assign/change character/factor level using characters

Question

I have a dataframe and select one row to obtain one row but maintain the levels. I need this row to use predict() with a model later one.

I try to change one value of the row by assigning a characters (?) to a factor level like this:

df <- data.frame(some_amazing_column = c("yes", "no"))
levels(df$some_amazing_column)

row <- head(df, 1)
char <- "no"

head(row)
levels(row$some_amazing_column)
row$some_amazing_column = char
levels(row$some_amazing_column)

This loses the factor levels. Am I changing the value (factor level) correctly? What is the right way to change, in this case, yes to no whilst maintaining the levels. please not that the assigned value is a character coming from a front-end.

PS:

The accepted answer's hint results in this working code:

df <- data.frame(some_amazing_column = c("yes", "no"))
levels(df$some_amazing_column)

row <- head(df, 1)
char <- "no"

head(row)
levels(row$some_amazing_column)
row$some_amazing_column <- factor(char, levels = levels(row$some_amazing_column))
head(row)
levels(row$some_amazing_column)

Answer 1

To change the level, one way is to convert the 'char' to factor with same levels as in the original dataset and assign

row$some_amazing_column <- factor(char, levels = levels(row$some_amazing_column))

levels(row$some_amazing_column)
#[1] "no"  "yes"

row
#  some_amazing_column
#1                  no