Abstract: A method of extending the functionality of an enterprise software suite is disclosed. A request is received from a client system to modify a programming object on a productive system deployed in the cloud environment. A logical unit of programming objects is identified on the productive system, the logical unit including the programming object. Copies of each of the programming objects in the logical unit of programming objects are created in a sandbox of a combined development and test system deployed in the cloud environment, the copies including a copy of the programming object. The copy of the programming object is modified in the sandbox. A result of the modifying of the copy of the programming object in the sandbox is communicated for presentation in a client system without modifying the programming object on the productive system.

Abstract: Techniques are provided for identifying and removing one or more layers of a software appliance. Layers may be identified and removed quickly and efficiently, without disrupting the distribution or provisioning of the software appliance, and without disrupting desired operations of the software appliance as a whole. In various implementations, it may be necessary or desirable to re-package the same or modified layer(s) within the software appliance at a later point in the distribution cycle of the software appliance. For example, one or more operating system layers may be identified and removed from a software appliance. Then, when the software appliance is distributed and provisioned in a cloud or other network context, a same or different version of the operating system layer(s) may be utilized in re-packaging the software appliance for one or more customers.

Abstract: Each of a plurality of database transactions are logged (i.e., recorded) in a log. Concurrent with the logging, one or more characteristics of the log are monitored. Thereafter, a savepoint is triggered when a pre-defined condition is met as indicated by the monitoring. The triggered savepoint can override or accelerate a savepoint that would have otherwise been triggered based on pre-specified parameters.

Abstract: Various techniques are described for precisely locking database objects within a database during a system upgrade with the use of database freeze triggers. Some database freeze triggers can be utilized to lock transport requests while other database freeze triggers can be utilized to lock database objects such as exchange tables. For transport requests, techniques describe the use of a whitelist identify transport objects or transport object types which shall remain unblocked. As a result, a database freeze trigger can be generated for transport objects or transport object types that are not included in the whitelist. For database objects, techniques describe the identification of a tool or service which has been allowed during the upgrade. Database objects which can be modified during execution of the tool or service can be identified. Database freeze triggers can be generated for the database objects besides those identified.

Abstract: A method includes obtaining first text information in a first format, converting the first text information to an offset value from a base time in a second format, generating an encryption key, encrypting the offset value with the encryption key to produce an encrypted offset value in the second format, converting the encrypted offset value to second text information in the first format, and outputting the second text information as encrypted text corresponding to the first text information.

Abstract: In an example embodiment, first user input including handwriting input and non-alphanumeric symbolic input is detected. The non-alphanumeric symbolic input is input into a first machine learning model trained to output a set of possible actions corresponding to the non-alphanumeric symbolic input and a probability score assigned to each action in the set of possible actions. A combination of the action having the highest probability score and textual input from the handwriting input is input into a second machine learning model trained to select a service from a plurality of services based on the textual input and the selected action by referencing a service model corresponding to each service in the plurality of services. The combination of the textual input and the selected action is transformed into a native request for the selected service based on the service model for the selected service.

Abstract: A generic state machine concept can be used to decouple specific state machine functionality from any specific consensus protocol. A generic state machine can prepare a local log record and generate a local log index in response to a data update received at a computing node of a plurality of computing nodes in a distributed data storage application. The generic state machine can pass replication information to one or more other computing nodes in the plurality of computing nodes in accordance with the consensus protocol and can mutate a state of the generic state machine upon the plurality of computing nodes achieving a quorum.

Abstract: A technology for path determination using robust optimization is provided. In accordance with one aspect, a network graph of a network is generated. The network graph comprises nodes corresponding to points in the network, and edges which connect the nodes. Costs for each edge of the network are determined and modeled using reference point, upper bound and lower bound parameters. A user input which includes a source node, destination node, and cost target may be received from a client device. A resultant path connecting the source node and destination node are determined by solving a target-oriented robust optimization problem, which optimizes a cost of the resultant path based on the modeled costs of the edges. The resultant path is displayed on a user interface of the client device.

Abstract: In one aspect, a computer-implemented method includes calculating cardinality differences and combined sizes for one or more data block pairs in a relational database. The method also includes selecting a data block pair from the one or more data block pairs based on a calculated cardinality difference and a calculated combined size associated with the selected data block pair. The method further includes merging the selected data block pair to form a merged data block, and replacing the selected data block pair in the relational database by the merged data block.

Abstract: Methods, systems, and computer-readable storage media for receiving a request to execute a custom process over a target system, the custom process including one or more operations, determining, for each operation of the custom process, a set of entities implicated by a respective operation based on an operations graph of the custom process, and relations paths of the respective operation, determining, for each set of entities, a sub-set of entities based on one or more constraints of the respective operation, providing, for each operation of the custom process, an operation process that is to be executed over one or more entities of a respective sub-set of entities, combining operation processes into a single custom process, and initiating execution of the single custom process.

Abstract: Systems and methods are provided for enforcing security requirement rules of an application. A security guard retrieves security annotations associated with an application from a server. Each security annotation is mapped to a security requirement rule. Based on the evaluated application compliance with the security requirement rules, an executable interpretation of the corresponding security requirement rule is implemented.

Abstract: The disclosure generally describes methods, software, and systems, including a method for optimizing queries referencing large object (LOB) information. A structured query language (SQL) query containing an LOB field is identified. LOB replacement information for the LOB field is obtained from a data model. The SQL query is optimized including substituting, using the obtained LOB replacement information, the LOB field with replacement language to create an updated SQL query, and extending the updated SQL query with a join of LOB field language to create an extended SQL query.

Abstract: Systems and methods are provided for analyzing a script file to generate a list of commands associated with access of one or more tables in at least one database and analyzing the generated list of commands to determine the one or more tables to be accessed. The system and methods further provide for generating a grouping of the commands for each of the one or more tables to be accessed such that each table of the one or more tables has one or more associated commands, distributing each grouping of commands to be processed in parallel, executing the grouped commands in parallel to determine column names to be accessed for each of the one or more tables, and generating a result list comprising a list of table names and a list of column names for each table in the list of table names to be accessed by the script file.

Abstract: Refresh requests are received by a data source that each request a snapshot of current members of one of a plurality of dynamically changing groups and dynamically changing rules corresponding to such group. Thereafter, the data source queues the received plurality of refresh requests for selective execution or deletion into a new request queue. In addition, real-time execution of refresh jobs are initiated for all of queued refresh requests if a number of refresh requests in both of the new request queue and a waiting requests queue is below a pre-defined threshold. Alternatively, a job framework schedules execution of task jobs for a subset of the queued requests in the new request queue and the waiting requests queue if certain conditions are met.

Abstract: Techniques and solutions are described for training and using a classifier based on a machine learning model to analyze performance information to assist in correcting a software bug or performance issue. The performance information can be processed prior to submission to a trained classifier, such as to remove, modify, or format data. A classification result provided by the classifier can be compared with a database to determine whether a solution or target is associated with the classification result. User feedback can be used to provide more accurate suggestions of solutions or targets, as well as to improve the accuracy of the classifier.

Abstract: Universal, automatic file conversion may be provided by a universal file conversion system or application. An input file may be received by the universal file conversion system. An input file type for the input file and a recipient of the input file may be determined. Programs available to the recipient for accessing a file may be determined. A target file type accessible to the recipient may be determined for converting the input file. A sequence of file conversions to convert the input file to the target file type may be determined. The input file may be converted to the target file type based on the sequence of file conversions. The converted file may be provided to the recipient. The recipient may return the converted file, and the converted file may be automatically converted back to the original input file type and provided to the original source of the input file.

Abstract: Techniques and solutions are described for providing integration objects that can be used to enhance data retrieval, and can be used to provide shared data access for multiple applications, and to allow access to application functionality through an interface. The integration object includes a plurality of integration data objects, at least a portion of which are mapped to data objects stored in one or more data sources. The integration data objects can hold values stored in correspondingly mapped locations of the data objects, such as in a value field. A reference field of the integration data objects can hold reference information regarding a location of a value in a respectively mapped data source. At least a portion of the integration data objects can include a defined sequencing with one or more other integration data objects.

Abstract: Embodiments of the present disclosure pertain to character recognition using neural networks. In one embodiment, the present disclosure includes a computer implemented method comprising processing a plurality of characters using a first recurrent machine learning algorithm. The first recurrent machine learning algorithm sequentially produces a first plurality of internal arrays of values. The first plurality of internal arrays of values are stored to form a stored plurality of arrays of values. The stored plurality of arrays of values are multiplied by a plurality of attention weights to produce a plurality of selection values. An attention array of values is generated from the stored arrays based on the selection values. The attention array of values is processed using a second recurrent machine learning algorithm, the second recurrent machine learning algorithm produces a recognized character sequence.

Abstract: In one respect, there is provided a system that may include a processor and a memory. The memory may be configured to store instructions that results in operations when executed by the processor. The operations may include: processing an image set with a convolutional neural network configured to detect, in the image set, a first feature and a second feature; determining a respective effectiveness of the first feature and the second feature in enabling the convolutional neural network to classify images in the image set; determining, based at least on the respective effectiveness of the first feature and the second feature, a first initial weight for the first feature and a second initial weight for the second feature; and initializing the convolutional neural network for training, the initialization of the convolutional neural network comprising configuring the convolutional neural network to apply the first initial weight and the second initial weight.

Abstract: Execution of a plurality of application services corresponding to a plurality of user sessions are initiated in a distributed computing environment. Thereafter, all functional units of each application service are executed as part of a node of the distributed computing environment in a separate service environment and without remote calls to other services. Subsequently, each user session is executed in a separate and distinct node of the distributed computing environment. Related apparatus, systems, techniques and articles are also described.