Function polars_core::frame::_duplicate_err

pub fn _duplicate_err(name: &str) -> PolarsResult<()>

Examples found in repository

src/frame/mod.rs (line 256)

    pub fn new<S: IntoSeries>(columns: Vec<S>) -> PolarsResult<Self> {
        let mut first_len = None;

        let shape_err = |s: &[Series]| {
            let msg = format!(
                "Could not create a new DataFrame from Series. \
            The Series have different lengths. \
            Got {s:?}",
            );
            Err(PolarsError::ShapeMisMatch(msg.into()))
        };

        let series_cols = if S::is_series() {
            // Safety:
            // we are guarded by the type system here.
            #[allow(clippy::transmute_undefined_repr)]
            let series_cols = unsafe { std::mem::transmute::<Vec<S>, Vec<Series>>(columns) };
            let mut names = PlHashSet::with_capacity(series_cols.len());

            for s in &series_cols {
                match first_len {
                    Some(len) => {
                        if s.len() != len {
                            return shape_err(&series_cols);
                        }
                    }
                    None => first_len = Some(s.len()),
                }
                let name = s.name();

                if names.contains(name) {
                    _duplicate_err(name)?
                }

                names.insert(name);
            }
            // we drop early as the brchk thinks the &str borrows are used when calling the drop
            // of both `series_cols` and `names`
            drop(names);
            series_cols
        } else {
            let mut series_cols = Vec::with_capacity(columns.len());
            let mut names = PlHashSet::with_capacity(columns.len());

            // check for series length equality and convert into series in one pass
            for s in columns {
                let series = s.into_series();
                match first_len {
                    Some(len) => {
                        if series.len() != len {
                            return shape_err(&series_cols);
                        }
                    }
                    None => first_len = Some(series.len()),
                }
                // we have aliasing borrows so we must allocate a string
                let name = series.name().to_string();

                if names.contains(&name) {
                    _duplicate_err(&name)?
                }

                series_cols.push(series);
                names.insert(name);
            }
            drop(names);
            series_cols
        };

        Ok(DataFrame {
            columns: series_cols,
        })
    }

    /// Creates an empty `DataFrame` usable in a compile time context (such as static initializers).
    ///
    /// # Example
    ///
    /// ```rust
    /// use polars_core::prelude::DataFrame;
    /// static EMPTY: DataFrame = DataFrame::empty();
    /// ```
    pub const fn empty() -> Self {
        DataFrame::new_no_checks(Vec::new())
    }

    /// Removes the last `Series` from the `DataFrame` and returns it, or [`None`] if it is empty.
    ///
    /// # Example
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// let s1 = Series::new("Ocean", &["Atlantic", "Indian"]);
    /// let s2 = Series::new("Area (km²)", &[106_460_000, 70_560_000]);
    /// let mut df = DataFrame::new(vec![s1.clone(), s2.clone()])?;
    ///
    /// assert_eq!(df.pop(), Some(s2));
    /// assert_eq!(df.pop(), Some(s1));
    /// assert_eq!(df.pop(), None);
    /// assert!(df.is_empty());
    /// # Ok::<(), PolarsError>(())
    /// ```
    pub fn pop(&mut self) -> Option<Series> {
        self.columns.pop()
    }

    /// Add a new column at index 0 that counts the rows.
    ///
    /// # Example
    ///
    /// ```
    /// # use polars_core::prelude::*;
    /// let df1: DataFrame = df!("Name" => &["James", "Mary", "John", "Patricia"])?;
    /// assert_eq!(df1.shape(), (4, 1));
    ///
    /// let df2: DataFrame = df1.with_row_count("Id", None)?;
    /// assert_eq!(df2.shape(), (4, 2));
    /// println!("{}", df2);
    ///
    /// # Ok::<(), PolarsError>(())
    /// ```
    ///
    /// Output:
    ///
    /// ```text
    ///  shape: (4, 2)
    ///  +-----+----------+
    ///  | Id  | Name     |
    ///  | --- | ---      |
    ///  | u32 | str      |
    ///  +=====+==========+
    ///  | 0   | James    |
    ///  +-----+----------+
    ///  | 1   | Mary     |
    ///  +-----+----------+
    ///  | 2   | John     |
    ///  +-----+----------+
    ///  | 3   | Patricia |
    ///  +-----+----------+
    /// ```
    pub fn with_row_count(&self, name: &str, offset: Option<IdxSize>) -> PolarsResult<Self> {
        let mut columns = Vec::with_capacity(self.columns.len() + 1);
        let offset = offset.unwrap_or(0);

        let mut ca = IdxCa::from_vec(
            name,
            (offset..(self.height() as IdxSize) + offset).collect(),
        );
        ca.set_sorted(false);
        columns.push(ca.into_series());

        columns.extend_from_slice(&self.columns);
        DataFrame::new(columns)
    }

    /// Add a row count in place.
    pub fn with_row_count_mut(&mut self, name: &str, offset: Option<IdxSize>) -> &mut Self {
        let offset = offset.unwrap_or(0);
        let mut ca = IdxCa::from_vec(
            name,
            (offset..(self.height() as IdxSize) + offset).collect(),
        );
        ca.set_sorted(false);

        self.columns.insert(0, ca.into_series());
        self
    }

    /// Create a new `DataFrame` but does not check the length or duplicate occurrence of the `Series`.
    ///
    /// It is advised to use [Series::new](Series::new) in favor of this method.
    ///
    /// # Panic
    /// It is the callers responsibility to uphold the contract of all `Series`
    /// having an equal length, if not this may panic down the line.
    pub const fn new_no_checks(columns: Vec<Series>) -> DataFrame {
        DataFrame { columns }
    }

    /// Aggregate all chunks to contiguous memory.
    #[must_use]
    pub fn agg_chunks(&self) -> Self {
        // Don't parallelize this. Memory overhead
        let f = |s: &Series| s.rechunk();
        let cols = self.columns.iter().map(f).collect();
        DataFrame::new_no_checks(cols)
    }

    /// Shrink the capacity of this DataFrame to fit its length.
    pub fn shrink_to_fit(&mut self) {
        // Don't parallelize this. Memory overhead
        for s in &mut self.columns {
            s.shrink_to_fit();
        }
    }

    /// Aggregate all the chunks in the DataFrame to a single chunk.
    pub fn as_single_chunk(&mut self) -> &mut Self {
        // Don't parallelize this. Memory overhead
        for s in &mut self.columns {
            *s = s.rechunk();
        }
        self
    }

    /// Aggregate all the chunks in the DataFrame to a single chunk in parallel.
    /// This may lead to more peak memory consumption.
    pub fn as_single_chunk_par(&mut self) -> &mut Self {
        if self.columns.iter().any(|s| s.n_chunks() > 1) {
            self.columns = self.apply_columns_par(&|s| s.rechunk());
        }
        self
    }

    /// Estimates of the DataFrames columns consist of the same chunk sizes
    pub fn should_rechunk(&self) -> bool {
        let hb = RandomState::default();
        let hb2 = RandomState::with_seeds(392498, 98132457, 0, 412059);
        !self
            .columns
            .iter()
            // The idea is that we create a hash of the chunk lengths.
            // Consisting of the combined hash + the sum (assuming collision probability is nihil)
            // if not, we can add more hashes or at worst case we do an extra rechunk.
            // the old solution to this was clone all lengths to a vec and compare the vecs
            .map(|s| {
                s.chunk_lengths().map(|i| i as u64).fold(
                    (0u64, 0u64, s.n_chunks()),
                    |(lhash, lh2, n), rval| {
                        let mut h = hb.build_hasher();
                        rval.hash(&mut h);
                        let rhash = h.finish();
                        let mut h = hb2.build_hasher();
                        rval.hash(&mut h);
                        let rh2 = h.finish();
                        (
                            _boost_hash_combine(lhash, rhash),
                            _boost_hash_combine(lh2, rh2),
                            n,
                        )
                    },
                )
            })
            .all_equal()
    }

    /// Ensure all the chunks in the DataFrame are aligned.
    pub fn rechunk(&mut self) -> &mut Self {
        if self.should_rechunk() {
            self.as_single_chunk_par()
        } else {
            self
        }
    }

    /// Get the `DataFrame` schema.
    ///
    /// # Example
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// let df: DataFrame = df!("Thing" => &["Observable universe", "Human stupidity"],
    ///                         "Diameter (m)" => &[8.8e26, f64::INFINITY])?;
    ///
    /// let f1: Field = Field::new("Thing", DataType::Utf8);
    /// let f2: Field = Field::new("Diameter (m)", DataType::Float64);
    /// let sc: Schema = Schema::from(vec![f1, f2].into_iter());
    ///
    /// assert_eq!(df.schema(), sc);
    /// # Ok::<(), PolarsError>(())
    /// ```
    pub fn schema(&self) -> Schema {
        Schema::from(self.iter().map(|s| s.field().into_owned()))
    }

    /// Get a reference to the `DataFrame` columns.
    ///
    /// # Example
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// let df: DataFrame = df!("Name" => &["Adenine", "Cytosine", "Guanine", "Thymine"],
    ///                         "Symbol" => &["A", "C", "G", "T"])?;
    /// let columns: &Vec<Series> = df.get_columns();
    ///
    /// assert_eq!(columns[0].name(), "Name");
    /// assert_eq!(columns[1].name(), "Symbol");
    /// # Ok::<(), PolarsError>(())
    /// ```
    #[inline]
    pub fn get_columns(&self) -> &Vec<Series> {
        &self.columns
    }

    #[cfg(feature = "private")]
    #[inline]
    pub fn get_columns_mut(&mut self) -> &mut Vec<Series> {
        &mut self.columns
    }

    /// Iterator over the columns as `Series`.
    ///
    /// # Example
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// let s1: Series = Series::new("Name", &["Pythagoras' theorem", "Shannon entropy"]);
    /// let s2: Series = Series::new("Formula", &["a²+b²=c²", "H=-Σ[P(x)log|P(x)|]"]);
    /// let df: DataFrame = DataFrame::new(vec![s1.clone(), s2.clone()])?;
    ///
    /// let mut iterator = df.iter();
    ///
    /// assert_eq!(iterator.next(), Some(&s1));
    /// assert_eq!(iterator.next(), Some(&s2));
    /// assert_eq!(iterator.next(), None);
    /// # Ok::<(), PolarsError>(())
    /// ```
    pub fn iter(&self) -> std::slice::Iter<'_, Series> {
        self.columns.iter()
    }

    /// # Example
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// let df: DataFrame = df!("Language" => &["Rust", "Python"],
    ///                         "Designer" => &["Graydon Hoare", "Guido van Rossum"])?;
    ///
    /// assert_eq!(df.get_column_names(), &["Language", "Designer"]);
    /// # Ok::<(), PolarsError>(())
    /// ```
    pub fn get_column_names(&self) -> Vec<&str> {
        self.columns.iter().map(|s| s.name()).collect()
    }

    /// Get the `Vec<String>` representing the column names.
    pub fn get_column_names_owned(&self) -> Vec<String> {
        self.columns.iter().map(|s| s.name().to_string()).collect()
    }

    /// Set the column names.
    /// # Example
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// let mut df: DataFrame = df!("Mathematical set" => &["ℕ", "ℤ", "𝔻", "ℚ", "ℝ", "ℂ"])?;
    /// df.set_column_names(&["Set"])?;
    ///
    /// assert_eq!(df.get_column_names(), &["Set"]);
    /// # Ok::<(), PolarsError>(())
    /// ```
    pub fn set_column_names<S: AsRef<str>>(&mut self, names: &[S]) -> PolarsResult<()> {
        if names.len() != self.columns.len() {
            return Err(PolarsError::ShapeMisMatch("the provided slice with column names has not the same size as the DataFrame's width".into()));
        }
        let unique_names: AHashSet<&str, ahash::RandomState> =
            AHashSet::from_iter(names.iter().map(|name| name.as_ref()));
        if unique_names.len() != self.columns.len() {
            return Err(PolarsError::SchemaMisMatch(
                "duplicate column names found".into(),
            ));
        }

        let columns = mem::take(&mut self.columns);
        self.columns = columns
            .into_iter()
            .zip(names)
            .map(|(s, name)| {
                let mut s = s;
                s.rename(name.as_ref());
                s
            })
            .collect();
        Ok(())
    }

    /// Get the data types of the columns in the DataFrame.
    ///
    /// # Example
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// let venus_air: DataFrame = df!("Element" => &["Carbon dioxide", "Nitrogen"],
    ///                                "Fraction" => &[0.965, 0.035])?;
    ///
    /// assert_eq!(venus_air.dtypes(), &[DataType::Utf8, DataType::Float64]);
    /// # Ok::<(), PolarsError>(())
    /// ```
    pub fn dtypes(&self) -> Vec<DataType> {
        self.columns.iter().map(|s| s.dtype().clone()).collect()
    }

    /// The number of chunks per column
    pub fn n_chunks(&self) -> usize {
        match self.columns.get(0) {
            None => 0,
            Some(s) => s.n_chunks(),
        }
    }

    /// Get a reference to the schema fields of the `DataFrame`.
    ///
    /// # Example
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// let earth: DataFrame = df!("Surface type" => &["Water", "Land"],
    ///                            "Fraction" => &[0.708, 0.292])?;
    ///
    /// let f1: Field = Field::new("Surface type", DataType::Utf8);
    /// let f2: Field = Field::new("Fraction", DataType::Float64);
    ///
    /// assert_eq!(earth.fields(), &[f1, f2]);
    /// # Ok::<(), PolarsError>(())
    /// ```
    pub fn fields(&self) -> Vec<Field> {
        self.columns
            .iter()
            .map(|s| s.field().into_owned())
            .collect()
    }

    /// Get (height, width) of the `DataFrame`.
    ///
    /// # Example
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// let df0: DataFrame = DataFrame::default();
    /// let df1: DataFrame = df!("1" => &[1, 2, 3, 4, 5])?;
    /// let df2: DataFrame = df!("1" => &[1, 2, 3, 4, 5],
    ///                          "2" => &[1, 2, 3, 4, 5])?;
    ///
    /// assert_eq!(df0.shape(), (0 ,0));
    /// assert_eq!(df1.shape(), (5, 1));
    /// assert_eq!(df2.shape(), (5, 2));
    /// # Ok::<(), PolarsError>(())
    /// ```
    pub fn shape(&self) -> (usize, usize) {
        match self.columns.as_slice() {
            &[] => (0, 0),
            v => (v[0].len(), v.len()),
        }
    }

    /// Get the width of the `DataFrame` which is the number of columns.
    ///
    /// # Example
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// let df0: DataFrame = DataFrame::default();
    /// let df1: DataFrame = df!("Series 1" => &[0; 0])?;
    /// let df2: DataFrame = df!("Series 1" => &[0; 0],
    ///                          "Series 2" => &[0; 0])?;
    ///
    /// assert_eq!(df0.width(), 0);
    /// assert_eq!(df1.width(), 1);
    /// assert_eq!(df2.width(), 2);
    /// # Ok::<(), PolarsError>(())
    /// ```
    pub fn width(&self) -> usize {
        self.columns.len()
    }

    /// Get the height of the `DataFrame` which is the number of rows.
    ///
    /// # Example
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// let df0: DataFrame = DataFrame::default();
    /// let df1: DataFrame = df!("Currency" => &["€", "$"])?;
    /// let df2: DataFrame = df!("Currency" => &["€", "$", "¥", "£", "₿"])?;
    ///
    /// assert_eq!(df0.height(), 0);
    /// assert_eq!(df1.height(), 2);
    /// assert_eq!(df2.height(), 5);
    /// # Ok::<(), PolarsError>(())
    /// ```
    pub fn height(&self) -> usize {
        self.shape().0
    }

    /// Check if the `DataFrame` is empty.
    ///
    /// # Example
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// let df1: DataFrame = DataFrame::default();
    /// assert!(df1.is_empty());
    ///
    /// let df2: DataFrame = df!("First name" => &["Forever"],
    ///                          "Last name" => &["Alone"])?;
    /// assert!(!df2.is_empty());
    /// # Ok::<(), PolarsError>(())
    /// ```
    pub fn is_empty(&self) -> bool {
        self.columns.is_empty()
    }

    pub(crate) fn hstack_mut_no_checks(&mut self, columns: &[Series]) -> &mut Self {
        for col in columns {
            self.columns.push(col.clone());
        }
        self
    }

    /// Add multiple `Series` to a `DataFrame`.
    /// The added `Series` are required to have the same length.
    ///
    /// # Example
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// fn stack(df: &mut DataFrame, columns: &[Series]) {
    ///     df.hstack_mut(columns);
    /// }
    /// ```
    pub fn hstack_mut(&mut self, columns: &[Series]) -> PolarsResult<&mut Self> {
        let mut names = PlHashSet::with_capacity(self.columns.len());
        for s in &self.columns {
            names.insert(s.name());
        }

        let height = self.height();
        // first loop check validity. We don't do this in a single pass otherwise
        // this DataFrame is already modified when an error occurs.
        for col in columns {
            if col.len() != height && height != 0 {
                return Err(PolarsError::ShapeMisMatch(
                    format!("Could not horizontally stack Series. The Series length {} differs from the DataFrame height: {height}", col.len()).into()));
            }

            let name = col.name();
            if names.contains(name) {
                return Err(PolarsError::Duplicate(
                    format!("Cannot do hstack operation. Column with name: {name} already exists",)
                        .into(),
                ));
            }
            names.insert(name);
        }
        drop(names);
        Ok(self.hstack_mut_no_checks(columns))
    }

    /// Add multiple `Series` to a `DataFrame`.
    /// The added `Series` are required to have the same length.
    ///
    /// # Example
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// let df1: DataFrame = df!("Element" => &["Copper", "Silver", "Gold"])?;
    /// let s1: Series = Series::new("Proton", &[29, 47, 79]);
    /// let s2: Series = Series::new("Electron", &[29, 47, 79]);
    ///
    /// let df2: DataFrame = df1.hstack(&[s1, s2])?;
    /// assert_eq!(df2.shape(), (3, 3));
    /// println!("{}", df2);
    /// # Ok::<(), PolarsError>(())
    /// ```
    ///
    /// Output:
    ///
    /// ```text
    /// shape: (3, 3)
    /// +---------+--------+----------+
    /// | Element | Proton | Electron |
    /// | ---     | ---    | ---      |
    /// | str     | i32    | i32      |
    /// +=========+========+==========+
    /// | Copper  | 29     | 29       |
    /// +---------+--------+----------+
    /// | Silver  | 47     | 47       |
    /// +---------+--------+----------+
    /// | Gold    | 79     | 79       |
    /// +---------+--------+----------+
    /// ```
    pub fn hstack(&self, columns: &[Series]) -> PolarsResult<Self> {
        let mut new_cols = self.columns.clone();
        new_cols.extend_from_slice(columns);
        DataFrame::new(new_cols)
    }

    /// Concatenate a `DataFrame` to this `DataFrame` and return as newly allocated `DataFrame`.
    ///
    /// If many `vstack` operations are done, it is recommended to call [`DataFrame::rechunk`].
    ///
    /// # Example
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// let df1: DataFrame = df!("Element" => &["Copper", "Silver", "Gold"],
    ///                          "Melting Point (K)" => &[1357.77, 1234.93, 1337.33])?;
    /// let df2: DataFrame = df!("Element" => &["Platinum", "Palladium"],
    ///                          "Melting Point (K)" => &[2041.4, 1828.05])?;
    ///
    /// let df3: DataFrame = df1.vstack(&df2)?;
    ///
    /// assert_eq!(df3.shape(), (5, 2));
    /// println!("{}", df3);
    /// # Ok::<(), PolarsError>(())
    /// ```
    ///
    /// Output:
    ///
    /// ```text
    /// shape: (5, 2)
    /// +-----------+-------------------+
    /// | Element   | Melting Point (K) |
    /// | ---       | ---               |
    /// | str       | f64               |
    /// +===========+===================+
    /// | Copper    | 1357.77           |
    /// +-----------+-------------------+
    /// | Silver    | 1234.93           |
    /// +-----------+-------------------+
    /// | Gold      | 1337.33           |
    /// +-----------+-------------------+
    /// | Platinum  | 2041.4            |
    /// +-----------+-------------------+
    /// | Palladium | 1828.05           |
    /// +-----------+-------------------+
    /// ```
    pub fn vstack(&self, other: &DataFrame) -> PolarsResult<Self> {
        let mut df = self.clone();
        df.vstack_mut(other)?;
        Ok(df)
    }

    /// Concatenate a DataFrame to this DataFrame
    ///
    /// If many `vstack` operations are done, it is recommended to call [`DataFrame::rechunk`].
    ///
    /// # Example
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// let mut df1: DataFrame = df!("Element" => &["Copper", "Silver", "Gold"],
    ///                          "Melting Point (K)" => &[1357.77, 1234.93, 1337.33])?;
    /// let df2: DataFrame = df!("Element" => &["Platinum", "Palladium"],
    ///                          "Melting Point (K)" => &[2041.4, 1828.05])?;
    ///
    /// df1.vstack_mut(&df2)?;
    ///
    /// assert_eq!(df1.shape(), (5, 2));
    /// println!("{}", df1);
    /// # Ok::<(), PolarsError>(())
    /// ```
    ///
    /// Output:
    ///
    /// ```text
    /// shape: (5, 2)
    /// +-----------+-------------------+
    /// | Element   | Melting Point (K) |
    /// | ---       | ---               |
    /// | str       | f64               |
    /// +===========+===================+
    /// | Copper    | 1357.77           |
    /// +-----------+-------------------+
    /// | Silver    | 1234.93           |
    /// +-----------+-------------------+
    /// | Gold      | 1337.33           |
    /// +-----------+-------------------+
    /// | Platinum  | 2041.4            |
    /// +-----------+-------------------+
    /// | Palladium | 1828.05           |
    /// +-----------+-------------------+
    /// ```
    pub fn vstack_mut(&mut self, other: &DataFrame) -> PolarsResult<&mut Self> {
        if self.width() != other.width() {
            if self.width() == 0 {
                self.columns = other.columns.clone();
                return Ok(self);
            }

            return Err(PolarsError::ShapeMisMatch(
                format!("Could not vertically stack DataFrame. The DataFrames appended width {} differs from the parent DataFrames width {}", self.width(), other.width()).into()
            ));
        }

        self.columns
            .iter_mut()
            .zip(other.columns.iter())
            .try_for_each::<_, PolarsResult<_>>(|(left, right)| {
                can_extend(left, right)?;
                left.append(right).expect("should not fail");
                Ok(())
            })?;
        Ok(self)
    }

    /// Does not check if schema is correct
    pub(crate) fn vstack_mut_unchecked(&mut self, other: &DataFrame) {
        self.columns
            .iter_mut()
            .zip(other.columns.iter())
            .for_each(|(left, right)| {
                left.append(right).expect("should not fail");
            });
    }

    /// Extend the memory backed by this [`DataFrame`] with the values from `other`.
    ///
    /// Different from [`vstack`](Self::vstack) which adds the chunks from `other` to the chunks of this [`DataFrame`]
    /// `extend` appends the data from `other` to the underlying memory locations and thus may cause a reallocation.
    ///
    /// If this does not cause a reallocation, the resulting data structure will not have any extra chunks
    /// and thus will yield faster queries.
    ///
    /// Prefer `extend` over `vstack` when you want to do a query after a single append. For instance during
    /// online operations where you add `n` rows and rerun a query.
    ///
    /// Prefer `vstack` over `extend` when you want to append many times before doing a query. For instance
    /// when you read in multiple files and when to store them in a single `DataFrame`. In the latter case, finish the sequence
    /// of `append` operations with a [`rechunk`](Self::rechunk).
    pub fn extend(&mut self, other: &DataFrame) -> PolarsResult<()> {
        if self.width() != other.width() {
            return Err(PolarsError::ShapeMisMatch(
                format!("Could not extend DataFrame. The DataFrames extended width {} differs from the parent DataFrames width {}", self.width(), other.width()).into()
            ));
        }

        self.columns
            .iter_mut()
            .zip(other.columns.iter())
            .try_for_each::<_, PolarsResult<_>>(|(left, right)| {
                can_extend(left, right)?;
                left.extend(right).unwrap();
                Ok(())
            })?;
        Ok(())
    }

    /// Remove a column by name and return the column removed.
    ///
    /// # Example
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// let mut df: DataFrame = df!("Animal" => &["Tiger", "Lion", "Great auk"],
    ///                             "IUCN" => &["Endangered", "Vulnerable", "Extinct"])?;
    ///
    /// let s1: PolarsResult<Series> = df.drop_in_place("Average weight");
    /// assert!(s1.is_err());
    ///
    /// let s2: Series = df.drop_in_place("Animal")?;
    /// assert_eq!(s2, Series::new("Animal", &["Tiger", "Lion", "Great auk"]));
    /// # Ok::<(), PolarsError>(())
    /// ```
    pub fn drop_in_place(&mut self, name: &str) -> PolarsResult<Series> {
        let idx = self.check_name_to_idx(name)?;
        Ok(self.columns.remove(idx))
    }

    /// Return a new `DataFrame` where all null values are dropped.
    ///
    /// # Example
    ///
    /// ```no_run
    /// # use polars_core::prelude::*;
    /// let df1: DataFrame = df!("Country" => ["Malta", "Liechtenstein", "North Korea"],
    ///                         "Tax revenue (% GDP)" => [Some(32.7), None, None])?;
    /// assert_eq!(df1.shape(), (3, 2));
    ///
    /// let df2: DataFrame = df1.drop_nulls(None)?;
    /// assert_eq!(df2.shape(), (1, 2));
    /// println!("{}", df2);
    /// # Ok::<(), PolarsError>(())
    /// ```
    ///
    /// Output:
    ///
    /// ```text
    /// shape: (1, 2)
    /// +---------+---------------------+
    /// | Country | Tax revenue (% GDP) |
    /// | ---     | ---                 |
    /// | str     | f64                 |
    /// +=========+=====================+
    /// | Malta   | 32.7                |
    /// +---------+---------------------+
    /// ```
    pub fn drop_nulls(&self, subset: Option<&[String]>) -> PolarsResult<Self> {
        let selected_series;

        let mut iter = match subset {
            Some(cols) => {
                selected_series = self.select_series(cols)?;
                selected_series.iter()
            }
            None => self.columns.iter(),
        };

        // fast path for no nulls in df
        if iter.clone().all(|s| !s.has_validity()) {
            return Ok(self.clone());
        }

        let mask = iter
            .next()
            .ok_or_else(|| PolarsError::NoData("No data to drop nulls from".into()))?;
        let mut mask = mask.is_not_null();

        for s in iter {
            mask = mask & s.is_not_null();
        }
        self.filter(&mask)
    }

    /// Drop a column by name.
    /// This is a pure method and will return a new `DataFrame` instead of modifying
    /// the current one in place.
    ///
    /// # Example
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// let df1: DataFrame = df!("Ray type" => &["α", "β", "X", "γ"])?;
    /// let df2: DataFrame = df1.drop("Ray type")?;
    ///
    /// assert!(df2.is_empty());
    /// # Ok::<(), PolarsError>(())
    /// ```
    pub fn drop(&self, name: &str) -> PolarsResult<Self> {
        let idx = self.check_name_to_idx(name)?;
        let mut new_cols = Vec::with_capacity(self.columns.len() - 1);

        self.columns.iter().enumerate().for_each(|(i, s)| {
            if i != idx {
                new_cols.push(s.clone())
            }
        });

        Ok(DataFrame::new_no_checks(new_cols))
    }

    pub fn drop_many<S: AsRef<str>>(&self, names: &[S]) -> Self {
        let names = names.iter().map(|s| s.as_ref()).collect();
        fn inner(df: &DataFrame, names: Vec<&str>) -> DataFrame {
            let mut new_cols = Vec::with_capacity(df.columns.len() - names.len());
            df.columns.iter().for_each(|s| {
                if !names.contains(&s.name()) {
                    new_cols.push(s.clone())
                }
            });

            DataFrame::new_no_checks(new_cols)
        }
        inner(self, names)
    }

    fn insert_at_idx_no_name_check(
        &mut self,
        index: usize,
        series: Series,
    ) -> PolarsResult<&mut Self> {
        if series.len() == self.height() {
            self.columns.insert(index, series);
            Ok(self)
        } else {
            Err(PolarsError::ShapeMisMatch(
                format!(
                    "Could not add column. The Series length {} differs from the DataFrame height: {}",
                    series.len(),
                    self.height()
                )
                .into(),
            ))
        }
    }

    /// Insert a new column at a given index.
    pub fn insert_at_idx<S: IntoSeries>(
        &mut self,
        index: usize,
        column: S,
    ) -> PolarsResult<&mut Self> {
        let series = column.into_series();
        self.check_already_present(series.name())?;
        self.insert_at_idx_no_name_check(index, series)
    }

    fn add_column_by_search(&mut self, series: Series) -> PolarsResult<()> {
        if let Some(idx) = self.find_idx_by_name(series.name()) {
            self.replace_at_idx(idx, series)?;
        } else {
            self.columns.push(series);
        }
        Ok(())
    }

    /// Add a new column to this `DataFrame` or replace an existing one.
    pub fn with_column<S: IntoSeries>(&mut self, column: S) -> PolarsResult<&mut Self> {
        fn inner(df: &mut DataFrame, mut series: Series) -> PolarsResult<&mut DataFrame> {
            let height = df.height();
            if series.len() == 1 && height > 1 {
                series = series.new_from_index(0, height);
            }

            if series.len() == height || df.is_empty() {
                df.add_column_by_search(series)?;
                Ok(df)
            }
            // special case for literals
            else if height == 0 && series.len() == 1 {
                let s = series.slice(0, 0);
                df.add_column_by_search(s)?;
                Ok(df)
            } else {
                Err(PolarsError::ShapeMisMatch(
                    format!(
                        "Could not add column. The Series length {} differs from the DataFrame height: {}",
                        series.len(),
                        df.height()
                    )
                        .into(),
                ))
            }
        }
        let series = column.into_series();
        inner(self, series)
    }

    fn add_column_by_schema(&mut self, s: Series, schema: &Schema) -> PolarsResult<()> {
        let name = s.name();
        if let Some((idx, _, _)) = schema.get_full(name) {
            // schema is incorrect fallback to search
            if self.columns.get(idx).map(|s| s.name()) != Some(name) {
                self.add_column_by_search(s)?;
            } else {
                self.replace_at_idx(idx, s)?;
            }
        } else {
            self.columns.push(s);
        }
        Ok(())
    }

    pub fn _add_columns(&mut self, columns: Vec<Series>, schema: &Schema) -> PolarsResult<()> {
        for (i, s) in columns.into_iter().enumerate() {
            // we need to branch here
            // because users can add multiple columns with the same name
            if i == 0 || schema.get(s.name()).is_some() {
                self.with_column_and_schema(s, schema)?;
            } else {
                self.with_column(s.clone())?;
            }
        }
        Ok(())
    }

    /// Add a new column to this `DataFrame` or replace an existing one.
    /// Uses an existing schema to amortize lookups.
    /// If the schema is incorrect, we will fallback to linear search.
    pub fn with_column_and_schema<S: IntoSeries>(
        &mut self,
        column: S,
        schema: &Schema,
    ) -> PolarsResult<&mut Self> {
        let mut series = column.into_series();

        let height = self.height();
        if series.len() == 1 && height > 1 {
            series = series.new_from_index(0, height);
        }

        if series.len() == height || self.is_empty() {
            self.add_column_by_schema(series, schema)?;
            Ok(self)
        }
        // special case for literals
        else if height == 0 && series.len() == 1 {
            let s = series.slice(0, 0);
            self.add_column_by_schema(s, schema)?;
            Ok(self)
        } else {
            Err(PolarsError::ShapeMisMatch(
                format!(
                    "Could not add column. The Series length {} differs from the DataFrame height: {}",
                    series.len(),
                    self.height()
                )
                    .into(),
            ))
        }
    }

    /// Get a row in the `DataFrame`. Beware this is slow.
    ///
    /// # Example
    ///
    /// ```
    /// # use polars_core::prelude::*;
    /// fn example(df: &mut DataFrame, idx: usize) -> Option<Vec<AnyValue>> {
    ///     df.get(idx)
    /// }
    /// ```
    pub fn get(&self, idx: usize) -> Option<Vec<AnyValue>> {
        match self.columns.get(0) {
            Some(s) => {
                if s.len() <= idx {
                    return None;
                }
            }
            None => return None,
        }
        // safety: we just checked bounds
        unsafe { Some(self.columns.iter().map(|s| s.get_unchecked(idx)).collect()) }
    }

    /// Select a `Series` by index.
    ///
    /// # Example
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// let df: DataFrame = df!("Star" => &["Sun", "Betelgeuse", "Sirius A", "Sirius B"],
    ///                         "Absolute magnitude" => &[4.83, -5.85, 1.42, 11.18])?;
    ///
    /// let s1: Option<&Series> = df.select_at_idx(0);
    /// let s2: Series = Series::new("Star", &["Sun", "Betelgeuse", "Sirius A", "Sirius B"]);
    ///
    /// assert_eq!(s1, Some(&s2));
    /// # Ok::<(), PolarsError>(())
    /// ```
    pub fn select_at_idx(&self, idx: usize) -> Option<&Series> {
        self.columns.get(idx)
    }

    /// Select a mutable series by index.
    ///
    /// *Note: the length of the Series should remain the same otherwise the DataFrame is invalid.*
    /// For this reason the method is not public
    fn select_at_idx_mut(&mut self, idx: usize) -> Option<&mut Series> {
        self.columns.get_mut(idx)
    }

    /// Select column(s) from this `DataFrame` by range and return a new DataFrame
    ///
    /// # Examples
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// let df = df! {
    ///     "0" => &[0, 0, 0],
    ///     "1" => &[1, 1, 1],
    ///     "2" => &[2, 2, 2]
    /// }?;
    ///
    /// assert!(df.select(&["0", "1"])?.frame_equal(&df.select_by_range(0..=1)?));
    /// assert!(df.frame_equal(&df.select_by_range(..)?));
    /// # Ok::<(), PolarsError>(())
    /// ```
    pub fn select_by_range<R>(&self, range: R) -> PolarsResult<Self>
    where
        R: ops::RangeBounds<usize>,
    {
        // This function is copied from std::slice::range (https://doc.rust-lang.org/std/slice/fn.range.html)
        // because it is the nightly feature. We should change here if this function were stable.
        fn get_range<R>(range: R, bounds: ops::RangeTo<usize>) -> ops::Range<usize>
        where
            R: ops::RangeBounds<usize>,
        {
            let len = bounds.end;

            let start: ops::Bound<&usize> = range.start_bound();
            let start = match start {
                ops::Bound::Included(&start) => start,
                ops::Bound::Excluded(start) => start.checked_add(1).unwrap_or_else(|| {
                    panic!("attempted to index slice from after maximum usize");
                }),
                ops::Bound::Unbounded => 0,
            };

            let end: ops::Bound<&usize> = range.end_bound();
            let end = match end {
                ops::Bound::Included(end) => end.checked_add(1).unwrap_or_else(|| {
                    panic!("attempted to index slice up to maximum usize");
                }),
                ops::Bound::Excluded(&end) => end,
                ops::Bound::Unbounded => len,
            };

            if start > end {
                panic!("slice index starts at {start} but ends at {end}");
            }
            if end > len {
                panic!("range end index {end} out of range for slice of length {len}",);
            }

            ops::Range { start, end }
        }

        let colnames = self.get_column_names_owned();
        let range = get_range(range, ..colnames.len());

        self.select_impl(&colnames[range])
    }

    /// Get column index of a `Series` by name.
    /// # Example
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// let df: DataFrame = df!("Name" => &["Player 1", "Player 2", "Player 3"],
    ///                         "Health" => &[100, 200, 500],
    ///                         "Mana" => &[250, 100, 0],
    ///                         "Strength" => &[30, 150, 300])?;
    ///
    /// assert_eq!(df.find_idx_by_name("Name"), Some(0));
    /// assert_eq!(df.find_idx_by_name("Health"), Some(1));
    /// assert_eq!(df.find_idx_by_name("Mana"), Some(2));
    /// assert_eq!(df.find_idx_by_name("Strength"), Some(3));
    /// assert_eq!(df.find_idx_by_name("Haste"), None);
    /// # Ok::<(), PolarsError>(())
    /// ```
    pub fn find_idx_by_name(&self, name: &str) -> Option<usize> {
        self.columns.iter().position(|s| s.name() == name)
    }

    /// Select a single column by name.
    ///
    /// # Example
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// let s1: Series = Series::new("Password", &["123456", "[]B$u$g$s$B#u#n#n#y[]{}"]);
    /// let s2: Series = Series::new("Robustness", &["Weak", "Strong"]);
    /// let df: DataFrame = DataFrame::new(vec![s1.clone(), s2])?;
    ///
    /// assert_eq!(df.column("Password")?, &s1);
    /// # Ok::<(), PolarsError>(())
    /// ```
    pub fn column(&self, name: &str) -> PolarsResult<&Series> {
        let idx = self
            .find_idx_by_name(name)
            .ok_or_else(|| PolarsError::NotFound(name.to_string().into()))?;
        Ok(self.select_at_idx(idx).unwrap())
    }

    /// Selected multiple columns by name.
    ///
    /// # Example
    ///
    /// ```rust
    /// # use polars_core::prelude::*;
    /// let df: DataFrame = df!("Latin name" => &["Oncorhynchus kisutch", "Salmo salar"],
    ///                         "Max weight (kg)" => &[16.0, 35.89])?;
    /// let sv: Vec<&Series> = df.columns(&["Latin name", "Max weight (kg)"])?;
    ///
    /// assert_eq!(&df[0], sv[0]);
    /// assert_eq!(&df[1], sv[1]);
    /// # Ok::<(), PolarsError>(())
    /// ```
    pub fn columns<I, S>(&self, names: I) -> PolarsResult<Vec<&Series>>
    where
        I: IntoIterator<Item = S>,
        S: AsRef<str>,
    {
        names
            .into_iter()
            .map(|name| self.column(name.as_ref()))
            .collect()
    }

    /// Select column(s) from this `DataFrame` and return a new `DataFrame`.
    ///
    /// # Examples
    ///
    /// ```
    /// # use polars_core::prelude::*;
    /// fn example(df: &DataFrame) -> PolarsResult<DataFrame> {
    ///     df.select(["foo", "bar"])
    /// }
    /// ```
    pub fn select<I, S>(&self, selection: I) -> PolarsResult<Self>
    where
        I: IntoIterator<Item = S>,
        S: AsRef<str>,
    {
        let cols = selection
            .into_iter()
            .map(|s| s.as_ref().to_string())
            .collect::<Vec<_>>();
        self.select_impl(&cols)
    }

    fn select_impl(&self, cols: &[String]) -> PolarsResult<Self> {
        self.select_check_duplicates(cols)?;
        let selected = self.select_series_impl(cols)?;
        Ok(DataFrame::new_no_checks(selected))
    }

    pub fn select_physical<I, S>(&self, selection: I) -> PolarsResult<Self>
    where
        I: IntoIterator<Item = S>,
        S: AsRef<str>,
    {
        let cols = selection
            .into_iter()
            .map(|s| s.as_ref().to_string())
            .collect::<Vec<_>>();
        self.select_physical_impl(&cols)
    }

    fn select_physical_impl(&self, cols: &[String]) -> PolarsResult<Self> {
        self.select_check_duplicates(cols)?;
        let selected = self.select_series_physical_impl(cols)?;
        Ok(DataFrame::new_no_checks(selected))
    }

    fn select_check_duplicates(&self, cols: &[String]) -> PolarsResult<()> {
        let mut names = PlHashSet::with_capacity(cols.len());
        for name in cols {
            if !names.insert(name.as_str()) {
                _duplicate_err(name)?
            }
        }
        Ok(())
    }