README.md

In JohannPopp/epidatR: Reading EpiData Epx-Files and Converting Them to a Data.frame

epidatR

This package reads EpiData epx-files directly into R. So far it contains one major function read.EpiData(). The other functions in this repository are helpers.

Compared to the allready existing package epxToR, read.EpiData() not only transfers epx-files into an R-data.frame. Additionally, values defined as missing in EpiData will be set to NA, defined value and variable labels will be applied and variables will be converted to appropriate R-classes according to their field type in EpiData.

If this conversion is not wanted or causes some trouble, the original raw codes can be transferred into the data.frame as character-variables, by setting the argument convert="none".

If the epx-file consists of one data set only, the output will be a data.frame. If the epx-file contains a relational data base with multiple data sets, it will be a list of multiple data.frames.

Variable labels, study information and information about the relation of data sets are given as attributes.

Helper functions are:

• epx.extract() extracts the xml-like data from the epx-file.
• epx.read() transfers the extracted data into a data.frame.
• epx.missing() sets defined missings to NA.
• epx.class() applies value labels and variable classes.
• epx.example() returns the path of the example files in the attached package.

The function works fine with the example files "SomeFakeData.epx", "marathon.epx", "sample.v3.epx", "Clinical_Example.epx" and "OneLine.epx".

The function can not jet read encrypted epx-files. I would need much more expertise in cryptography to implement this feature. Perhaps there are some experts to help me out. For experimentation I have added the files "sample.v3_password.epx" with simple (symetric) encryption (Password = "password") and "sample.v3_ExtendedAccess.epx" with asymetric encryption produeced by the Extended Access module (Login = "admin", Password = "password").

The package is not jet available from CRAN. You can install it from here, using the devtools-package.

devtools::install_github("https://github.com/JohannPopp/epidatR")

Still To Do List

| Date | To do | Solved | Notes | |------------|-------------------------|--------|-----------------------------| | 2023-11-28 | Reading encrypted files | | complicated | | 2023-11-28 | Setting the time zone | 2023-11-28 | generally set to "UTC" | | 2023-11-28 | Writing epx files | | lot of work. Do we need it? |

JohannPopp/epidatR documentation built on Jan. 29, 2024, 1:23 a.m.