Search code examples
rdataframer-highcharter

Error "Columns `x`, `y` must be 1d atomic vectors or lists" in hchart

I'm trying to draw a barplot in R using the highcharter library

My dataframe mostly_used looks something like this:

        word    n
1        sir 8484
2       time 7339
3       miss 5954
4       dear 5422
5       hand 5305
6       head 4978
7      night 4240
8        day 4124
9       eyes 4040
10     house 4011

and I use the following line of code:

hchart(mostly_used, x = word, y = n, type = "column", name = "word count"
       , color = "blue")  %>% hc_add_theme(hc_theme_null())

I get the error Error: Columns`x`, `y` must be 1d atomic vectors or lists

can anyone explain why this is?

edit:

> dput(mostly_used)
structure(list(word = c("sir", "time", "miss", "dear", "hand", 
"head", "night", "day", "eyes", "house"), n = c(8484L, 7339L, 
5954L, 5422L, 5305L, 4978L, 4240L, 4124L, 4040L, 4011L)), .Names = c("word", 
"n"), row.names = c(NA, -10L), class = c("tbl_df", "tbl", "data.frame"
))
Solution

  • I looked at the documentation and see the examples didn't use the syntax you tried. It appeared more functional. (My efforts preceded your edit that included the data example, so my mostly_used was just an ordinary dataframe, hence my use of as.character to coerce what I thought would be a factor. That turned out to be unnecessary but harmless.)

    I took the line-type example from the vignette on my machine: http://localhost:13297/library/highcharter/doc/replicating-highcharts-demos.html and replaced the corresponding values:

    highchart() %>% 
  hc_chart(type = "column") %>% 
  hc_title(text = "Counts of Mostly Used") %>% 
    hc_xAxis(categories = as.character(mostly_used$word)) %>% 
  hc_yAxis(title = list(text = "N")) %>% 
  hc_plotOptions(line = list(
    dataLabels = list(enabled = TRUE),
    enableMouseTracking = FALSE)
    ) %>% 
  hc_series(
    list(  name = "Used",
      data = mostly_used$n
    )
  )

    This is a screenshot of what appeared in my Chrome session:

    enter image description here

    Your comment with the citation to a different bit of documentation showed that there was an hcaes-function wrapped around the x and y argument assignments. This worked for me:

    hchart(mostly_used ,type = "column", title="Counts of Mostly Used",
                     hcaes( x = word, y=   n)    )

    The "why" is the need to respect that package's handling of non-standard evaluation. It emulates the ggplot2 package's strategy of using an "aes"-function to define the column names using real R names, i.e. unquoted tokens that are evaluated in the context of the data argument.