useR! 2024

Learn how to streamline the creation of R client libraries for RESTful APIs by leveraging the OpenAPI specification (v2 or v3) and a customizable, template-based code generation tool, OpenAPI Generator (https://openapi-generator.tech/). This talk introduces the code generation workflow, demonstrates the functionality of the resulting R package, and explores customization options for the generated R code and documentation. We explore an efficient approach to leveraging openapi.json (swagger) files as a starting point for creating tailored R client libraries for seamless integration.

Exploring the usage of R and SQL Server, elucidating the benefits, challenges, and practical applications of harnessing R's statistical computing capabilities directly within the SQL Server environment.

By leveraging R scripts and functions seamlessly within Power BI, data professionals can gain access to a powerful toolkit for advanced analytics, predictive modelling, and machine learning.

Both integrations facilitates the execution of complex statistical analyses directly on large datasets stored in SQL Server databases or in Power BI (Vertipax), eliminating the need for data movement and enabling additional insights.

This talk presents 'rcpptimer' [1]. 'rcpptimer' is a novel R package that provides Rcpp bindings for 'cpptimer' [2], a simple tic-toc timer class for benchmarking C++ code. This sleek tic-toc timer supports overlapping timers and OpenMP parallelism. It boasts a microsecond-level time resolution. We did not find any overhead of the timer itself at this resolution. Results (with summary statistics) are automatically passed back to R as a data frame.

I will demonstrate the versatile application of 'rcpptimer' in diverse Rcpp projects. Furthermore, we'll delve into the intricacies of its implementation, shedding light on its evolution from 'rcppclock' [3]. I will highlight key features such as OpenMP parallelism and automatically returning results to R. Additionally, I will discuss the rationale behind the decision to separate the core components of 'rcpptimer' into the standalone project 'cpptimer'.

This talk addresses 'Rcpp' beginners. Basic knowledge of C++ classes is needed to follow every implementation aspect.

[1] https://cran.r-project.org/web/packages/rcpptimer/index.html
[2] https://github.com/BerriJ/cpptimer
[3] https://github.com/zdebruine/RcppClock