Abstract: In-place XML schema evolution occurs by evaluating an existing XML schema in a database system to see if it is compatible with a new XML schema. To determine if the old schema is compatible with the new schema, a lock-step traversal is performed on the two schemas. Each instruction in the old schema is compared to instructions in the new schema. Detected differences between the two schemas are evaluated to determine if existing XML documents are still compatible with the new schema. If they are, then an in-place schema evolution operation can take place. The in-place schema operation basically, involves appending the updated object information to the underlying object type and preserving the order of elements in a document by storing element mapping information on disk.

Abstract: A method and apparatus are described for sharing a session to access a database. A database server receives, in a session, a session context identifier and a command. The session context identifier identifies a session context to use for the session. The session context is a set of information or commands that plug into a session state and specify how commands in the session are to be performed for a particular user or privilege level. In response to receiving the identifier, the database server associates the session context with the database session for the connection. The database server uses the session context to process the command. The session context may then be detached from the session, allowing another user to attach to the session via another session context.

Abstract: A method and apparatus for facilitating the management of metadata is disclosed, specifically by associating metadata with an XML schema. The database system provides a method that allows end users to maintain additional information describing resources. User-defined metadata attributes are associated as an XML schema which in turn is mapped to a table, referred to as the user metadata table. The XML schema contains user-defined information specifying the metadata that the end-user wants the system to store and maintain. Additionally, a reference is created between an entry in the user metadata table and its corresponding resource entry in the resource table. The resources are accessed directly or by issuing queries against the user metadata table. The querying of the user metadata table gains the benefit of fast and efficient query results and access time.

Abstract: A platform (referred to herein as Extensible Application Platform, or XAP) for developing extensible business applications. In one set of embodiments, XAP can include an Extensible Markup Language (XML) based application model that is designed to support a high degree of application extensibility. For example, the XAP application model can support declarative definition of application semantics, separation of application data and metadata, and a flexible type system that allows constraints and calculations (e.g., XQuery constraints and calculations) for a data object instance (e.g., an XML document) to change over time. In certain embodiments, the XAP application model can be implemented via a server-side database/application engine that is configured to interoperate with a spreadsheet-like, client-side user interface. Using the spreadsheet-like UI, a non-programmer can build, customize, and run an application workflow that operates on top of a shared database like a traditional business application.

Abstract: A method and storage media for performing access resolution using ACL types is provided. Under an AND semantic, an intersection set formed from the types of multiple ACLs protecting a resource may be utilized to efficiently determine whether a request for a privilege to access the resource is granted or denied. If the privilege is not a member of the intersection set, the privilege cannot be granted. A union set may be used for an OR semantic. A global ACL type may represent all privileges system-wide or application-wide. A global ACL may represent a system-wide or application-wide access policy. A conjunction of a global ACL and a regular ACL may be stored in a cache. The union set, intersection set, or access resolution may also be cached for subsequent request processing.

Abstract: A method and apparatus for sharing state information among a plurality of servers is provided. A first server receives a request to perform a file operation, such as a NFS file system operation. Thereafter, the first server updates state information to reflect a change in state associated with processing the request. The first server then causes state update information, which identifies the change in state that was made at the first server, to be propagated to a second server. The second server updates state information, stored at the second server, to reflect the change in state associated with processing the request at the first server. Advantageously, a client may send a series of stateful file operations to either the first server or the second server, even if processing any one of the stateful file operation requests requires knowledge of the state of processing prior requests in the series of requests.

Abstract: Efficient evaluation of resource version history queries is enabled by using a data container storing records corresponding to links between successive versions of a resource. The records include: (a) a record identifier, (b) a predecessor identifier, (c) a successor identifier, (d) a predecessor rowset, which comprises the record identifier of each record in which the predecessor identifier in this record is the successor identifier, and (e) a successor rowset, which comprises the record identifier of each record in which the successor identifier in this record is the predecessor identifier. Depending on the nature of a request, a record is identified in which a particular version is identified by either the predecessor identifier or the successor identifier. Either the predecessor or successor rowsets are recursively read, while the predecessor or successor identifiers corresponding to each record read are added to a result set, thereby forming the ancestor or descendant version history.

Abstract: A method and apparatus for reverting a resource to a prior state in time is provided. Changes are committed to a resource at a particular point in time. After the particular point in time, a request, which may be a file system operation request, to revert the resource to a state prior to the particular point in time is received at a database server. In response to the request, the resource is reverted to the state prior to the particular point in time. The current state of the resource ceases to reflect the plurality of changes. Advantageously, if a requestor creates an incorrect version of a resource, the changes made to a schema-based resource are not compatible with the schema, or the changes of multiple requesters are not compatible with each other, the resource may be reverted to the earlier point in time.

Abstract: A method, mechanism, and computer program product for storing, accessing, and managing XML data is disclosed. The approach supports efficient evaluation of XPath queries and also improves the performance of data/fragment extraction. The approach can be applied to schema-less documents. The approach is applicable to all database systems and other servers which support storing and managing XML content. In addition, the approach can be applied to store, manage, and retrieve other types of unstructured or semi-structured data in a database system.

Abstract: A method and apparatus for processing path-based database operations is provided. According to one aspect, a “parent row ID” column is added to a directory links table. For each row in the table, the value of the “parent row ID” column indicates the row ID, in the table, of the node that is that row's node's parent in the file hierarchy. To determine whether a particular node is within a specified path of the file hierarchy, the particular node's pathname is derived by visiting child-to-parent links indicated in the table and prepending, to the front of a pathname being constructed for the particular node, node names indicated in the visited rows. Each parent node's row in the directory links table is located using the “parent row ID” column of the child node's row. The resulting pathname is compared with the specified path to determine whether the particular node exists therein.

Abstract: A method and apparatus for processing a file system operation at a database server is provided. A request to perform a file system operation on a resource stored in a database is received at a database system. The request may be implemented using the NFS protocol. The request may include state identification data that identifies state information associated with the request. State information associated with the request is retrieved within the database system based on the state identification data. State information is information that describes the operational state of the requestor for a particular file. The request is then processed based, at least in part, on the state identification. File system operations may be processed a database management system to access any data, such as a file, relational data, and object-relational data.

Abstract: Techniques are provided for responding to file requests received through a file system interface. According to one technique, an association is established between a file identifier and a query. A request for a file associated with the file identifier is received through the file system interface. In response to the request, the following steps are performed: executing the query to generate a set of data; generating contents of the file based on the set of data; and providing the file through the file system interface.

Abstract: A method and system of handling document operation requests on documents with large collections and constrained memory is disclosed. In some embodiments, when one or more collection elements of a document are required to be in memory for a document operation, each of the one or more required collection elements not already in the memory are loaded into a collection partition in the memory before the document operation is executed.

Abstract: A method and apparatus for performing file system operations involving a link is provided. A request to perform a file system operation involving a link is received from a client at a database server. The link may establish a relationship a source resource has to a target resource. The source resource and the target resource may be stored within a repository implemented by a storage medium, such as a database. The database server may perform the file system operation by storing, in the database, one or more link records that identifies the link between the first resource and the second resource. File system operations, such as a NFS file system operation, which involve links may be performed at a database management system.

Abstract: An XPath expression is converted into a tree-based representation where each node represents an operation to be performed and the links between nodes in the tree represent the flow of data between operations. The conversion may involve creating a parse tree for the XPath expression, and then converting the parse tree into an operator tree. The operator tree is constructed in such a way that execution of the XPath expression begins at the leaf nodes of the operator tree, and the results are then passed up the tree. After each node is executed, the results are either (1) passed to a filter that filters the results with a predicate statement or (2) passed to another node to be operated upon. This occurs until no nodes remain to be executed.

Abstract: In a multi-tier data server system, data from the first tier is cached in a mid-tier cache of the middle tier. Access control information from the first tier for the data is also cached within the mid-tier cache. Caching the security information in the middle tier allows the middle tier to make access control decisions regarding requests for data made by clients in the outer tier.

Abstract: A method and apparatus for providing file system operation locks at a database server is provided. A database server may employ database locks and file system operation locks in servicing requests from consistent requestors and inconsistent requesters. A database lock is a lock that is obtained in response to performing a database operation, and the database lock is released when the database operation has successfully completed. A file system operation lock is a lock that is obtained in response to performing an OPEN file system operation, and the file system operation lock is released when a CLOSE file system operation is performed. The database server may use a temporary copy of the resource, which reflects all the current changes that have been made to the resource by database operations, in servicing consistent requestors, and may use the original version of the resource in servicing inconsistent requesters.

Abstract: Evaluation of workspace-local queries on a repository that supports versioning is enabled by tracking which versions of resources that are stored in the repository are associated with particular workspaces. A mapping is created and maintained for each workspace and each workspace-enabled table, i.e., a repository table that supports multiple versions of resources in the table. The mapping, such as a bit vector, is indexed by resource identifier. A set bit indicates that a particular resource version that corresponds to the bit is associated with the particular workspace. Therefore, the mapping indicates which rows in the workspace-enabled table are in the particular workspace. Index-based and functional evaluation schemes may utilize the per workspace, per table mapping to evaluate queries on the versioned repository. A new operator (IN_WORKSPACE) is introduced, through which an evaluation routine is invoked to determine whether or not a particular row maps to a particular workspace.

Abstract: A shared workspace is associated with a group of users who share the workspace, which isolates a set of shared resources from other resources stored in a repository that supports versioning of resources. The content of a shared workspace is maintained using a set of shared references, to the set of resources associated with the shared workspace. A virtual workspace is created for a user by generating a personal mapping for the user, where the personal mapping indicates which resources from the set of shared resources the user has made personal, e.g., by checking out such resources from the repository. If the user submits a query on the repository, the particular version of each resource involved in the query is readily determinable from the shared workspace and the user's virtual workspace.

Abstract: Techniques are provided for performing file operations. According to one technique, a file system interface is exposed to applications. The file system interface includes routines for saving and retrieving files. Calls to perform a plurality of file operations are received through the file system interface. The plurality of file operations are performed as a single transaction by performing the steps of: if all file operations of the plurality of file operations are completed without a failure, then making permanent all changes made by the plurality of file operations; and if any file operations of the plurality of file operations fail, then undoing all changes made by all of the plurality of file operations.