-
Notifications
You must be signed in to change notification settings - Fork 15
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Merge pull request #508 from posit-dev/feature/parse-data
Add `ParseData` type
- Loading branch information
Showing
20 changed files
with
704 additions
and
171 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,178 @@ | ||
| use crate::vector::Vector; | ||
| use crate::List; | ||
| use crate::RObject; | ||
|
|
||
| /// Typed data frame | ||
| /// | ||
| /// Type guarantees: | ||
| /// - Storage type is list. | ||
| /// - Class is `"data.frame"`. | ||
| /// - All columns are vectors and the same size as the data frame. | ||
| #[derive(Debug)] | ||
| pub struct DataFrame { | ||
| pub list: List, | ||
| pub obj: RObject, | ||
|
|
||
| pub names: Vec<String>, | ||
| pub nrow: usize, | ||
| pub ncol: usize, | ||
| } | ||
|
|
||
| impl DataFrame { | ||
| pub fn new(sexp: libr::SEXP) -> harp::Result<Self> { | ||
| let list = List::new(sexp)?; | ||
| harp::assert_class(sexp, "data.frame")?; | ||
|
|
||
| let dim = unsafe { harp::df_dim(list.obj.sexp) }?; | ||
| let nrow = dim.num_rows as usize; | ||
| let ncol = list.obj.length() as usize; | ||
|
|
||
| let Some(names) = list.obj.names() else { | ||
| return Err(harp::anyhow!("Data frame must have names")); | ||
| }; | ||
| let Ok(names) = harp::assert_non_optional(names) else { | ||
| return Err(harp::anyhow!("Data frame can't have missing names")); | ||
| }; | ||
|
|
||
| // Validate columns | ||
| for obj in list.iter() { | ||
| let obj = RObject::view(obj); | ||
|
|
||
| if unsafe { libr::Rf_isVector(obj.sexp) == 0 } { | ||
| return Err(harp::anyhow!("Data frame column must be a vector")); | ||
| } | ||
|
|
||
| if obj.length() as usize != nrow { | ||
| return Err(harp::anyhow!( | ||
| "Data frame column must be the same size as the number of rows" | ||
| )); | ||
| } | ||
| } | ||
|
|
||
| // SAFETY: Protected by `list` | ||
| let obj = RObject::view(sexp); | ||
|
|
||
| Ok(Self { | ||
| list, | ||
| obj, | ||
| names, | ||
| nrow, | ||
| ncol, | ||
| }) | ||
| } | ||
|
|
||
| pub fn col(&self, name: &str) -> harp::Result<RObject> { | ||
| let Some(idx) = self.names.iter().position(|n| n == name) else { | ||
| return Err(harp::Error::MissingColumnError { name: name.into() }); | ||
| }; | ||
|
|
||
| self.list | ||
| // FIXME: `get()` should take `usize` | ||
| .get(idx as isize)? | ||
| .ok_or_else(|| harp::unreachable!("missing column")) | ||
| } | ||
| } | ||
|
|
||
| #[cfg(test)] | ||
| mod tests { | ||
| use crate::assert_match; | ||
| use crate::r_alloc_integer; | ||
| use crate::r_chr_poke; | ||
| use crate::r_list_poke; | ||
| use crate::test::r_test; | ||
| use crate::vector::Vector; | ||
| use crate::DataFrame; | ||
| use crate::List; | ||
| use crate::RObject; | ||
|
|
||
| #[test] | ||
| fn test_data_frame_structure() { | ||
| r_test(|| { | ||
| let df = harp::parse_eval_base("data.frame(x = 1:2, y = 3:4)").unwrap(); | ||
| let df = DataFrame::new(df.sexp).unwrap(); | ||
|
|
||
| assert_match!(df.list, List { .. }); | ||
| assert_match!(df.obj, RObject { .. }); | ||
|
|
||
| assert_eq!(df.names, vec![String::from("x"), String::from("y")]); | ||
| assert_eq!(df.nrow, 2); | ||
| assert_eq!(df.ncol, 2); | ||
| }) | ||
| } | ||
|
|
||
| #[test] | ||
| fn test_data_frame_no_names() { | ||
| r_test(|| { | ||
| let df = harp::parse_eval_base("data.frame(x = 1:2, y = 3:4)").unwrap(); | ||
| df.set_attr("names", RObject::null().sexp); | ||
| let df = DataFrame::new(df.sexp); | ||
|
|
||
| assert_match!(df, harp::Result::Err(err) => { | ||
| assert!(format!("{err}").contains("must have names")) | ||
| }); | ||
| }) | ||
| } | ||
|
|
||
| #[test] | ||
| fn test_data_frame_bad_names() { | ||
| r_test(|| { | ||
| let df = harp::parse_eval_base("data.frame(x = 1:2, y = 3:4)").unwrap(); | ||
| let nms = df.attr("names").unwrap(); | ||
| unsafe { | ||
| r_chr_poke(nms.sexp, 0, libr::R_NaString); | ||
| } | ||
| let df = DataFrame::new(df.sexp); | ||
|
|
||
| assert_match!(df, harp::Result::Err(err) => { | ||
| assert!(format!("{err}").contains("missing names")) | ||
| }); | ||
| }) | ||
| } | ||
|
|
||
| #[test] | ||
| fn test_data_frame_bad_column_type() { | ||
| r_test(|| { | ||
| let df = harp::parse_eval_base("data.frame(x = 1:2, y = 3:4)").unwrap(); | ||
| r_list_poke(df.sexp, 0, RObject::null().sexp); | ||
|
|
||
| let df = DataFrame::new(df.sexp); | ||
|
|
||
| assert_match!(df, harp::Result::Err(err) => { | ||
| assert!(format!("{err}").contains("must be a vector")) | ||
| }); | ||
| }) | ||
| } | ||
|
|
||
| #[test] | ||
| fn test_data_frame_bad_column_size() { | ||
| r_test(|| { | ||
| let df = harp::parse_eval_base("data.frame(x = 1:2, y = 3:4)").unwrap(); | ||
| let bad_col = r_alloc_integer(3); | ||
| r_list_poke(df.sexp, 0, bad_col); | ||
|
|
||
| let df = DataFrame::new(df.sexp); | ||
|
|
||
| assert_match!(df, harp::Result::Err(err) => { | ||
| assert!(format!("{err}").contains("must be the same size")) | ||
| }); | ||
| }) | ||
| } | ||
|
|
||
| #[test] | ||
| fn test_data_frame_col() { | ||
| r_test(|| { | ||
| let df = harp::parse_eval_base("data.frame(x = 1:2, y = 3:4)").unwrap(); | ||
| let df = DataFrame::new(df.sexp).unwrap(); | ||
|
|
||
| let col_y = df.col("y").unwrap(); | ||
| assert_eq!(col_y.sexp, df.list.get_value(1).unwrap().sexp); | ||
|
|
||
| assert_match!(df.col("foo"), harp::Result::Err(err) => { | ||
| assert_match!(err, harp::Error::MissingColumnError { ref name } => { | ||
| assert_eq!(name, "foo"); | ||
| }); | ||
| assert!(format!("{err}").contains("Can't find column `foo` in data frame")) | ||
| }); | ||
| }) | ||
| } | ||
| } |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.