Shiny app runs fine locally, fails when deployed. Issue was rename function

shiny
1

Hi there,

I'm a beginner and this is also my first request for help with R . If I've not followed all guidelines for posting please let me know.

Edit #2: there's a lot of text and code below. Apologies. I think my core request is if anyone can look at my shiny script (first chunk of code) and my use of observe, if, and else if in the server and see if there's anything that would stop the app from deploying online? It runs fine from my computer.

Edit to explain what I'm trying to do with this app: I want users to select a year level (education context), and for that choice to determine what classes are available for selection in the second selectInput. So, if they choose 'year 8', only year 8 classes (8A, 8B etc) appear as options for their second choice Depending on their selections, student growth data for that year level or class is rendered End edit

The following shiny app runs locally but fails when deployed online. I'm very new to Shiny but have written a couple of apps before this one. Another had the same issue and I was able to debug it (problem was renaming a horrendously named column that formatted differently when the csv was read by the shinyapps server).
This time I'm stumped.

An earlier version of this app deployed fine but had some other problems so I rewrote it (moved conditionality from UI to server). As such, I don't think it's the data or the 'pat_growth.r' script the app calls (also included below) that is the issue. I think it's the conditional UI but it's my first go at this sort of conditionality so I'm unsure what the problem might be.

Shiny code:

library(ggiraph)
library(tidyverse)
library(janitor)

#conflicted::conflict_prefer("filter", "dplyr")

source("pat_growth.R")


# Define User Interface for app ----

ui <- fluidPage(
  
  # App title ----
  titlePanel("PAT-R Growth: <school name> Nov2020 - Feb2021"),
  
  # Sidebar layout with input and output definitions ----
  sidebarLayout(
    sidebarPanel(
      
      # Input: selector for year level
      
      selectInput("year", "Current Year Level", 
                  choices = list("All",
                                 "8", 
                                 "9", 
                                 "10"),
                  selected = "All"
      ),
      
      selectInput("tags", "Current English Class",
                  choices = "All")
      
      
      
      
      
    ),
    
    #Main panel for displaying outputs-----
    mainPanel(
      
      #Output plot ----------------
      girafeOutput("growth_plot_test")
    )
  )
)



# Define server logic required to draw a histogram
server <- function(input, output, session) {
  
  
  ## Render the conditional parts of the UI
  observe({
    
      
    if (input$year == "All") {
      updateSelectInput(session, "tags", "Current English Class",
                        choices = list("All")
      )
    }
    
    if (input$year == "8") {
      updateSelectInput(session, "tags", "Current English Class",
                        choices = list("All", 
                                       "8A",
                                       "8B",
                                       "8C", 
                                       "8D",
                                       "8F",
                                       "8G",
                                       "8H",
                                       "8K",
                                       "Lit Support 8")
      )
    }
    
    if (input$year == "9") {
      updateSelectInput(session, "tags", "Current English Class",
                        choices = list("All", 
                                       "9A",
                                       "9B",
                                       "9C",
                                       "9D",
                                       "9E",
                                       "9F",
                                       "9G",
                                       "9H",
                                       "9I",
                                       "Lit Support 9&10")
      )
    }
    
    
    if (input$year == "10") {
      updateSelectInput(session, "tags", "Current English Class",
                        choices = list("All", 
                                       "10EA01",
                                       "10EA02",
                                       "10EA03",
                                       "10EA04",
                                       "10EA05",
                                       "10EB02",
                                       "Lit Support 9&10")
      )
    }
    })
  
  
  
  output$growth_plot_test <- renderGirafe({
    
    year <- switch(input$year,
                   "All"="All years",
                   "8"= "Year 8",
                   "9" = "Year 9",
                   "10" = "Year 10")
    
    class <- switch(input$tags,
                    "All"="All classes",
                    "8A"="08A",
                    "8B"="08B",
                    "8C"="08C", 
                    "8D"="08D",
                    "8F"="08F",
                    "8G"="08G",
                    "8H"="08H",
                    "8K"="8K",
                    "9A"="09A",
                    "9B"="09B",
                    "9C"="09C",
                    "9D"="09D",
                    "9E"="09E",
                    "9F"="09F",
                    "9G"="09G",
                    "9H"="09H",
                    "9I"="09I",
                    "10EA01"="EA01",
                    "10EA02"="EA02",
                    "10EA03"="EA03",
                    "10EA04"="EA04",
                    "10EA05"="EA105",
                    "10EB02"="EB02",
                    "Lit Support 8"="Lit Support 8",
                    "Lit Support 9&10"="Lit Support 910"
    )
    
    
    if (input$year=="All") {
      
      growth_plot_all <- 
        growth_data_years_n %>%
        ggplot(aes(yl_two, growth_years))+
        theme_classic()+
        geom_smooth_interactive(alpha=0.1, colour="grey", se=TRUE)+
        geom_jitter_interactive(aes(colour=factor(year_level), 
                                    data_id=(year_level), 
                                    tooltip=paste0(name.x, ", growth: ", growth_years, ", Year 7 NAPLAN:", band_description)
        ), 
        size=point_size, alpha=point_alpha)+
        scale_x_continuous(breaks = seq(0,11, by=1))+
        geom_hline(yintercept = 0, linetype = "dashed", alpha = 0.3)+
        geom_vline(xintercept = 7, linetype = "dashed", alpha = 0.3)+
        geom_vline(xintercept = 10, linetype = "dashed", alpha = 0.3)+
        labs(x="Approx Year level (March 2021)", y="Approx growth in years", title="PAT-R March 2021, whole school", subtitle="Snapshot of student scores and growth since last test", colour="Year level")
      
      girafe(ggobj = growth_plot_all, width_svg = 7, height_svg = 4, options = list(
        opts_hover_inv(css = "opacity:0.1;")))
      
      
      
    }
    
    else if (input$year=="8"){
      growth_function_8("Year 8", class)
      
    }
    
    else if (input$year=="9"){
      growth_function_9("Year 9", class)
      
    }
    
    else if (input$year=="10"){
      growth_function_10("Year 10", class)
      
    }
    })
  }
  

# Run the application 
shinyApp(ui, server)

Here's the r script that contains growth_function_8/9/10':

## Making functions for PAT growth, adding in NAPLAN data when available


##Presets and making required objects

#presets

point_size <- 3
point_alpha <- 0.4

## Read in the longitudinal data & clean

growth_data <- read.csv("data/cc_patr.csv") %>%
  row_to_names(3) %>% 
  clean_names() %>% 
  unite(name, c(given_name, family_name), sep = " ",remove = FALSE) %>%
  rename(scale_one = 11,
         scale_two = 15) %>% 
  mutate(scale_two=as.numeric(scale_two), scale_one=as.numeric(scale_one)) %>%
  drop_na()


## Read in the year level conversion chart & clean

scale_to_year <- read.csv("data/patR_range_to_year.csv") %>% 
  clean_names() %>% 
  rename(range_start=i_range_start_score)


## Add approx year level via scale_to_year

growth_data_years <- growth_data %>% 
  mutate(yl_two = map_dbl(scale_two, .f = function(x) {
    scale_to_year %>% 
      filter(x >= range_start, x <= range_end) %>% 
      pull(range_level)
  }
  ),
  yl_one = map_dbl(scale_one, .f = function(x) {
    scale_to_year %>% 
      filter(x >= range_start, x <= range_end) %>% 
      pull(range_level)
  }
  ),
  growth_scale=(scale_two-scale_one), 
  growth_years=(yl_two-yl_one))

## Add NAPLAN reading bands (years 9 & 10 only)

## Year 9
NAP_scales_9 <- read.csv("data/reading_2019_7_cc.csv") %>% 
  clean_names() %>% 
  unite(name, c(first_name, surname), sep = " ",remove = FALSE) %>% 
  rename(reading = reading_nb, username = cases_id) %>% 
  mutate(band = case_when(between(reading, -44, 417.1) ~ 4,
                          between(reading, 417.2, 484.5) ~ 5,
                          between(reading, 484.6, 534.8) ~ 6, 
                          between(reading, 534.9, 583.7) ~ 7,
                          between(reading, 583.8, 634.6) ~ 8, 
                          between(reading, 634.7, 1000) ~ 9 
  )) %>% 
  mutate(band_description = case_when(
    between(band, 4, 5) ~ "Bottom Two Bands",
    between(band, 6, 7) ~ "Middle Two Bands",
    between(band, 8, 9) ~ "Top Two Bands"))


## Year 10
NAP_scales_10 <- read.csv("data/reading_2018_7_cc.csv") %>% 
  clean_names() %>% 
  unite(name, c(first_name, surname), sep = " ",remove = FALSE) %>% 
  rename(reading = reading_nb, username = cases_id) %>% 
  mutate(band = case_when(between(reading, -200, 416.7) ~ 4,
                          between(reading, 416.8, 468.7) ~ 5,
                          between(reading, 468.8, 523.1) ~ 6, 
                          between(reading, 523.2, 574.2) ~ 7,
                          between(reading, 574.3, 626.7) ~ 8, 
                          between(reading, 626.8, 1000) ~ 9
  )) %>% 
  mutate(band_description = case_when(
    between(band, 4, 5) ~ "Bottom Two Bands",
    between(band, 6, 7) ~ "Middle Two Bands",
    between(band, 8, 9) ~ "Top Two Bands"))


## Join two NAP tables together
All_NAP <- NAP_scales_10 %>% 
  bind_rows(NAP_scales_9)

## joining PAT growth to NAPLAN band

growth_data_years_n <- growth_data_years %>% 
  left_join(All_NAP, by="username")


## Making the function: Year 10

##Answers the question: 'Where are our students at and how have they grown?' 
growth_function_10 <- 
  function(year, class) {
    data_wall_10 <- 
      growth_data_years_n %>%
      filter(year_level==year | year=="All years",
             tags==class | class=="All classes") %>% 
      ggplot(aes(yl_two, growth_years))+
      theme_classic()+
      geom_smooth_interactive(alpha=0.1, colour="grey", se=TRUE)+
      geom_jitter_interactive(aes(colour=(factor(band_description)), data_id=(band_description),
                                  tooltip=paste0(name.x, ", growth: ", growth_years,
                                                 ", Year 7 NAPLAN:", band_description)),
                              size=point_size, alpha=point_alpha)+
      scale_x_continuous(breaks = seq(0,11, by=1))+
      geom_hline(yintercept = 0, linetype = "dashed", alpha = 0.3)+
      geom_vline(xintercept = 7, linetype = "dashed", alpha = 0.3)+
      geom_vline(xintercept = 10, linetype = "dashed", alpha = 0.3)+
      labs(x="Approx Year level (March 2021)", 
           y="Approx growth in years", 
           title=paste("PAT-R March 2021,", year, class), 
           subtitle="Snapshot of student scores and growth since previous test", 
           colour="Year 7 NAPLAN Band")
    
    girafe(ggobj = data_wall_10, width_svg = 7, height_svg = 4, options = list(
      opts_hover_inv(css = "opacity:0.1;")))
    
  }

## Year 9
##Answers the question: 'Where are our students at and how have they grown?'

growth_function_9 <- 
  function(year, class) {
    data_wall_9 <- 
      growth_data_years_n %>%
      filter(year_level==year | year=="All years",
             tags==class | class=="All classes") %>% 
      ggplot(aes(yl_two, growth_years))+
      theme_classic()+
      geom_smooth_interactive(alpha=0.1, colour="grey", se=TRUE)+
      geom_jitter_interactive(aes(colour=(factor(band_description)), data_id=(band_description),
                                  tooltip=paste0(name.x, ", growth: ", growth_years,
                                                 ", Year 7 NAPLAN:", band_description)),
                              size=point_size, alpha=point_alpha)+
      scale_x_continuous(breaks = seq(0,11, by=1))+
      geom_hline(yintercept = 0, linetype = "dashed", alpha = 0.3)+
      geom_vline(xintercept = 7, linetype = "dashed", alpha = 0.3)+
      geom_vline(xintercept = 10, linetype = "dashed", alpha = 0.3)+
      labs(x="Approx Year level (March 2021)", 
           y="Approx growth in years", 
           title=paste("PAT-R March 2021,", year, class), 
           subtitle="Snapshot of student scores and growth since previous test", 
           colour="Year 7 NAPLAN Band")
    
    girafe(ggobj = data_wall_9, width_svg = 7, height_svg = 4, options = list(
      opts_hover_inv(css = "opacity:0.1;")))
    
  }


## Making the function: Year 8

##Answers the question: 'Where are our students at and how have they grown?' 
growth_function_8 <- 
  function(year, class) {
    data_wall_8 <- 
      growth_data_years_n %>%
      filter(year_level==year | year=="All years",
             tags==class | class=="All classes") %>% 
      ggplot(aes(yl_two, growth_years))+
      theme_classic()+
      geom_smooth_interactive(alpha=0.1, colour="grey", se=TRUE)+
      geom_jitter_interactive(aes(colour=factor(tags), data_id=(tags), 
                                  tooltip=paste0(name.x, ", growth: ", growth_years)),
                              size=point_size, alpha=point_alpha)+
      scale_x_continuous(breaks = seq(0,11, by=1))+
      geom_hline(yintercept = 0, linetype = "dashed", alpha = 0.3)+
      geom_vline(xintercept = 7, linetype = "dashed", alpha = 0.3)+
      geom_vline(xintercept = 10, linetype = "dashed", alpha = 0.3)+
      labs(x="Approx Year level (March 2021)", 
           y="Approx growth in years", 
           title=paste("PAT-R March 2021,", year, class), 
           subtitle="Snapshot of student scores and growth since previous test", 
           colour="Class")
    
    girafe(ggobj = data_wall_8, width_svg = 7, height_svg = 4, options = list(
      opts_hover_inv(css = "opacity:0.1;")))
    
  }

There's a bit of probably clunky stuff in my code as I'm a beginner but given my previous version launched online I don't think it's the problem.

Here's the logs from shiny. I've checked other posts from people with this same general issue and often the logs provide a hint (sometimes a very loud one) of the problem but I can't get anything from this myself:

2021-07-11T22:37:59.555637+00:00 shinyapps[4286832]: Running on host: 467471711e16
2021-07-11T22:37:59.563606+00:00 shinyapps[4286832]: Server version: 1.8.6.1
2021-07-11T22:37:59.563619+00:00 shinyapps[4286832]: LANG: en_AU.UTF-8
2021-07-11T22:37:59.563620+00:00 shinyapps[4286832]: R version: 4.1.0
2021-07-11T22:37:59.563644+00:00 shinyapps[4286832]: httpuv version: 1.6.1
2021-07-11T22:37:59.563669+00:00 shinyapps[4286832]: rmarkdown version: 2.9
2021-07-11T22:37:59.563636+00:00 shinyapps[4286832]: shiny version: 1.6.0
2021-07-11T22:37:59.563683+00:00 shinyapps[4286832]: knitr version: 1.33
2021-07-11T22:37:59.563717+00:00 shinyapps[4286832]: RJSONIO version: (none)
2021-07-11T22:37:59.563689+00:00 shinyapps[4286832]: jsonlite version: 1.7.2
2021-07-11T22:37:59.563895+00:00 shinyapps[4286832]: Using pandoc: /opt/connect/ext/pandoc/2.11
2021-07-11T22:37:59.563729+00:00 shinyapps[4286832]: htmltools version: 0.5.1.1
2021-07-11T22:37:59.767243+00:00 shinyapps[4286832]: Using jsonlite for JSON processing
2021-07-11T22:37:59.770504+00:00 shinyapps[4286832]: 
2021-07-11T22:37:59.770506+00:00 shinyapps[4286832]: Starting R with process ID: '23'
2021-07-11T22:38:00.786814+00:00 shinyapps[4286832]: ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.1 ──
2021-07-11T22:38:00.792087+00:00 shinyapps[4286832]: ✔ ggplot2 3.3.4     ✔ purrr   0.3.4
2021-07-11T22:38:00.792089+00:00 shinyapps[4286832]: ✔ tibble  3.1.2     ✔ dplyr   1.0.6
2021-07-11T22:38:00.792090+00:00 shinyapps[4286832]: ✔ readr   1.4.0     ✔ forcats 0.5.1
2021-07-11T22:38:00.792090+00:00 shinyapps[4286832]: ✔ tidyr   1.1.3     ✔ stringr 1.4.0
2021-07-11T22:38:00.879246+00:00 shinyapps[4286832]: ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
2021-07-11T22:38:00.895498+00:00 shinyapps[4286832]: 
2021-07-11T22:38:00.879249+00:00 shinyapps[4286832]: ✖ dplyr::filter() masks stats::filter()
2021-07-11T22:38:00.879250+00:00 shinyapps[4286832]: ✖ dplyr::lag()    masks stats::lag()
2021-07-11T22:38:00.895500+00:00 shinyapps[4286832]: Attaching package: ‘janitor’
2021-07-11T22:38:00.895501+00:00 shinyapps[4286832]: 
2021-07-11T22:38:00.896165+00:00 shinyapps[4286832]: The following objects are masked from ‘package:stats’:
2021-07-11T22:38:00.896167+00:00 shinyapps[4286832]: 
2021-07-11T22:38:00.971114+00:00 shinyapps[4286832]: Error in value[[3L]](cond) : 
2021-07-11T22:38:00.971118+00:00 shinyapps[4286832]: Execution halted
2021-07-11T22:38:00.896168+00:00 shinyapps[4286832]:     chisq.test, fisher.test
2021-07-11T22:38:00.971117+00:00 shinyapps[4286832]: Calls: local ... tryCatch -> tryCatchList -> tryCatchOne -> <Anonymous>
2021-07-11T22:38:00.896168+00:00 shinyapps[4286832]: 
2021-07-11T22:38:13.154676+00:00 shinyapps[4286832]: Running on host: 467471711e16
2021-07-11T22:38:13.162661+00:00 shinyapps[4286832]: Using pandoc: /opt/connect/ext/pandoc/2.11
2021-07-11T22:38:13.162389+00:00 shinyapps[4286832]: Server version: 1.8.6.1
2021-07-11T22:38:13.162485+00:00 shinyapps[4286832]: RJSONIO version: (none)
2021-07-11T22:38:13.162398+00:00 shinyapps[4286832]: LANG: en_AU.UTF-8
2021-07-11T22:38:13.352490+00:00 shinyapps[4286832]: Using jsonlite for JSON processing
2021-07-11T22:38:13.162456+00:00 shinyapps[4286832]: shiny version: 1.6.0
2021-07-11T22:38:13.355629+00:00 shinyapps[4286832]: Starting R with process ID: '41'
2021-07-11T22:38:13.162432+00:00 shinyapps[4286832]: R version: 4.1.0
2021-07-11T22:38:13.355628+00:00 shinyapps[4286832]: 
2021-07-11T22:38:13.162470+00:00 shinyapps[4286832]: httpuv version: 1.6.1
2021-07-11T22:38:13.162480+00:00 shinyapps[4286832]: rmarkdown version: 2.9
2021-07-11T22:38:13.162480+00:00 shinyapps[4286832]: knitr version: 1.33
2021-07-11T22:38:13.162481+00:00 shinyapps[4286832]: jsonlite version: 1.7.2
2021-07-11T22:38:13.162510+00:00 shinyapps[4286832]: htmltools version: 0.5.1.1
2021-07-11T22:38:14.360531+00:00 shinyapps[4286832]: ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.1 ──
2021-07-11T22:38:14.365837+00:00 shinyapps[4286832]: ✔ ggplot2 3.3.4     ✔ purrr   0.3.4
2021-07-11T22:38:14.365838+00:00 shinyapps[4286832]: ✔ tibble  3.1.2     ✔ dplyr   1.0.6
2021-07-11T22:38:14.365839+00:00 shinyapps[4286832]: ✔ tidyr   1.1.3     ✔ stringr 1.4.0
2021-07-11T22:38:14.365839+00:00 shinyapps[4286832]: ✔ readr   1.4.0     ✔ forcats 0.5.1
2021-07-11T22:38:14.451571+00:00 shinyapps[4286832]: ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
2021-07-11T22:38:14.451574+00:00 shinyapps[4286832]: ✖ dplyr::lag()    masks stats::lag()
2021-07-11T22:38:14.451573+00:00 shinyapps[4286832]: ✖ dplyr::filter() masks stats::filter()
2021-07-11T22:38:14.467414+00:00 shinyapps[4286832]: 
2021-07-11T22:38:14.467416+00:00 shinyapps[4286832]: Attaching package: ‘janitor’
2021-07-11T22:38:14.468071+00:00 shinyapps[4286832]: The following objects are masked from ‘package:stats’:
2021-07-11T22:38:14.468073+00:00 shinyapps[4286832]:     chisq.test, fisher.test
2021-07-11T22:38:14.467416+00:00 shinyapps[4286832]: 
2021-07-11T22:38:14.542881+00:00 shinyapps[4286832]: Execution halted
2021-07-11T22:38:14.468073+00:00 shinyapps[4286832]: 
2021-07-11T22:38:14.468072+00:00 shinyapps[4286832]: 
2021-07-11T22:38:14.542881+00:00 shinyapps[4286832]: Calls: local ... tryCatch -> tryCatchList -> tryCatchOne -> <Anonymous>
2021-07-11T22:38:14.542879+00:00 shinyapps[4286832]: Error in value[[3L]](cond) :

If anyone can help I'd be eternally grateful (I'm planning on having a team of teachers access this app tomorrow night--eek!)

Alexander

2

Hi there, if you do a search on the forum you’ll find several posts with the same error you have received while trying to deploy your app. Maybe one of them will help? Here is one of those posts:

3

Thanks Michael. I've done many searches and not just here. One thing that is stumping me is that in the example you provided, there is a hint to what the issue is. In the case you've pointed to: "Error in value[3L] : Cannot open data source

My logs do not have anything after "Error in value[3L] : " so I don't know where to start looking for the error.

4

You’re right I see now that you said that, sorry about that. I don’t have experience with shinyapps.io, so this is a total guess, but what if for testing purposes you move your functions into the app file itself, and not source it like you’re doing after loading the packages. Maybe the error is coming up blank because it lies in the file being sourced and doesn’t know where to find it? Again, that’s a big guess but probably my next step in debugging…

5

I appreciate you having a look :slight_smile:

I know there's a lot in my post but mainly I'm after someone to take a look at my use of

observe
if
else if

in the server and to see if there's anything obvious that could be the problem ..?

Thanks again.

6

Okay--for anyone who comes across this common issue of apps running locally but spitting out the following error when deployed to shinyapps.io:

Error in value[3L] :

(Note that there is no indicator for what the error might be, whereas there often is)

I've lost many hours to this problem on two different apps and BOTH TIMES the issue has been the rename function. Trying to rename janky columns by reproducing that title might be a n00b mistake but it's one I've made twice now. Hopefully if anyone else is searching (or if my future self goes for the hat trick), this solution may help. The example above was resolved by changing:

rename(range_start = i_range_start_score)

to

rename(range_start = 1)
7

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.