sebatrag / datahandling

About

Lecture materials for the SEPS-HSG (University of St. Gallen) course 'Data Handling: Import, Cleaning and Visualisation' (BA, Major in Economics). The materials consist of two parts:

• Lecture notes: materials/notes/
• Lecture slides: materials/slides/

Additional materials related to notes and slides:

• Sourcecode of examples used in slides/notes: materials/sourcecode/
• Images used in slides/notes: materials/img
• References used in slides/notes: materials/references

Code and data:

• code/ contains r-scripts with additional examples as well as utility functions to curate/maintain and compile the lecture materials.
• data/ contains data files used in the code examples.

How to contribute

• Open issues:
  • report bugs and typos
  • suggest enhancements
• Resolve open issues with a pull request
• Contribute to the wiki

Prerequisites

install.packages("tidyverse")
install.packages("knitr")
install.packages("bookdown")
install.packages("pacman")
install.packages("qdapRegex")

As well as all the packages used in the code examples.

# get a list of all rmd files (slides and notes)
notes_files <- list.files("materials/notes", pattern = "\\.Rmd", full.names =TRUE)
slides_files <- list.files("materials/slides", pattern = "\\.Rmd", full.names = TRUE)
all_files <- c(notes_files, slides_files)

# parse the rmds, extract a list of package dependencies
rmds <- lapply(all_files, readLines)
to_install <- lapply(rmds, qdapRegex::rm_between, 
                     left = c("library(", "require("),
                     right = c(")", ")"),
                     extract = TRUE)
to_install <-  unique(na.omit(unlist(to_install)))
to_install <- to_install[! to_install %in% c("PACKAGE-NAME",  "<PACKAGE NAME>")]

# install all missing packages
pacman::p_load(char = to_install)

In addition to the standard pandoc installation (should be installed with RStudio) you will have to install the pandocfilters library (in OSX: pip install --user pandocfilters).

Compilation of materials

Run the following in the terminal:

sh makeall_datahandling.sh

This tests all the R code, compiles notes as html (materials/notes/html) and pdf (materials/notes/pdf), slides as html/ioslides (materials/slides/html), and outputs the sourcecode of code examples in the notes to an R script (materials/sourcecode).

Error messages

In case you are confronted with an error message of the type "pandoc document conversion failed" and you still see a tex file that is generated before the knitting process stopped, then do the following: open the tex file in RStudio, run it and install the latex packages that are suggested.