From ea6912c3f7f59ebe404a35bbede12abf3bf64a61 Mon Sep 17 00:00:00 2001
From: Fernando Herrera <fernando.j.herrera@gmail.com>
Date: Tue, 21 Dec 2021 23:35:02 +0000
Subject: [PATCH] missing commands (#549)

---
 .../src/dataframe/eager/drop_duplicates.rs    | 106 ++++++++++++++++++
 crates/nu-command/src/dataframe/eager/mod.rs  |   2 +
 2 files changed, 108 insertions(+)
 create mode 100644 crates/nu-command/src/dataframe/eager/drop_duplicates.rs

diff --git a/crates/nu-command/src/dataframe/eager/drop_duplicates.rs b/crates/nu-command/src/dataframe/eager/drop_duplicates.rs
new file mode 100644
index 0000000000..4d61a95e11
--- /dev/null
+++ b/crates/nu-command/src/dataframe/eager/drop_duplicates.rs
@@ -0,0 +1,106 @@
+use nu_engine::CallExt;
+use nu_protocol::{
+    ast::Call,
+    engine::{Command, EngineState, Stack},
+    Category, Example, PipelineData, ShellError, Signature, Span, SyntaxShape, Value,
+};
+
+use super::super::values::utils::convert_columns_string;
+use super::super::values::{Column, NuDataFrame};
+
+#[derive(Clone)]
+pub struct DropDuplicates;
+
+impl Command for DropDuplicates {
+    fn name(&self) -> &str {
+        "dfr drop-duplicates"
+    }
+
+    fn usage(&self) -> &str {
+        "Drops duplicate values in dataframe"
+    }
+
+    fn signature(&self) -> Signature {
+        Signature::build(self.name())
+            .optional(
+                "subset",
+                SyntaxShape::Table,
+                "subset of columns to drop duplicates",
+            )
+            .switch("maintain", "maintain order", Some('m'))
+            .category(Category::Custom("dataframe".into()))
+    }
+
+    fn examples(&self) -> Vec<Example> {
+        vec![Example {
+            description: "drop duplicates",
+            example: "[[a b]; [1 2] [3 4] [1 2]] | dfr to-df | dfr drop-duplicates",
+            result: Some(
+                NuDataFrame::try_from_columns(vec![
+                    Column::new(
+                        "a".to_string(),
+                        vec![Value::test_int(1), Value::test_int(3)],
+                    ),
+                    Column::new(
+                        "b".to_string(),
+                        vec![Value::test_int(2), Value::test_int(4)],
+                    ),
+                ])
+                .expect("simple df for test should not fail")
+                .into_value(Span::test_data()),
+            ),
+        }]
+    }
+
+    fn run(
+        &self,
+        engine_state: &EngineState,
+        stack: &mut Stack,
+        call: &Call,
+        input: PipelineData,
+    ) -> Result<PipelineData, ShellError> {
+        command(engine_state, stack, call, input)
+    }
+}
+
+fn command(
+    engine_state: &EngineState,
+    stack: &mut Stack,
+    call: &Call,
+    input: PipelineData,
+) -> Result<PipelineData, ShellError> {
+    let columns: Option<Vec<Value>> = call.opt(engine_state, stack, 0)?;
+    let (subset, col_span) = match columns {
+        Some(cols) => {
+            let (agg_string, col_span) = convert_columns_string(cols, call.head)?;
+            (Some(agg_string), col_span)
+        }
+        None => (None, call.head),
+    };
+
+    let df = NuDataFrame::try_from_pipeline(input, call.head)?;
+
+    let subset_slice = subset.as_ref().map(|cols| &cols[..]);
+
+    df.as_ref()
+        .drop_duplicates(call.has_flag("maintain"), subset_slice)
+        .map_err(|e| {
+            ShellError::SpannedLabeledError(
+                "Error dropping duplicates".into(),
+                e.to_string(),
+                col_span,
+            )
+        })
+        .map(|df| PipelineData::Value(NuDataFrame::dataframe_into_value(df, call.head), None))
+}
+
+#[cfg(test)]
+mod test {
+    use super::super::super::test_dataframe::test_dataframe;
+    use super::*;
+
+    #[test]
+    fn test_examples() {
+        test_dataframe(vec![Box::new(DropDuplicates {})])
+    }
+}
diff --git a/crates/nu-command/src/dataframe/eager/mod.rs b/crates/nu-command/src/dataframe/eager/mod.rs
index 493779916b..1177bd2059 100644
--- a/crates/nu-command/src/dataframe/eager/mod.rs
+++ b/crates/nu-command/src/dataframe/eager/mod.rs
@@ -4,6 +4,7 @@ mod column;
 mod command;
 mod describe;
 mod drop;
+mod drop_duplicates;
 mod drop_nulls;
 mod dtypes;
 mod dummies;
@@ -36,6 +37,7 @@ pub use column::ColumnDF;
 pub use command::Dataframe;
 pub use describe::DescribeDF;
 pub use drop::DropDF;
+pub use drop_duplicates::DropDuplicates;
 pub use drop_nulls::DropNulls;
 pub use dtypes::DataTypes;
 pub use dummies::Dummies;