--- title: "Specifications" output: rmarkdown::html_vignette vignette: > %\VignetteIndexEntry{Specifications} %\VignetteEngine{knitr::rmarkdown} %\VignetteEncoding{UTF-8} --- ```{r, include = FALSE} knitr::opts_chunk$set( collapse = TRUE, comment = "#>" ) ``` ```{r setup} suppressPackageStartupMessages(library(sap)) ``` This vignette provides example specifications for the `sap` application (written in the [Gherkin](https://cucumber.io/docs/gherkin/) syntax) ## Movies App Specifications **Example:** The application should source movie review data from platforms like IMDB or Rotten Tomatoes ### Features **Feature:** Movie Review Dataset Variables ***As a*** user ***I want to*** have a dataset with variables from IMDB and Rotten Tomatoes ***In order to*** provide comprehensive movie reviews ### Background ***Given*** the application has access to data from the IMDB and Rotten Tomatoes APIs ## Functional Requirements (i.e., Scenarios) **Scenario**: Movie Review Continuous and Categorical Variables ***When*** the application loads from IMDB and Rotten Tomatoes movie review data ***Then*** the dataset should include a continuous critic ratings variable ***And*** the dataset should include a continuous audience ratings variable ***And*** the dataset should include a categorical mpaa ratings variable ***And*** the dataset should include a categorical genres variable ## Traceability matrix +-----------------------------------------------------------------------------------------------------+-----------------------------------------------------------------------------+----------------------------------------------------------------------------------+------+ | Specifications | Features | Requirements | Test | +=====================================================================================================+=============================================================================+==================================================================================+======+ | **S1**: The application should source movie review data from platforms like IMDB or Rotten Tomatoes | **F1.1**: | **Scenario**: Movie Review Continuous and Categorical Variables | ? | | | | | | | | **Feature:** Movie Review Dataset Variables | ***When*** the application loads from IMDB and Rotten Tomatoes movie review data | | | | | | | | | ***As a*** user | ***Then*** the dataset should include a continuous critic ratings variable | | | | | | | | | ***I want to*** have a dataset with variables from IMDB and Rotten Tomatoes | ***And*** the dataset should include a continuous audience ratings variable | | | | | | | | | ***In order to*** provide comprehensive movie reviews | ***And*** the dataset should include a categorical mpaa ratings variable | | | | | | | | | **Background**: | ***And*** the dataset should include a categorical genres variable | | | | | | | | | ***Given*** the application has access to IMDB and Rotten Tomatoes APIs | | | +-----------------------------------------------------------------------------------------------------+-----------------------------------------------------------------------------+----------------------------------------------------------------------------------+------+