Files
tantivy/src/query/phrase_query/phrase_query.rs
2022-11-09 11:20:41 +01:00

137 lines
4.6 KiB
Rust

use super::PhraseWeight;
use crate::query::bm25::Bm25Weight;
use crate::query::{EnableScoring, Query, Weight};
use crate::schema::{Field, IndexRecordOption, Term};
/// `PhraseQuery` matches a specific sequence of words.
///
/// For instance the phrase query for `"part time"` will match
/// the sentence
///
/// **Alan just got a part time job.**
///
/// On the other hand it will not match the sentence.
///
/// **This is my favorite part of the job.**
///
/// [Slop](PhraseQuery::set_slop) allows leniency in term proximity
/// for some performance tradeof.
///
/// Using a `PhraseQuery` on a field requires positions
/// to be indexed for this field.
#[derive(Clone, Debug)]
pub struct PhraseQuery {
field: Field,
phrase_terms: Vec<(usize, Term)>,
slop: u32,
}
impl PhraseQuery {
/// Creates a new `PhraseQuery` given a list of terms.
///
/// There must be at least two terms, and all terms
/// must belong to the same field.
/// Offset for each term will be same as index in the Vector
pub fn new(terms: Vec<Term>) -> PhraseQuery {
let terms_with_offset = terms.into_iter().enumerate().collect();
PhraseQuery::new_with_offset(terms_with_offset)
}
/// Creates a new `PhraseQuery` given a list of terms and their offsets.
///
/// Can be used to provide custom offset for each term.
pub fn new_with_offset(terms: Vec<(usize, Term)>) -> PhraseQuery {
PhraseQuery::new_with_offset_and_slop(terms, 0)
}
/// Creates a new `PhraseQuery` given a list of terms, their offsets and a slop
pub fn new_with_offset_and_slop(mut terms: Vec<(usize, Term)>, slop: u32) -> PhraseQuery {
assert!(
terms.len() > 1,
"A phrase query is required to have strictly more than one term."
);
terms.sort_by_key(|&(offset, _)| offset);
let field = terms[0].1.field();
assert!(
terms[1..].iter().all(|term| term.1.field() == field),
"All terms from a phrase query must belong to the same field"
);
PhraseQuery {
field,
phrase_terms: terms,
slop,
}
}
/// Slop allowed for the phrase.
///
/// The query will match if its terms are separated by `slop` terms at most.
/// By default the slop is 0 meaning query terms need to be adjacent.
pub fn set_slop(&mut self, value: u32) {
self.slop = value;
}
/// The [`Field`] this `PhraseQuery` is targeting.
pub fn field(&self) -> Field {
self.field
}
/// `Term`s in the phrase without the associated offsets.
pub fn phrase_terms(&self) -> Vec<Term> {
self.phrase_terms
.iter()
.map(|(_, term)| term.clone())
.collect::<Vec<Term>>()
}
/// Returns the [`PhraseWeight`] for the given phrase query given a specific `searcher`.
///
/// This function is the same as [`Query::weight()`] except it returns
/// a specialized type [`PhraseWeight`] instead of a Boxed trait.
pub(crate) fn phrase_weight(
&self,
enable_scoring: EnableScoring<'_>,
) -> crate::Result<PhraseWeight> {
let schema = enable_scoring.schema();
let field_entry = schema.get_field_entry(self.field);
let has_positions = field_entry
.field_type()
.get_index_record_option()
.map(IndexRecordOption::has_positions)
.unwrap_or(false);
if !has_positions {
let field_name = field_entry.name();
return Err(crate::TantivyError::SchemaError(format!(
"Applied phrase query on field {:?}, which does not have positions indexed",
field_name
)));
}
let terms = self.phrase_terms();
let bm25_weight_opt = match enable_scoring {
EnableScoring::Enabled(searcher) => Some(Bm25Weight::for_terms(searcher, &terms)?),
EnableScoring::Disabled(_) => None,
};
let mut weight = PhraseWeight::new(self.phrase_terms.clone(), bm25_weight_opt);
if self.slop > 0 {
weight.slop(self.slop);
}
Ok(weight)
}
}
impl Query for PhraseQuery {
/// Create the weight associated with a query.
///
/// See [`Weight`].
fn weight(&self, enable_scoring: EnableScoring<'_>) -> crate::Result<Box<dyn Weight>> {
let phrase_weight = self.phrase_weight(enable_scoring)?;
Ok(Box::new(phrase_weight))
}
fn query_terms<'a>(&'a self, visitor: &mut dyn FnMut(&'a Term, bool)) {
for (_, term) in &self.phrase_terms {
visitor(term, true);
}
}
}