v0.0.3

release notes

• Fix rlang warning. (#1, thanks @mvuorre)

• poly_add_columns() gains an na_values argument for how to handle missing values. Default is "error" to raise an error. Other options include "warn" to ignore the missing values but raise a warning and "allow" to silently accept the missing values. (#2)

demo of na_values

By default, poly_add_columns() will "error" when it encounters na_values.

data <- mtcars
data$wt[c(1, 2, 6, 9)] <- NA_real_

polypoly::poly_add_columns(data, wt, 3)
#> Error in `polypoly::poly_add_columns()`:
#> ! NA values found in `wt`.
#> ℹ `stats::poly()` cannot handle missing values.

#> Backtrace:
#>     ▆
#>  1. └─polypoly::poly_add_columns(data, wt, 3)
#>  2.   └─rlang::abort(...)

polypoly::poly_add_columns(data, wt, 3, na_values = "error")
#> Error in `polypoly::poly_add_columns()`:
#> ! NA values found in `wt`.
#> ℹ `stats::poly()` cannot handle missing values.

#> Backtrace:
#>     ▆
#>  1. └─polypoly::poly_add_columns(data, wt, 3, na_values = "error")
#>  2.   └─rlang::abort(...)

poly_add_columns() can "warn": We get a warning that missing values are present, but polynomials for the non-missing values are added.

data2 <- polypoly::poly_add_columns(data, wt, 3, na_values = "warn")
#> Warning: NA values found in `wt`.
#> ℹ `stats::poly()` cannot handle missing values, so those values will be ignored.
#> ℹ Derived columns like `wt1` will contain missing values.

# the three `wt1`, `wt2`, `wt3` have been added
colnames(data2)
#>  [1] "mpg"  "cyl"  "disp" "hp"   "drat" "wt"   "qsec" "vs"   "am"   "gear"
#> [11] "carb" "wt1"  "wt2"  "wt3"

# and they have NAs
round(data2$wt1, 3)
#>  [1]     NA     NA -0.167  0.000  0.042     NA  0.066 -0.005     NA  0.042
#> [11]  0.042  0.159  0.096  0.105  0.378  0.411  0.396 -0.189 -0.298 -0.257
#> [21] -0.140  0.057  0.041  0.116  0.117 -0.238 -0.200 -0.317 -0.009 -0.083
#> [31]  0.066 -0.081

Finally, poly_add_columns() can silently "allow" missing values.

data3 <- polypoly::poly_add_columns(data, wt, 3, na_values = "allow")
round(data3$wt1, 3)
#>  [1]     NA     NA -0.167  0.000  0.042     NA  0.066 -0.005     NA  0.042
#> [11]  0.042  0.159  0.096  0.105  0.378  0.411  0.396 -0.189 -0.298 -0.257
#> [21] -0.140  0.057  0.041  0.116  0.117 -0.238 -0.200 -0.317 -0.009 -0.083
#> [31]  0.066 -0.081

^{Created on 2022-10-20 with reprex v2.0.2}

First CRAN release

• Same features as before: Tidying, plotting, rescaling, column-binding of polynomial matrices.
• Added poly_plot_data() which returns the dataframe that would be plotted by poly_plot().

First version

First initial version.

Main features:

• plotting
• tidying
• column-binding
• rescaling