spsValidate {spsComps}

Validate expressions

Description

this function is used on server side to usually validate input dataframe or some expression. The usage is similar to shiny::validate but is not limited to shiny render functions and provides better user notification and server-end logging (dual-end logging).

Usage

spsValidate(
  expr,
  vd_name = "my validation",
  pass_msg = glue("validation: '{vd_name}' passed"),
  shiny = TRUE,
  verbose = spsOption("verbose"),
  prefix = ""
)

Arguments

expr

the expression to validate data or other things. Use stop("your message") or generate some errors inside to fail the validation. If there is no error, it will return TRUE and display pass_msg on both console and shiny app if verbose = TRUE or global SPS option verbose is TRUE.

If the expression fails, it will block the code following this function within the same reactive domain to continue, similar to shinyCatch().

vd_name

validate title

pass_msg

string, if pass, what message do you want to show

shiny

bool, show message on console but hide from users? see shinyCatch() for more details

verbose

bool, show pass message? Default follows global verbose setting, use spsUtil::spsOption to set up the value spsOption("verbose, TRUE") to turn on and spsOption("verbose, FALSE") to turn off and spsOption("verbose") to check current setting, see examples.

prefix

see prefix in shinyCatch()

Details

• Since spsComps 0.3.1 to have the message displayed on shiny UI, you don't need to attach the dependencies manually by adding spsDepend("spsValidate") or spsDepend("toastr") (old name) on UI. This becomes optional, only in the case that automatic attachment is not working.

Value

If expression fails, block the code following this validation function and no final return, else TRUE.

Examples

if(interactive()){
    ui <- fluidPage(
        spsDepend("spsValidate"), # optional
        column(
            4,
            h3("click below to make the plot"),
            p("this button will succeed, verbose on"),
            actionButton("vd1", "make plot 1"),
            plotOutput("p1")
        ),
        column(
            4,
            h3("click below to make the plot"),
            p("this button will succeed, verbose off"),
            actionButton("vd2", "make plot 2"),
            plotOutput("p2")
        ),
        column(
            4,
            h3("click below to make the plot"),
            p("this button will fail, no plot will be made"),
            actionButton("vd3", "make plot 3"),
            plotOutput("p3")
        ),
        column(
            4,
            h3("click below to make the plot"),
            p("this button will fail, but the message is hidden from users"),
            actionButton("vd4", "make plot 4"),
            plotOutput("p4")
        )
    )
    server <- function(input, output, session) {
        mydata <- datasets::iris
        observeEvent(input$vd1, {
            spsOption("verbose", TRUE) # use global sps verbose setting
            spsValidate({
                is.data.frame(mydata)
            }, vd_name = "Is dataframe")
            output$p1 <- renderPlot(plot(iris$Sepal.Length, iris$Sepal.Width))
        })
        observeEvent(input$vd2, {
            spsValidate({
                is.data.frame(mydata)
            },
            vd_name = "Is dataframe",
            verbose = FALSE) # use in-function verbose setting
            output$p2 <- renderPlot(plot(iris$Sepal.Length, iris$Sepal.Width))
        })
        observeEvent(input$vd3, {
            spsValidate({
                is.data.frame(mydata)
                if(nrow(mydata) <= 200) stop("Input needs more than 200 rows")
            })
            print("other things blocked")
            output$p3 <- renderPlot(plot(iris$Sepal.Length, iris$Sepal.Width))
        })
        observeEvent(input$vd4, {
            spsValidate({
                is.data.frame(mydata)
                if(nrow(mydata) <= 200) stop("Input needs more than 200 rows")
            }, shiny = FALSE)
            print("other things blocked")
            output$p4 <- renderPlot(plot(iris$Sepal.Length, iris$Sepal.Width))
        })
    }
    shinyApp(ui, server)
}
# outside shiny example
mydata2 <- list(a = 1, b = 2)
spsValidate({(mydata2)}, "Not empty")
try(spsValidate(stopifnot(is.data.frame(mydata2)), "is dataframe?"), silent = TRUE)