Passing params to shiny rmarkdown - html

Question

My first question here, though I've been reading the site for a long time.
I've been struggling (to put it mildly) for a few days to import the data into a shiny rmarkdown file.
I'm open to other methods, but, ultimately, it needs to be linked to in a website.
The file that the data comes from would be on the users computer.
My latest attempt is this..
It gets the error:
"Error: cannot open the connection" when I click on the link.
html:
<input type="file" id="files" name="files[]" multiple />
<output id="list">
<a href="https://<myName>.shinyapps.io/<myShinyRmarkdownFile>/?"id>linkText</a>
</output>
Bits from Rmarkdown file:
---
title: "x"
author: "x"
date: "x"
output: html_document
runtime: shiny
params:
filename: "<link on desktop which works when running in desktop rstudio>"
---
```{r echo=FALSE, comment=NA}
print(paste("x",params$filename,sep=" "))
A<-read.delim(params$filename,header=T,sep="\t")
.
.
.

Answer 1

I found out how to do what I wanted, though the title of the question was probably misleading. I wanted to get data into shiny, that I could then analyse and make a pdf with.
The following helped me get the data into shiny: youtube video and corresponding server.R file.
https://www.youtube.com/watch?v=HPZSunrSo5M
https://github.com/aagarw30/R-Shinyapp-Tutorial/blob/master/fileinput/server.R
thanks,
John
Edit:
The server.R file is below:
library(shiny)
# use the below options code if you wish to increase the file input limit, in this example file input limit is increased from 5MB to 9MB
# options(shiny.maxRequestSize = 9*1024^2)
shinyServer(function(input,output){
# This reactive function will take the inputs from UI.R and use them for read.table() to read the data from the file. It returns the dataset in the form of a dataframe.
# file$datapath -> gives the path of the file
data <- reactive({
file1 <- input$file
if(is.null(file1)){return()}
read.table(file=file1$datapath, sep=input$sep, header = input$header, stringsAsFactors = input$stringAsFactors)
})
# this reactive output contains the summary of the dataset and display the summary in table format
output$filedf <- renderTable({
if(is.null(data())){return ()}
input$file
})
# this reactive output contains the summary of the dataset and display the summary in table format
output$sum <- renderTable({
if(is.null(data())){return ()}
summary(data())
})
# This reactive output contains the dataset and display the dataset in table format
output$table <- renderTable({
if(is.null(data())){return ()}
data()
})
# the following renderUI is used to dynamically generate the tabsets when the file is loaded. Until the file is loaded, app will not show the tabset.
output$tb <- renderUI({
if(is.null(data()))
h5("Powered by", tags$img(src='RStudio-Ball.png', heigth=200, width=200))
else
tabsetPanel(tabPanel("About file", tableOutput("filedf")),tabPanel("Data", tableOutput("table")),tabPanel("Summary", tableOutput("sum")))
})
})
The "important bits":
data <- reactive({
file1 <- input$file
if(is.null(file1)){return()}
read.table(file=file1$datapath, sep=input$sep, header = input$header, stringsAsFactors = input$stringAsFactors)
# [ john's edit: more code can be put here to alter data before passed to data(). eg:
# o=read.delim(file=file1$datapath)
# convertData(o)
#]
})
Reactive meaning, it waits for input$file (ie, the file chooser), then reads the file you've chosen. Personally, I use read.delim(....) rather than read.table(...). The data (dataframe) is now stored in "data()" and is referred to as such. Hence:
output$table <- renderTable({
if(is.null(data())){return ()}
data()
})
Note the above checks that there is data in data() before displaying.
data() can be passed into functions in the same was as a dataframe:
new.dataframe=do.something.with.the.data(data())
Another issue that had me stumped for a long time is how to export/render a pdf.
output$download_pdf <- downloadHandler(
filename = function() {
paste('pdf_name', sep = '.','pdf')
},
content = function(file) {
src <- normalizePath('rmarkdownFile.Rmd')
print(paste("src",src,sep=": "))
owd <- setwd(tempdir())
on.exit(setwd(owd))
file.copy(src, 'rmarkdownFile.Rmd')
df.A=data.frame(functionA(data()))
df.B=data.frame(functionB(data()))
out=rmarkdown::render('rmarkdownFile.Rmd',pdf_document())
file.rename(out,file)
}
)
I've not got it all worked out, eg, you're supposed to copy the file first to rename it, but that seems to cause problems, I've left it for now.
df.A and df.B are dataframes passed to the rmarkdown file. Nothing else needs to be done. Within rmarkdown just use them as normal, no params etc..
Using on your home computer (with the settings above) the pdf is saved within the project directory. Using as an uploaded app in shinyapps.io a file save location window pops up.
I hope the above helps.