Seva Safris

The SpecialAgent is software that attaches to Java applications and automatically instruments 3rd-party libraries within it. The architecture was specifically designed to include contributions from the community, whereby its platform automates the installation of OpenTracing integrations written by individual contributors. In addition to Integrations, the SpecialAgent also supports Trace Exporters, which connect an instrumented runtime to OpenTracing-compliant tracer vendors, such as LightStep, Wavefront, or Jaeger. At its core, the SpecialAgent is itself nothing more than an engine that abstracts the functionality for the automatic installation of Integrations, and their connection to Trace Exporters. A benefit of this approach is that the SpecialAgent intrinsically embodies and encourages community involvement.

JSONx is a lightweight framework based on an XSD JSONx schema that allows one to create a schema for JSON classes. To its disadvantage, JavaScript is compiler-less and a highly non-cohesive language that results in errors being realized in runtime. As there does not exist a formalized schema specification for JSON, developers often make repeated mistakes when designing JSON messages, encoding JSON objects, and decoding JSON strings. jJB presents a schema model that can bring errors to edit-time and compile-time, greatly reducing the risk of the same errors appearing in runtime.

JSONx uses a json.jsonx file that conforms to the JSONx Schema to generate Java beans to bind the JSON classes to Java. The generated classes are strongly typed and offer the full benefits of a cohesive interface to JSON objects in Java. The generated classes can be used to parse and marshal JSON messages, confident that all messages conform to the definition in the JSONx.

Toptal's Blockchain specialization evaluation includes a deep dive into blockchain technology as the final stage of the process. This two-week-long stage is modeled to resemble a real project defined by a client. The requirements of the project express specific features to be present in the blockchain, and the specification allows the developer to either fork an existing codebase or create one from scratch. In the interest of gaining the most exposure to the detailed considerations embodied in blockchain technology, I chose to complete the project with a blockchain I created from scratch. This project allowed me to gain firsthand experience with the nuances of blockchain at every level.

TestingBot provides a simple-to-use system for cross-browser testing. In 2018, I contributed to the project by implementing support for upstream proxy relay of HTTP traffic with Jetty's non-blocking socket APIs.

Why keep reinventing the wheel? The authentication pattern has been done over and over again. Get a jump start on implementing your business logic immediately instead of wasting time making your own.

This is the back-end component of the angular-auth-seed project. Together, they provide a fully functional platform for authenticated applications based on AngularJS and Java's JAX-RS v2.1 specification. It is a server implementation of the Basic authentication methodology for RESTful applications, written in Java and 100% JAX-RS-compliant. It can be run on servers that implement the JAX-RS 2.1 specification, including Jersey, JBoss RESTEasy, and Restlet. Though any JAX-RS server vendor can be used, I recommend the XRS server, which is a CohesionFirst™ alternative to the JAX-RS 2.1 spec. The XRS server is designed to be ultra-lightweight and is configured to run in the Jetty servlet container.

This is a complete solution, a bug-free and instant "plug-and-play" base to get you started fast and easy. Preconfigured to install the Angular framework, development prerequisites, and testing tools for instant web deployment gratification, this solution can be used to quickly bootstrap your Angular project and dev environment.

Why do we keep reinventing the wheel? The authentication pattern has been done and done again thousands and thousands of times. Instead of spending the first two weeks of development on the authentication requirement, get a jump-start on your business logic immediately.

Created with the highest development standard in mind, this project uses best practices and patterns to empower the developer to write code that is clear, cohesive, and easily testable.

The angular-auth-seed project is a complete solution intended to be bug-free and an instant "plug-and-play" base to get you started quickly and easily. Preconfigured to install the Angular framework, development prerequisites, and testing tools for instant web deployment gratification, this solution can be used to bootstrap your Angular project and dev environment quickly.

XSB is a Java framework for binding to an XML schema. Based on the CohesionFirst™ approach, the XSB framework provides a complete solution for the cohesive integration of Java applications to XML documents via the XML schema. The framework provides an XSB Java source code generator and an XSB runtime to parse and marshal XML documents to and from Java objects. The XSB framework is a complete implementation and covers the entirety of the XSD 1.1 specification.

XSB supports all directives of the XSD specification and generates bindings with the highest degree of cohesion possible with the Java language.

A Java framework for binding to RDBMS data models. Based on the CohesionFirst™ approach, the RDB framework provides a strongly typed and cohesive approach to connect the DB tier to the business tier.

jSQL is an extension to DDLx that offers a lightweight object-relational mapping (ORM) solution that runs on the JDBC v4.1 API. The jSQL framework provides strongly typed semantics for the SQL language and a cohesive binding to user data models. jSQL uses a SQL schema defined in a DDLx file to create a one-to-one, object-model-to-data-model API that is vendor agnostic.

In addition to generating Java classes that bind to a DDL, the jSQL framework offers an API for strongly typed DML semantics. These APIs come in the form of method invocations that resemble a non-cohesive, string-based SQL alternative.

Strongly-typed DML semantics are powerful because they extend the power of the Java compiler to realize errors in edit-time or compile-time. Alternatively, if non-cohesive, string-based SQL is used, errors are only presented in runtime upon execution by the application to the database. In addition to binding Java classes to the DDL, jSQL provides a strongly typed approach for the construction of SQL DML.

JetRS is an implementation of the JAX-RS v2.1 specification that runs in a servlet container. This project was inspired by the goal of creating a better, simpler, easier-to-use, reliable, and debuggable JAX-RS implementation using the cohesion-first approach.

XRS was built to implement the JAX-RS v2.1 specification while keeping the internal complexity as low as feasible. Static state is used in only one use case, so debugging a request or a response is never a challenge.

Existing solutions such as Jersey, JBoss RESTEasy, Restlet, Apache CXF, Apache Wink, and others are challenging to work with because they are buggy, difficult to debug, and not pure to the JAX-RS specification. Many people experience unnecessary pains using existing JAX-RS implementations. Debugging JAX-RS servers is especially difficult because of their high internal complexities.

A common pattern used in JAX-RS implementations is dynamic method invocation. Dynamic method invocation is powerful, but it comes at a cost: debugability. It results in stack traces that lack information about the specific execution path that led to the exception.

A one-keystroke way to navigate to the next/previous occurrence of selected text (like in Eclipse) without raising the "You've reached the end of the document" popup. This plugin supplements the default actions available with IntelliJ IDEA.

OpenJAX DBCP allows a developer to configure a connection pool with a standardized XML Schema, which is used by a consumer class to initiate the connection pool. OpenJAX DBCP uses the JAXB framework to significantly reduce boilerplate code, thus providing a lean API with support for all possible connection pool configuration variations.

OpenJAX DBCP is based on an XML schema used to specify the format of XML documents accepted by the configuration consumer. The XML schema is designed to use the full power of XML validation to allow a developer to quickly determine errors in his draft. Once a dbcp.xml passes the validation checks, it is almost guaranteed to properly initialize the connection pool configured by the file.

OpenJAX CLI allows a developer the full range of variations of the command line arguments pattern and removes the unnecessary boilerplate code present in other solutions. OpenJAX CLI uses the JAXB framework to create an OpenJAX CLI specification based on an XSD specification, which can be used to create a custom set of CLI options and arguments for any application.

OpenJAX CLI is based on an XML schema that is used to specify the format of XML documents that describe the command line options and arguments accepted by an application. The XML schema is designed to use the full power of XML validation to allow a developer to quickly determine errors in his draft. Once a cli.xml passes the validation checks, it thereafter provides a clear and simple API to access the options and arguments in the code.

A Cordova/PhoneGap plugin to hide the status bar on the startup of the application by adding a declaration to the Info.plist file that would set the status bar as hidden by default, allowing the developer more control over the application UI.

A startup script used in GCP's Compute Engine instances to automatically authorize an auto-scale instance group to a Google Cloud SQL database. This script was inspired by a growing desire from the community for such a feature that was not available from GCP itself.

A lightweight interpolator that replaces delimited keys in a string with values from a properties map. This module was created as a simple alternative for Angular's $interpolate provider. A common use of this tool is for a template interpolation use-case that does not require $parse, $interpolate, or $compile.

In 2008, I worked with a small team of electronic music enthusiasts to develop a novel application to extend the functionality of live DJ equipment. We developed a beat-tracking algorithm that was able to identify the tempo and phase of electronic music by performing live DSP on the audio signal. The algorithm thereafter extrapolated the immediate future beats with the confidence factor, which was realized during the DSP analysis. The algorithm was thereafter interfaced with a MIDI-out controller to send the metronome of the beats in the audio signal. The MIDI metronome allowed MIDI-compliant devices to use the signal for higher-layer applications. Examples of such applications are automatic tempo and beat synchronization, audial and visual effect overlays, and seamless sample looping, amongst others.