Pipeline throws an error when trying to evaluate ":=" using dplyr

[husamrahman]

The package throws an error if the function it is referencing contains ":=" in one of its operations.

Simply example below where df is any data frame. Running the code below in a normal R console works, however when wrapped in a function and used in the pipeline it throws an error.

Example Cod:
df <- df %>%
mutate(!!sym("this_is_a_test") := 1)

Error:

ERROR [2019-02-27 15:46:58] || EXCEPTION OCCURED WHILE RUNNING THE PIPELINE FUNCTION WITH PROVIDED PARAMETERS: The LHS of := must be a string or a symbol ||
ERROR [2019-02-27 15:46:58] Error in genericPipelineException(error = structure(list(message = "The LHS of := must be a string or a symbol", : || EXCEPTION OCCURED WHILE RUNNING THE PIPELINE FUNCTION WITH PROVIDED PARAMETERS: The LHS of := must be a string or a symbol ||

[naren1991]

There does seem to be an issue using Non-Standard Evaluation, though it seems to relate to the !! operator.

However, what you are trying to accomplish can be done like so:

df <- df %>% mutate(this_is_a_test := 1)

This works within a pipeline.

The problem occurs when a dynamic column name is required which would need you to use the !! operator.

[husamrahman]

Thanks for the response. The function I was using lets the user define their own variable name, hence the !! operator. I can make a work around but thanks for looking into it!