How to parse a text table in rmarkdown code chunks

Question

There is an rmarkdown file with a markdown table that will be regularly updated. The content should be parsed in a code chunk so that e.g. ggplot could be used. I don't want to maintain the table in a code chunk or a separate file.

How can I read the table from the code chunk?

You can find as a starter rmarkdown code with a markdown table below.

---
title: "Parse tables"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(message = FALSE, warning = FALSE)
```

# Step 1: Create markdown table as text

That table will be manually updated directly in the markdown file.

Table: Project Timeline

| date       | description |
|------------|-------------|
| 2020-05-11 | Milestone 1 |
| 2020-07-11 | Milestone 2 |
| 2020-07-20 | Milestone 3 |


# Step 2: Parse the table above

The table should be maintained as a markdown table. That seems to be more easy than working directly with
`tibble` or `tribble`. How can I read the table from the code chunk? 

```{r}
library(tidyverse)
df <- tibble(date = c("2020-05-11", "2020-07-11", "2020-07-20"), 
             description = c("Milestone 1", "Milestone 2", "Milestone 3"))
df
```

I don't understand your question. Do you mean you want to output `df` from Step 2 so it looks like the table in Step 1? Or read the table from Step 1, which is not in a code chunk, so it can become a tibble? — LC-datascientist, Jul 11 '20 at 21:29
Read the table from step 1, which is not a code chunk, so it can become a `tibble`. — Birger, Jul 12 '20 at 06:39

Stéphane Laurent · Accepted Answer · 2020-07-12T05:43:35.643

In a code chunk, apply readLines to your Rmd file to get the lines of this file in a vector:

allLines <- readLines("yourFile.Rmd")

Select the lines which start and end with |, and remove the second one (which is the separator line "|-----|-----|"):

tableLines <- allLines[grep("^\\|.*\\|$", allLines)][-2]

Then with the code below, you get the table as a matrix, whose first line contains the column names:

tableAsMatrix <- t(sapply(strsplit(tableLines, "\\|"), function(pieces){
  stringr::str_trim(pieces[-1])
}))

Finally convert this matrix deprived of its first line to a dataframe, and use its first line to set the column names:

setNames(as.data.frame(tableAsMatrix[-1,,drop = FALSE]), tableAsMatrix[1,])

Full code

---
title: "Parse tables"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(message = FALSE, warning = FALSE)
```

# Step 1: Create markdown table as text

That table will be manually updated directly in the markdown file.

Table: Project Timeline

| date       | description |
|------------|-------------|
| 2020-05-11 | Milestone 1 |
| 2020-07-11 | Milestone 2 |
| 2020-07-20 | Milestone 3 |


# Step 2: Parse the table above

The table should be maintained as a markdown table. How can I read the table from the code chunk? 

```{r}
allLines <- readLines("ParseTable.Rmd")

tableLines <- allLines[grep("^\\|.*\\|$", allLines)][-2]

tableAsMatrix <- t(sapply(strsplit(tableLines, "\\|"), function(pieces){
  stringr::str_trim(pieces[-1])
}))

df <- setNames(as.data.frame(tableAsMatrix[-1,,drop = FALSE]), tableAsMatrix[1,])

df
```

How to parse a text table in rmarkdown code chunks

1 Answers1

Full code