Merge pull request #175 from promised-ai/feature/added-not-in-date-error

feat!: Added `DataParseError::DataFrameMissingColumn` variant for missing column in DataFrame when it is present in the metadata.

CHANGELOG.md

@@ -9,6 +9,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Added
 - `DataParseError::CodebookAndDataRowsMismatch` variant for when the number of rows in the codebook and the number of rows in the data do not match.
+- `DataParseError::DataFrameMissingColumn` variant for when a column is in the codebook but not in the initial dataframe.
 
 ### Fixed
 - Initializing an engine with a codebook that has a different number of rows than the data will result in an error instead of printing a bunch on nonsense.

lace/src/data/init.rs

@@ -224,7 +224,11 @@ pub fn df_to_col_models<R: rand::Rng>(
         .iter()
         .enumerate()
         .map(|(id, colmd)| {
-            let srs = srss[colmd.name.as_str()];
+            let srs = srss.get(colmd.name.as_str()).ok_or_else(|| {
+                DataParseError::DataFrameMissingColumn {
+                    column: colmd.name.clone(),
+                }
+            })?;
             let col_model = match &colmd.coltype {
                 ColType::Continuous { hyper, prior } => continuous_col_model(
                     id,

lace/src/interface/engine/error.rs

@@ -42,6 +42,10 @@ pub enum DataParseError {
         n_codebook_rows: usize,
         n_data_rows: usize,
     },
+    #[error(
+        "The dataframe does not contain the column `{column}` listed in the codebook"
+    )]
+    DataFrameMissingColumn { column: String },
 }
 
 /// Errors that can arise when creating a new engine