Extract value from tibble in R markdown inline code

Question

I'm looking to access a value from a tibble using inline code in an R Markdown document.

The following code gives me a 1 x 2 tibble.

#library(tidyverse)
#library(knitr)

# Minimal working example
df <- tibble(
  Event = c("Swimming","Swimming","Swimming","Camping","Hiking","Hiking")
)
df

# Creates 1 row x 2 column
df %>% count(Event) %>% slice(which.max(n))

If I create an object in a code chunk, I can select [1,1] and report that inline:

temp <- df %>% count(Event) %>% slice(which.max(n))

# The most popular event was `r temp[1,1]`.

# which becomes "The most popular event was Swimming."

Is there a way to do that within my dplyr pipes without creating a new object? Ideally, I would be able to do it all via inline code, something like:

`r df %>% count(Event) %>% slice(which.max(n)) %>% df[1,1]`  # fails

Stepping back, I'm looking for a generalizable technique to use to select particular rows and cells from within a series of dplyr commands without creating new objects along the way.

I think this question can help you out: stackoverflow.com/questions/37345268/inline-dplyr-in-r-markdown — mmyoung77
– mmyoung77, Commented Jun 22, 2018 at 20:53
Yes, very helpful. Let's see if I can make it apply to this case... — Daniel
– Daniel, Commented Jun 22, 2018 at 21:15
you might try r df %>% count(Event) %>% slice(which.max(n)) %>% {.[1,1]} . Well, if you include this in a sentence, you might want to add %>% tolower — lebatsnok
– lebatsnok, Commented Jun 22, 2018 at 21:17

Calum You · Accepted Answer · 2018-06-22 21:05:13Z

3

Your line of code that you want to display inline doesn't run even when not in R Markdown, try it in the console. When using special operators like [, [[, $ with the pipe, you need to specify them in the prefix rather than infix form like other functions, like this. Note the comparison of your attempted code, and alternatively using [[ to drill down and select just the text instead of returning a tibble.

library(tidyverse)
df <- tibble(
  Event = c("Swimming","Swimming","Swimming","Camping","Hiking","Hiking")
)

df %>% count(Event) %>% slice(which.max(n)) %>% `[`(1,1)
#> # A tibble: 1 x 1
#>   Event   
#>   <chr>   
#> 1 Swimming
df %>% count(Event) %>% slice(which.max(n)) %>% `[[`("Event")
#> [1] "Swimming"

However, this doesn't quite solve your problem because R Markdown uses backticks to denote the inline code, and it's complicated to change the parsing rules that knitr uses to find and evaluate inline code, so here it would think the code ended at `[`. So you can instead use the magrittr aliases for these functions, extract and extract2 in inline code:

Most popular event was `r df %>% count(Event) %>% slice(which.max(n)) %>% magrittr::extract2("Event")`

which correctly prints Most popular event was Swimming in the rendered markdown.

edited Jun 22, 2018 at 21:05

answered Jun 22, 2018 at 20:59

Calum You

15.2k4 gold badges27 silver badges46 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

Daniel Over a year ago

This is really helpful. Do you know of a more generalizable solution that would allow me to select, say, [2,1] or [2,2] from a tibble using inline code? Your elegant solution works well when the result I want is in [1,1]. I'm trying to think about other use cases...

Calum You Over a year ago

Just change the indices in extract2! If you want a different column, use that number/name. Here because you did which.max it's a one row table, so selecting a column is also selecting a single element. If there were more rows, add another [ to the end of the pipe to get the element you want.

Daniel · Accepted Answer · 2018-06-22 21:29:42Z

2

I'm learning several ways to do this - thanks all!

1. Following @labatnok:

r df %>% count(Event) %>% slice(which.max(n)) %>% {.[1,1]}

2. Following @mmyoung77:

`r df %>% count(Event) %>% arrange(desc(n)) %>% slice(1) %$% Event`

# result
[1] "Swimming"

This approach avoids using "which.max." The "1" in slice(1) indicates the row, then %$% Event indicates the column (by name).

It's not quite as easy to follow what is happening as base R's [1,1], but it gets the job done.

3. Following @Mo Alfaifi:

(df %>% count(Event) %>% arrange(desc(n)))[1,1]

This approach makes it easier to tell which row/column you are selecting. It also avoids needing to name the column.

4. Following @Colum You:

df %>% count(Event) %>% slice(which.max(n)) %>% magrittr::extract2("Event")

edited Jun 22, 2018 at 21:29

answered Jun 22, 2018 at 21:21

Daniel

4551 gold badge6 silver badges16 bronze badges

Comments

Moh · Accepted Answer · 2018-06-22 21:16:30Z

1

Is this what you are after?

df %>% count(Event)

# A tibble: 3 x 2
#  Event        n
# 1 Camping      1
# 2 Hiking       2
# 3 Swimming     3

# To get the first row
(df %>% count(Event))[1,]

# Event       n
# 1 Camping   1

answered Jun 22, 2018 at 21:16

Moh

1988 bronze badges

1 Comment

Daniel Over a year ago

Close to this. This was a good idea - I elaborated on your answer in mine.

Collectives™ on Stack Overflow

Extract value from tibble in R markdown inline code

3 Answers 3

2 Comments

Comments

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

2 Comments

Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related