Dynamic UiOutput Causing Issue when Adding Inputs to a DataFrame Shiny - r

Question

I have an application where the user selects the stocks she want's to analyze. Depending on the number of stocks chosen, the app will render equal amounts of UIOuputs where the user can choose the weight for each stock. So for example, if you choose 6 stocks to analyze, 6 uioutputs will render each asking to select a weight.
The problem I am having is, I would like to create a data frame with the inputs. So if a user selects AAPl and MSFT with weights .50 and .50. I would like to create a df:
Ticker Weight
AAPL .50
MSFT .50
However, when I try and create the dataframe I get an error inputs are not of the same length. I believe this is because of how shiny reactivity works (not ordered). Any inputs would be greatly appreciated. Below is the app.
library(shiny)
library(purrr)
library(tidyverse)
library(DT)
tickers = c("SPY", "IWM", "QQQ", "TLT", "AGG", "GLD", "SLV")
ui <- fluidPage(
# Application title
titlePanel("Portfolio Builder"),
#select the stocks you want to analyze
mainPanel(
selectizeInput("mult", "chose stock", choices = tickers, selected = "SPY", multiple = T),
uiOutput("plo"),
dataTableOutput("dataTab")
)
)
# Define server logic required to draw a histogram
server <- function(input, output) {
output$plo = renderUI({
z = length(input$mult)
name = input$mult
map2(seq(z), name, ~ numericInput(inputId = paste0("hey",.x), label = paste("weight", .y), value = 10))
})
weights = reactiveValues()
observe({weights$current = map(seq(length(input$mult)),~input[[paste0("hey",.x)]]) %>% unlist()})
mat = reactive({
#if(length(weights$current) == length(input$mult)){
df = data.frame(ticker = input$mult, weight = weights$current) %>% mutate(weightPct = weights$current/sum(weights$current))
# }else{NULL}
})
output$dataTab = renderDataTable({
mat()
})
observe(print(weights$current))
observe(print(input$mult))
}

Answer 1

I converted your observe() and reactiveValues() to a single reactive() object. This way it reacts to changes without the complexity you had. The other big difference is that I converted the weights object to a list, but I think it should still be easy to follow. The data frame error persisted as the length of the user inputs and weights momentarily mismatch, so I returned the length check you already had:
library(shiny)
library(purrr)
library(tidyverse)
library(DT)
tickers = c("SPY", "IWM", "QQQ", "TLT", "AGG", "GLD", "SLV")
suppressWarnings()
ui <- fluidPage(
# Application title
titlePanel("Portfolio Builder"),
#select the stocks you want to analyze
mainPanel(
selectizeInput("mult", "chose stock", choices = tickers, selected = "SPY", multiple = T),
uiOutput("plo"),
dataTableOutput("dataTab")
)
)
# Define server logic required to draw a histogram
server <- function(input, output) {
output$plo = renderUI({
z = length(input$mult)
name = input$mult
map2(seq(z), name, ~ numericInput(inputId = paste0("hey",.x), label = paste("weight", .y), value = 10))
})
weights = reactive({
req(input$mult)
list(current = map(seq(length(input$mult)),~input[[paste0("hey",.x)]]) %>% unlist())
})
mat = reactive({
req(weights()$current)
if(length(weights()$current) == length(input$mult)){
df = data.frame(ticker = input$mult, weight = weights()$current) %>% mutate(weightPct = weights()$current/sum(weights()$current))
}
})
output$dataTab = renderDataTable({
req(mat())
mat()
})
}
shinyApp(ui, server)