Shiny to Markdown PDF Fails on Second and Subsequent Runs - r

Question

I've been trying to follow this tutorial: generating reports
I've been able to get the suggested code to run successfully and produces PDFs without issue. When I try to run the Shiny app using my own .Rmd file, it works fine the first time and produces a PDF where desired. On the second and any additional runs, the file generation fails with ! LaTeX Error: Unknown float option 'H'.
I've tried mentioning the {float} package in the YAML header of the markdown file without any improvement. The only success seems to be removing all of the r code blocks, which leaves my document looking pretty sparse.
Here is app.R:
shinyApp(
ui = fluidPage(
sliderInput("slider", "Slider", 1, 100, 50),
dateInput("dateinput", "Select a date"),
downloadButton("report.pdf", "Generate report")
),
server = function(input, output) {
output$report.pdf <- downloadHandler(
filename = "report.pdf",
content = function(file) {
tempReport <- file.path(tempdir(), "deviants.Rmd")
file.copy("deviants.Rmd", tempReport, overwrite = TRUE)
params <- list(n = input$slider, d = input$dateinput)
print(class(params$n))
print(class(params$d))
print(params)
rmarkdown::render(tempReport, output_file = file,
params = params,
envir = new.env(parent = globalenv())
)
}
)
}
)
and deviants.Rmd:
---
output: pdf_document
always_allow_html: yes
params:
n: NA
d: NA
header-includes:
- \usepackage{booktabs}
- \usepackage{sectsty} \subsectionfont{\centering}
- \usepackage{sectsty} \sectionfont{\centering}
- \usepackage{float{
---
```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
library(grid)
library(png)
library(dplyr)
```
```{r echo=FALSE, results='asis'}
cat(params$n)
cat(params$d)
```
```{r echo=FALSE}
sampleMetrics <- data.frame(
Sample = c("Threshold","A","2","4","11","C","DEF"),
Length = c(">= 6", 10, 11, 11, 11, 11, 12),
Experiment1 = c(">= 10000",5696,8006,6675,9477,5028,7093),
Experiment2 = c(">= 10000", 21223, 27890, 34623, 24152, 25716, 45187),
Sum = c(">=20000", 28409, 41895, 46181, 34129, 12244, 51910),
Total1 = c("N/A", 41382, 132670, 78271, 89930, 98788, 13015),
Total2 = c("N/A", 43170, 53280, 57568, 46584, 51156, 55045),
stringsAsFactors = FALSE
)
super_cell_spec <- function(data, threshold) {
cell_spec(as.numeric(data), "latex", background = ifelse(
as.numeric(data) >= threshold, "#45ff41", "#ff4242"))
}
sampleMetrics$Length[-1] <- super_cell_spec(sampleMetrics$Length[-1], 6)
sampleMetrics$Experiment1[-1] <- super_cell_spec(
sampleMetrics$Experiment1[-1], 10000)
sampleMetrics$Experiment2[-1] <- super_cell_spec(
sampleMetrics$Experiment2[-1], 10000)
sampleMetrics$Sum[-1] <- super_cell_spec(sampleMetrics$Sum[-1], 20000)
sampleMetrics[1,] <- cell_spec(sampleMetrics[1,], "latex",
background = "#afafaf")
sampleMetrics%>%
kable("latex", booktabs = F, escape = F,
col.names = linebreak(c("Sample",
"Length", "Experiment 1",
"Experiment 2","Sum",
"Total\n1", "Total\n2"),
align = "c")) %>%
kable_styling(latex_options = "scale_down")
```
I definitely wouldn't be as confused if it didn't work flawlessly the first time around. Maybe something to do with the envir = new.env(parent = globalenv()) in app.R? Thanks for any thoughts.

Answer 1

Managed to fix this by adding additional packages, mostly through the trial and error of ! Undefined control sequence errors. It would report an issue with multirow and I would look up the right latex package to include.
In the end, my YAML Rmarkdown header looked like:
---
output: pdf_document
always_allow_html: yes
params:
d: NA
st: NA
header-includes:
- \usepackage[table]{xcolor}
- \usepackage{makecell, tabularx, float, multirow, pgfplots}
- \usepackage{booktabs}
- \usepackage{sectsty} \subsectionfont{\centering}
- \usepackage{sectsty} \sectionfont{\centering}
---
This fixed the problem.