Struct ValidatingExtractor 

pub struct ValidatingExtractor<'a> { /* private fields */ }

An Extract implementation that extracts entities and provides TwitterTextParseResults validation data.

Implementations

impl<'a> ValidatingExtractor<'a>

pub fn new(configuration: &Configuration) -> ValidatingExtractor<'_>

Create a new Extractor. ValidatingExtractor::prep_input must be called prior to extract.

pub fn prep_input(&mut self, s: &str) -> String

Initialize the ValidatingExtractor text length data.

pub fn new_with_nfc_input( configuration: &'a Configuration, s: &str, ) -> ValidatingExtractor<'a>

Create a new Extractor from text that is already nfc-normalized. There’s no need to call ValidatingExtractor::prep_input for this text.

Trait Implementations

impl<'a> Extract<'a> for ValidatingExtractor<'a>

type T = ExtractResult<'a>

The result type returned from the various extract methods.

type Mention = MentionResult<'a>

The result type returned from the various mention extract methods.

fn get_extract_url_without_protocol(&self) -> bool

Get whether the extractor will detect URLs without schemes, such as “example.com”.

fn set_extract_url_without_protocol( &mut self, extract_url_without_protocol: bool, )

Set whether the extractor will detect URLs without schemes, such as “example.com”.

fn extract(&self, s: &'a str, r_match: fn(Rule) -> bool) -> Self::T

Extract entities from the source text that match rules allowed by r_match.

fn create_result( &self, s: &'a str, count: usize, scanned: &mut Vec<UnprocessedEntity<'a>>, ) -> ExtractResult<'a>

Create the result type. The concrete type varies by implementation.

fn extract_reply_username(&self, s: &'a str) -> MentionResult<'a>

Create the mention result type. The concrete type varies by implementation.

fn mention_result( &self, s: &'a str, pair: Option<Pair<'a, Rule>>, ) -> MentionResult<'a>

Create a mention result type from a pest::Pair.

fn empty_result(&self) -> ExtractResult<'a>

Returns an empty result. Used when the input is invalid.

fn extract_impl(&self, s: &'a str, r_match: fn(Rule) -> bool) -> Self::T

fn extract_urls_with_indices(&self, s: &'a str) -> Self::T

Extract all URLs from the text, subject to value returned by Extract::get_extract_url_without_protocol.

fn extract_hashtags(&self, s: &'a str) -> Self::T

Extract all Hashtags from the text

fn extract_cashtags(&self, s: &'a str) -> Self::T

Extract all Cashtags from the text

fn extract_mentioned_screennames(&self, s: &'a str) -> Self::T

Extract all usernames from the text. The same as Extract::extract_mentioned_screennames_with_indices, but included for compatibility.

fn extract_mentioned_screennames_with_indices(&self, s: &'a str) -> Self::T

Extract all usernames from the text.

fn extract_mentions_or_lists_with_indices(&self, s: &'a str) -> Self::T

Extract all usernames and lists from the text.

fn extract_reply_username_impl(&self, s: &'a str) -> Self::Mention

Extract a “reply”–a username that appears at the beginning of a tweet.

fn extract_entities_with_indices(&self, s: &'a str) -> Self::T

Extract all entities from the text (Usernames, Lists, Hashtags, Cashtags, and URLs).

fn extract_scan(&self, s: &'a str) -> Self::T

Parse the text without extracting any entities.

fn entity_from_pair( &self, ue: UnprocessedEntity<'a>, start: i32, end: i32, ) -> Option<Entity<'a>>