[go: up one dir, main page]
Skip to contents

rdataretriever

R interface to the Data Retriever.

The rdataretriever provides access to cleaned versions of hundreds of commonly used public datasets with a single line of code.

These datasets come from many different sources and most of them require some cleaning and restructuring prior to analysis. The rdataretriever uses a set of actively maintained recipes for downloading, cleaning, and restructing these datasets using a combination of the Frictionless Data Specification and custom data cleaning scripts.

The rdataretriever also facilitates the automatic storage of these datasets in a choice of database management systems (PostgreSQL, SQLite, MySQL, MariaDB) or flat file formats (CSV, XML, JSON) for later use and integration with large data analysis pipelines.

The rdatretriever also facilitates reproducibile science by providing tools to archive and rerun the precise version of a dataset and associated cleaning steps that was used for a specific analysis.

The rdataretriever handles the work of cleaning, storing, and archiving data so that you can focus on analysis, inference and visualization.

Table of Contents

Installation

The rdataretriever is an R wrapper for the Python package, Data Retriever. This means that Python and the retriever Python package need to be installed first.

Basic Installation

If you just want to use the Data Retriever from within R follow these instuctions run the following commands in R. This will create a local Python installation that will only be used by R and install the needed Python package for you.

install.packages('reticulate') # Install R package for interacting with Python
reticulate::install_miniconda() # Install Python
reticulate::py_install('retriever') # Install the Python retriever package
install.packages('rdataretriever') # Install the R package for running the retriever
rdataretriever::get_updates() # Update the available datasets

After running these commands restart R.

Advanced Installation for Python Users

If you are using Python for other tasks you can use rdataretriever with your existing Python installation (though the basic installation above will also work in this case by creating a separate miniconda install and Python environment).

Install the retriever Python package

Install the retriever Python package into your prefered Python environment using either conda (64-bit conda is required):

conda install -c conda-forge retriever

or pip:

pip install retriever

Select the Python environment to use in R

rdataretriever will try to find Python environments with retriever (see the reticulate documentation on order of discovery for more details) installed. Alternatively you can select a Python environment to use when working with rdataretriever (and other packages using reticulate).

The most robust way to do this is to set the RETICULATE_PYTHON environment variable to point to the preferred Python executable:

Sys.setenv(RETICULATE_PYTHON = "/path/to/python")

This command can be run interactively or placed in .Renviron in your home directory.

Alternatively you can do select the Python environment through the reticulate package for either conda:

library(reticulate)
use_conda('name_of_conda_environment')

or virtualenv:

library(reticulate)
use_virtualenv("path_to_virtualenv_environment")

You can check to see which Python environment is being used with:

py_config()

Install the rdataretriever R package 

install.packages("rdataretriever") # latest release from CRAN
remotes::install_github("ropensci/rdataretriever") # development version from GitHub

Installing Tabular Datasets 

library(rdataretriever)

# List the datasets available via the Retriever
rdataretriever::datasets()

# Install the portal into csv files in your working directory
rdataretriever::install_csv('portal')

# Download the raw portal dataset files without any processing to the
# subdirectory named data
rdataretriever::download('portal', './data/')

# Install and load a dataset as a list
portal = rdataretriever::fetch('portal')
names(portal)
head(portal$species)

Installing Spatial Datasets

Set-up and Requirements

Tools

  • PostgreSQL with PostGis, psql(client), raster2pgsql, shp2pgsql, gdal,

The rdataretriever supports installation of spatial data into Postgres DBMS.

  1. Install PostgreSQL and PostGis

    To install PostgreSQL with PostGis for use with spatial data please refer to the OSGeo Postgres installation instructions.

    We recommend storing your PostgreSQL login information in a .pgpass file to avoid supplying the password every time. See the .pgpass documentation for more details.

    After installation, Make sure you have the paths to these tools added to your system’s PATHS. Please consult an operating system expert for help on how to change or add the PATH variables.

    For example, this could be a sample of paths exported on Mac:

    #~/.bash_profile file, Postgres PATHS and tools.
export PATH="/Applications/Postgres.app/Contents/MacOS/bin:${PATH}"
export PATH="$PATH:/Applications/Postgres.app/Contents/Versions/10/bin"

  2. Enable PostGIS extensions

    If you have Postgres set up, enable PostGIS extensions. This is done by using either Postgres CLI or GUI(PgAdmin) and run

    For psql CLI shell psql -d yourdatabase -c "CREATE EXTENSION postgis;" psql -d yourdatabase -c "CREATE EXTENSION postgis_topology;"

    For GUI(PgAdmin)

    CREATE EXTENSION postgis;
CREATE EXTENSION postgis_topology

    For more details refer to the PostGIS docs.

Sample commands

rdataretriever::install_postgres('harvard-forest') # Vector data
rdataretriever::install_postgres('bioclim') # Raster data

# Install only the data of USGS elevation in the given extent
rdataretriever::install_postgres('usgs-elevation', list(-94.98704597353938, 39.027001800158615, -94.3599408119917, 40.69577051867074))

Provenance

To ensure reproducibility the rdataretriever supports creating snapshots of the data and the script in time.

Use the commit function to create and store the snapshot image of the data in time. Provide a descriptive message for the created commit. This is comparable to a git commit, however the function bundles the data and scripts used as a backup.

With provenace, you will be able to reproduce the same analysis in the future.

Commit a dataset

By default commits will be stored in the provenance directory .retriever_provenance, but this directory can be changed by setting the environment variable PROVENANCE_DIR.

rdataretriever::commit('abalone-age',
                       commit_message='A snapshot of Abalone Dataset as of 2020-02-26')

You can also set the path for an individual commit:

rdataretriever::commit('abalone-age',
                       commit_message='Data and recipe archive for Abalone Data on 2020-02-26',
                       path='.')

View a log of committed datasets in the provenance directory

rdataretriever::commit_log('abalone-age')

Install a committed dataset

To reanalyze a committed dataset, rdataretriever will obtain the data and script from the history and rdataretriever will install this particular data into the given back-end. For example, SQLite:

rdataretriever::install_sqlite('abalone-age-a76e77.zip')

Datasets stored in provenance directory can be installed directly using hash value

rdataretriever::install_sqlite('abalone-age', hash_value='a76e77')

Using Docker Containers

To run the image interactively

docker-compose run --service-ports rdata /bin/bash

To run tests

docker-compose run rdata Rscript load_and_test.R

Release

Make sure you have tests passing on R-oldrelease, current R-release and R-devel

To check the package

R CMD Build #build the package
R CMD check  --as-cran --no-manual rdataretriever_[version]tar.gz

To Test

setwd("./rdataretriever") # Set working directory
# install all deps
# install.packages("reticulate")
library(DBI)
library(RPostgreSQL)
library(RSQLite)
library(reticulate)
library(RMariaDB)
install.packages(".", repos = NULL, type="source")
roxygen2::roxygenise()
devtools::test()

To get citation information for the rdataretriever in R use citation(package = 'rdataretriever')

Acknowledgements

A big thanks to Ben Morris for helping to develop the Data Retriever. Thanks to the rOpenSci team with special thanks to Gavin Simpson, Scott Chamberlain, and Karthik Ram who gave helpful advice and fostered the development of this R package. Development of this software was funded by the National Science Foundation as part of a CAREER award to Ethan White.