Unable to knit .Rmd to html; problem with plotly and interactive plots

Hello! I wonder if anyone here is able to point me to how to get further with my troubleshooting of an Rmd document that used to knit before but recently has failed to do so. I have isolated the problem to a specific chunk that calls a plotly function that includes some interactive features.

When I run the chunks I do not get any warnings of problems with the function. It's only when I try to knit that it fails to produce the html document with the correct styling of the left sidebar and it throws an error:

Error in if (title == x2) return(head) : the condition has length > 1
Calls: <Anonymous> ... split_chapters -> build -> sub -> is.factor -> prepend_chapter_title
Execution halted

It seems to me a problem that is slightly more advanced than what I normally deal with. I provide a minimum reprex here together with the session info. I would appreciate any suggestions. Thank you!

title: "Supplementary Information"
author: "Tan, M., Xie, X., Jaeger, T. F."
date: "`r Sys.Date()`"
  - \usepackage{booktabs}
  - \usepackage{siunitx}
  - \usepackage{tabto}
  - \usepackage{soul}
  - \usepackage{xcolor}
  - \usepackage{placeins}
  - \usepackage{lscape}
  - \newcommand{\blandscape}{\begin{landscape}}
  - \newcommand{\elandscape}{\end{landscape}}
  - \makeatletter\renewcommand{\fps@table}{!ht}\makeatother
  - \setstcolor{red}
  - \usepackage{sectsty}
  - \sectionfont{\color{red}} 
  - \subsectionfont{\color{red}}
  - \subsubsectionfont{\color{red}}
    split_by: rmd
    self_contained: true
        collapse: subsection
        scroll_highlight: yes
        before: null
        after: null
        position: static
      edit : null
      download: null
      search: yes
documentclass: book
always_allow_html: true
citation_package: BibTex
bibliography: [refs.bib]
link-citations: yes
csl: frontiers-in-psychology.csl


<style type="text/css">
div.main-container {
  max-width: 1800px;
  margin-left: auto;
  margin-right: auto;
knitr::opts_chunk$set(echo = FALSE, 
                      message=FALSE, warning=TRUE, error=TRUE, cache = FALSE)


library(plotly)         # 3D plotting 
library(shiny)          # tabs

# set plot themes
  theme_bw() + 
    theme(panel.border = element_blank(),
          axis.title.x = element_text(margin = margin(t = 20)),
          axis.title.y = element_text(margin = margin(r = 20)))

colors.Accent = colors.Group
colors.Category = c("blue", "orange")
colors.Language = c("cyan", "pink")
shapes.Category = c("circle","square")


# make dataframe
d.stims.long <- 
  NativeLanguage = rep("English", 20),
  Accent = rep("native", 20),
  Used_for = rep("training", 20),
  Item.Word = rep(c("pet", "bet", "pound", "bound", "but", "bud", "white", "wide", "gilt", "gild"), 2),
  Sound = rep(c("/t/", "/d/"), 10),
  ContextCorrected = rep(c("uncorrected", "step 1"), each = 10),
  vowel = c(rnorm(10, 220, 90), rnorm(10, 20, 90)),
  closure = c(rnorm(10, 185, 97), rnorm(10, 18, 75)),
  burst = c(rnorm(10, 165, 92), rnorm(10, 12, 80)))

# function to make 3d plot with interactive slider
plot_corrected_cues <- function(.data, title = "Correction step") {
  plot_ly(data = .data %>% 
          frame= ~ContextCorrected,
          ids= ~Item.Word, 
          x= ~vowel,
          y= ~closure,
          z= ~burst,
          text= ~Item.Word,
          color= ~Sound,
          colors = colors.Category,
          opacity = .35, 
          symbol= ~Used_for,
          symbols = shapes.Category,
          hoverinfo = "text", showlegend = T,
          type="scatter3d", mode="markers") %>%
      currentvalue = list(prefix = paste0(title, ": "), 
                          font = list(color="black"))

# call function to make 3D plot
  d.stims.long %>% 
    filter(NativeLanguage == "English", 
                 Accent == "native"))

Session info

options(width = 120)

Please can you format your example correctly ? Thanks


Thanks for reading this post. I wasn't sure which part was confusing but I've tried my best to clean it up and hopefully it's clearer now. Please let me know if this is still unclear.