How to point function arguments directly to your read_excel file columns

[hhpe0709]

raw <- read_excel('discharge_data_1.xlsx')
df <- raw[,c(colnames(raw[1]),colnames(raw[2]))]
df %>%  ptd_spc(value_field = "Actual pre 12pm discharges", date_field = "Date", improvement_direction = "decrease")`

I’m trying to understand why I can’t run using the below instead:
df %>% ptd_spc(value_field = colnames(df[2]) , date_field = colnames(df[1]), improvement_direction = "decrease")

I get the error: value_field: 'colnames(df[2])' must be a valid column name in the data frame. Is there a way that I can create a variable for these fields which will be pulled directly from read_excel if the data changes (assume the columns of interest will always be the first two columns in the excel file)

[AlexisW]

Do you mean colnames(df)[2]?

[hhpe0709]

I’m trying to pass the second column name of my excel file into the first argument value_field and the first column name of my excel file into the second argument date_field for the function ptd_spc(). Both colnames(df)[1] or colnames(df[1]) results in the error:
Error in check("value_field") :
value_field: 'colnames(df)[2]' must be a valid column name in the data frame.
e.g.
df %>% ptd_spc(value_field = colnames(df[2]) , date_field = colnames(df[1]), improvement_direction = "decrease")

[AlexisW]

Oh indeed that wasn't the problem, sorry. To understand, you need to check the help of the function, I guess it's this one. The help for value_field and date_field says:

Field name can be specified using non-standard evaluation (i.e. no quotation marks).

So, that's a bit complicated, you can find details here, but the idea is you can use the special adverb .data as if it was the data frame itself:

library(tidyverse)
library(NHSRplotthedots)

df <- data.frame(`Actual pre 12pm discharges` = rnorm(4),
                  Date = as.Date(c("1jan1960", "2jan1960", "31mar1960", "30jul1960"), "%d%b%Y"),
                 check.names = FALSE)

x1 <- df %>%
  ptd_spc(value_field = "Actual pre 12pm discharges",
          date_field = "Date",
          improvement_direction = "decrease")
#> Warning in ptd_add_short_group_warnings(.): Some groups have 'n < 12'
#> observations. These have trial limits, which will be revised with each
#> additional observation until 'n = fix_after_n_points' has been reached.

x2 <- df %>%
  ptd_spc(value_field = .data[[colnames(df[1])]],
          date_field = .data[[colnames(df[2])]],
          improvement_direction = "decrease")
#> Warning in ptd_add_short_group_warnings(.): Some groups have 'n < 12'
#> observations. These have trial limits, which will be revised with each
#> additional observation until 'n = fix_after_n_points' has been reached.

waldo::compare(x1, x2)
#> ✔ No differences

^{Created on 2022-12-19 by the reprex package (v2.0.1)}

[hhpe0709]

Great thanks AlexisW