I'm trying to build an automated report that will have three charts right underneath each other without margin space between each other.
I've mocked up my problem with the following Rmd script
---
output: pdf_document
---
```{r setup, include=FALSE}
library(gridExtra)
```
```{r, echo=FALSE}
car_tbl <- tableGrob(mtcars[1:10,])
grid.arrange(car_tbl, car_tbl, car_tbl)
```
You can see how the tables overlap each other. There seems like there are actually a few issues comprising my problem.
How do I use the options of tableGrob and grid.arrange to keep the tables from overlapping.
How do I make sure nothing is cut off? In other words, how do I set the graphic to take the whole page if I need it too?
How can I re-actively shrink the text of the plot to fit on one page?
How can I set the size of the page to whatever size I want? Are there options set the knitr document to print to a pdf page of any size I want? Perhaps poster size if I need it to?
There are different angles to approach your question. Here's a working example with three tables fitting in one chunk placed on a pdf of size A2.
---
output: pdf_document
geometry: paper=a2paper
---
```{r setup, include=FALSE}
library(gridExtra)
```
```{r, echo=FALSE, fig.height=10}
car_tbl <- tableGrob(mtcars[1:10,])
grid.arrange(car_tbl, car_tbl, car_tbl)
```
To address your questions specifically,
1. How do I use the options of tableGrob and grid.arrange to keep the tables from overlapping.
Their overlap had more to do with the default fig.height set by knitr.
2. How do I make sure nothing is cut off? In other words, how do I set the graphic to take the whole page if I need it too?
Make sure the figure height and page are big enough for the table.
3. How can I re-actively shrink the text of the plot to fit on one page?
It's not trivial, but I guess it could be done. I doubt it's a good idea though (for typographical reasons): typically the font size should be set, and if the table doesn't fit, it should be split into several pages (there's an example in a link I provided).
3. How can I set the size of the page to whatever size I want? Are there options set the knitr document to print to a pdf page of any size I want? Perhaps poster size if I need it to?
It's a latex question, which in the context of Rmd documents means that you could pass these layout options to latex via rmarkdown and pandoc, typically as options for the geometry package.
Related
I'm trying to create a summary table with the package vtable. I do get a table but somehow the font size is super large and not the whole table is shown? I'm really at a loss here. I have spent hours on it, but could not find a solution... Can please someone tell me how to get a smaller table (font size and the whole table in a pdf document?
---
title: "Example"
date: "22 March 2021"
output:
pdf_document
---
{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
library(vegan)
library(vtable)
{r dune}
data(dune.env)
{r plot1, echo = F,results='asis',out.width='10%',message=F, fig.show='hold'}
st(dune.env,summ=c('mean(x)','median(x)','min(x)','max(x)'),group="Management",group.long = TRUE,out="latex")
By default, sumtable with out = 'latex' applies a LaTeX \resizebox{\textwidth} to the table to make it fit horizontally on the page. When the table is very narrow and tall, though, like this one, the resizebox can scale it to be bigger (making the text big).
You can control the width of the table with the fit.page option, which scales the horizontal width of the table. By default it's '\\textwidth', but you could do, for example, .6\\textwidth to only fill 60% of the horizontal space. This is what I get with fit.page='.5\\textwidth':
Now, this makes the text tiny, but it does fit. This is a consequence of LaTeX giving tall tables a hard time, and having no built-in support for multi-page tables. So it just fills up a page and the rest isn't shown. This is a LaTeX thing, not sumtable. There isn't currently support for multi-page LaTeX tables in sumtable/vtable. However, you could add keep_tex: true to your YAML, then after the fact go in and edit the TeX file to use the longtable LaTeX package, which adds support for multi-page tables, and turn the \begin{table} into \begin{longtable}, etc..
I'm having a problem with assigning LaTeX environments within an RMarkdown for-loop code-chunk.
In short, I've written an R Markdown document and a series of R-scripts to automatically generate PDF reports at the end of a long data analysis pipeline. The main section of the report can have a variable number of sections that I'm generating using a for-loop, with each section containing a \subsection heading, a datatable and plot generated by ggplot. Some of these sections will be very long (spanning several pages) and some will be very short (~1/4 of a page).
At the moment I'm just inserting a \pagebreak at the end of each for-loop iteration, but that leaves a lot of wasted space with the shorter sections, so I'm trying to "group" each section (i.e. the heading, table and chart) so that there can be several per page, but they will break to a new page if the whole section won't fit.
I've tried using a figure or minipage environment, but for some reason those commands are printed as literal text when the plot is included; these work as expected with the heading and data table, but aren't returned properly in the presence of the image.
I've also tried to create a LaTeX samepage environment around the whole subsection (although not sure this will behave correctly with multi-page sections?) and then it appears that the Markdown generated for the plot is not interpreted correctly somewhere along the way (Pandoc?) when it's within that environment and throws an error when compiling the TeX due to the raw Markdown ![]... image tag.
Finally, I've also tried implementing \pagebreak[x] and \nopagebreak[y] hints at various points in the subsection but can't seem get these to be produce the desired page breaking behaviour.
I've generated an MWE that reproduces my issues below.
I'd be really grateful for any suggestions on how to get around this, or better ways of approaching "grouping" of elements that are generated in a dynamic fashion like this?
---
title: "Untitled"
author: "I don't know what I'm doing"
date: "26/07/2020"
output:
pdf_document:
latex_engine: xelatex
---
```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = FALSE, dev = "cairo_pdf")
```
```{r cars, results='asis'}
for (i in 1:5){
cat("\\begin{figure}")
cat(paste0("\\subsection{This is subsection ",i,"}"))
cat("\\Huge Here's some bulk text that would represent a data table... kasvfkwsvg fiauwe grfiwgiu iudaldbau iausbd ouasbou asdbva asdbaisd i iuahihai hiuh iaiuhqijdblab ihlibljkb liuglugu h uhi uhi uhqw iuh qoijhoijoijoi qwegru wqe grouw egq\\newline")
plot(mtcars$wt,mtcars[,i])
cat("\\end{figure}")
}
```
Edit to add: interestingly these figure and minipage environments seems to work as expected when executing the same example in an .Rnw using knitr... so does that narrow it down to an issue with Pandoc? Again, any help much appreciated!
What happens is that the raw TeX commands are not treated as TeX when going through Markdown. You can fix that by explicitly marking the relevant snippets as LaTeX:
for (i in 1:5){
cat("`\\begin{figure}`{=latex}")
cat(paste0("\\subsection{This is subsection ",i,"}"))
cat("\\Huge Here's some bulk text that would represent a data table... kasvfkwsvg fiauwe grfiwgiu iudaldbau iausbd ouasbou asdbva asdbaisd i iuahihai hiuh iaiuhqijdblab ihlibljkb liuglugu h uhi uhi uhqw iuh qoijhoijoijoi qwegru wqe grouw egq\\newline")
plot(mtcars$wt,mtcars[,i])
cat("`\\end{figure}`{=latex}")
}
See the generic raw attribute section in the pandoc manual for details.
I am using RMarkdown to create a html file (my preferred format). I then use the officer package to create a powerpoint (PPT) (everyone else's preferred format) that reads in the .png images that are automatically created and saved when I knit the document (I believe this is default when the fig.path is specified).
To get consistent fontsize throughout the PPT figures, I have specified in each knitr chunk fig.width, out.width, etc to be equal to the relevant PPT placeholder dimensions. E.G if the PPT placeholder is 5.29in high x 5.89in wide, then in the knitr chunk I specify , out.height="5.29in", out.width="5.89in", fig.height=5.29, fig.width=5.89). This seems to work in terms of the PPT file, however it results in extremely small figures for the HTML file. Is there a way to get the knitr code to work for both html and PPT, without needing to specifically save the image, using, for example ggsave()?
Following is some auto-generated test code:
---
title: "Test_Figure_Size"
output:
html_document
---
```{r setup, include=FALSE}
knitr::opts_chunk$set(fig.path = "Delete_Me/", echo=FALSE)
```
When you include out.height, out.width the figure is extremely small in the knitted html document, but perfectly sized as a .png file.
```{r pressure, echo=FALSE, out.height="5.29in", out.width="5.89in", fig.height=5.29, fig.width=5.89}
plot(pressure)
```
When you say out.height="5.29in", out.width="5.89in", those values are written into the HTML output as height and width attributes of the img. But height and width are expressed in pixels, so you'll end up with a figure about 5 pixels square.
I don't use Powerpoint, but is there a way to specify the image size in pixels, e.g. "589px" or something similar? Then it will use the same scale as the browser, and things should be consistent.
I am trying to insert a pdf image into an r markdown file. I know it is possible to insert jpg or png images. I was just wondering if it is also possible to insert a pdf image. Thanks very much!
If you are just trying to insert an image that has been exported from, for example, some R analysis into a pdf image, you can also use the standard image options from the knitr engine.
With something like:
```{r, out.width="0.3\\linewidth", include=TRUE, fig.align="center", fig.cap=c("your caption"), echo=FALSE}
knitr::include_graphics("./images/imagename.pdf")
```
Unfortunately you can't specify the initial dimensions of your image output (fig.width and fig.height), which you would need to pre-define in your initial output, but you can specify the ultimate size of the image in your document (out.width). As noted below, however, this is limited to scaling down.
You could also of course leave out the initial directory specification if your files are in the same working directory. Just be aware of operating system differences in specifying the path to the image.
An alternative method is to use Markdown syntax noted by #hermestrismegistus on this post:
{width=65%}
This can also be collected for multiple images side-by side:
{width=33%}{width=33%}{width=33%}
Edit:
After working more extensively with in-text referencing, I have found that using r chunks and the include_graphics option to be most useful. Also because of the flexibility in terms of image alignment (justification).
As an example:
```{r image-ref-for-in-text, echo = FALSE, message=FALSE, fig.align='center', fig.cap='Some cool caption', out.width='0.75\\linewidth', fig.pos='H'}
knitr::include_graphics("./folder/folder/plot_file_name.pdf")
```
The reference can later be used in-text, for example, Figure \#ref(fig:image-ref-for-in-text) illustrates blah blah.
Some important things to note using this format:
You can only expand PDF images via a code chunk up to the out.width and out.height conditions set in the original .pdf file. So I would recommend setting them slightly on the larger side in your original image (just note that any chart text will scale accordingly).
The in-text reference code (in this case image-ref-for-in-text) CANNOT contain any underscores (_) but can contain dashes (-). You will know if you get this wrong by an error message stating ! Package caption Error: \caption outside float.
To stop your plots drifting to the wrong sections of your document, but in a way that unfortunately will generate some white space, the above example includes fig.pos='H'. Where H refers to "hold" position. The same can be achieved for the former Markdown option by placing a full-stop (period .) immediately after the last curly bracket.
Example:
{width=75%}.
Unfortunately, this latter option results in some unsightly full-stops. Another reason I prefer the include_graphics option.
Sorry, I found that there is a similar post before:
Add pdf file in Rmarkdown file
Basically, I can use something like below works well for the html output:
<img src="myFirstAlignment2.pdf" alt="some text" width="4200" height="4200">
And something like below works well for the pdf output:
(1)possible solution
\begin{center} <br>
\includegraphics[width=8in]{myFirstAlignment2.pdf} <br>
\end{center}
(2)possible solution
The myFirstAlignment2.pdf should be replaced with path\myFirstAlignment2.pdf if the pdf file is not in your working directory.
In relation to the comment of the best answer, there is a way to use the second option, and the output not come out tiny.
Use the following syntax below with the height being a large number. Having text in the brackets is necessary for it to work.
{width=100% height=400}
None of the answers outlined worked well for me in terms of sizing the pdf, so adding another answer using the code chunk options for out.height and out.width to control the size:
```{r out.height = "460px", out.width='800px', echo=F}
knitr::include_graphics("./images/imagename.pdf")
```
I'm producing a solutions manual for a book, using .Rmd files with the following YAML header:
---
title: "DDAR: Solutions and Hints for Exercises"
date: "`r Sys.Date()`"
output:
word_document:
reference_docx: solutions-setup.docx
---
where I control the general layout of the document with the reference_docx to get an output Word document.
There will be many figures, and I'd like to set some global graphics parameters to give relatively tight bounding boxes and reasonable font sizes
in the figures without having to tweak each one from what I see in a PDF document.
I tried the following, but the par() setting doesn't seem to have any effect:
{r setup, echo=FALSE}
options(digits=4)
par(mar=c(5,4,1,1)+.1)
Instead I get images like the following in my document with larger bounding boxes than I would like and with much larger font sizes than I would like.
I know how to control all this in .Rnw files produced with LaTeX, but I
can't find how to do it in .Rmd -> Word. Is there a chunk hook I could
use? I don't think that there is an out.width chunk option that re-scales
a figure as in LaTeX.
#scoa's answer shows how to use a hook to set some graphical parameters at the beginning of each chunk. This is necessary because "by default, knitr opens a new graphics device to record plots and close it after evaluating the code, so par() settings will be discarded", i.e. graphical parameters for later chunks cannot be set in an early setup-chunk but need to be set for each chunk separately.
If this behavior is not wanted, the package option global.par = TRUE can be used:
opts_knit$set(global.par = TRUE)
Finding the correct values for the margins is sometimes quite painful. In these cases, hook_pdfcrop can help. In all chunks where the option crop = TRUE, white margins will be removed. To apply this to all chunks, use
library(knitr)
knit_hooks$set(crop = hook_pdfcrop)
opts_chunk$set(crop = TRUE)
This works for docx output as well because "when the plot format is not PDF (e.g. PNG), the program convert in ImageMagick is used to trim the white margins" (from ?hook_pdfcrop).
Note that under some circumstances, cropping plots has the side effect of sometimes apparently different "zoom" factors of plots: This happens in cases where we start with identical sized elements on two plots but larger white margins around one of the plots. If then both are resized to a fixed output width after cropping, elements on the plot with larger margins look larger. However, this is not relevant for docx output because out.width/out.height cannot be used in that case.
The knitr documentation for hooks actually uses small margins as an example of what you can do with hooks. Here is a solution (adapted from this documentation).
---
output: word_document
---
```{r setup, echo=FALSE}
library(knitr)
knit_hooks$set(small.mar = function(before, options, envir) {
if (before) par(mar=c(5,4,1,1)+.1) # smaller margin on top and right
})
opts_chunk$set(small.mar=TRUE)
```
```{r}
plot(iris$Sepal.Length)
```
Using opts_chunk$set(small.mar=TRUE) is a way to avoid passing it to every chunk in the document.
The margin appears fixed (screenshot from the docx output in libreoffice with default reference-docx).