Struct polars_arrow::array::Utf8Array

pub struct Utf8Array<O: Offset> { /* private fields */ }

A Utf8Array is arrow’s semantic equivalent of an immutable Vec<Option<String>>. Cloning and slicing this struct is O(1).

Example

use polars_arrow::bitmap::Bitmap;
use polars_arrow::buffer::Buffer;
use polars_arrow::array::Utf8Array;
let array = Utf8Array::<i32>::from([Some("hi"), None, Some("there")]);
assert_eq!(array.value(0), "hi");
assert_eq!(array.iter().collect::<Vec<_>>(), vec![Some("hi"), None, Some("there")]);
assert_eq!(array.values_iter().collect::<Vec<_>>(), vec!["hi", "", "there"]);
// the underlying representation
assert_eq!(array.validity(), Some(&Bitmap::from([true, false, true])));
assert_eq!(array.values(), &Buffer::from(b"hithere".to_vec()));
assert_eq!(array.offsets().buffer(), &Buffer::from(vec![0, 2, 2, 2 + 5]));

Generic parameter

The generic parameter Offset can only be i32 or i64 and tradeoffs maximum array length with memory usage:

• the sum of lengths of all elements cannot exceed Offset::MAX
• the total size of the underlying data is array.len() * size_of::<Offset>() + sum of lengths of all elements

Safety

The following invariants hold:

• Two consecutives offsets casted (as) to usize are valid slices of values.
• A slice of values taken from two consecutives offsets is valid utf8.
• len is equal to validity.len(), when defined.

Implementations

impl<O: Offset> Utf8Array<O>

pub fn try_new( data_type: ArrowDataType, offsets: OffsetsBuffer<O>, values: Buffer<u8>, validity: Option<Bitmap> ) -> PolarsResult<Self>

Returns a Utf8Array created from its internal representation.

Errors

This function returns an error iff:

• The last offset is not equal to the values’ length.
• the validity’s length is not equal to offsets.len().
• The data_type’s crate::datatypes::PhysicalType is not equal to either Utf8 or LargeUtf8.
• The values between two consecutive offsets are not valid utf8

Implementation

This function is O(N) - checking utf8 is O(N)

pub fn from_slice<T: AsRef<str>, P: AsRef<[T]>>(slice: P) -> Self

Returns a Utf8Array from a slice of &str.

A convenience method that uses Self::from_trusted_len_values_iter.

pub fn from<T: AsRef<str>, P: AsRef<[Option<T>]>>(slice: P) -> Self

Returns a new Utf8Array from a slice of &str.

A convenience method that uses Self::from_trusted_len_iter.

pub fn iter(&self) -> ZipValidity<&str, Utf8ValuesIter<'_, O>, BitmapIter<'_>>

Returns an iterator of Option<&str>

pub fn values_iter(&self) -> Utf8ValuesIter<'_, O>

Returns an iterator of &str

pub fn non_null_values_iter(&self) -> NonNullValuesIter<'_, Utf8Array<O>>

Returns an iterator of the non-null values `&str.

pub fn len(&self) -> usize

Returns the length of this array

pub fn value(&self, i: usize) -> &str

Returns the value of the element at index i, ignoring the array’s validity.

Panic

This function panics iff i >= self.len.

pub unsafe fn value_unchecked(&self, i: usize) -> &str

Returns the value of the element at index i, ignoring the array’s validity.

Safety

This function is safe iff i < self.len.

pub fn get(&self, i: usize) -> Option<&str>

Returns the element at index i or None if it is null

Panics

iff i >= self.len()

pub fn data_type(&self) -> &ArrowDataType

Returns the ArrowDataType of this array.

pub fn values(&self) -> &Buffer<u8>

Returns the values of this Utf8Array.

pub fn offsets(&self) -> &OffsetsBuffer<O>

Returns the offsets of this Utf8Array.

pub fn validity(&self) -> Option<&Bitmap>

The optional validity.

pub fn slice(&mut self, offset: usize, length: usize)

Slices this Utf8Array.

Implementation

This function is O(1).

Panics

iff offset + length > self.len().

pub unsafe fn slice_unchecked(&mut self, offset: usize, length: usize)

Slices this Utf8Array.

Implementation

This function is O(1)

Safety

The caller must ensure that offset + length <= self.len().

pub fn sliced(self, offset: usize, length: usize) -> Self

Returns this array sliced.

Implementation

This function is O(1).

Panics

iff offset + length > self.len().

pub unsafe fn sliced_unchecked(self, offset: usize, length: usize) -> Self

Returns this array sliced.

Implementation

This function is O(1).

Safety

The caller must ensure that offset + length <= self.len().

pub fn with_validity(self, validity: Option<Bitmap>) -> Self

Returns this array with a new validity.

Panic

Panics iff validity.len() != self.len().

pub fn set_validity(&mut self, validity: Option<Bitmap>)

Sets the validity of this array.

Panics

This function panics iff values.len() != self.len().

pub fn boxed(self) -> Box<dyn Array>

Boxes this array into a Box<dyn Array>.

pub fn arced(self) -> Arc<dyn Array>

Arcs this array into a std::sync::Arc<dyn Array>.

pub fn into_inner( self ) -> (ArrowDataType, OffsetsBuffer<O>, Buffer<u8>, Option<Bitmap>)

Returns its internal representation

pub fn into_mut(self) -> Either<Self, MutableUtf8Array<O>>

Try to convert this Utf8Array to a MutableUtf8Array

pub fn new_empty(data_type: ArrowDataType) -> Self

Returns a new empty Utf8Array.

The array is guaranteed to have no elements nor validity.

pub fn new_null(data_type: ArrowDataType, length: usize) -> Self

Returns a new Utf8Array whose all slots are null / None.

pub fn default_data_type() -> ArrowDataType

Returns a default ArrowDataType of this array, which depends on the generic parameter O: DataType::Utf8 or DataType::LargeUtf8

pub unsafe fn new_unchecked( data_type: ArrowDataType, offsets: OffsetsBuffer<O>, values: Buffer<u8>, validity: Option<Bitmap> ) -> Self

Creates a new Utf8Array without checking for offsets monotinicity nor utf8-validity

Panic

This function panics (in debug mode only) iff:

• The last offset is not equal to the values’ length.
• the validity’s length is not equal to offsets.len().
• The data_type’s crate::datatypes::PhysicalType is not equal to either Utf8 or LargeUtf8.

Safety

This function is unsound iff:

• The values between two consecutive offsets are not valid utf8

Implementation

This function is O(1)

pub fn new( data_type: ArrowDataType, offsets: OffsetsBuffer<O>, values: Buffer<u8>, validity: Option<Bitmap> ) -> Self

Creates a new Utf8Array.

Panics

This function panics iff:

• The last offset is not equal to the values’ length.
• the validity’s length is not equal to offsets.len().
• The data_type’s crate::datatypes::PhysicalType is not equal to either Utf8 or LargeUtf8.
• The values between two consecutive offsets are not valid utf8

Implementation

This function is O(N) - checking utf8 is O(N)

pub fn from_trusted_len_values_iter<T: AsRef<str>, I: TrustedLen<Item = T>>( iterator: I ) -> Self

Returns a (non-null) Utf8Array created from a TrustedLen of &str.

Implementation

This function is O(N)

pub fn from_iter_values<T: AsRef<str>, I: Iterator<Item = T>>( iterator: I ) -> Self

Creates a new Utf8Array from a Iterator of &str.

pub unsafe fn from_trusted_len_iter_unchecked<I, P>(iterator: I) -> Self

where P: AsRef<str>, I: Iterator<Item = Option<P>>,

Creates a Utf8Array from an iterator of trusted length.

Safety

The iterator must be TrustedLen. I.e. that size_hint().1 correctly reports its length.

pub fn from_trusted_len_iter<I, P>(iterator: I) -> Self

where P: AsRef<str>, I: TrustedLen<Item = Option<P>>,

Creates a Utf8Array from an iterator of trusted length.

pub unsafe fn try_from_trusted_len_iter_unchecked<E, I, P>( iterator: I ) -> Result<Self, E>

where P: AsRef<str>, I: IntoIterator<Item = Result<Option<P>, E>>,

Creates a Utf8Array from an falible iterator of trusted length.

Safety

The iterator must be TrustedLen. I.e. that size_hint().1 correctly reports its length.

pub fn try_from_trusted_len_iter<E, I, P>(iter: I) -> Result<Self, E>

where P: AsRef<str>, I: TrustedLen<Item = Result<Option<P>, E>>,

Creates a Utf8Array from an fallible iterator of trusted length.

pub fn apply_validity<F: FnOnce(Bitmap) -> Bitmap>(&mut self, f: F)

Applies a function f to the validity of this array.

This is an API to leverage clone-on-write

Panics

This function panics if the function f modifies the length of the Bitmap.

pub fn to_binary(&self) -> BinaryArray<O>

Trait Implementations

impl<O: Offset> Array for Utf8Array<O>

fn as_any(&self) -> &dyn Any

Converts itself to a reference of Any, which enables downcasting to concrete types.

fn as_any_mut(&mut self) -> &mut dyn Any

Converts itself to a mutable reference of Any, which enables mutable downcasting to concrete types.

fn len(&self) -> usize

The length of the Array. Every array has a length corresponding to the number of elements (slots).

fn data_type(&self) -> &ArrowDataType

The ArrowDataType of the Array. In combination with Array::as_any, this can be used to downcast trait objects (dyn Array) to concrete arrays.

fn slice(&mut self, offset: usize, length: usize)

Slices this Array.

unsafe fn slice_unchecked(&mut self, offset: usize, length: usize)

Slices the Array.

fn to_boxed(&self) -> Box<dyn Array>

Clone a &dyn Array to an owned Box<dyn Array>.

fn validity(&self) -> Option<&Bitmap>

The validity of the Array: every array has an optional Bitmap that, when available specifies whether the array slot is valid or not (null). When the validity is None, all slots are valid.

fn with_validity(&self, validity: Option<Bitmap>) -> Box<dyn Array>

Clones this Array with a new new assigned bitmap.

fn is_empty(&self) -> bool

whether the array is empty

fn null_count(&self) -> usize

The number of null slots on this Array.

fn is_null(&self, i: usize) -> bool

Returns whether slot i is null.

unsafe fn is_null_unchecked(&self, i: usize) -> bool

Returns whether slot i is null.

fn is_valid(&self, i: usize) -> bool

Returns whether slot i is valid.

fn sliced(&self, offset: usize, length: usize) -> Box<dyn Array>

Returns a slice of this Array.

unsafe fn sliced_unchecked( &self, offset: usize, length: usize ) -> Box<dyn Array>

Returns a slice of this Array.

impl<'a, O: Offset> ArrayAccessor<'a> for Utf8Array<O>

type Item = &'a str

unsafe fn value_unchecked(&'a self, index: usize) -> Self::Item

Safety

fn len(&self) -> usize

impl<T: StrIntoBytes> ArrayFromIter<Option<T>> for Utf8Array<i64>

fn arr_from_iter<I: IntoIterator<Item = Option<T>>>(iter: I) -> Self

fn arr_from_iter_trusted<I>(iter: I) -> Self

where I: IntoIterator<Item = Option<T>>, I::IntoIter: TrustedLen,

fn try_arr_from_iter<E, I: IntoIterator<Item = Result<Option<T>, E>>>( iter: I ) -> Result<Self, E>

fn try_arr_from_iter_trusted<E, I: IntoIterator<Item = Result<Option<T>, E>>>( iter: I ) -> Result<Self, E>

impl<T: StrIntoBytes> ArrayFromIter<T> for Utf8Array<i64>

fn arr_from_iter<I: IntoIterator<Item = T>>(iter: I) -> Self

fn arr_from_iter_trusted<I>(iter: I) -> Self

where I: IntoIterator<Item = T>, I::IntoIter: TrustedLen,

fn try_arr_from_iter<E, I: IntoIterator<Item = Result<T, E>>>( iter: I ) -> Result<Self, E>

fn try_arr_from_iter_trusted<E, I: IntoIterator<Item = Result<T, E>>>( iter: I ) -> Result<Self, E>

impl<O: Offset> Arrow2Arrow for Utf8Array<O>

fn to_data(&self) -> ArrayData

Convert this Array into [ArrayData]

fn from_data(data: &ArrayData) -> Self

Create this Array from [ArrayData]

impl<O: Clone + Offset> Clone for Utf8Array<O>

fn clone(&self) -> Utf8Array<O>

Returns a copy of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl<O: Offset> Debug for Utf8Array<O>

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl<O: Offset> Default for Utf8Array<O>

fn default() -> Self

Returns the “default value” for a type.

impl<'a, O: Offset> From<GrowableUtf8<'a, O>> for Utf8Array<O>

fn from(val: GrowableUtf8<'a, O>) -> Self

Converts to this type from the input type.

impl<O: Offset> From<MutableUtf8Array<O>> for Utf8Array<O>

fn from(other: MutableUtf8Array<O>) -> Self

Converts to this type from the input type.

impl<O: Offset> From<MutableUtf8ValuesArray<O>> for Utf8Array<O>

fn from(other: MutableUtf8ValuesArray<O>) -> Self

Converts to this type from the input type.

impl FromDataUtf8 for Utf8Array<i64>

unsafe fn from_data_unchecked_default( offsets: Buffer<i64>, values: Buffer<u8>, validity: Option<Bitmap> ) -> Self

Safety

impl<O: Offset, P: AsRef<str>> FromIterator<Option<P>> for Utf8Array<O>

fn from_iter<I: IntoIterator<Item = Option<P>>>(iter: I) -> Self

Creates a value from an iterator.

impl<O: Offset> GenericBinaryArray<O> for Utf8Array<O>

fn values(&self) -> &[u8]

The values of the array

fn offsets(&self) -> &[O]

The offsets of the array

impl<'a, O: Offset> IntoIterator for &'a Utf8Array<O>

type Item = Option<&'a str>

The type of the elements being iterated over.

type IntoIter = ZipValidity<&'a str, ArrayValuesIter<'a, Utf8Array<O>>, BitmapIter<'a>>

Which kind of iterator are we turning this into?

fn into_iter(self) -> Self::IntoIter

Creates an iterator from a value.

impl ParameterFreeDtypeStaticArray for Utf8Array<i64>

fn get_dtype() -> ArrowDataType

impl<O: Offset> PartialEq<&(dyn Array + 'static)> for Utf8Array<O>

fn eq(&self, other: &&dyn Array) -> bool

This method tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

This method tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl<O: Offset> PartialEq<Utf8Array<O>> for &dyn Array

fn eq(&self, other: &Utf8Array<O>) -> bool

This method tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

This method tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl<O: Offset> PartialEq for Utf8Array<O>

fn eq(&self, other: &Self) -> bool

This method tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

This method tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl StaticArray for Utf8Array<i64>

type ValueT<'a> = &'a str

type ZeroableValueT<'a> = Option<&'a str>

type ValueIterT<'a> = ArrayValuesIter<'a, Utf8Array<i64>>

unsafe fn value_unchecked(&self, idx: usize) -> Self::ValueT<'_>

Safety

fn values_iter(&self) -> Self::ValueIterT<'_>

fn iter( &self ) -> ZipValidity<Self::ValueT<'_>, Self::ValueIterT<'_>, BitmapIter<'_>>

fn with_validity_typed(self, validity: Option<Bitmap>) -> Self

fn full_null(length: usize, dtype: ArrowDataType) -> Self

fn get(&self, idx: usize) -> Option<Self::ValueT<'_>>

unsafe fn get_unchecked(&self, idx: usize) -> Option<Self::ValueT<'_>>

Safety

fn last(&self) -> Option<Self::ValueT<'_>>

fn value(&self, idx: usize) -> Self::ValueT<'_>

fn as_slice(&self) -> Option<&[Self::ValueT<'_>]>

fn from_vec(v: Vec<Self::ValueT<'_>>, dtype: ArrowDataType) -> Self

fn from_zeroable_vec( v: Vec<Self::ZeroableValueT<'_>>, dtype: ArrowDataType ) -> Self

impl Utf8FromIter for Utf8Array<i64>

fn from_values_iter<I, S>( iter: I, len: usize, size_hint: usize ) -> Utf8Array<i64>

where S: AsRef<str>, I: Iterator<Item = S>,

impl ValueSize for Utf8Array<i64>

fn get_values_size(&self) -> usize

Get the values size that is still “visible” to the underlying array. E.g. take the offsets into account.

impl ArrowArray for Utf8Array<i64>