Struct vaporetto::Sentence

pub struct Sentence { /* fields omitted */ }

Sentence with boundary annotations.

Implementations

impl Sentence

pub fn from_raw<S>(text: S) -> Result<Self> where
    S: Into<String>, 

Creates a new Sentence from a given string.

Arguments

• text - A raw string without any annotation.

Returns

A new Sentence.

Errors

If the given text is empty, an error variant will be returned.

Examples

use vaporetto::Sentence;

let s = Sentence::from_raw("How are you?");
assert!(s.is_ok());

let s = Sentence::from_raw("");
assert!(s.is_err());

pub fn to_raw_string(&self) -> &str

Gets a string without any annotation.

Returns

A reference to the string.

Examples

use vaporetto::Sentence;

let s = Sentence::from_raw("How are you?").unwrap();
assert_eq!("How are you?", s.to_raw_string());

pub fn from_tokenized<S>(tokenized_text: S) -> Result<Self> where
    S: AsRef<str>, 

Creates a new Sentence from a tokenized string.

Arguments

• tokenized_text - A tokenized string containing whitespaces for word boundaries.

Returns

A new Sentence.

Errors

This function will return an error variant when:

• tokenized_text is empty.
• tokenized_text starts/ends with a whitespace.
• tokenized_text contains consecutive whitespaces.

Examples

use vaporetto::Sentence;

let s = Sentence::from_tokenized("How are you?");
assert!(s.is_ok());

let s = Sentence::from_tokenized("How  are you?");
assert!(s.is_err());

pub fn to_tokenized_string(&self) -> Result<String>

Generates a string with whitespaces for word boundaries.

Returns

A newly allocated string containing whitespaces for word boundaries.

Errors

If the sentence contains unknown boundary, an error variant will be returned.

Examples

use vaporetto::Sentence;

let s = Sentence::from_tokenized("How are you?").unwrap();
assert_eq!("How are you?", s.to_tokenized_string().unwrap());

pub fn to_tokenized_vec(&self) -> Result<Vec<String>>

Generates a vector of words.

Returns

A newly allocated vector of words.

Errors

If the sentence contains unknown boundaries, an error variant will be returned.

Examples

use vaporetto::Sentence;

let s = Sentence::from_tokenized("How are you ?").unwrap();
assert_eq!(vec![
    "How".to_string(),
    "are".to_string(),
    "you".to_string(),
    "?".to_string(),
], s.to_tokenized_vec().unwrap());

pub fn from_partial_annotation<S>(labeled_text: S) -> Result<Self> where
    S: AsRef<str>, 

Creates a new Sentence from a string with partial annotations.

Arguments

• labeled_text - A string with partial annotations.

Returns

A new Sentence.

Errors

This function will return an error variant when:

• labeled_text is empty.
• The length of lsbeled_text is even numbers.
• labeled_text contains invalid boundary characters.

Examples

use vaporetto::Sentence;

let s = Sentence::from_partial_annotation("g-o-o-d|i-d e-a");
assert!(s.is_ok());

let s = Sentence::from_partial_annotation("b-a-d/i-d-e-a");
assert!(s.is_err());

pub fn to_partial_annotation_string(&self) -> String

Generates a string with partial annotations.

Returns

A newly allocated string with partial annotations.

Examples

use vaporetto::Sentence;

let s = Sentence::from_tokenized("How are you ?").unwrap();
assert_eq!("H-o-w|a-r-e|y-o-u|?", &s.to_partial_annotation_string());

pub fn boundaries(&self) -> &[BoundaryType]

Gets a reference to the boundary information.

Returns

A reference to the boundary information.

Examples

use vaporetto::{BoundaryType, Sentence};

let s = Sentence::from_partial_annotation("a|b-c d").unwrap();
assert_eq!(&[
    BoundaryType::WordBoundary,
    BoundaryType::NotWordBoundary,
    BoundaryType::Unknown,
], s.boundaries());

pub fn boundaries_mut(&mut self) -> &mut [BoundaryType]

Gets a mutable reference to the boundary information.

Returns

A mutable reference to the boundary information.

pub fn char_types(&self) -> &[u8]

Gets a reference to the character type information.

Returns

A reference to the character type information.

Examples

use vaporetto::Sentence;

let s = Sentence::from_raw("A1あエ漢?").unwrap();
assert_eq!(&[b'R', b'D', b'H', b'T', b'K', b'O',], s.char_types());

pub fn boundary_scores(&self) -> Option<&[f64]>

Gets a reference to the boundary score information.

Returns

If the predictor inserted, the boundary score information is returned. Otherwise, None.

pub fn get_char_pos(&self, index: usize) -> Result<usize>

Gets a character position in the code point unit.

Returns

A position in the code point unit.

Errors

index must be a valid position.

Trait Implementations

impl Clone for Sentence

fn clone(&self) -> Sentence

Returns a copy of the value.

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

Performs copy-assignment from source.

impl Debug for Sentence

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

Formats the value using the given formatter.

impl PartialEq<Sentence> for Sentence

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

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

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

This method tests for !=.

impl StructuralPartialEq for Sentence