Reading a delimited file – the format that declares nothing, and therefore the one with the most to say.
One constructor per format: each takes exactly the settings its format has, so an argument from another format is rejected by R as unused, and each format's own enums and defaults can be documented against the class they belong to.
Usage
setup_DelimitedIngest(
columns = NULL,
character2factor = TRUE,
clean_colnames = TRUE,
remove_duplicates = FALSE,
reader = "data.table",
sep = NULL,
quote = "\"",
na_strings = ""
)Arguments
- columns
Optional Named vector: Declared column types, keyed by column name, in the
profile/r/v1vocabulary –c(Age = "integer"). A column named here is converted to that type after reading, so a known type takes precedence over the reader's inference.- character2factor
Logical: If TRUE, read character columns as factors.
- clean_colnames
Logical: If TRUE, normalize column names.
- remove_duplicates
Logical: If TRUE, drop rows repeating an earlier row.
- reader
Character {"data.table", "vroom", "duckdb", "arrow"}: Library used to read the file.
- sep
Optional Character: Field separator. If
NULL, the reader detects it.- quote
Character: Quote character.
- na_strings
Character vector: Strings to read as missing.
Examples
setup_DelimitedIngest(sep = ";")#> <DelimitedIngestConfig>
#>
#> format: <chr> delimited
#> columns: <NUL> NULL
#> character2factor: <lgc> TRUE
#> clean_colnames: <lgc> TRUE
#> remove_duplicates: <lgc> FALSE
#> reader: <chr> data.table
#> sep: <chr> ;
#> quote: <chr> "
#> na_strings: <chr>
#>