README.md

In UBC-MDS/LibRely-R-: Meta analysis and documentation of R scripts

LibRely (R)

Latest Update Date: 2019 March

Collaborators

| Name | GitHub Handle | | ---- | ------ | | Alex Hope | @ehhope | | Zixin Zhang | @zxzzhangg | | Aaron Quinton | @aaronquinton |

Releases

Milestone 1 Release

Milestone 3 Release

Overview

The LibRely package is designed to provide a meta analysis of R scripts to aid in the workflow for a data science or software development project. The functionalities in this package interpret scripts to compile library/module dependencies, document specific functions imported in the script and measure the length of the script by the number of lines and characters.

The user can use LibRely to better understand the packages used in a script authored by someone else. Alternatively, a user can take advantage of the functionalities when preparing the ReadMe documentation in their project. The following functions are included in the LibRely package.

lib_search()

• Input parameter(string): a file path to the script
• Output parameter(list): a list of dependencies
• Function purpose: find and list the libraries/modules used in the script passed as an argument

lib_count()

• Input parameter(string): a file path to the script
• Output parameter(tibble): a dataframe that summarizes package, corresponding functions and number of usage
• Function purpose: count the functions used and detail their corresponding libraries/packages

lib_lines()

• Input parameter(string): a file path to the script
• Output parameter(vector): a vector of two number where the first is the number of line and the second is the number of characters
• Function purpose: count the number of lines and characters of the script

Installation

You can install the released version of LibRely from CRAN in R with:

devtools::install_github("UBC-MDS/librely")

Usage

This is a simple usage case of our package in R under the LibRely-R- working directory.

> devtools::load_all()
> library(librely)   

> lib_search("tests/testthat/file/script_for_test1.R")
[1] "dplyr"  "tibble"

> lib_count("tests/testthat/file/script_for_test1.R")
# A tibble: 4 x 3
  package_name function_name count
  <chr>        <chr>         <int>
1 dplyr        mutate            1
2 dplyr        summarize_all     1
3 dplyr        tibble            1
4 tibble       tibble            1

> lib_lines("tests/testthat/file/script_for_test1.R")
$num_line
[1] 7

$num_char
[1] 193

Branch Coverage Test

We use covr package for branch test. You can go to LibRely.Rproj and run the following command lines

install.packages("covr") # if you don't have it installed
library(covr)
report()

The results are shown below.

R Ecosystems

• R: We were not able to find any R software packages that address the same functionality as we intend to with LibRely. Our package fits the R ecosystem because it analyzes a .R script and all these functions return common return types, like a vector, a list and a tibble.

UBC-MDS/LibRely-R- documentation built on May 7, 2019, 7:14 p.m.