wherein the step of converting said stored data from said first format to said second format includes the step of:

comparing attributes specified in said first format with attributes specified in said second format;

if an attribute exists in said first format and said second format, then copying data for said attribute from said stored data to said target instance;

if an attribute exists in said second format but not said first format, then storing a default value for said attribute in said target instance;

if an attribute exists in said first format but not in said second format, then not copying data that corresponds to said attribute from said stored data to said target instance.

10. The method of claim 9 wherein the step of copying data for said attribute comprises the steps of:

determining whether said attribute has an identical format in said first format and said second format;

if said attribute has an identical format in said first format and said second format, then copying data for said attribute from said stored data to said target instance without performing any conversion operation; and

if said attribute does not have an identical format in said first format and said second format, then

reading data for said attribute from said stored data;

performing a conversion operation on said data for said attribute to create reformatted data; and

storing said reformatted data for said attribute in said target instance.

11. The method of claim 9 wherein said step of comparing attributes specified in said first format with attributes specified in said second format includes comparing a first set of attribute identifiers that uniquely identify attributes in said first format with a second set of attribute identifiers that uniquely identify attributes in said second format.

12. The method of claim 7 wherein:

the data type includes an embedded data type; and

the step of reading data that describes said first format from format data contained in said first schema version record includes reading data that describes a structure of said embedded data type.

13. A computer-readable medium carrying one or more sequences of one or more instructions for supplying data to an application, the one or more sequences of one or more instructions including instructions which, when executed by one or more processors, cause the one or more processors to perform the steps of:

receiving a request initiated by the application for stored data to be supplied to the application;

determining a first format, said first format being the format in which said stored data is stored, and

determining a second format, said second format being the format in which said application expects to receive said stored data;

creating a target instance of said stored data by converting said stored data from said first format to said second format without said application that requested the stored data or users of the application that requested the stored data knowing the format in which the stored data is stored; and

supplying said target instance to the application that requested the stored data.

14. The computer-readable medium of claim 13, wherein the step of determining the second format comprises the steps of:

determining a data type that corresponds to said stored data;

determining a first version identifier that identifies a version of said data type that said application uses;

searching for a first schema version record that