The pumilioR package has functions to get data and sound files from the web-based sound archive system Pumilio (http://ljvillanueva.github.io/pumilio). This introduction will provide a brief overview of setting up Pumilio and examples of how to query the system and extract both data and soundfiles.
pumilioR requires R version 2.14, or newer, and depends on the packages XML and RCurl.
The source of the package is kept at Github: https://github.com/ljvillanueva/pumilioR
The website of the package is http://ljvillanueva.github.io/pumilioR/
To allow the use of the package, log in as an administrator to the Pumilio system. Click the Admin menu link (on the top right edge of the page). On the main settings page of the Admin menu, under “System behavior”, there is an option:
Set it to Yes. Now pumilioR, and other programs, can read data from the archive using XML.
To limit the access to valid users registered in the Pumilio system, set the option:
to Users. To allow anyone to query the system set it to Any.
Before querying the system, first check that it is properly configured and that you can access it, for example:
#Load the package
library(pumilioR)
#Check that the system can use pumilioR, returns TRUE if it is allowed and the version supports it
access <- checkVersion(pumilio_URL = "http://pumilio.coquipr.com/pumilio/")
access
To query the system, use the functions getCollections() or getSites() to get a data frame with the collections and the sites, respectively, in the system. Then, you can get the sounds for that collection or site using the function getSounds(). For example:
#Get all collections
cols <- getCollections(pumilio_URL = "http://pumilio.coquipr.com/pumilio/")
#Show the first row
cols[1,]
#Get all sites
sites <- getSites(pumilio_URL = "http://pumilio.coquipr.com/pumilio/")
#Show the first row
sites[1,]
#Get all the sounds in the site with the SiteID = 102
sounds <- getSounds(pumilio_URL="http://pumilio.coquipr.com/pumilio/", SiteID=102, type="site")
#Show the first row
sounds[1,]
Once a result is obtained from the system, the audio file can be downloaded using the function getFile(). This function needs the result of either getCollections() or getSites() and the SoundID of the desired file.
#First, get a list of results
results <- getSounds(pumilio_URL="http://pumilio.coquipr.com/pumilio/", SiteID=111, type="site")
results[1,]
Then, get the file from the first row, identified with the SoundID 4078. The audio file is saved in the working directory.
getFile(results, SoundID = 4078)
Basic authentication can be used on the Apache webserver as a way to limit access to the data by requiring a username and password. This is outside the scope of this vignette. Since version 1.2 of pumilioR these sites can be accessed providing the username and password, separated with a colon, to the argument credentials in each function.
sites <- getSites(pumilio_URL = "http://pumilio.coquipr.com/pumilio2/", credentials = "pumilio:pumilio")
sites[1,]
Since version 1.2 of pumilioR and version 2.7.0 of Pumilio, the system can be accessed providing a valid username and password, separated with a colon, already registered in the system to the argument pumiliologin in each function.
library(pumilioR)
sites <- getSites(pumilio_URL = "http://pumilio.coquipr.com/pumilio3/", pumiliologin = "pumilio:pumilio")
sites[1,]
Please visit the package website for more information and updates: http://ljvillanueva.github.io/pumilioR/
For suggestions or to report bugs or problems: http://github.com/ljvillanueva/pumilioR/issues
The package page in CRAN is https://cran.r-project.org/package=pumilioR
Vignette “Setting up Pumilio and Usage of the pumilioR Package” by LJ Villanueva-Rivera
Version 1.0.1 (22 November 2016)