How to automate nested sections in Rmds which include text, maps and tables

Question

I found several examples (How to extract the content of SQL-Files using R, Use loop to generate section of text in rmarkdown, Automating the generation of preformated text in Rmarkdown using R), but none works in my case with nested sections including figures. Expected output:

It seems, I need some way to handle the included code chunks (e.g. cat("```{r, fig.width=8.4, fig.height=7}\n") ... cat("```\n") properly, but I can't find a way.

Edit: I found this here, but still can't make it work

kexpand <- function(chunkname, fig.width, fig.height) {
  cat(knitr::knit(text = knitr::knit_expand(text = 
     sprintf("```{r %s, fig.width=%s, fig.height=%s}\n.chunk_code\n", 
             chunkname, fig.width, fig.height))))
}

and then in the chunks:

  .chunk_code <- leaflet() %>%
    addTiles() %>%
    addProviderTiles(providers$OpenStreetMap, group = "OSM") %>%
    addProviderTiles(providers$Stamen.TonerLite, group = "Toner Lite") %>%
    addLayersControl(baseGroups = c("OSM", "Toner Lite")) %>%
    addCircleMarkers(lat = 47, lng = 9, popup = ranges[i])
  kexpand(i, 8.4, 7)

Edit 2: There is a related question. Although it seems to be close, it still does not work. See also my approach here which I use because I have nested list - I do not want all figures next to each other.

All sections include the same type of figures, but the content changes of course. Below you find a

the best I got so far and
reference Rmd which shows the expeted output but with copy-paste chunks

Examples:

---
title: "The best I got"
author: "SQC"
date: "11 Juli 2019"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = FALSE, fig.width=8.4, fig.height=7)
library(leaflet)
ranges <- paste0("section_", c(1:3))
```

# Title level 1

This is an R Markdown document. 

## Title level 2 (tab) {.tabset}

### Title level 3 (tab) {.tabset}

```{r, echo = FALSE, results='asis'}
for (i in 1:length(ranges)) {
  cat(paste0("#### Title level 4 `r ", 
             ranges[i], "` vs. `r ", ranges[i - 1], "` {.tabset}\n"))
  cat("##### Map\n")

  # code //
  cat("```{r, fig.width=8.4, fig.height=7}\n")
  leaflet() %>%
    addTiles() %>%
    addProviderTiles(providers$OpenStreetMap, group = "OSM") %>%
    addProviderTiles(providers$Stamen.TonerLite, group = "Toner Lite") %>%
    addLayersControl(baseGroups = c("OSM", "Toner Lite")) %>%
    addCircleMarkers(lat = 47, lng = 9, popup = ranges[i])
  cat("```\n")
  # //

  cat("##### Tabelle\n")

  # code //
  cat("```{r}\n")
  DT::datatable(data.frame(x = 1, y = ranges[i], stringsAsFactors = FALSE),
                rownames = FALSE)
  cat("```\n")
  # //
}
```

### Title level 3 (tab) {.tabset}

```{r, echo = FALSE, results='asis'}
for (i in 1:length(ranges)) {
  cat(paste0("#### `r ", ranges[i], "` vs. `r ", ranges[i - 1], "` {.tabset}\n"))
  cat("##### Karte\n")

  # code //
  cat("```{r, fig.width=8.4, fig.height=7}\n")
  leaflet() %>%
    addTiles() %>%
    addProviderTiles(providers$OpenStreetMap, group = "OSM") %>%
    addProviderTiles(providers$Stamen.TonerLite, group = "Toner Lite") %>%
    addLayersControl(baseGroups = c("OSM", "Toner Lite")) %>%
    addCircleMarkers(lat = 47, lng = 9, popup = ranges[i])
  cat("```\n")
  # //

  cat("##### Tabelle\n")

  # code //
  cat("```{r}\n")
  DT::datatable(data.frame(x = 1, y = ranges[i], stringsAsFactors = FALSE),
                rownames = FALSE)
  cat("```\n")
  # // 
}
```

---
title: "Reference"
author: "SQC"
date: "11 Juli 2019"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = FALSE)
library(leaflet)
ranges <- paste0("section_", c(1:4))
```

# Title level 1

This is an R Markdown document. 

## Title level 2 (tab) {.tabset}

### Title level 3 (tab) {.tabset}

#### Title level 4 `r ranges[2]` vs. `r ranges[1]` {.tabset}

##### Karte
```{r, fig.width=8.4, fig.height=7}
leaflet() %>%
  addTiles() %>%
  addProviderTiles(providers$OpenStreetMap, group = "OSM") %>%
  addProviderTiles(providers$Stamen.TonerLite, group = "Toner Lite") %>%
  addLayersControl(baseGroups = c("OSM", "Toner Lite")) %>%
  addCircleMarkers(lat = 47, lng = 9, popup = ranges[2])
```

##### Tabelle
```{r}
DT::datatable(data.frame(x = 1, y = ranges[2], stringsAsFactors = FALSE),
              rownames = FALSE)
```

#### Title level 4 `r ranges[3]` vs. `r ranges[2]` {.tabset}

##### Karte 
```{r, fig.width=8.4, fig.height=7}
leaflet() %>%
  addTiles() %>%
  addProviderTiles(providers$OpenStreetMap, group = "OSM") %>%
  addProviderTiles(providers$Stamen.TonerLite, group = "Toner Lite") %>%
  addLayersControl(baseGroups = c("OSM", "Toner Lite")) %>%
  addCircleMarkers(lat = 47, lng = 9, popup = ranges[3])
```

##### Tabelle 
```{r}
DT::datatable(data.frame(x = 1, y = ranges[3], stringsAsFactors = FALSE),
              rownames = FALSE)
```

#### Title level 4 `r ranges[4]` vs. `r ranges[3]` {.tabset}

##### Karte 
```{r, fig.width=8.4, fig.height=7}
leaflet() %>%
  addTiles() %>%
  addProviderTiles(providers$OpenStreetMap, group = "OSM") %>%
  addProviderTiles(providers$Stamen.TonerLite, group = "Toner Lite") %>%
  addLayersControl(baseGroups = c("OSM", "Toner Lite")) %>%
  addCircleMarkers(lat = 47, lng = 9, popup = ranges[4])
```

##### Tabelle 
```{r}
DT::datatable(data.frame(x = 1, y = ranges[4], stringsAsFactors = FALSE),
              rownames = FALSE)
```

### Title level 3 (tab) {.tabset}

#### Title level 4 `r ranges[2]` vs. `r ranges[1]` {.tabset}

##### Karte
```{r, fig.width=8.4, fig.height=7}
leaflet() %>%
  addTiles() %>%
  addProviderTiles(providers$OpenStreetMap, group = "OSM") %>%
  addProviderTiles(providers$Stamen.TonerLite, group = "Toner Lite") %>%
  addLayersControl(baseGroups = c("OSM", "Toner Lite")) %>%
  addCircleMarkers(lat = 47, lng = 9, popup = ranges[2])
```

##### Tabelle
```{r}
DT::datatable(data.frame(x = 1, y = ranges[2], stringsAsFactors = FALSE),
              rownames = FALSE)
```

#### Title level 4 `r ranges[3]` vs. `r ranges[2]` {.tabset}

##### Karte 
```{r, fig.width=8.4, fig.height=7}
leaflet() %>%
  addTiles() %>%
  addProviderTiles(providers$OpenStreetMap, group = "OSM") %>%
  addProviderTiles(providers$Stamen.TonerLite, group = "Toner Lite") %>%
  addLayersControl(baseGroups = c("OSM", "Toner Lite")) %>%
  addCircleMarkers(lat = 47, lng = 9, popup = ranges[3])
```

##### Tabelle 
```{r}
DT::datatable(data.frame(x = 1, y = ranges[3], stringsAsFactors = FALSE),
              rownames = FALSE)
```

#### Title level 4 `r ranges[4]` vs. `r ranges[3]` {.tabset}

##### Karte 
```{r, fig.width=8.4, fig.height=7}
leaflet() %>%
  addTiles() %>%
  addProviderTiles(providers$OpenStreetMap, group = "OSM") %>%
  addProviderTiles(providers$Stamen.TonerLite, group = "Toner Lite") %>%
  addLayersControl(baseGroups = c("OSM", "Toner Lite")) %>%
  addCircleMarkers(lat = 47, lng = 9, popup = ranges[4])
```

##### Tabelle 
```{r}
DT::datatable(data.frame(x = 1, y = ranges[4], stringsAsFactors = FALSE),
              rownames = FALSE)
```

From xfun::session_info('knitr')

R version 3.5.1 (2018-07-02)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows >= 8 x64 (build 9200), RStudio 1.1.453

Locale:
  LC_COLLATE=German_Switzerland.1252  LC_CTYPE=German_Switzerland.1252    LC_MONETARY=German_Switzerland.1252
  LC_NUMERIC=C                        LC_TIME=German_Switzerland.1252    

Package version:
  evaluate_0.14   glue_1.2.0      graphics_3.5.1  grDevices_3.5.1 highr_0.7       knitr_1.23      magrittr_1.5   
  markdown_0.8    methods_3.5.1   mime_0.5        stats_3.5.1     stringi_1.2.3   stringr_1.3.1   tools_3.5.1    
  utils_3.5.1     xfun_0.7        yaml_2.1.19

Take a look at https://stackoverflow.com/questions/63534247/recommended-way-to-initialize-js-renderer-in-asis-markdown-chunck/63557622?noredirect=1#comment112417672_63557622 — atusy, Oct 19 '20 at 12:27
@atusy Do you mean the part `knitr::knit_print() %>% attr('knit_meta') %>% knitr::knit_meta_add() %>% invisible()` which I could use with `leaflet() %>% ...`? — Christoph, Oct 19 '20 at 12:35
Yes, it is required to add JS dependencies. There are some alternative solutions. Maybe easy way is the one I just posted as an answer. — atusy, Oct 19 '20 at 13:24

score 1 · Answer 1 · edited Oct 20 '20 at 06:55

Try something like

---
title: "The best I got"
author: "SQC"
date: "11 Juli 2019"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = FALSE, fig.width=8.4, fig.height=7)
library(leaflet)
ranges <- paste0("section_", c(1:3))
leaflet()
DT::datatable(data.frame())
```

# Title level 1

This is an R Markdown document. 

## Title level 2 (tab) {.tabset}

### Title level 3 (tab) {.tabset}

```{r, echo = FALSE, results='asis'}
for (i in 1:length(ranges)) {
  cat(paste0("#### Title level 4 `r ", 
             ranges[i], "` vs. `r ", ranges[i - 1], "` {.tabset}\n"))
  cat("##### Map\n")

  # code //
  cat("```{r, fig.width=8.4, fig.height=7}\n")
  leaflet() %>%
    addTiles() %>%
    knitr::knit_print() %>%
    cat()
  cat("```\n")
  # //

  cat("##### Tabelle\n")

  # code //
  cat("```{r}\n")
  DT::datatable(data.frame(x = 1, y = ranges[i], stringsAsFactors = FALSE),
                rownames = FALSE) %>%
    knitr::knit_print() %>%
    cat()
  cat("```\n")
  # //
}
```

Sorry, if I run your example, only the first title is displayed correctly. The for-loop with titles and sections does not work. I tried some variants of `knitr::knit_print(), cat(), attr('knit_meta'), knitr::knit_meta_add(), invisible()` but without success. Any ideas? I can add a screenshot if you like... — Christoph, Oct 20 '20 at 06:58

score 0 · Answer 2 · answered May 04 '21 at 03:24

0

I was trying to do this with plotly, but above codes did not work for me.

However, I look up the R-markdown cookbook and found the knit_child function work as expected.

```{r, results='asis'}
res <- knitr::knit_child(text = c(
  '```{r}',
  'plotly::plot_ly(x = rnorm(100))',
  '```'
), envir = environment(), quiet = TRUE)
cat(unlist(res), sep = '\n')
```

answered May 04 '21 at 03:24

PWbio

41
2
5

I don't understand how this helps to automate nested sections. Are you able to make my example work? – Christoph Jul 20 '21 at 13:23

score 0 · Accepted Answer · answered Mar 03 '23 at 09:35

0

Somehow this questions was answered here

The trick is to use tagList(html)

answered Mar 03 '23 at 09:35

Christoph

6,841
4
37
89

How to automate nested sections in Rmds which include text, maps and tables

3 Answers3

Linked