rmarkdown does not render PDFs as batch - r-markdown

No solution offered yet and cross-posted to RStudio community:
https://community.rstudio.com/t/rmarkdown-does-not-render-pdfs-as-batch/136316
I have a rmarkdown file that renders well in RStudio with the Knitr button or with rmarkdown() but does not render when rendered as part of a batch with purrr::walk() and being passed a parameter. The batch issue only occurs when fig.show="hold", out.width="50%" is included in a section of the rmarkdown file.
library(purrr)
library(tinytex)
SurveySet <- c(20,19,68,79,30,18,42)
purrr::walk(
.x = SurveySet,
~ rmarkdown::render(
input = "Test.Rmd",
output_file = glue::glue("REPORTS/Report Fails {.x}.pdf"),
params = list(Unit = {.x})
)
)
purrr::walk(
.x = SurveySet,
~ rmarkdown::render(
input = "TestWorks.Rmd",
output_file = glue::glue("REPORTS/Report Works {.x}.pdf"),
params = list(Unit = {.x})
)
)
Test.Rmd as
---
title: "`r params$Unit`"
output: pdf_document
params:
Unit: 68
---
```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = FALSE)
```
## R Markdown
This Document renders with RStudio Knitr button but not with batch
```{r figures-side, fig.show="hold", out.width="50%"}
par(mar = c(4, 4, .1, .1))
plot(cars)
plot(mpg ~ hp, data = mtcars, pch = 19)
```
TestWorks.Rmd as
---
title: "`r params$Unit`"
output: pdf_document
params:
Unit: 68
---
```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = FALSE)
```
## R Markdown
Without fig.show="hold", out.width="50%" this document renders individually and as batch
```{r figures}
par(mar = c(4, 4, .1, .1))
plot(cars)
plot(mpg ~ hp, data = mtcars, pch = 19)
```

Related

Dynamically generating figures with captions in Word output from Rmarkdown document

I'm trying to generate a Word document with figures. I use {officedown} and {officer} packages. I need to do it dynamically, in a loop since I don't know how many figures there will be. This is my Rmarkdown code:
---
output:
officedown::rdocx_document:
plots:
caption:
style: Table Caption
---
```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = FALSE, message = FALSE, warning = FALSE)
```
```{r}
library(dplyr)
library(flextable)
library(officer)
```
```{r}
block_toc(seq_id = "fig")
```
```{r results='asis'}
ttp <- c(3, 7)
test_items <- c("item A", "item B", "item C")
fpars <- lapply(test_items, function(ti) {
fpar(
ftext("Application of "),
ftext(ti),
ftext(" Variable text - describe any test item-related effects"),
ftext(" (see "),
run_reference("fig:results1"),
ftext(")."),
fp_p = fp_par(padding.bottom = 12)
)
})
do.call(block_list, fpars)
```
```{r}
titles <- lapply(seq_len(length(ttp)), function(i) {
sprintf(
"My custom figure caption with %s, having %s side effects",
paste(test_items, collapse = ", "),
ttp[i]
)
})
```
```{r}
tmps <- lapply(seq_len(length(ttp)), function(i) {
tmp <- tempfile(fileext = ".png")
png(tmp, width = 6, height = 5, units = "in", res = 120)
plot(iris[sample(1:150, 30), i + 1:2])
dev.off()
return(tmp)
})
```
```{r}
fpars <- lapply(seq_len(length(ttp)), function(i) {
fpar(
run_autonum(
seq_id = "fig",
pre_label = "Figure ",
bkm = paste0("fig:results", i),
bkm_all = TRUE,
prop = fp_text(bold = TRUE, font.size = 12)
),
titles[[i]],
external_img(src = tmps[[i]], width = 6, height = 5)
)
})
do.call(block_list, fpars)
```
The problem is when I generate the table of figures in the rendered document. It looks like this:
An entry is kept together with the image itself, I don't know why.
I save temporary png files to be able to use them inside fpar function. Using plot function directly inside fpar causes bad effects. Maybe there's another/better way?
I found this construction useful, but unfortunately it puts captions under the figures by default. My goal is figure captions behave more like table captions, i.e. a caption is above a figure.
```{r fig.cap=unlist(titles)}
plot(iris[1:10, 1:2])
```
How can I generate the plots with captions dynamically with {officedown} or {officer} packages?
The only issue is about the last lapply, you need to put the image in a separate paragraph, so a for loop will be easier to stack 2 paragraphs into a list instead of 1 for each iteration.
---
output:
officedown::rdocx_document:
plots:
caption:
style: Table Caption
---
```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = FALSE, message = FALSE, warning = FALSE)
```
```{r}
library(dplyr)
library(flextable)
library(officer)
library(officedown)
```
```{r}
block_toc(seq_id = "fig")
```
```{r}
ttp <- c(3, 7)
test_items <- c("item A", "item B", "item C")
fpars <- lapply(test_items, function(ti) {
fpar(
ftext("Application of "),
ftext(ti),
ftext(" Variable text - describe any test item-related effects"),
ftext(" (see "),
run_reference("fig:results1"),
ftext(")."),
fp_p = fp_par(padding.bottom = 12)
)
})
do.call(block_list, fpars)
```
```{r}
titles <- lapply(seq_len(length(ttp)), function(i) {
sprintf(
"My custom figure caption with %s, having %s side effects",
paste(test_items, collapse = ", "),
ttp[i]
)
})
```
```{r}
tmps <- lapply(seq_len(length(ttp)), function(i) {
tmp <- tempfile(fileext = ".png")
png(tmp, width = 6, height = 5, units = "in", res = 120)
plot(iris[sample(1:150, 30), i + 1:2])
dev.off()
return(tmp)
})
```
```{r}
fpars <- list()
for (i in seq_along(ttp)) {
fpars[[length(fpars)+1]] <- fpar(
run_autonum(
seq_id = "fig",
pre_label = "Figure ",
bkm = paste0("fig:results", i),
bkm_all = TRUE,
prop = fp_text(bold = TRUE, font.size = 12)
),
titles[[i]]
)
fpars[[length(fpars)+1]] <- fpar(
external_img(src = tmps[[i]], width = 6, height = 5)
)
}
do.call(block_list, fpars)
```
This is an answer for "how to do it with officer" from scratch.
But the following is much more simple. There is an option fig.topcaption that do the same thing.
---
output:
officedown::rdocx_document:
plots:
caption:
style: Table Caption
---
```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = FALSE, message = FALSE, warning = FALSE)
```
```{r}
library(dplyr)
library(flextable)
library(officer)
library(officedown)
```
```{r}
block_toc(seq_id = "fig")
```
```{r fig.cap = "miaou", fig.topcaption=TRUE}
plot(cars)
```
```{r fig.cap = "ouaf", fig.topcaption=TRUE}
plot(cars)
```

How do you filter a data frame in a shiny document and display a datatable?

I'm trying to filter a data frame and then do some simple ggplots off of the data. I've tried to leverage the R studio example on Shiny documents along with the following SO post on the subject:
Reactively filtering/subsetting a data frame in shiny
Here is my code.
---
title: "Shiny Filter Test"
author: "Novice"
date: "12/13/2019"
output: html_document
runtime: shiny
---
```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```
```{r}
library(tidyverse)
library(shiny)
inputPanel(
selectInput("n_break", label = "Number of bins:",
choices = c(10, 20, 35, 50), selected = 10)
)
cdat <- reactive({
data <- tibble(x = c(10,20,35), y = c("a","b","c"))
data %>%
filter(x %in% input$n_break)
output$table <- DT::renderDT({
cdat()
}, options = list(scrollX = TRUE))
})
```
Can anyone point out where I'm going wrong? When I run the code I get my dropdown box, but that is all. No errors. Just no filtered datatable.
Thanks.
The closing brackets of your reactive are at the wrong place. They should close once you have filtered the data.
---
title: "Shiny Filter Test"
author: "Novice"
date: "12/13/2019"
output: html_document
runtime: shiny
---
```{r setup}
knitr::opts_chunk$set(
echo = FALSE
)
```
```{r}
library(tidyverse)
library(shiny)
inputPanel(
selectInput("n_break", label = "Number of bins:",
choices = c(10, 20, 35, 50), selected = 10)
)
cdat <- reactive({
data <- tibble(x = c(10,20,35), y = c("a","b","c"))
data %>% filter(x %in% input$n_break)
})
DT::renderDT({
cdat()
}, options = list(scrollX = is ))
```
A remark on the reactive: if you plan to extend this futher, such that the filtered data is used elsewhere, it makes sense to do the filtering in a reactive function. However, if this is not the case I would just do the filtering inside the renderDT:
---
title: "Shiny Filter Test"
author: "Novice"
date: "12/13/2019"
output: html_document
runtime: shiny
---
```{r setup}
knitr::opts_chunk$set(
echo = FALSE
)
```
```{r}
library(tidyverse)
library(shiny)
data <- tibble(x = c(10,20,35), y = c("a","b","c"))
inputPanel(
selectInput("n_break", label = "Number of bins:",
choices = c(10, 20, 35, 50), selected = 10)
)
DT::renderDT({
data %>% filter(x %in% input$n_break)
}, options = list(scrollX = TRUE))
```

rmarkdown flexdashboard images folder deletes itself

My images folder deletes itself after first knitting of the document.
I've tried self_contained = TRUE in the YAML header, doesn't work.
(I'm not sure if this makes a difference but Shiny is embedded in the dashboard)
Below is my code :
---
title : app demo
author : yeshipants
output :
flexdashboard::flex_dashboard:
orientation: rows
self_contained : TRUE
source_code: embed
runtime: shiny
---
```{r setup}
knitr::opts_chunk$set(cache = FALSE)
```
```{r loadPackages, cache = TRUE}
setwd("C:/Users/user/Desktop/Training/OCR")
library(magick)
```
Column {.sidebar data-width=350}
-------------------------------------
### Input & Parameters
```{r inputImages, cache = TRUE}
selectInput("imagesToChoose",
label = "Choose an image to process",
choices = c("Language example 1",
"Language example 2",
"Jounal example"),
selected = "Language example 1")
```
Row {.tabset}
-------------------------------------
### Original Image
```{r displayImage, cache = FALSE}
renderImage({
if (input$imagesToChoose == "Language example 1"){
list(src = "images/receipt.png", height = 240, width = 300)
}
else if(input$imagesToChoose == "Language example 2"){
list(src = "images/french.JPG", height = 240, width = 300)
}
else if(input$imagesToChoose == "Jounal example"){
list(src = "images/journal.jpg", height = 240, width = 300)
}
})
```
I think you have to do:
renderImage({
......
}, deleteFile = FALSE)

Create slide with render dynamically in Rmarkdown shiny presentation

I want to produce slide dynamically in my Rmarkdown presentation with shiny
Here below the code works.
But as soon as I uncomment the for loop it doesn't.
---
title: "dynamic_plots"
output: ioslides_presentation
runtime: shiny
---
```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = FALSE)
```
```{r, echo = FALSE, results = "asis"}
number_of_bins_selected <- c(20,30)
i <- 1 # If for loop commented
#for (i in 1:length(number_of_bins_selected)) {
cat("## New slide \n")
input1_id <- paste(i, "nbreaks", sep = "_")
input2_id <- paste(i, "bw_adjust", sep = "_")
inputPanel(
selectInput(input1_id, label = "Number of bins:",
choices = c(10, 20, 35, 50), selected = number_of_bins_selected[i]),
sliderInput(input2_id, label = "Bandwidth adjustment:",
min = 0.2, max = 2, value = 1, step = 0.2)
)
renderPlot({
hist(faithful$eruptions, probability = TRUE, breaks = as.numeric(input[[input1_id]]),
xlab = "Duration (minutes)", main = "Geyser eruption duration")
dens <- density(faithful$eruptions, adjust = input[[input2_id]])
lines(dens, col = "blue")
})
cat(" \n\n") # End the slide
#}
```
Isn't it possible to create slides on a for loop ?

Xtable grey rows overwritting vertical lines

---
title: "Title"
author: ''
date: ''
output:
pdf_document:
template: default.tex
geometry: top=0.5cm, bottom=0.5cm, left=0.5cm, right=0.5cm
header-includes: null
fontsize: 4pt
classoption: portrait
sansfont: Calibri Light
---
#Name1: `r "Name1"`
#Name2: `r "Name2"`
```{r, echo=FALSE, message=FALSE, warning=FALSE, results='asis'}
df <- mtcars
n = nrow(df)
hlines=c(-1,0,(n-1),n)
my_align = "c|c|c|ccccc|ccc|c|"
rws <- seq(1, (n-1), by = 2)
col <- rep("\\rowcolor[gray]{.90} ", length(rws))
xtable::print.xtable(xtable(df
, align = my_align)
, add.to.row = list(pos = as.list(rws), command = col)
, booktabs = F
, hline.after = hlines, type = "latex")
```
I am using an Rmarkdown to print a table which has a lot of formatting. When I add the add.to.rwo part to get grey and white alternate rows the vertical lines are removed in the grey rows.
How do I correct this? It is very difficult to create a reproducible example but hopefully the same problem will apply to any df (with the correct Latex packages behind it)
Thanks :)
Try comparing these two tables. The first is your table as you coded it, the second is done by pixiedust with the hhline option set to TRUE.
---
title: "Title"
author: ''
date: ''
output:
pdf_document:
geometry: top=0.5cm, bottom=0.5cm, left=0.5cm, right=0.5cm
header-includes:
- \usepackage{amssymb}
- \usepackage{arydshln}
- \usepackage{caption}
- \usepackage{graphicx}
- \usepackage{hhline}
- \usepackage{longtable}
- \usepackage{multirow}
- \usepackage[dvipsnames,table]{xcolor}
fontsize: 4pt
classoption: portrait
sansfont: Calibri Light
---
#Name1: `r "Name1"`
#Name2: `r "Name2"`
```{r, echo=FALSE, message=FALSE, warning=FALSE, results='asis'}
library(xtable)
df <- mtcars
n = nrow(df)
hlines=c(-1,0,(n-1),n)
my_align = "c|c|c|ccccc|ccc|c|"
rws <- seq(1, (n-1), by = 2)
col <- rep("\\rowcolor[gray]{.90} ", length(rws))
xtable::print.xtable(xtable(df
, align = my_align)
, add.to.row = list(pos = as.list(rws), command = col)
, booktabs = F
, hline.after = hlines, type = "latex")
```
```{r}
library(pixiedust)
dust(df,
hhline = TRUE,
keep_rownames = TRUE) %>%
medley_bw() %>%
sprinkle_colnames(.rownames = "") %>%
sprinkle(cols = c(".rownames", "mpg", "cyl", "qsec", "gear", "carb"),
border = "right") %>%
sprinkle(rows = nrow(mtcars),
border = "top") %>%
sprinkle(bg_pattern_by = "rows")
```