[This article was first published on Econometrics by Simulation, and kindly contributed to R-bloggers]. (You can report issue about the content on this page here)

---

Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.

In serveral previous posts I have posted shiny applications which temporarily store data on shiny servers such as hit counters or the survey tool which I created,  These do not work in the long term since shiny will restart its servers without warning when needed.  In addition, saving data to a shiny server is not an ideal method since special database specific commands should be set up to handle the simultaneous write requirements of web applications.

In this post I will show how to add an effective hit counter to shiny applications using a remote database server (MongoHQ).  Much of my code follows the MongoHQ package demo found at http://docs.mongohq.com/languages/r.html

Start and account with MongoHQ. A Sandbox free database account with 512 MB of memory should be more than sufficient.

Once you have started an account you need to log into app.mongohq.com and start a database as well as a collection.  Within a database you will need to select the admin tab as well in order to create a user id which you can use to log into the collection.

The following code is what I use to create a hit counter.

# Load the CRAN library
library(rmongodb)
 
# You can find the host information for the collection under the admin tab.
host <- "myarea.mongohq.com:myport"
username <- "mycreateduser"
password <- "mycreatedpassword"
db <- "mydatabase"
 
mongo <- mongo.create(host=host , db=db, username=username, password=password)
 
# Load the collection.  In this case the collection is.
collection <- "OLS-app"
namespace <- paste(db, collection, sep=".")
 
# Insert a simple entry into the collection at the time of log in
# listing the date that the collection was accessed.
b <- mongo.bson.from.list(list(platform="MongoHQ",
                    app="counter", date=toString(Sys.Date())))
ok <- mongo.insert(mongo, namespace, b)
 
# Now we query the database for the number of hits
buf <- mongo.bson.buffer.create()
mongo.bson.buffer.append(buf, "app", "counter")
query <- mongo.bson.from.buffer(buf)
counter <- mongo.count(mongo, namespace, query)
 
# I am not really sure if this is a good way of doing this
# at all.
 
# I send the number of hits to the shiny counter as a renderText
# reactive function
paste0("Hits: ", counter)

The now database run hit counter can be seen at:
http://econometricsbysimulation.shinyapps.io/OLS-App/

You can find the updated code at github
https://github.com/EconometricsBySimulation/OLS-demo-App/blob/master/server.R

---

R-bloggers.com offers daily e-mail updates about R news and tutorials about learning R and many other topics. Click here if you're looking to post or find an R/data-science job.

---

Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.