Rmarkdown Shiny Limits - r-markdown

Question

I have a Rmarkdown document with an embedded shiny application (runtime: shiny) which I'd like to upload to shinyapps.io (eventually). When I build the document locally, it fails to completely build, as in the page stops halfway through the document. I've confirmed that if I remove some large leaflet objects in the middle of the document then the build finishes.
I'm working on making the leaflet objects smaller, but I've seen that there is a memory limit on Shiny apps that can be reconfigured (options(shiny.maxRequestSize=30*1024^2) for 30 MB). Supposedly this is supposed to go in the server section of an app, but if the entire document is an app, does this go in the yaml, or in a setup chunk, or somewhere else?
I was able to make an MWE that illustrates my basic environment, though it does not reproduce the error. The maps chunk shows a leaflet map of census tracts for each of the 50 states and DC, and then there's a true shiny app following.
---
title: "Test RMD"
output: html_document
runtime: shiny
---
```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
library(leaflet)
library(shiny)
library(tigris)
library(htmltools)
library(RColorBrewer)
options(shiny.maxRequestSize=30*1024^2)
```
# leaflet maps
```{r maps, echo=T,results='asis'}
us_states <- unique(fips_codes$state)[1:51] # for small, set to 2
createMaps <- function(state){
stmap <- tracts(state, cb = TRUE)
leaflet(stmap) %>% addTiles() %>% addPolygons()
}
htmltools::tagList(lapply(us_states, function(x) createMaps(x) ))
```
# Shiny application
```{r tabsets, echo=FALSE}
shinyApp(
ui = bootstrapPage(
tags$style(type = "text/css", "html, body {width:100%;height:100%}"),
leafletOutput("map", width = "100%", height = "100%"),
absolutePanel(top = 10, right = 10,
sliderInput("range", "Magnitudes", min(quakes$mag), max(quakes$mag),
value = range(quakes$mag), step = 0.1
),
selectInput("colors", "Color Scheme",
rownames(subset(brewer.pal.info, category %in% c("seq", "div")))
),
checkboxInput("legend", "Show legend", TRUE)
)
),
server = function(input, output, session) {
# Reactive expression for the data subsetted to what the user selected
filteredData <- reactive({
quakes[quakes$mag >= input$range[1] & quakes$mag <= input$range[2],]
})
# This reactive expression represents the palette function,
# which changes as the user makes selections in UI.
colorpal <- reactive({
colorNumeric(input$colors, quakes$mag)
})
output$map <- renderLeaflet({
# Use leaflet() here, and only include aspects of the map that
# won't need to change dynamically (at least, not unless the
# entire map is being torn down and recreated).
leaflet(quakes) %>% addTiles() %>%
fitBounds(~min(long), ~min(lat), ~max(long), ~max(lat))
})
# Incremental changes to the map (in this case, replacing the
# circles when a new color is chosen) should be performed in
# an observer. Each independent set of things that can change
# should be managed in its own observer.
observe({
pal <- colorpal()
leafletProxy("map", data = filteredData()) %>%
clearShapes() %>%
addCircles(radius = ~10^mag/10, weight = 1, color = "#777777",
fillColor = ~pal(mag), fillOpacity = 0.7, popup = ~paste(mag)
)
})
# Use a separate observer to recreate the legend as needed.
observe({
proxy <- leafletProxy("map", data = quakes)
# Remove any existing legend, and only if the legend is
# enabled, create a new one.
proxy %>% clearControls()
if (input$legend) {
pal <- colorpal()
proxy %>% addLegend(position = "bottomright",
pal = pal, values = ~mag
)
}
})
}
)
```
I guess my main question is if the options() call is going in a place where Shiny can see it. It's also possible that if I made the application itself bigger that it would cause problems; I can try to get to that this evening.