crime-tables-r is a Reproducible Analytical Pipeline (RAP) built in R for producing the Nature of Crime tables from the Crime Survey for England and Wales.
crime-tables-r uses gptables and the official guidance on good practice spreadsheets. It advocates a strong adherence to the guidance by restricting the range of operations possible.
crime-tables-r is developed and maintained by the Centre for Crime and Justice in the Office for National Statistics, UK.
For more information on this code please contact crimestatistics@ons.gov.uk
For development, clone the package from the ONS GitLab using:
git clone http://np2rvlapxx507/Crime_Stats/crime-tables-r.git
In R studio open the file crime-tables-r.proj.
When opened in the R project crime-tables-r.proj the autotable package (in the directory autotable) needs to be set as the working directory to use build tools, using
setwd("./autotable")
The package can be installed using the package devtools by running:
devtools::install()
NOTE: RStudio exposes similar functionality in the Build menu and in the Build pane via the Install and Restart button.
You will need to build the documentation using devtools:document().
At the moment, the pipeline has examples of writing outputs to Excel using the R package xlsxwriter and the Python package gptables which is called from R using the reticulate package. Reticulate is the preferred option for the future.
The pipeline uses a package called autotable which contains the functions used in the pipeline. This follows a standard R package layout. For more details read R Packages
The pipeline can be tested using the package testthat. You can run the tests using:
devtools::test().