The goal of rvisidata is to combine the power of Visidata, "A Swiss Army Chainsaw for Data", with R.
You need to have Visidata installed. Please follow the official installation instructions provided here.
If you already have python3 installed, the easiest way to install it is with pip3: pip3 install visidata.
To verify visidata is installed correctly, vd --version should print out the installed version.
rvisidata can be installed with devtools from the official GitHub repo:
devtools::install_github('paulklemm/rvisidata')vd(iris)Internally, rvisidata writes the data frame as a csv file to a temporary folder and then loads it with visidata. Be therefore careful with very large data frames, because the writing process can take a while. It will delete the temporary file after the visidata session is closed.
I do not claim that this implementation is efficient or very elegant. But it works. Suggestions for making it faster are very welcome.
This package supports integration with tmux.
The default behaviour is to open vd in a vertical pane above the current pane, if R is running in a tmux session.
To disable this, set:
options(rvisidata.tmux = FALSE)To customize the way tmux sets up this workspace (For example, opening vd in a new window rather than pane), you
may supply your own shell script that invokes vd on your dataset.
Set:
options(rvisidata.tmux = "path/to/your/script")Your script should take one argument, the filename to be opened.
For example, the following will open your dataframe in a vertical split above the R console window, and return focus back to R:
#!/bin/bash
filename=$1
tmux split-window -vb -p 70 "vd $filename" \; \
select-pane -lRight now this tool only opens the data frame and discards any changes you make for two reasons: (1) you should do your data manipulation in R to have a reproducible analysis workflow and (2) it is easier to implement this way 😇.
Visidata is designed and developed by Saul Pwanson (official GitHub repository).