R users, and all programmers,
I would like to ask a it of help for my first shiny application. Since I do not have computer science background, my question is probably trivial to many users out there. But if anybody can provide some clues, that would be much appreciated.
I am trying to draw interactive graphics for average temperature in London, Paris, and Berlin. I downloaded the data from www.wunderground.com using weatherData package. I am using examples from Chris Beeley's book and Rstudio in order to design my own application.
In my server.R, I upload three data files first. Then, I have sidebar with controls to select a dataset. I also have date range in the sidebar. Once users choose a location and time range, I am asking R to do some data arrangement and pass the object, passData for the next operation. By the time, R reaches renderplot( ), I am assuming that R has a right data frame and produce a graphic using ggplot2. But, I receive the following error message.
Error in print(theGraph) : object 'theGraph' not found
This makes me think that R may not have a right data frame to generate an output graphic. I wonder if anybody can spot what I am doing wrong here. I also wonder if arranging data in reactive() is a good thing to do. Thank you very much for your attention and support.
I leave my codes here.
### ui.R
library(shiny)
shinyUI(pageWithSidebar(
headerPanel("Europe temperature 2013"),
sidebarPanel(
selectInput("dataset", "Choose a location",
choices = c("Berlin Tigel Airport",
"London City Airport",
"Paris Charles De Gaulle")
),
dateRangeInput("daterange", "Date Range",
start = "2013-01-01",
end = "2013-12-31",
min = "2013-01-01",
max = "2013-12-31"
)
),
mainPanel(
plotOutput("theGraph")
)
))
### Weather server.R
library(shiny)
library(weatherData)
library(ggplot2)
library(scales)
library(plyr)
### load weather data.
berlin <- read.csv("berlin.csv", header = T)
london <- read.csv("london.csv", header = T)
paris <- read.csv("paris.csv", header = T)
shinyServer(function(input, output){
# Return the requested dataset
datasetInput <- reactive({
switch(input$dataset,
"Berlin Tigel Airport" = berlin,
"London City Airport" = london,
"Paris Charles De Gaulle" = paris)
})
# Prepare data once and then pass around the program.
passData <- reactive({
foo <- datasetInput()
foo$shortdate <- strftime(foo$Time, format = "%Y-%m-%d")
foo$shortdate <- as.Date(foo$shortdate, format = "%Y-%m-%d")
foo <- foo[foo$shortdate %in%
seq.Date(input$daterange[1],
input$daterange[2], by = 1), ]
foo
})
output$theGraph <- renderPlot({
graphdata <- ddply(passData(), .(shortdate), summarize, mean_C = mean(TemperatureC))
if(input$dataset == "berlin"){
theGraph <- ggplot(graphdata(), aes(shortdate, mean_C)) +
geom_line() +
scale_x_date(labels = date_format("%Y-%m-%d")) +
xlab("") +
ylab("Mean Temperature (C)") +
ggtitle("2013 Average Daily Temperature in Berlin")
}
if(input$dataset == "london"){
theGraph <- ggplot(graphdata(), aes(shortdate, mean_C)) +
geom_line() +
scale_x_date(labels = date_format("%Y-%m-%d")) +
xlab("") +
ylab("Mean Temperature (C)") +
ggtitle("2013 Average Daily Temperature in London")
}
if(input$dataset == "paris"){
theGraph <- ggplot(graphdata(), aes(shortdate, mean_C)) +
geom_line() +
scale_x_date(labels = date_format("%Y-%m-%d")) +
xlab("") +
ylab("Mean Temperature (C)") +
ggtitle("2013 Average Daily Temperature in Paris")
}
print(theGraph)
})
})
### The files look like this. Three columns (Time, Temperature C, TemperatureF)
Time TemperatureC TemperatureF
2013-01-01 01:00:00 6 NA
Sincerely, Kota
If you have problems, simplify and make the example self-contained. I have reduced the example to 1 station, used a preloaded sample set to make the example self-contained, and corrected errors, e.g graphdata instead of graphdata(). Use this to restart with additional locations.
server.R
# server.R
### Weather server.R
library(shiny)
library(weatherData)
library(ggplot2)
library(scales)
library(plyr)
### load weather data.
data(Mumbai2013) # we could not reproduce your example
shinyServer(function(input, output){
# Return the requested dataset
datasetInput <- reactive({
switch(input$dataset,
"Mumbai" = Mumbai2013)
})
# Prepare data once and then pass around the program.
passData <- reactive({
foo <- datasetInput()
foo$shortdate <- strftime(foo$Time, format = "%Y-%m-%d")
foo$shortdate <- as.Date(foo$shortdate, format = "%Y-%m-%d")
foo <- foo[foo$shortdate %in%
seq.Date(input$daterange[1],
input$daterange[2], by = 1), ]
foo
})
output$theGraph <- renderPlot({
graphdata <- ddply(passData(), .(shortdate), summarize, mean_C = mean(Temperature))
theGraph = NULL
theGraph <- ggplot(graphdata, aes(shortdate, mean_C)) +
geom_line() +
scale_x_date(labels = date_format("%Y-%m-%d")) +
xlab("") +
ylab("Mean Temperature (C)") +
ggtitle("2013 Average Daily Temperature in Mumbai")
if (!is.null(theGraph))
print(theGraph)
})
})
ui.R
# ui.R
library(shiny)
shinyUI(pageWithSidebar(
headerPanel("Europe temperature 2013"),
sidebarPanel(
selectInput("dataset", "Choose a location",
choices = c("Mumbai")
),
dateRangeInput("daterange", "Date Range",
start = "2013-01-01",
end = "2013-12-31",
min = "2013-01-01",
max = "2013-12-31"
)
),
mainPanel(
plotOutput("theGraph")
)
))