US20100332526A1 - Method(s) For Updating Database Object Metadata - Google Patents

Abstract

Two computers are respectively programmed to generate descriptions of database objects (such as tables) in a first database, and compare the generated descriptions to corresponding schema objects in a second database. Objects in the second database are changed to conform to the first database if differences are found during comparison. The just-described differences in objects may arise due to changes being made to a software program that uses data contained in the objects.

Description

CROSS-REFERENCE TO PARENT APPLICATION
• This application is a continuation application of U.S. application Ser. No. 10/966,250 filed on Oct. 15, 2004 by Sarita Brahmandam Nori et al, entitled “METHOD(S) FOR UPDATING DATABASE OBJECT METADATA” which is incorporated by reference here in its entirety.
BACKGROUND
• Distribution of software upgrades and source code changes has become a familiar routine that provides customers of a software product with increased features, bug fixes, new drivers and other updates to the software product (which may be, for example, application software such as “ORACLE FINANCIALS” that is commonly used by customers to implement finance-related functions in their computers). An object through which software upgrades are distributed is typically referred to as a software patch.
• For application software used in large enterprises (such as a publicly-owned corporation), a database is a quintessential element that provides scalable, performance-oriented data storage. Patches for portions of application software that interface to the database may require corresponding modifications to the database itself. Specifically, changes may need to be made to one or more attributes of tables, indexes, or views, e.g. a new column for holding an employee's middle initial may need to be added to a table and/or an existing column (or combination of columns) may need to be indexed to support patches being made to application software.
• Unlike code changes related to software written in high level languages (e.g. Java/C/C++), a change to a database needs to be handled carefully to not adversely impact or in any manner corrupt data that is already existing in the database at the time of the change. Enterprise Applications such as Oracle E-Business Suite Applications rarely enforce a rigid patch level (e.g. identified by a release number and/or a version number) for their customers. In other words, customers have the flexibility to adopt different patch levels for products used by them, as per their implementation and configuration. Such scenarios pose additional patching challenges for changes to database components.
• Keeping in mind the criticality of preserving existing data, patching of a database currently requires extensive potential impact analysis, including consideration of database object revision history with a goal of correctly upgrading the customer's environment per their patch level. Database patch creation process that is traditionally performed involves human inefficiencies and redundancies, as several Release and Development personnel for different products that constitute the application (such as Oracle's E-Business Suite Applications) perform repetitive analysis. Additionally, in spite of laborious analysis, the results are error prone and are likely to cause failures during patching, on account of missing and/or overlooked object revisions, or failed handling of object dependencies. Yet another reason is a failure in the just-described analysis to anticipate different target database versions that customers may have implemented.
• A prior art utility for patching databases, called “ODF”, can be used by application development personnel to capture and propagate metadata definitions of four types of database objects—tables, indexes, sequences and views. A listing of the output generated by the ODF utility is shown inFIG. 6A asmetadata601 for an example of a simple database containing a single object of the type table, which is called “TEST_TABLE” and which has two columns. Note that the listing inFIGS. 6A and 6B is expressed in a non-standard syntax which is proprietary and requires a custom parser specific to the syntax.
• Note that the ODF utility capturesmetadata601 for only four types of objects identified in the previous paragraph, and for no other objects (e.g. triggers, constraints, queues and policies were not supported). Applicants note that the failure of ODF to support all types of database objects makes it an incomplete solution. Furthermore, themetadata601 that is captured by ODF for even just these four types of objects is incomplete for another reason. Specifically, Applicants note that many important properties of these database objects (for e.g. Table properties like Partitioned/non-partitioned, Global temporary v/s Permanent, Index organized) were not supported.
• Moreover, the above-describedpartial metadata601 is extracted by ODF from a set of application-specific tables. Applicants note that the use of application-specific tables, results in at least two issues with ODF. First, the prior art ODF utility additionally requires a CASE (Computer-Aided Software Engineering) data-model repository (such as Oracle Designer) and also other snapshot utilities to propagatemetadata definitions601 from the CASE data-model repository to the database. Second,metadata601 captured by ODF has the risk of being inaccurate, as it is not captured from the source of truth (system data dictionary) but instead from an intermediate repository.
• Also, ODF captures metadata601 (FIG. 6A) for only a list of those individual objects that are manually identified by a developer of an application for which a patch is being developed. Applicants note that requiring the developer to list all objects whose metadata is to be captured makes ODF prone to human error and results in one or more object definitions being missed. Furthermore, ODF captures the metadata of objects in a non-standard format shown inFIG. 6A. Applicants note that failure to use XML prevents easy data exchange and also does not allow any data transformation/manipulation. Moreover, ODF's stores the metadata for a huge number of mostly unrelated objects in a very large single file. Applicants have found that such bundling of unrelated objects into a single file has an adverse impact on performance because each and every object in the single file is compared to the target database during patching of the target database. Finally, ODF did not use any sophisticated approaches in comparingmetadata601 to the target database.
• An alternative to ODF is to manually prepare SQL scripts. There is no concept of a metadata definition file in the context of SQL Scripts. Instead, the SQL scripts execute specific, hard-coded DDL (Data Definition Language) commands like DROP TABLE, CREATE TABLE, etc on the source database to make the object metadata definitions match. It is extremely error-prone as it is difficult for developers to know and predict object metadata definitions, in light of varying patch levels and potential customizations applied at the target database. It is very inefficient, time and resource intensive as it is not a generic solution, and requires repeated analysis and the same DDL commands to be used in multiple SQL scripts for different hard-coded objects. For example, a developer may prepare a SQL script file called “create_test_table.sql” which contains specific, hard-coded DDL (Data Definition Language) commands like DROP TABLE TEST_TABLE, CREATE TABLE TEST_TABLE to create the TEST_TABLE. A listing of SQL is shown inFIG. 6B for the above-described example of table object called “TEST_TABLE” which has two columns.
• The drawbacks of SQL scripts are as follows. First, there is no concept of a metadata definition file in the context of SQL Scripts. Therefore, use of SQL scripts does not allow for the archiving and versioning of object metadata definitions in a source code control system. Second, the SQL is not generic because it cannot accept as input an object whose metadata is to be patched. Instead, prior art SQL contains hard-coded references to individual object names (such as “TEST_TABLE” in the example ofFIG. 6B) that are used in a specific database. Third, this method is inefficient, time and resource intensive as it requires extensive analysis to be repeated for many sql scripts that handle different database objects. Fourth, this method is very error-prone as it is often difficult to predict the object metadata at a target database on account of varying patch levels and customizations present at the target database.
• U.S. Pat. No. 6,785,673 granted to Fernandez et al. on Aug. 31, 2004 and entitled “Method For Converting Relational Data Into XML” is hereby incorporated by reference herein in its entirety. This patent describes converting relational data into an XML document, to construct materialized XML views of relational databases.
• See also U.S. Pat. No. 6,732,095 granted to Warshaysky et al. on May 4, 2004 and entitled “Method And Apparatus For Mapping Between XML And Relational Representations” that is also incorporated by reference herein in its entirety. This patent describes a method to convert data between a relational format and an XML document, by creating a set of XML Mapping Definition from metadata; selecting relational data from a relational application database, and converting the relational data to the XML document using the set of XML Mapping Definition.
SUMMARY
• Computers in accordance with the invention are programmed to automatically generate complete metadata that is descriptive of an object in a first database, compare the generated metadata to a corresponding object in a second database (either directly or indirectly through hashcodes), and change the second database object as may be necessary to conform to the generated metadata if certain differences are found during comparison. The just-described change in the second database is also referred to herein as “patching”.
• In several embodiments, the first database and the second database are relational databases, and the object being compared has at least one table which in turn has columns of predetermined data types. In many applications of the just-described embodiments, differences in tables between the two databases arise from changes being made to one or more software programs (also called “applications”) that use such databases, for example in upgrading the software program(s) to fix defects and/or provide additional functions, and such changes are propagated. If differences in the tables being compared are due to customizations of the second database that are permitted, then the second database is left unchanged. Therefore, automatic database patching in accordance with the invention, as described in the previous paragraph, supports application upgrade.
BRIEF DESCRIPTION OF THE FIGURES
• FIG. 1 illustrates, in a block diagram, a source computer programmed with a generation utility and a target computer programmed with a comparison utility in accordance with the invention.
• FIG. 2 illustrates, in a flow chart, acts performed in some embodiments of a generation utility of the type illustrated inFIG. 1.
• FIG. 3 illustrates, in a flow chart, acts performed in some embodiments of a comparison utility of the type illustrated inFIG. 1.
• FIGS. 4A-4G illustrate portions of a file that is generated by the generation utility ofFIG. 2 and used by the comparison utility ofFIG. 3 in certain embodiments of the invention.
• FIG. 5A illustrates, in a block diagram, portions of a source computer and of a target computer of the type illustrated inFIG. 1, andFIG. 5B illustrates, in a block diagram, software inmemory506 ofFIG. 5A.
• FIGS. 6A and 6B illustrate files used in two prior art methods for patching a database, namely output of a utility called “ODF” and SQL scripts respectively.
DETAILED DESCRIPTION
• Databases are automatically patched in some embodiments of the invention by use of twoutilities111 and131 that are respectively executed in asource computer110 and atarget computer130, as illustrated inFIG. 1.Source computer110 is coupled to a storage medium (such as a hard drive) containing asource database112 used by an application being maintained by a human (hereinafter “developer”)101.Developer101 identifies to sourcecomputer110 one or more objects (which are structures for storage of data, such as tables and views) used by the application, so that thetarget database132 can be patched when the application is changed intarget computer130. In many of the embodiments described herein,databases112 and132 are both relational databases, that are accessed through an identical database management system (DBMS). Note, however, that the invention may also be implemented in embodiments wherein the databases are not relational and/or wherein the databases (that are being kept in conformance) are accessed through different DBMS.
• Source computer110, when executing utility111 (also called “generation utility”) automatically extracts, fromsource database112, metadata that completely describes an object identified bydeveloper101, for example by querying a data dictionary contained insource database112. The data dictionary holds information about the structure ofdatabase112, such as (1) names of objects (e.g. tables, views, snapshots, indexes, clusters, synonyms, sequences, procedures, functions, packages, triggers, and so on), (2) information about integrity constraints, (3) default values for columns, (4) how much space has been allocated for and is currently used by the objects in the database, (5) auditing information, such as who has accessed or updated various objects.
• In many embodiments of the type described herein, a “schema” is a collection of logical structures of data, or schema objects. A schema in some such embodiments is owned by a database user and has the same name as that user, and each user owns a single schema. Schema objects in some embodiments can be created and manipulated with SQL and include the following types of objects: Clusters, Database links, Database triggers, Dimensions, External procedure libraries, Indexes and index types, Java classes, Java resources, and Java sources, Materialized views and materialized view logs, Object tables, object types, and object views, Operators, Sequences, Stored functions, procedures, and packages, Synonyms, Tables and index-organized tables, and Views.
• In several of the embodiments described in the previous paragraph, other types of objects are also stored in the database and can be created and manipulated with SQL but are not contained in a schema: Contexts, Directories, Profiles, Roles, Tablespaces, and Users. In such embodiments, schema objects are logical data storage structures. Schema objects of such embodiments do not have a one-to-one correspondence to physical files on disk that store their information. However, in these embodiments, the relational database stores a schema object logically within a tablespace of the database. The data of each object in such embodiments is physically contained in one or more of the tablespace's datafiles. For some objects, such as tables, indexes, and clusters, the embodiments may allow a developer to specify how much disk space the relational database allocates for the object within the tablespace's datafiles. Also, in many embodiments, there is no relationship between schemas and tablespaces: a tablespace can contain objects from different schemas, and the objects for a schema can be contained in different tablespaces.
• Note that the metadata being extracted byutility111 of some embodiments is all of the information that the data dictionary insource database112 maintains for a given object. Thereforeutility111 captures any change to the object that may have been made insource database112 by anyone for any reason.Source computer110 uses the extracted metadata to automatically prepare (when programmed by utility111) atext file120 that describes the developer-identified schema object, including its property values. For example, if the schema object identified by the developer is a table, thentext file120 contains at least (a) name, (b) owner, (c) data type of each column in the table, and (d) default data value of each column in the table.
• As noted at the beginning of the previous paragraph, many embodiments ofgeneration utility111 capture the values of all properties that schema objects of certain types may have. For example, schema objects that are tables have one or more of the following properties: partitioned/non-partitioned, type of partition, number of partitions, number of subpartitions, global temporary v/s permanent, index organized table (IOT), clustered or not, and tablespace. All such property values that are held in a data dictionary of a database for the developer-identified object are automatically captured in some embodiments ofutility111, for example by use of a Metadata API of the Oracle database management system. Therefore,utility111 of these embodiments eliminates the use of an enumerated set of properties that each object type has. Hence,utility111 does not need to be changed in any manner, even when one or more new properties are created for an object type during a future enhancement.
• Furthermore, in some embodiments,generation utility111 also captures metadata of one or more objects (called “secondary objects”) that are in some manner related to (e.g. depend on) developer-identified schema objects of certain types. For example, developer-identified objects that are tables could have, associated therewith, one or more of the following types of objects: indexes, policies, triggers and constraints. Such object types are identified in some embodiments in a predetermined set (called “containment list”) that is uniquely associated with (and hence identified by) each type of schema object that a developer may identify (also called “primary object’). As noted in the previous paragraph, for each secondary object (that is associated with a developer-identified object), all its property values are captured byutility111, based on the data dictionary, and hence file120 provides a complete description of a developer-identified object. Specifically, containment lists111 R (FIG. 1) are used with a data dictionary insource database112 to see if any information identified by the just-described list is available insource database112 for objects of the type (called primary objects) identified by the developer. Each object type, such as a table or a materialized view, has itsown containment list111R as shown inFIG. 1.
• Note that an object type is primary if its identification by the developer is accepted byutility111 for generation of acorresponding text file120, and otherwise the object type is secondary. Note also that secondary object types are normally present in one or more containment lists and hence they are not identified explicitly by the developer. Note further that each primary object type has an associated containment list (which may be empty in some cases). Depending on the embodiment, an object type that is primary in one embodiment may be secondary in another embodiment and vice versa.Utility111 contains an identification of the types of objects that are primary, and checks the developer's input to confirm that a primary object is being identified before generation of thecorresponding text file120.
• Note that although the word “list” is used in referring to a data structure that holds an identification of all secondary object types that may be associated with a primary object type, any other data structure (such as an array) may be used depending on the embodiment. One or more containment lists are used bygeneration utility111 of some embodiments to extract from the data dictionary insource database112 the metadata on all secondary objects (their properties) associated with a developer-identified object. The extracted metadata (in whole or in part, depending on the embodiment) of a given developer-identified object is written to thecorresponding text file120. Note thattext file120 is just one of a number of text files that are generated byutility111, e.g. one text file for each developer-identified object.
• Use of containment lists as discussed herein eliminates the need for a developer to explicitly identify toutility111 all objects whose metadata is to be captured, becauseutility111 automatically identifies all the secondary objects of a developer-specified primary object. Therefore,utility111 eliminates errors that arise from a developer overlooking one or more objects that may be present in a data dictionary and whose metadata needs to be captured to reflect a change in the database. Moreover,utility111 obtains metadata from the data dictionary which is the source of truth, thereby (1) eliminating errors that arise from the use of intermediate data (e.g. if the intermediate data is not up to date) and (2) eliminating the need for a CASE data-model repository and a snapshot utility. Such use of containment lists also results in storage of all metadata of all related objects in an single data structure (such as a file), which can improve performance during comparison with a target database, as discussed below.
• File120 is referred to herein as a “text” file when it contains character sequences, e.g. in any predetermined format such as ASCII or EBCDIC. File120 of all embodiments is a text file that can be edited by use of a simple text editor, and requires no special or proprietary editing tools. In contrast, binary files are not text-based and thus are neither editable nor accessible to text-oriented tools. Note that in some embodiments, file120 is expressed in an XML file format, which defines an industry standard syntax for data exchange, because XML allows for easy data transformation and manipulation unlike proprietary data file formats.
• Specifically, file120 of several embodiments is expressed in a predetermined language that contains tags, e.g. in extensible markup language XML or any language in conformance with SGML. XML and SGML are well known industry standards, as described at, for example, the website obtained by replacing dashes with dots in the following www-w3-org. Use of an industry standard syntax infile120 enables the expression of hierarchical relationships within the metadata infile120. For example, an industry standard syntax (e.g. as per XML) allows identification of any number of properties and the values associated with such identified properties for any object in the database.
• Regardless of which format (whether binary or text) and which language (XML or SGML) is used, file120 that has been prepared bysource computer110 is supplied to targetcomputer130.File120 may be supplied tocomputer130 in any conventional manner (e.g. by transmission over a network, or by storage to an optical storage medium that is physically moved betweencomputers110 and130).
• Next,target computer130 retrieves, from target database122, metadata that describes a version of the objects infile120, and compares the retrieved metadata to file120.Target computer130 changes the schema object version intarget database132 to conform to file120, depending at least partially on an outcome of comparing.
• Note that the mere fact that a schema object differs betweendatabases112 and132 does not automatically result in changing oftarget database132 in most embodiments, because the difference may arise from customizations that are allowed to be made to targetdatabase132. For example, a table intarget database132 may have been customized by addition of a new column, and/or by addition of a new index and if these are the only differences then the comparison utility leavesdatabase132 unchanged in the target computer (although it's structure is different from database112).
• Note that the above-described acts can be performed repeatedly, for each of a number of schema objects that are identified bydeveloper101.Developer101 may identify all schema objects that are used in an application being changed, or alternatively identify only those schema objects thatdeveloper101 knows are being changed.
• Note further that in many embodiments of the type described above,application developer101 also changes an application incomputer110 that usessource database112, and the changed application is supplied tocomputer130.Computer130 executes the changed application, ontarget database132, after the target database is changed as described above. In this manner, the above-described database patching supports application upgrades.
• In some embodiments,databases112 and132 are Oracle databases that are accessible through RDBMS Release 9.2, and an application programming interface (API) called Meta-data API is used by each ofcomputers110 and130 to extract metadata from therespective databases112 and132.
• Note thatcomputer130's comparison of metadata retrieved fromtarget database131 to file120 may be performed either directly (e.g. metadata is itself compared) or indirectly (e.g. a hash code is prepared using the metadata, and it is the hash codes that are compared). Specifically, in many embodiments of the invention,source computer110 prepares and stores infile120 one or more hash code(s) to be used bytarget computer130 in comparing with corresponding hash code(s) prepared from the schema object's version intarget database132. Use of hash codes during comparison results in performance optimization which was not achieved in any prior art system known to Applicants.
• If the two hash codes from the two databases match, then no change is made ontarget database132. If the two hash codes do not match, then the two metadata versions of each property and/or each secondary object are compared (which comparison can also be done either directly or indirectly). Specifically, in some embodiments,source computer110 prepares and stores in file120 a hash code for each and every secondary object whose metadata is being stored infile120. Therefore,target computer130 compares these hash codes with corresponding hash codes prepared from corresponding secondary objects of the schema object's version intarget database132.
• The following description is of an illustrative embodiment of the invention for Oracle E-Business Suite Application upgrade and/or patching process. In this embodiment, an XML-based database object definition (XDF) file is prepared and used in the context of an Oracle Database. The above-described two utilities are implemented as command line executables with the generation utility named as FNDXDFGEN and the comparison utility named as FNDXDFCMP. Note that in this embodiment, certain attributes are “secondary” objects, whereas schema objects that result in an XDF file are “primary” objects. Note that in this embodiment, there is one XDF file for each primary object. The following object types are supported in this embodiment.

• 	TABLE A
  			Primary/
  	Object Type	Details	Dependent

  1	Tables	These include normal tables, partitioned tables,	Primary
  		IOTs (index-organized tables), tables with
  		collection datatypes like VARRAYS and nested
  		table type,spatial datatypes
  2	Indexes	These include B*tree, Bitmap, Hash,	Dependent, as they are based on
  		Domain indexes, partitioned indexes.	specific columns in specific tables.
  3	Views		Primary
  4	Sequences		Optional - could be treated as per
  			user choice as either Primary/
  			secondary object
  5	Triggers		Optional - could be treated as per
  			user choice as either Primary/
  			secondary object
  6	Constraints	These include - Primary Key, Foreign Key,	Dependent, as they are based
  		Not null, Unique and Check Constraints	on tables.
  7	User Data-		Primary
  	Types (UDTs)
  8	Application		Primary
  	Context
  9	Queue		Primary
  	Tables
  10	Queus		Primary
  11	Policies		Dependent
  12	Materialized		Primary
  	Views on
  	Materialized
  	Views or
  	Base Tables
  13	Materialized		Primary
  	view logs on
  	base tables
  14	Materialized		Dependent
  	View logs on
  	Materialized
  	Views

• In this embodiment, the generation utility200 (seeFIG. 2) aids in creating XDF files containing database object definitions in XML file format. Thegeneration utility200 uses the Metadata API available in Oracle RDBMS v9i to capture the database object definition in XML and augments the XML to enable its use in patching. The generation utility captures object definitions for all specified objects in XDF files. Thegeneration utility200 extracts schema object definitions into xml format in files that end with the extension “.xdf”. The xdf extension is to distinguish XML files containing object definitions from other XML files containing data.
• The format of each xdf file is such that a primary object along with all its secondary object(s) are clubbed together in one file. For example—the definition of a table along with all its indexes, constraints and triggers are clubbed together in one single xdf file.
• Thegeneration utility200 is written in PERL (xdfgen) and Java (FndXdfGen). The PERL xdfgen utility is a simplified wrapper over the standalone java utility—FndXdfGen. Thegeneration utility200 augments the base definition generated by the Metadata API and has the following features. Thegeneration utility200 also: Automatically generates XDF filenames based on the name of primary object; Generates hash codes based on object definitions to enable optimum comparison at the target databases. The generated hashcode gets compared with the hashcode generated for the object in the target database. DBMS_UTILITY.GET_VALUE( ). Thegeneration utility200 uses a standard, more granular XDF structure where definition of a primary object gets clubbed along with relevant secondary objects. This allows for a more optimized comparison and enables the developer to easily make change/maintain the object definitions.
• Thegeneration utility200 also provides support for features like deferred Indexes, Drop/Creation of Constraints, CustomDefault. The former two features apply only in the context of Xdf Comparison utility (FndXdfCmp) invoked during application patching. The deferred indexes feature is applicable when extensive data-loading into tables is anticipated in a patch, subsequent to index creation. The deferred indexes feature defers the creations of the ‘flagged indexes’ to a subsequent phase after the data-loading to optimize the index rebuild. The Drop/Create Constraint feature is also useful in the data-loading context (similar to deferred indexes) or while making data-model changes that involves changes to relationships between tables.
• Note that the CustomDefault feature can be used even while running comparison utility300 in standalone mode. This is a very useful feature when a constraint on a column is being changed from NULL to NOT NULL, or while adding a new column with NOT NULL constraint. In such cases, this parameter which can either contain a value or SQL clause can be used. Also, hint generation (which results in a description) for all objects in the XDF files makes it easier for any calling program, e. g. a patching utility (called “ADPATCH”) to invoke comparison utility300.
• Generation utility200 also: captures database comments on tables, view, materialized view and their columns; generates tags for Oracle Applications specific metadata such as: Tablespace Classification, Logical Primary Keys & PK Columns, Logical Foreign Keys & FK Columns, Histogram Information, and Translation related information (e.g. whether the data is translatable into a another language, such as German).
• Comparison utility300 (seeFIG. 3) aids in propagating the definitions of schema objects from a source database to target databases by using compare-merge techniques and can be invoked by any patching utility to process XDF files. In this manner, XDF utilities ofFIGS. 2 and 3 facilitate efficient propagation of database component changes with application changes.
• In general, the XDF utilities ofFIGS. 2 and 3 help to streamline the Oracle E-Business Suite Application patching process in the context of database component changes. They also make all Oracle E-Business Suite Applications patches more efficient, performance optimized and easy to debug/diagnose with respect to database component changes. By converging all database component changes into XML format, they enable customers to proactively review and analyze changes that are to be made to their database on application of a patch in a single and consistent manner. Note that this happens when a debug feature of the comparison utility is used in some embodiments.
• Theutilities200 and300 ofFIGS. 2 and 3 are generic and applicable for use in patches/patching process of any product outside of Oracle that uses Oracle database components. For example, any other Enterprise Applications product or Suite of Applications based on the Oracle Database can use the generation and comparison utilities illustrated inFIGS. 2 and 3 in their patches/patching process.
• Generation utility200 ofFIG. 2 includes two components, aprogram210 written in Java that performs acts212-214, and anotherprogram220 written in a language for stored programs (e.g. PL/SQL) that performs acts221-225. Specifically, input from a human (also called “user” or “developer”) is provided inact201 to identify the database and to identify one or more primary objects whose metadata is to be written into the XDF file.Program210 validates the received information inact212, e.g. to confirm that the user who provided the information is the owner, etc.Program210 also checks inact212 that the identified object is a primary object (as per Table “A” above).
• If the user's input is found to be valid inact212, thenprogram220 is invoked.Program220 performsact221 to extract the metadata for a primary object identified in the user's input, as peract221. As noted above, the Metadata API of Oracle RDBMS is used in some embodiments to obtain the metadata of the primary object from the source database. Note that the functions of the Metadata API access the system data dictionary. Next, inact222,program220 retrieves a containment list of secondary object types that the primary object is likely to have. The just-described list of secondary object types is defined in this particular embodiment in the form of containment rules, as shown in Table “B”.

• 	TABLE B
  	Primary
  	Object		Secondary	Containment in
  	Types	Details	objects	XDF

  1	Tables	These include Global Temporary tables,	Indexes,	One Table & all its secondary
  		IOTs (index organized tables), Queue	triggers	object definitions in an
  		Tables, normal tables, Tables with	(optional),	XDF file.
  		types (eg. VARRAYS and nested tables	constraints,
  		types, object types, spatial	sequences,
  		data-types)Partitioned tables	policies
  2	Materialized		Indexes,	One MV & its secondary object
  	Views on		triggers	definitions in an XDF
  	Materialized		(optional),
  	Views		Materialized
  			view logs.
  3	Materialized		Indexes,	One MV & its secondary object
  	Views on		triggers	definitions in an XDF
  	Base		(optional)
  	Tables
  4	Materialized		Indexes,	One MV Log & its secondary object
  	View		triggers	definitions in an XDF
  	Logs on		(optional),
  	base		constraints
  	Tables
  5	Views			One View definition perXDF
  6	Application			One Context definition perXDF
  	Context
  7	Queue			One Queue Table definition
  	Tables			per XDF
  8	Queue			One Queue definition per XDF
  9	User Data			One User Data Type per XDF
  	Types
  10	Sequences	Optional - could be treated as per user		One or more sequences per XDF
  		choice as either Primary/secondary
  		object
  11	Triggers	Optional - could be treated as per user		One or more triggers per XDF
  		choice as either Primary/secondary
  		object

  
  In the above table, “MV” is an abbreviation of “materialized view”, which is further described elsewhere herein.
• Referring toFIG. 2, afteract222, act223 is performed to check whether or not, based on the containment rules for the current object, there are any secondary objects existing in the source database. If the outcome is no, then act224 is performed byprogram220 to generate metadata, followed by act213 inprogram210. In act213, XSL transformation is applied to replace schema references by parsing the object definition. Specifically, schema references for an object (including owner of the object and any other schema references that could be part of the object definition) in the source database are likely to differ at the target database. For example, a customer's requirement could be that same object be created under another schema owner in the target database. For this reason, additional identifiers are added to the XDF file (in a process called “remapping”), to facilitate schema replacement at the target database, to support portability.
• Next, as peract225, a hash code is prepared for each object in the metadata (regardless of the object's type being primary or dependent). While a hash code may be prepared in any conventional manner, in some embodiments the hash code is obtained by invoking an API in the Oracle database called DBMS_UTILITY.GETHASHVALUE( ). The hash code being prepared is for use in comparing the corresponding definitions of object types (like Tables, Indexes, Queues, Queue Tables, Sequences) in a target database. In case of objects like Tables, there is a base hashcode generated for the table itself. There are additional hashcodes generated for certain properties (e.g. column in a table). This is because, in case of differences in the object being compared, there are additional properties that are compared to quickly identify the change (e.g. column level properties like Column name, Data-types, Data-type length, Constraints, Default Values may be individually compared if a specific column's hash codes are found to be different during comparison).
• In an illustrative example, the following table's metadata is extracted from the data dictionary:

• SQL> DESC TEST_TABLE

  	Name	Null?	Type
  	COL1		NUMBER(15)
  	COL2		DATE

  
  To obtain an individual hashcode, for each column in a table,utility200 constructs a concatenated string of that column's properties such as name, Data-types, Data-type length, Constraints, Default Values, etc. Hence in the above example, to get the base hashcode for COL1—pass a concatenated string—‘COL1NUMBER15’ to DBMS_UTILITY.GET_HASH_VALUE, which returns the hashcode for COL1. Similarly get hashcodes for COL2-COL4. To get the base hashcode for TEST_TABLE—pass a concatenated string containing Table name, table properties like ‘Cache’/“NoCache’/‘Logging’/‘Nologging’/etc and individual hashcodes for each column to DBMS_UTILITY.GET_HASH_VALUE, which returns hashcode for TEST_TABLE.
• After hash codes are obtained, they are written to the xdf file, followed by performance ofact214 inFIG. 2. Specifically, inact214, invocation hints are added to the xdf file, followed by writing of the file to a medium to be used to transfer the xdf file(s) to the target computer.Act214 is the last act inprogram210 which therefore concludes, thereby to complete the execution ofgeneration utility200.
• Note that inact223, if a secondary object is found to exist in the source database (i.e. the outcome is yes), then act226 is performed to extract metadata of the secondary object, followed byact227 in which the secondary objects metadata is physically appended to metadata previously extracted, to form valid XML. Thereafter, act223 is again performed, until there are no secondary objects of any of the types defined by object containment rules in Table “B” listed above.
• Note thatact223 of some embodiments may incorporate multiple checks, e.g. an initial check as to whether an object of the type currently being processed has a containment rule, and if so whether any object types are identified by the containment rule for this type of object, and if so whether any actual objects are defined in this database that are of the object type identified in the containment rule. All such checks are performed automatically, to enablegeneration utility200 to extract the metadata of any object that a developer may identify, as well as any dependent objects that the developer-identified object may have.
• Comparison utility300 (FIG. 3) parses thefile120 prepared by thegeneration utility200, for object definitions that reside in it and compares each object definition with that of a corresponding object present in the target database. If a corresponding object is not present, utility300 attempts to create the object by extracting a create DDL from thefile120, and otherwise, it applies delta differences (Alter DDLs) to make the object definitions in thefile120 match to thetarget database132.
• Comparison utility300 of some embodiments also handles differences in database DDL syntax with respect to new features available in specific versions. Comparison utility300 of several embodiments also has a debug mode which when invoked reports all the changes that are intended to be made based on aspecific file120 and aspecific target database132, without actually making the changes.
• Comparison utility300 of one embodiment has the following advantages: 1) unlike prior (non-generic) SQL script-based solutions, no manual analysis is required for each database component change, including revision history of the database object; 2) less error-prone and therefore improves patch success rates; 3) enables easy diagnosis and analysis in case of failures; 4) provides support for all Oracle database components; 5) easier to maintain, evolve and avoid costs in terms of resources and time; 6) optimized to improve time taken during patching; 7) eliminates myriad different files involved previously & replaces them with XML file format, which is the Industry standard
• A comparison utility300 of this embodiment is written in PERL and Java, and it compares database object definitions contained in thefile120 with the actual database object definitions in the target database and then applies the changes to the object in the target database accordingly. This comparison utility300 also contains the following features: Supports Customization to object storage that is frequent occurrence at most customer databases; supports both Dictionary and locally managed Tablespaces available in Oracle Databases; supports propagating database comments for tables/views/materialized views at the target database; and supports automatic drop/recreate of Indexes and materialized view logs on materialized views. When a XDF materialized view query differs from the query in the target database the MV needs to be dropped and re-created. Once a materialized view is dropped the dependent indexes and materialized view logs are automatically dropped. XDF handles the re-creation of these secondary objects automatically.
• Comparison utility300 of this embodiment also supports object creation in parallel, grants privileges and creates synonyms for base schema objects to the appropriate APPS (application) schema(s) at the time it creates the base schema objects. It verifies the synonyms in the same way that it verifies the base schema objects themselves. This comparison utility300 also updates AOL (Application Object Library) Data Dictionary for schema object definition related changes. These tables are automatically updated by XDF with the relevant data. Finally, this comparison utility300 also support for new and old tablespace models & integration with the AD (Applications Database) Tablespace API.
• AOL (Application Object Library) Data Dictionary tables updated by the XDF Comparison utility are as shown in Table “C” below.

• TABLE C
  FND_TABLES	FND_COLUMNS
  FND_FOREIGN_KEYS	FND_PRIMARY_KEYS
  FND_FOREIGN_KEY_COLUMNS	FND_PRIMARY_KEY_COLUMNS
  FND_VIEWS	FND_HISTOGRAM_COLS
  FND_OBJECT_TABLESPACES

• Referring toFIG. 3, on receipt ofinput301 from a human, act312 is performed to validate the input. Specifically, a database connection is established in act312 and an xdf file identified by the human is checked to confirm that its syntax is valid. Note that in many embodiments of the type described herein, the xdf file being checked is the same xdf file which is prepared by the generation utility discussed above. Note, however, that the xdf file may be an altered version of the generated xdf file, depending on the embodiment. For example, because the xdf file is expressed in XML format, a simple text editor can be used by a human to modify the generated xdf file and the modified file can be supplied to the comparison utility300. Note further that since the xdf file is in XML format a source code control system (such as SCCS available for the Unix operating system) is used in some embodiments to version control the xdf file. Such version control of the xdf file is used in several such embodiments with version control of application source code, so that changes being made to a database (e.g. addition of a new column to an existing table) are in conformity with changes being made to application source code that refers to the database (e.g. references to the new column).
• After act312, the comparison utility300 performs anact313 wherein metadata is extracted from the xdf file. The metadata is being extracted for all the objects in the file. Next, inact314 the target database is checked for existence of the object whose metadata is extracted inact313. If the object doesn't exist, then act315 is performed. Inact315, XSLT is applied to 1) replace schema references and storage properties for the object and 2) build the Create DDL (data definition language) for the object. XSLT is the XSL Transform language being used to apply remap of schema and storage for the object, as these could differ at the target database.
• Act315 of the comparison utility300 builds the Create DDL. Therefore, afteract315 is completed, act316 is performed to execute a CREATE/ALTER DDL command on the target database. Next, inact317, a check is made to see if metadata for any additional secondary object is still present in the xdf file, and if so then control returns to act313. Inact317 if no additional metadata is present, then act318 is performed, wherein APPS (Applications) specific metadata present in the AOL (Application Object Library) Data Dictionary is extracted from the xdf file and updated in the AOL (Application Object Library) Data Dictionary at the target database. Next, logs are generated inact319 which record the results of all the Xdf comparison acts including any errors, warnings and finally reports either success/failure of the metadata propagation to the target database. Applications System Administrators and Database Administrators will monitor the logs to gauge the success/failure of the patch.
• Inact314, if the object is found to exist in the target database, then act321 is performed wherein hash codes are compared. If the hash codes match inact322, then act317 is performed as discussed above. If there is no match in the hash codes, then act323 is performed wherein the object metadata is extracted from the target database. Next, act324 is performed wherein the object metadata in the xdf file is compared with the metadata from the target database. All differences between metadata are resolved in favor of the xdf file metadata. Next, act316 is performed, as discussed above.
• FIGS. 4A-4G illustrate the output of one embodiment of the above-describedgeneration utility200 in the form of a single file test_table.xdf which uses the above-described TEST_TABLE example. In view of the descriptive annotations provided to test_table.xdf, the tags shown inFIGS. 4A-4G are self-explanatory to a skilled artisan, as these tags and the metadata representation in the XDF directly correspond to the system data dictionary in an Oracle Database. Note that all of the text inFIGS. 4A-4G is located one after another in a single file of extension “xdf”, although shown separated in the individual figures for convenience. The very beginning of this file contains the text inFIG. 4A which includes ahint401 to assist a user of a target computer in invocation of the comparison utility.
• Followinghint401 is alist402 of all objects whose metadata is present in this file. Followinglist402 is a set of tags411 (FIGS. 4B and 4C) that contain application specific metadata. Following the applicationspecific metadata tags411 is the set of tags412 (FIGS.4D-4F) that contain metadata of the TEST_TABLE object. Note that a hash code421 (FIG. 4D) is included in this file for the entire TEST_TABLE object. Thishash code421 is generated from not only property values of the table, such as name, flags inFIG. 4D and blocks, groups inFIG. 4E, but also additional hash codes as noted next. Specifically, infile120, a hash code is also included for certain properties such as each column, as illustrated byhash codes422 and423 (FIG. 4F) for the two columns in this example. Note that thesehash codes422 and423, are used in preparation ofhash code421 for the table. All hash codes421-425 are sent over from the source computer to the target computer, at their appropriate locations infile120 to enable quick comparison.
• Note that after the metadata of the TEST_TABLE object, in this example file120 contains additional metadata as shown inFIG. 4G which is automatically extracted bygeneration utility200 as peract226 inFIG. 2, for a secondary object that is automatically identified in this example as being a constraint from a containment list for “table” type primary objects (because object TEST_TABLE is a table type primary object). Note that the metadata inFIG. 4G for the index is placed infile120 immediately after the metadata of TEST_TABLE.
• Utilities200 and300 are used to program a computer system500 of the type illustrated inFIG. 5 which is discussed next. Specifically, computer system500 includes a bus502 (FIG. 5) or other communication mechanism for communicating information, and aprocessor505 coupled withbus502 for processing information. Computer system500 also includes amain memory506, such as a random access memory (RAM) or other dynamic storage device, coupled tobus502 for storing information and instructions to be executed byprocessor505.
• Main memory506 also may be used for storing temporary variables or other intermediate information during execution of instructions to be executed byprocessor505. Computer system500 further includes a read only memory (ROM)508 or other static storage device coupled tobus502 for storing static information and instructions forprocessor505. Astorage device510, such as a magnetic disk or optical disk, is provided and coupled tobus502 for storing information and instructions.
• Computer system500 may be coupled viabus502 to adisplay512, such as a cathode ray tube (CRT), for displaying information to a computer user. Aninput device514, including alphanumeric and other keys, is coupled tobus502 for communicating information and command selections toprocessor505. Another type of user input device is cursor control516, such as a mouse, a trackball, or cursor direction keys for communicating direction information and command selections toprocessor505 and for controlling cursor movement ondisplay512. This input device typically has two degrees of freedom in two axes, a first axis (e.g., x) and a second axis (e.g., y), that allows the device to specify positions in a plane.
• As described elsewhere herein, automated database patching is provided by computer system500 in response toprocessor505 executing one or more sequences of one or more instructions contained inmain memory506. Such instructions may be read intomain memory506 from another computer-readable medium, such asstorage device510. Execution of the sequences of instructions contained inmain memory506 causesprocessor505 to perform the process steps described herein. In alternative embodiments, hard-wired circuitry may be used in place of or in combination with software instructions to implement the invention. Thus, embodiments of the invention are not limited to any specific combination of hardware circuitry and software.
• The term “computer-readable medium” as used herein refers to any medium that participates in providing instructions toprocessor505 for execution. Such a medium may take many forms, including but not limited to, non-volatile media, volatile media, and transmission media. Non-volatile media includes, for example, optical or magnetic disks, such asstorage device510. Volatile media includes dynamic memory, such asmain memory506. Transmission media includes coaxial cables, copper wire and fiber optics, including the wires that comprisebus502. Transmission media can also take the form of acoustic or light waves, such as those generated during radio-wave and infra-red data communications.
• Common forms of computer-readable media include, for example, a floppy disk, a flexible disk, hard disk, magnetic tape, or any other magnetic medium, a CD-ROM, any other optical medium, punch cards, paper tape, any other physical medium with patterns of holes, a RAM, a PROM, and EPROM, a FLASH-EPROM, any other memory chip or cartridge, a carrier wave as described hereinafter, or any other medium from which a computer can read.
• Various forms of computer readable media may be involved in carrying one or more sequences of one or more instructions toprocessor505 for execution. For example, the instructions may initially be carried on a magnetic disk of a remote computer. The remote computer can load the instructions into its dynamic memory and send the instructions over a telephone line using a modem. A modem local to computer system500 can receive the data on the telephone line and use an infra-red transmitter to convert the data to an infra-red signal. An infra-red detector can receive the data carried in the infra-red signal and appropriate circuitry can place the data onbus502.Bus502 carries the data tomain memory506, from whichprocessor505 retrieves and executes the instructions. The instructions received bymain memory506 may optionally be stored onstorage device510 either before or after execution byprocessor505.
• Computer system500 also includes acommunication interface515 coupled tobus502.Communication interface515 provides a two-way data communication coupling to a network link520 that is connected to alocal network522.Local network522 may interconnect multiple computers (as described above). For example, communication interface518 may be an integrated services digital network (ISDN) card or a modem to provide a data communication connection to a corresponding type of telephone line. As another example,communication interface515 may be a local area network (LAN) card to provide a data communication connection to a compatible LAN. Wireless links may also be implemented. In any such implementation,communication interface515 sends and receives electrical, electromagnetic or optical signals that carry digital data streams representing various types of information.
• Network link520 typically provides data communication through one or more networks to other data devices. For example, network link520 may provide a connection throughlocal network522 to ahost computer525 or to data equipment operated by an Internet Service Provider (ISP)526.ISP526 in turn provides data communication services through the world wide packet data communication network528 now commonly referred to as the “Internet”.Local network522 and network528 both use electrical, electromagnetic or optical signals that carry digital data streams. The signals through the various networks and the signals on network link520 and through communication interface518, which carry the digital data to and from computer system500, are exemplary forms of carrier waves transporting the information.
• Computer system500 can send messages and receive data, including program code, through the network(s), network link520 and communication interface518. In the Internet example, aserver550 might transmit a requested code for an application that has been changed and file120 (describing objects in a changed database that are used by the application) through Internet528,ISP526,local network522 andcommunication interface515. In accordance with the invention, one such downloaded application and file120 provides for implementing automated patching of a database, as described herein.
• The received application and file120 may be executed byprocessor505 as it is received, and/or stored instorage device510, or other non-volatile storage for later execution. In this manner, computer system500 may obtain thefile120 and the related application in the form of a carrier wave.
• Note thatFIG. 5A is a very low-level representation of many hardware components of a computer system. Several embodiments have one or more additional software components inmain memory506 as shown inFIG. 5B: Operating System591 (e.g. Microsoft WINDOWS 2000), Database Server595 (e.g. Oracle Server v9i2 for the source computer; e.g. Oracle Server v8i for the target computer), Java Development Kit593 (e.g. JDK v118), Java XMLParser592 (e.g. xmlparser available from Oracle Corporation), and JDBC drivers594 (e.g. JDBC driver available from Oracle Corporation).
• Numerous modifications and adaptations of the embodiments described herein will become apparent to the skilled artisan in view of this disclosure. For example, although twocomputers110 and130 are shown inFIG. 1, all acts described above as being performed by these computers are performed by a singe computer in an alternative embodiment. In the alternative embodiment, the source database and the target database are both present in the single computer. In still other embodiments, generation offile120 and comparison with target database122 are performed in one computer while changing of the database is performed in another computer. Moreover, file120 as well as changes to an application that are supported by database changes documented infile120 may be packaged into a set of files that are distributed together (on a single storage medium or on the same carrier wave) to customers interested in upgrading their version of the application. In addition, the comparison utility may itself be packaged into the just-described distribution in some embodiments. Also, the above-described generation and comparison utilities can be used i) by internal development teams during creation and testing of object definitions and ii) for propagating these object definitions to customer databases during patching.
• Numerous modifications and adaptations of the embodiments described herein are encompassed by the scope of the invention.
• The following appendices A-E are integral parts of this detailed description and are incorporated by reference herein in their entirety. These appendices provide further detailed descriptions of an illustrative embodiment of the type shown inFIGS. 2 and 3.
APPENDIX AAssumptions
• The following assumptions have been made in the context of XDF Generation/Comparison Utility Design.
• The format of the xdf (xml file containing object definitions) file is such that the parent object and its secondary object(s) are clubbed together in one file. For Example—the table definition along with all its indexes, constraints and triggers are clubbed together in one single XML file. XDF utility creates the grants and synonyms for base schema objects to the appropriate APPS (applications) schema(s) at the time the base schema objects are created. This happens in the comparison utility. This utility verifies the synonyms in the same way that it verifies the base schema objects themselves, and only recreates the synonym if they are missing or incorrect. XDF utility creates a grant for a base object giving the corresponding APPS (applications) schema ‘grant all with grant option’ on that base object.
• Tables
• In addition to normal tables, XDF supports the following types of tables:
• • Global Temporary Tables.
  • Queue Tables.
  • Partitioned Tables.
  • Index Organized Tables (IOTs).
• 1. XDF utility supports both the database native default clause and also the custom default value. Product team that require the default clause in effect after the patch is applied will use the native default clause and product teams which require the default clause only during the time the patch is applied, typically for a column modification from NULL to NOT NULL will use the custom default clause.
• 2. Support for range and hash partitions during table creation will be supported.
• 3. No partition maintenance operations are supported.
• 4. XDF utility does not support partitioning of an existing non-partitioned table. This requires that the following steps be performed which are time consuming and may require manual intervention
• • Record grants and constraints on the table
  • Disable constraints
  • Rename table
  • Create table as partitioned
  • Transfer data from non-partitioned table to partitioned table, either by Exchanging partitions or DML
  • Enable constraints and recreate grants
• Hence XDF utility does not support partitioning of an existing non-partitioned table.
• 5. XDF utility provides support for creating tables with collection datatype like Varrays and nested Table as the datatype of a relational table. Here it is assumed that the corresponding Type reference already exists.
• 6. Global temporary tables cannot be altered or dropped if in use by another session. DDL statements on these tables must be done in a phase where no DML is using these tables.
• Constraints
• 1. XDF utility supports constraints—Primary Key, Foreign key, Not null, Unique and Check Constraints. By default these constraints are created along with the object creation unless the drop and recreate flag for the constraints has been defined. In that case the constraints are not created when the underlying table is created/modified. If the constraints are found to already exist, then they are dropped. The constraints are then created/recreated later in a different phase. The drop recreate flag concept is similar to the defer flag passed to FNDXDFGEN utility. This flag should be used only in cases like when patches change data structures and require a data upgrade and a constraint change and the data cannot be changed with the old constraint in place, and the old constraint cannot be replaced with the new one until the data has changed.
• 2. If a constraint has evolved then the existing constraint is dropped and recreated with the definition in the XDF file. If the constraint is Primary or Unique Key then the CASCADE option is used which drops any existing referential constraints. The onus is on the development team to make sure that any dependent foreign keys are released in the same patch as the evolved Primary or Unique Key.
• 3. The constraints are an integral part of the table XDF definition. There is no separate section in the XDF file dealing with constraints. Also any modification to the constraints is done along with the underlying table changes after the indexes on the tables have been created. This prevents the indexes from being created in the wrong tablespace.
• Views
• The query underlying the view should not contain any schema references. The view query should be using the synonym created for the base tables and should not refer the base tables directly.
• Materialized Views
• Materialized views are schema objects that can be used to summarize, precompute, replicate, and distribute data. They are suitable in various computing environments such as data warehousing, decision support, and distributed or mobile computing. Materialized views are used to precompute and store aggregated data, such as sums and averages. They can also be used to precompute joins with or without aggregations. Cost-based optimization can make use of materialized views to improve query performance by automatically recognizing when a materialized view can and should be used to satisfy a request. The optimizer transparently rewrites the request to use the materialized view. Queries are then directed to the materialized view and not to the underlying detail tables or views. Materialized views are similar to indexes in several ways: they consume storage space, they must be refreshed when the data in their master tables changes, and, when used for query rewrites, they improve the performance of SQL execution and their existence is transparent to SQL applications and users.
• Oracle maintains the data in materialized views by refreshing them after changes are made to their master tables. The refresh method can be incremental (fast refresh) or complete. For materialized views that use the fast refresh method, a materialized view log or direct loader log keeps a record of changes to the master tables.
• Materialized views can be refreshed either on demand or at regular time intervals. Alternatively, materialized views in the same database as their master tables can be refreshed whenever a transaction commits its changes to the master tables.
• A materialized view, when created, will create a table, an index and may be one or more views, all in the same schema as that of the materialized view. If you create a materialized view MV, a table (MV), an index (I_SNAP$_MV) will be created in the same schema.
• XDF supports the following Materialized View features
• • Partitioned Materialized Views
  • Hierarchical Materialized Views (i.e. Materialized Views on Materialized Views)
  • Other Materialized View features provided by the RDBMS.
• XDF does not support the following Materialized View features at this point
• • Creation of Refresh Groups
  • Materialized views in a distributed environment (for eg. Over DB links)
  • Subsequent alters to a Partitioned MV
• XDF recommends the following standards, applicable to Materialized Views creation and modification. Some of these are enforced
• • Pre-built Tables are not used in conjunction with Materialized Views and are not supported. It is stated—“Do not create MV with pre-built option. Pre-built does not impose the integrity check on the MV and it is up to the users to verify it, making it an unreliable option to use for DBI”—Enforced
  • Materialized Views should be created in the APPS schema. A Materialized View can reference multiple tables belonging to multiple products and only the APPS schema is guaranteed to have required privileges, on these tables.
  • Base schema tables, referenced in Fast-refresh MV query, must be qualified with schema name, as there are some issues with referencing the synonyms directly. This is required for fast refresh to work.—Enforced
  • The Materialized View should be initially created with deferred Refresh.
  • Refreshes are not handled by XDF other than the initial refresh during creation of the Materialized View.
  • Comments on the Materialized View and all columns in the Materialized View are required for future eTRM purposes
• XDF alters the Materialized view in the following cases
• • Storage parameters change
  • Refresh method, mode, or time change
  • Enable or disable query rewrite.
• XDF drops & re-creates the Materialized View in the following cases
• • Query change Note: Dropping and recreating Materialized Views is an expensive operation, requiring the handling of secondary objects like (MV Logs, I\ndexes). It requires a complete refresh of the MV and other dependent MVs that reference it.
• Materialized View Logs
• • A materialized view log is a schema object that records changes to a master table's data so that a materialized view defined on the master table can be refreshed incrementally.
  • Each materialized view log is associated with a single master table or a materialized view. The materialized view log resides in the same database and schema as its master table or materialized view.
• XDF supports the following Materialized View Log features
• • Creation of Partitioned Materialized View Logs
  • Materialized View Logs on Materialized Views (partitioned/non-partitioned) based.
• XDF alters the Materialized view Log in the following cases
• • Storage parameters change
  • Filter Column changes
• XDF drops & re-creates the Materialized View Log in the following cases
• • Rowid to Primary Key & vice versa
  • Filter Columns in Target DB, but not in XDF
• Sequences
• The ‘START WITH’ value of a sequence cannot be modified once the sequence has been already created because the data could be potentially already used.
• Indexes
• • 1. The physical attributes of a partition in a partitioned index are assumed to be the same as that of the underlying index. If different sets of values have been specified for different partitions then modification to these physical attributes is not supported byXDF phase 2.
  • 2. If the functional indexes use a user-defined function then the function must exist in the database i.e. the creation of function-based index should in a phase after the creation of PL/SQL packages.
  • 3. As per divisional standards for Function-based indexes—The packages are created in the APPS schema and in order to create a function-based index, the schema owning the table (BASE) must have EXECUTE privilege on the functions used in the index. Also, the packages are created with INVOKERS RIGHTS privileges. So the schema owning the table must have access to underlying objects.
• Types
• Object types are schema objects. They are abstractions of the real-world entities—for example, purchase orders—that application programs deal with. There are two categories of object datatypes:
• • Object types
  • Collection types
• Object datatypes use the built-in datatypes and other user-defined datatypes as the building blocks for datatypes that model the structure and behavior of data in applications.
• Each collection type describes a data unit made up of an indefinite number of elements, all of the same datatype. The collection types are array types and table types. The corresponding data units are called VARRAYs and nested tables.
• XDF alters the ADT in the following cases:
• • Storage parameters change
  • Filter Column
• XDF drops & re-creates the ADT in the following cases:
• • Rowid to Primary Key & vice versa
  • Filter Columns in Target DB, but not in XDF
APPENDIX BDesign Details
• Table and Constraints on Table
• XSL (kutable.xsl) is used to transform the XDF file containing the description of the table into a create DDL statement, if the object does not exist in the target database. If the object exists then a hashcode is generated for the object from the target database and is compared with the hashcode present in the XDF. If they do not match the individual columns are compared based on the column names. The column sequence of the column is not considered. Any difference in the column structure is modified with ‘ALTER TABLE’ clause.
• To add an additional column to the table the following syntax is used
• ALTER TABLE <table name> ADD (<column definition>)
• If the column to be added is a NOT NULL column then a database default value has to be provided. Otherwise an error is returned indicating that a NOT NULL column without any default value is being added to the existing table.
• To modify a column from NOT NULL to NULL
• ALTER TABLE <table name> MODIFY (<column name> NULL)
• To modify a column from NULL to NOT NULL
• ALTER TABLE <table name> MODIFY (<column name> NOT NULL)
• If the column contains NULL values then the above DDL statement fails and the corresponding error is returned.
• To modify other properties of a column
• ALTER TABLE <table name> MODIFY (<new column definition>)
• The datatype of any column can be changed only if all the rows for the column are null. The size of a column can be increased. But to decrease the size of column all the rows for the column must contain NULL. If the column to be modified is part of the table or index partitioning key or part of a domain index the definition of the column cannot be modified. Corresponding errors are returned for any of the above exceptions.
• To modify the storage parameters
• ALTER TABLE <table name> STORAGE (<storage clause> <value>)
• ALTER TABLE <table name> INITRANS <value> MAXTRANS <value> etc. . . .
• The storage parameters are modified only if the modified value is greater than the current existing default value. Error related to auto segment space management is ignored. Also the degree of parallelism is always set to NOPARALLEL if otherwise. The PCTINCREASE is also always set to 0 if otherwise and similarly MAXEXTENTS to UNLIMITED.
• To modify the integrity and referential constraints
• If the constraint already exists and has evolved then the constraint is first dropped using
• ALTER TABLE <tableName> DROP CONSTRAINT <constraintName>
• The keyword CASCADE is appended to the above ALTER statement if the constraint is a primary or unique key.
• The constraint definition is then constructed from the xdf file and created using the following statement
• ALTER TABLE <tableName> ADD CONSTRAINT <constraintName>
• [UNIQUE| PRIMARY KEY] (column list separated by a ‘,’)
• If the constraint is a foreign key constraint then the following DDL is used
• ALTER TABLE <tableName> ADD CONSTRAINT <constraintName>
• FOREIGN KEY (column list separated by a ‘,’) REFERENCES <tableName> (column list separated by a ‘,’)
• The referenced table name cannot exist in multiple schemas. The table schema should be uniquely identifiable.
• If the constraint is a check constraint then the following DDL is used.
• ALTER TABLE <tableName> ADD CONSTRAINT <constraintName> CHECK (check condition)
• Views
• If the view contained in the XDF file does not exist in the target database then a XSL file (kuview.xsl) is used to transform the XDF file into a create DDL statement. If the object exists then the underlying view query in the xdf file is compared with the query in the database and if different the view is re-created by transforming the XDF file into a create or replace DDL statement.
• ODF creates all views with FORCE option. The default behavior for existing views is “replace” (replace existing view using ‘CREATE OR REPLACE FORCE VIEW’).
• Sequences
• The ‘kuseq.xsl’ XSL file is used to transform the XDF file containing the description of the sequence into a create DDL statement, if the sequence does not exist in the target database. If the object exists in the target database the existing sequence is compared with the sequence description in the XDF file and any difference is modified with the ALTER SEQUENCE clause.
• An Error is returned if the existing START WITH value of the sequence is greater than MAXVALUE of the sequence in the XDF file.
• The DDL statements that can be issued to equate the sequence in the target database are
• ALTER SEQUENCE <sequenceName> INCREMENT BY <new Value>
• [MAXVALUE <integer>/NOMAXVALUE] <MINVALUE
• <integer>/NOMINVALUE] [CYCLE/NOCYCLE] [CACHE
• <integer>/NOCACHE] [ORDER/NOORDER]
• Indexes
• The ‘kuindex.xsl’ XSL file is used to transform the XML file containing the description of the index, into a create DDL statement, if the object does not exist in the target database. If the DDL statement execution fails because of “maximum number of extents exceeded” error, then the initial extent is set to the value of how much space the index used when it hit maxextents and the process is looped to maximum of ten times for the CREATE statement to succeed.
• If the create statement fails because of “column list already indexed” error then an attempt is made to rename the already existing index to the one specified in the XML file. But if the uniqueness of both the indexes is different then the existing index is dropped and recreated according to the definition in the XML file along with a warning message.
• An error message is displayed for any other unacceptable error in the creation of the index.
• To rename the index the following DDL statement is used:
• ALTER INDEX <existing index name> RENAME to <indexName>
• To alter storage parameters the following statements are used:
• ALTER INDEX <indexName> STORAGE (<storage clause> <value>)
• ALTER INDEX <indexName> INITRANS <value> MAXTRANS <value> etc. . . .
• To split or add to an existing partitioned index the following statement is used
• ALTER INDEX <indexName> SPLIT PARTITION <partition_name> INTO (PARTITION <partition_name1> PARTITION <partition_name2>)
• To rename an existing partition
• ALTER INDEX <indexName> RENAME PARTITION <partition name> TO <partition name1>
• If the index already exists in the database, then the index in the XML file and the index in the database are compared. If the difference is because of any of the following cases then the index is dropped and recreated:
• The table name or table owner is incorrect
• The uniqueness is wrong
• The database index has too few columns
• The database index has too many columns
• The database index has right number, wrong column names
• The database index has right number, right names, and wrong order
• The database index is a non-partitioned index and the file index is a partitioned index and vice versa
• The database index is a global index partition and the file index is a local index partitions and vice versa.
• The database index partition is a range partition and the file index is a hash partition
• The database index is bitmap index and file index is a non-bitmap index
• The following statement is used to drop the index:
• DROP INDEX <index name>
• Once the index is dropped then the XSL transformation is applied to the XML file to generate the CREATE INDEX statement.
APPENDIX CGenerator Utility Usage Notes
• The XDF generator program is the next generation version of the current odf utilities to generate database object definitions in XML. The specific file extension used for these xml files is .xdf (to distinguish xml files containing object definitions from others). The generator program currently requires Oracle 9iR2 database, since it uses the metadata API (dbms_metadata).
• Note: The xdf file contains the complete definition of the primary object and secondary object(s). After initial generation of the xdf file, subsequent changes, regardless of whether they are specific only to the primary object and/or secondary object(s) result in the complete definition of the primary and secondary object(s) being generated.
• Usage:

• java oracle.apps.fnd.odf2.FndXdfGen
  jdbc_conn_string=<jdbc tns info> object_name=<object name>
  xsl_directory=<xsl file Dir>
  [object_type=<object type>]
  [owner_app_shortname=<application short name of schema in which
  object is to be created in>]
  [apps_schema=<appId>]
  [apps_pwd= <appPassword>]
  [jdbc_protocol=<jdbc driver type>]
  [xdf_filename=<xdf file name>]
  [deferred=<all|<list>>]
  [Dropcr=<list>]
  [logfile=<filename>]
  [Custom_Defval=colName1=col1Value;colname2=col2Value]
  [IncludeTriggers=<n|<list>>]
  [IncludePolicy=<n|<list>>]
  [PolObjName=<object_name>]
  [PolObjOwner=<object_owner>]
  [Data_Sec_Vpd=<y|Y>]
  [Base_Partition=<partition_name>]
  [IncludeSequence=<list>]
  [Apply_DbVersion=<Database_Version>]
  [Comments_Only=<Y/N>]
  [obsolete=<all|none>]

• Mandatory Arguments
• jdbc_conn_string: jdbc tns information either in format of Net8 connect string enclosed in double quotes or hostname:port:oracle_sid format.
• object_name: The name of the database object whose XML object definition is being extracted. If the mode is trigger or sequence then the wildcard character ‘%’ can be used in the object name to retrieve a set of related sequences or triggers.
• xsl_directory: The xsl directory, which contains all the xsl files required for XSLT transformation. This is parameterized to facilitate any additional custom transformations that may be required for hosting or other purposes.
• Optional Parameters
• apps_schema: ORACLE schema name of current APPS(Applications) schema. Default=APPS, if not specified.
• apps_pwd: ORACLE schema password of current APPS(Applications) schema. Default=APPS, if not specified.
• jdbc_protocol: jdbc protocol (thin or oci8). This parameter is required to procure a connection from the JDBC Driver manager. Default=thin, if not specified.
• object_type: The Primary object type for which the xml definition is required—table, qtable, mview, mviewlog, sequence, type, trigger, view
• owner_app_shortname: This parameter refers to the Application short name of the application to which the object belongs. The owner_app_shortname is currently used as an informational parameter only. The Application information is used to populate the AOL (Applications object Library) Data Dictionary. Since the Database allows the same object name to be used in different schemas, this is required. This will be used to determine the schema in which to create the object, given that schema names can be different at customer sites. If object already exists, it defaults to the value in AOL(Applications object Library) data Dictionary. Otherwise, the base application (application with lowest application id will be picked).
• xdf_filename: The default behavior is that the xdf file name will be autogenerated based on the Primary Object Name. The object definition of the specified object is retrieved in the file mentioned. The file name will be automatically prefixed with “.xdf” extension. Default=<object_name>.xdf
• Deferred: This is support for the deferred creation of indexes until later. To specify which indexes to defer, use the option: deferred=<comma-separated list of indexes to defer> or if all indexes have to be deferred then deferred=all. This option is relevant in context of Applications Database Patching Utility. It is relevant for the standalone utility usage only when used in conjunction with constraints.
• DropCr: ODF supports dropping constraints and recreating them in a later phase. The list of constraints that have to be dropped and recreated are specified as dropCr=<comma separated list of constraints>.
• Logfile: The output of odfv2 generator program is written to standard out. Specify a logfile name if it has to be written to a log file.
• Custom_defval: This pertains only when the mode is equal to table. If a table column needs to be updated with a specific value when the table definition is modified (usually from null to not null, or adding a new not null column), then the custom_defval parameter can be used. The format of this parameter value is columnName=<value>. If more than one column in the table has to be updated, then the delimiter is semicolon ‘;’.
• IncludeTriggers: This is to specify a list of triggers to be included with the table definition. By default all triggers on a table are included in the XDF file along with the table definition. If trigger definitions are not to be included along with table definitions, this parameter needs to be used with ‘N’. Else, a comma-separated list of trigger names should be provided. The default value is IncludeTriggers=y
• Base_Partition: This is to specify the name of the partition to be included with the table definition. By default all the partitions on a table, is included in the XDF file along with the table definition. By specifying this parameter it is ensured that only the specific partition is included in the XDF file.
• IncludeSequence: This is to specify a list of sequences to be included with the table definition. By default the sequence, which drives the (logical) primary key on a table, is included in the XDF file along with the table definition. It is assumed to be of the format <table_name>_S. If additional sequence definitions are to be included along with table definition, this parameter needs to be used with a comma-separated list of sequence names.
• Apply_DbVersion: This is to specify the minimum version of the target database to apply the objects in the XDF file. By default all the objects in the XDF file are created/compared on the target database. By specifying this option you apply the objects in the XDF file if and only if the target database version is greater than or equal to the specified apply_dbversion.
• Comments_only: [Y/N], This option is used if, only the comments specified on the object in the XDF file have to be propagated. By default the value for this option is N i.e. all the objects specified in the XDF file are propagated.
• Obsolete: Currently, support to obsolete MVs & base table MV logs is provided. In case of MVs, the base MV will be dropped, obsoleting all its secondary objects like indexes, MV log, triggers, etc, In case of base table MV logs, the MV log is dropped, obsoleting secondary objects like indexes, triggers, etc. Partial obsoletion of secondary objects is an enhancement we are considering for the future. Permitted values are obsolete=all. The default value is obsolete=none.
Examples
• 1. Global Temporary Table—The following command is used to generate a XDF file with file name wf_local_users_temp.xdf for a table WF_LOCAL_USERS_TEMP, which belongs to a product with application short name FND.
• java oracle.apps.fnd.odf2.FndXdfGen
• jdbc_conn_string=ap105fam:1521:atg9idev
• object_name=WF_LOCAL_USERS_TEMP xsl_directory=/nfs/group/fnddev/fnd/11.5/patch/115/xdf/xsl
• 2. Table—The following command is used to generate a XDF file with file name jtftasksb.xdf in the directory /home/bhthiaga/odfjtatest/xml/ for a table JTF_TASKS_B which belongs to JTF(application short name). In this example, the optional parameter custom_defval is used (if you are changing a column definition from null to not null or adding a not null column and need to give a custom default value to update that column with the custom default value).
• java oracle.apps.fnd.odf2.FndXdfGen
• jdbc_conn_string=ap105fam:1521:atg9idev object_name=JTF_TASKS_B
• xsl_directory=/nfs/group/fnddev/fnd/11.5/patch/115/xdf/xsl
• custom_defval=“OBJECT_VERSION_NUMBER=1;”
• 3. Table—The following command is used to generate a XDF file with file name percon for table PER_CONTRACTS_F that belongs to PER(application short name of the product). In this example, the optional parameters deferred (for indexes) and dropcr (constraints) are used. This means that the indexes are deferred and constraints are dropped and recreated in a later phase. Usually this is required when there is a change in constraint/index definition and data upgrade is required to achieve that change.
• java oracle.apps.fnd.odf2.FndXdfGen
• jdbc_conn_string=ap105fam:1521:atg9idev
• object_name=PER_CONTRACTS_F xsl_directory=/nfs/group/fnddev/fnd/11.5/patch/115/xdf/xsl
• deferred=PER_CONTRACTS_F_PK dropCr=PER_CONTRACTS_F_PK
• 4. Materialized View—The following command is used to generate a XDF file with file name bilbymv.xdf for a materialized view BIL_BI_OPTY_T_MV.
• java oracle.apps.fnd.odf2.FndXdfGen jdbc_conn_string=“(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=ap513dbs)(PORT=1521))(CONNECT_DATA=(SID=OLAPDEV)))”
• object_name=BIL_BI_OPTY_T_MV xsl_directory=/nfs/group/fnddev/fnd/11.5/patch/115/xdf/xsl
• 5. View—The following command is used to generate a XDF for a view ABMBG_ACT_ACCT_DATA_VAR_V. The auto-generated xdf filename will be abmbg_act_acct_data_var_v.xdf.
• java oracle.apps.fnd.odf2.FndXdfGen
• jdbc_conn_string=ap105fam:1521:atg9idev
• object_name=ABMBG_ACT_ACCT_DATA_VAR_V
• xsl_directory=/nfs/group/fnddev/fnd/11.5/patch/115/xdf/xsl
• 6. Materialized View Log—The following command is used to generate a XDF file with file name ak_regions_mlog.xdf for a materialized view log on the table AK_REGIONS.
• java oracle.apps.fnd.odf2.FndXdfGen
• jdbc_conn_string=ap105fam:1521:atg9idev object_name=AK_REGIONS
• object_type=mviewlog
• xsl_directory=/nfs/group/fnddev/fnd/11.5/patch/115/xdf/xsl
APPENDIX DComparison Utility Usage Notes
• The Comparison Utility compares database object definitions contained in a file with extension “.xdf” (which is an XML file) with the database objects in an ORACLE schema, detects any differences in database structure, and executes SQL statements to propagate the differences, so that the objects in the schema will match the definitions in the xdf file.
• XDF (xml) files are generated by the XDF generator utility. The .xdf file extension is used to distinguish xml files containing object definitions from others.
• Usage:

• 	java oracle.apps.fnd.odf2.FndXdfCmp
  	<schemaId>
  	<schemaPassword>
  	<appId>
  	<appPassword>
  	<jdbc driver type>
  	<jdbc tns info>
  	<object_type>
  	<xdf file name>
  	<xsl file directory>
  	[deferred=<list> ]
  	[dropCr=<list>]
  	[logfile=<filename>]
  	[Data_Sec_Vpd=<y|Y>]
  	[Standalone=<y|Y>]
  	[changedb=<y|n>]

• Mandatory Arguments
• schema_name: ORACLE schema name in which the object exists or should be created.
• schema_Pwd: ORACLE schema password in which the object exists or should be created.
• apps_schema: ORACLE schema name of current APPS(Applications) schema.
• apps_pwd: ORACLE schema password of current APPS(Applications) schema.
• jdbc_protocol: jdbc protocol (thin or oci8).
• jdbc_connect_string: jdbc tns information either in format of Net8 connect string enclosed in double quotes or hostname:port:oracle_sid format.
• object_type: the object type information—table, qtable, mview, mviewlog, sequence, type, trigger, view, policy
• xdf_filename: The xdf file name, which contains the object definition.
• xsl_directory: The xsl directory, which contains all the xsl files required for XSLT transformation.
• Optional Parameters
• Deferred: XDF supports deferring creation of some indexes until later. The list of indexes that have to be deferred are specified as deferred=<comma-separated list of indexes to defer>
• Logfile: The output of XDF comparison utility is written to standard out. Specify a logfile name if it has to be written to a log file.
• ChangeDb: The default for changeDB=y. If the changedb is specified as ‘n’ then the SQL statements generated by the odf comparison utility are not executed but displayed on the standard output or a log file.
• DropCr: XDF supports dropping constraints and recreating them in a later phase. The list of constraints that have to be dropped and recreated are specified as dropCr=<comma separated list of constraints>.
• Data_Sec_Vpd: This is used to specify that a service security synonyms or view has to be created dynamically based on the database version. If the database version is 9 then a synonym is created else a view is created. Permitted values are data_sec_vpd=y
• Standalone: This option is used to execute the XDF comparison utility in a standalone mode. Permitted values are standalone=y
Examples
• 1. Global temporary Table—The following command is used to compare a table definition in the XDF file wf_local_users_temp.xdf with the definition in the database atgcore and run sql statements to remove the differences such that the object definition in the database matches the definition in the XDF file.
• java oracle.apps.fnd.odf2.FndXdfCmp applsys apps apps apps thin ap105fam:1521:atgcore table /home/sbrahman/wf_local_users_temp.xdf/fnddev/fnd/11.5/patch/115/xdf/xsl
• 2. Table—The following command is used to compare a table definition in the XDF file jtasyncmap.xdf with the definition in the database atgcore and run sql statements to remove the differences such that the object definition in the database matches the definition in the XDF file. In this example, the index JTA_SYNC_CONTACT_MAPPING_U1 on the table is deferred.
• java oracle.apps.fnd.odf2.FndXdfCmp jtf jtf apps apps thin
• ap112fam:1521:atgcore table jta_sync_contact_mapping.xdf/fnddev/fnd/11.5/patch/115/xdf/xsl
• deferred=JTA_SYNC_CONTACT_MAPPING_U1
• 3. Materialized View—The following command is used to compare a materialized view definition in the XDF file bil_bi_opty_t_mv.xdf with the definition in the database atgcore and run sql statements to remove the differences such that the object definition in the database matches the definition in the XDF file.
• java oracle.apps.fnd.odf2.FndXdfCmp apps apps apps apps thin
• ap105fam:1521:atgcore mview bil_bi_opty_t_mv.xdf/fnddev/fnd/11.5/patch/115/xdf/xsl
• 4. View—The following command is used to compare a view definition in the XDF file aview.xdf with the definition in the database atgcore and run sql statements to remove the differences such that the object definition in the database matches the definition in the XDF file.
• java oracle.apps.fnd.odf2.FndXdfCmp apps apps apps apps thin
• ap112fam:1511:atgcore view abmbg_act_acct_data_var_v.xdf/fnddev/fnd/11.5/patch/115/xdf/xsl

• APPENDIX E
  (TABLE Example)

  CREATE TABLE “TEST_TABLE”

  ( “COL1” NUMBER(15,0),

  “COL2” DATE

  ) STORAGE (INITIAL 4096 NEXT 40960MINEXTENTS 1

  MAXEXTENTS 505

  PCTINCREASE 50FREELISTS 1FREELIST GROUPS 1

  	BUFFER_POOL
  	DEFAULT)
  	CREATE INDEX “TEST_TABLE_I1” ON TA

Claims (6)

1. A computer-readable storage medium encoded with a file comprising a description of properties of a first object in a first database, the description being expressed in a text based markup language, the storage medium being further encoded with a patch to an application that uses said first object, the storage medium being further encoded with instructions to compare the description in said file with metadata of a second object in a second database, and to change the second database based on said file to generate a changed second database, depending at least partially on an outcome of said comparing, and to execute the application software as changed by said patch, to use the changed second database.

2. The computer-readable storage medium ofclaim 1 wherein:

said file comprises at least a first plurality of hash codes related to said first object, said plurality of hash codes comprising a base hash code and a group of additional hash codes, the base hash code being based on at least a portion of said description other than an identifier of the first object, and the group of additional hash codes being based on a plurality of properties of said first object; and

said instructions comprise instructions to prepare a second plurality of hash codes related to said second object, and to compare said base hash code in said first plurality of hash codes to a base hash code in said second plurality of hash codes and only in case of no match therebetween to further compare at least one additional hash code in the first plurality to a corresponding additional hash code in the second plurality and to perform said change to the second database depending on results of comparing hash codes.

4. A method for supporting an upgrade of application software, wherein the application software has an interface for accessing a first database in a first computer, the method comprising:

preparing a plurality of descriptions of a corresponding plurality of data structures of the first database in the first computer;

wherein each description in said plurality of descriptions is expressed, by said preparing, in a predetermined markup language;

comparing the plurality of descriptions to a second database in a second computer by use of hash codes; and

changing the second database in the second computer based on at least one description in said plurality of descriptions, if at least one difference is found by said comparing.

5. One or more non-transitory computer-readable storage media comprising instructions to propagate a plurality of changes made to application software that uses a first relational database, the instructions comprising:

instructions to receive an identification of a first table in the first relational-database;

wherein first data in the first table in the first relational database is used in the application software;

instructions to check if the first table is a primary object, by using a first data structure comprising a first plurality of predetermined types indicative of primary objects;

instructions to extract first metadata describing the first table from a first data dictionary;

wherein the first data dictionary is comprised in the first relational database;

instructions to use the first metadata to prepare a first description of properties of the first table;

instructions to check the first relational database, to see if said first table has associated thereto a first secondary object of any type identified by a second data structure comprising a second plurality of predetermined types indicative of secondary objects;

instructions, responsive to finding said first secondary object associated with said first table on execution of said instructions to check the first relational database, to extract from the first data dictionary second metadata describing the first secondary object in the first relational database, and to prepare a second description of properties of the first secondary object.

6. The computer-readable storage media ofclaim 5 further comprising:

Instructions to receive at least: said first description of said first metadata, said second description of said second metadata, and the plurality of changes to the application software;

wherein at least one change in the plurality of changes is related to use of first data in said first table.

7. The computer-readable storage media ofclaim 6 further comprising:

instructions to identify a structural difference between said first table in said first relational database and a second table in a second relational database, by comparing at least said first metadata of said first table to corresponding second metadata of said second table in a second data dictionary of the second relational-database;

wherein the second data in the second table is used in executing a copy of the application software;

instructions to structurally change the second table depending at least partially on an outcome of said comparing, to generate a changed second table; and

instructions to execute a changed version of the application software, obtained by changing said copy by applying said plurality of changes, to use the changed second table.

Images