2020-11-26 20:42:54 +01:00
|
|
|
use std::fmt::Debug;
|
2021-06-16 18:33:33 +02:00
|
|
|
use std::ops::Bound::{self, Excluded, Included};
|
2020-11-26 20:42:54 +01:00
|
|
|
|
2021-01-07 10:17:27 +01:00
|
|
|
use either::Either;
|
2021-04-07 11:57:16 +02:00
|
|
|
use heed::types::DecodeIgnore;
|
2020-11-26 20:42:54 +01:00
|
|
|
use log::debug;
|
2021-10-10 08:50:59 +02:00
|
|
|
use nom::error::{convert_error, VerboseError};
|
2020-11-26 20:42:54 +01:00
|
|
|
use roaring::RoaringBitmap;
|
|
|
|
|
2021-06-01 15:25:17 +02:00
|
|
|
use self::FilterCondition::*;
|
2021-09-28 11:50:15 +02:00
|
|
|
use super::filter_parser::{Operator, ParseContext};
|
2021-06-23 10:29:00 +02:00
|
|
|
use super::FacetNumberRange;
|
2021-09-16 11:56:18 +02:00
|
|
|
use crate::error::{Error, UserError};
|
2021-07-17 12:50:01 +02:00
|
|
|
use crate::heed_codec::facet::{
|
|
|
|
FacetLevelValueF64Codec, FacetStringLevelZeroCodec, FacetStringLevelZeroValueCodec,
|
|
|
|
};
|
2021-10-12 16:19:28 +02:00
|
|
|
use crate::{distance_between_two_points, CboRoaringBitmapCodec, FieldId, Index, Result};
|
2020-11-26 20:42:54 +01:00
|
|
|
|
|
|
|
#[derive(Debug, Clone, PartialEq)]
|
2021-06-01 15:25:17 +02:00
|
|
|
pub enum FilterCondition {
|
2021-05-03 11:45:45 +02:00
|
|
|
Operator(FieldId, Operator),
|
2020-11-26 20:42:54 +01:00
|
|
|
Or(Box<Self>, Box<Self>),
|
|
|
|
And(Box<Self>, Box<Self>),
|
2021-07-27 16:24:21 +02:00
|
|
|
Empty,
|
2020-11-26 20:42:54 +01:00
|
|
|
}
|
|
|
|
|
2021-09-15 12:57:18 +02:00
|
|
|
// impl From<std::>
|
|
|
|
|
|
|
|
//for nom
|
2021-06-01 15:25:17 +02:00
|
|
|
impl FilterCondition {
|
2021-01-07 10:17:27 +01:00
|
|
|
pub fn from_array<I, J, A, B>(
|
|
|
|
rtxn: &heed::RoTxn,
|
|
|
|
index: &Index,
|
|
|
|
array: I,
|
2021-06-14 16:46:19 +02:00
|
|
|
) -> Result<Option<FilterCondition>>
|
2021-06-16 18:33:33 +02:00
|
|
|
where
|
|
|
|
I: IntoIterator<Item = Either<J, B>>,
|
|
|
|
J: IntoIterator<Item = A>,
|
|
|
|
A: AsRef<str>,
|
|
|
|
B: AsRef<str>,
|
2021-01-07 10:17:27 +01:00
|
|
|
{
|
2021-09-28 11:50:15 +02:00
|
|
|
let mut ands: Option<FilterCondition> = None;
|
2021-01-07 10:17:27 +01:00
|
|
|
|
|
|
|
for either in array {
|
|
|
|
match either {
|
|
|
|
Either::Left(array) => {
|
|
|
|
let mut ors = None;
|
|
|
|
for rule in array {
|
2021-06-01 15:25:17 +02:00
|
|
|
let condition = FilterCondition::from_str(rtxn, index, rule.as_ref())?;
|
2021-01-07 10:17:27 +01:00
|
|
|
ors = match ors.take() {
|
|
|
|
Some(ors) => Some(Or(Box::new(ors), Box::new(condition))),
|
|
|
|
None => Some(condition),
|
|
|
|
};
|
|
|
|
}
|
|
|
|
|
|
|
|
if let Some(rule) = ors {
|
|
|
|
ands = match ands.take() {
|
|
|
|
Some(ands) => Some(And(Box::new(ands), Box::new(rule))),
|
|
|
|
None => Some(rule),
|
|
|
|
};
|
|
|
|
}
|
2021-06-16 18:33:33 +02:00
|
|
|
}
|
2021-01-07 10:17:27 +01:00
|
|
|
Either::Right(rule) => {
|
2021-06-01 15:25:17 +02:00
|
|
|
let condition = FilterCondition::from_str(rtxn, index, rule.as_ref())?;
|
2021-01-07 10:17:27 +01:00
|
|
|
ands = match ands.take() {
|
|
|
|
Some(ands) => Some(And(Box::new(ands), Box::new(condition))),
|
|
|
|
None => Some(condition),
|
|
|
|
};
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
Ok(ands)
|
|
|
|
}
|
2020-11-26 20:42:54 +01:00
|
|
|
pub fn from_str(
|
|
|
|
rtxn: &heed::RoTxn,
|
|
|
|
index: &Index,
|
|
|
|
expression: &str,
|
2021-09-15 12:57:18 +02:00
|
|
|
) -> Result<FilterCondition> {
|
|
|
|
let fields_ids_map = index.fields_ids_map(rtxn)?;
|
|
|
|
let filterable_fields = index.filterable_fields(rtxn)?;
|
|
|
|
let ctx =
|
|
|
|
ParseContext { fields_ids_map: &fields_ids_map, filterable_fields: &filterable_fields };
|
2021-09-16 11:56:18 +02:00
|
|
|
match ctx.parse_expression::<VerboseError<&str>>(expression) {
|
2021-09-15 12:57:18 +02:00
|
|
|
Ok((_, fc)) => Ok(fc),
|
2021-10-10 08:50:59 +02:00
|
|
|
Err(e) => {
|
2021-10-12 04:05:10 +02:00
|
|
|
let ve = match e {
|
|
|
|
nom::Err::Error(x) => x,
|
|
|
|
nom::Err::Failure(x) => x,
|
2021-10-10 08:50:59 +02:00
|
|
|
_ => unreachable!(),
|
2021-10-12 04:05:10 +02:00
|
|
|
};
|
2021-10-13 05:06:15 +02:00
|
|
|
Err(Error::UserError(UserError::InvalidFilter {
|
2021-10-12 04:05:10 +02:00
|
|
|
input: convert_error(expression, ve).to_string(),
|
|
|
|
}))
|
2021-10-10 08:50:59 +02:00
|
|
|
}
|
2021-09-15 12:57:18 +02:00
|
|
|
}
|
|
|
|
}
|
2021-09-28 11:50:15 +02:00
|
|
|
pub fn negate(self) -> FilterCondition {
|
2020-11-26 20:42:54 +01:00
|
|
|
match self {
|
2021-05-03 11:45:45 +02:00
|
|
|
Operator(fid, op) => match op.negate() {
|
|
|
|
(op, None) => Operator(fid, op),
|
|
|
|
(a, Some(b)) => Or(Box::new(Operator(fid, a)), Box::new(Operator(fid, b))),
|
2021-04-07 11:57:16 +02:00
|
|
|
},
|
2020-11-26 20:42:54 +01:00
|
|
|
Or(a, b) => And(Box::new(a.negate()), Box::new(b.negate())),
|
|
|
|
And(a, b) => Or(Box::new(a.negate()), Box::new(b.negate())),
|
2021-07-27 16:24:21 +02:00
|
|
|
Empty => Empty,
|
2020-11-26 20:42:54 +01:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2021-06-01 15:25:17 +02:00
|
|
|
impl FilterCondition {
|
2020-11-26 20:42:54 +01:00
|
|
|
/// Aggregates the documents ids that are part of the specified range automatically
|
|
|
|
/// going deeper through the levels.
|
2021-04-07 11:57:16 +02:00
|
|
|
fn explore_facet_number_levels(
|
|
|
|
rtxn: &heed::RoTxn,
|
|
|
|
db: heed::Database<FacetLevelValueF64Codec, CboRoaringBitmapCodec>,
|
2020-11-26 20:42:54 +01:00
|
|
|
field_id: FieldId,
|
|
|
|
level: u8,
|
2021-04-07 11:57:16 +02:00
|
|
|
left: Bound<f64>,
|
|
|
|
right: Bound<f64>,
|
2020-11-26 20:42:54 +01:00
|
|
|
output: &mut RoaringBitmap,
|
2021-06-16 18:33:33 +02:00
|
|
|
) -> Result<()> {
|
2020-11-26 20:42:54 +01:00
|
|
|
match (left, right) {
|
|
|
|
// If the request is an exact value we must go directly to the deepest level.
|
|
|
|
(Included(l), Included(r)) if l == r && level > 0 => {
|
2021-06-16 18:33:33 +02:00
|
|
|
return Self::explore_facet_number_levels(
|
|
|
|
rtxn, db, field_id, 0, left, right, output,
|
|
|
|
);
|
|
|
|
}
|
2020-11-26 20:42:54 +01:00
|
|
|
// lower TO upper when lower > upper must return no result
|
|
|
|
(Included(l), Included(r)) if l > r => return Ok(()),
|
|
|
|
(Included(l), Excluded(r)) if l >= r => return Ok(()),
|
|
|
|
(Excluded(l), Excluded(r)) if l >= r => return Ok(()),
|
|
|
|
(Excluded(l), Included(r)) if l >= r => return Ok(()),
|
|
|
|
(_, _) => (),
|
|
|
|
}
|
|
|
|
|
|
|
|
let mut left_found = None;
|
|
|
|
let mut right_found = None;
|
|
|
|
|
|
|
|
// We must create a custom iterator to be able to iterate over the
|
|
|
|
// requested range as the range iterator cannot express some conditions.
|
2021-06-23 10:29:00 +02:00
|
|
|
let iter = FacetNumberRange::new(rtxn, db, field_id, level, left, right)?;
|
2020-11-26 20:42:54 +01:00
|
|
|
|
|
|
|
debug!("Iterating between {:?} and {:?} (level {})", left, right, level);
|
|
|
|
|
|
|
|
for (i, result) in iter.enumerate() {
|
|
|
|
let ((_fid, level, l, r), docids) = result?;
|
|
|
|
debug!("{:?} to {:?} (level {}) found {} documents", l, r, level, docids.len());
|
2021-06-30 14:12:56 +02:00
|
|
|
*output |= docids;
|
2020-11-26 20:42:54 +01:00
|
|
|
// We save the leftest and rightest bounds we actually found at this level.
|
2021-06-16 18:33:33 +02:00
|
|
|
if i == 0 {
|
|
|
|
left_found = Some(l);
|
|
|
|
}
|
2020-11-26 20:42:54 +01:00
|
|
|
right_found = Some(r);
|
|
|
|
}
|
|
|
|
|
|
|
|
// Can we go deeper?
|
|
|
|
let deeper_level = match level.checked_sub(1) {
|
|
|
|
Some(level) => level,
|
|
|
|
None => return Ok(()),
|
|
|
|
};
|
|
|
|
|
|
|
|
// We must refine the left and right bounds of this range by retrieving the
|
|
|
|
// missing part in a deeper level.
|
|
|
|
match left_found.zip(right_found) {
|
|
|
|
Some((left_found, right_found)) => {
|
|
|
|
// If the bound is satisfied we avoid calling this function again.
|
|
|
|
if !matches!(left, Included(l) if l == left_found) {
|
|
|
|
let sub_right = Excluded(left_found);
|
2021-06-16 18:33:33 +02:00
|
|
|
debug!(
|
|
|
|
"calling left with {:?} to {:?} (level {})",
|
|
|
|
left, sub_right, deeper_level
|
|
|
|
);
|
|
|
|
Self::explore_facet_number_levels(
|
|
|
|
rtxn,
|
|
|
|
db,
|
|
|
|
field_id,
|
|
|
|
deeper_level,
|
|
|
|
left,
|
|
|
|
sub_right,
|
|
|
|
output,
|
|
|
|
)?;
|
2020-11-26 20:42:54 +01:00
|
|
|
}
|
|
|
|
if !matches!(right, Included(r) if r == right_found) {
|
|
|
|
let sub_left = Excluded(right_found);
|
2021-06-16 18:33:33 +02:00
|
|
|
debug!(
|
|
|
|
"calling right with {:?} to {:?} (level {})",
|
|
|
|
sub_left, right, deeper_level
|
|
|
|
);
|
|
|
|
Self::explore_facet_number_levels(
|
|
|
|
rtxn,
|
|
|
|
db,
|
|
|
|
field_id,
|
|
|
|
deeper_level,
|
|
|
|
sub_left,
|
|
|
|
right,
|
|
|
|
output,
|
|
|
|
)?;
|
2020-11-26 20:42:54 +01:00
|
|
|
}
|
2021-06-16 18:33:33 +02:00
|
|
|
}
|
2020-11-26 20:42:54 +01:00
|
|
|
None => {
|
|
|
|
// If we found nothing at this level it means that we must find
|
|
|
|
// the same bounds but at a deeper, more precise level.
|
2021-06-16 18:33:33 +02:00
|
|
|
Self::explore_facet_number_levels(
|
|
|
|
rtxn,
|
|
|
|
db,
|
|
|
|
field_id,
|
|
|
|
deeper_level,
|
|
|
|
left,
|
|
|
|
right,
|
|
|
|
output,
|
|
|
|
)?;
|
|
|
|
}
|
2020-11-26 20:42:54 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
Ok(())
|
|
|
|
}
|
|
|
|
|
2021-05-03 11:45:45 +02:00
|
|
|
fn evaluate_operator(
|
2021-04-07 11:57:16 +02:00
|
|
|
rtxn: &heed::RoTxn,
|
2020-11-26 20:42:54 +01:00
|
|
|
index: &Index,
|
2021-05-03 11:45:45 +02:00
|
|
|
numbers_db: heed::Database<FacetLevelValueF64Codec, CboRoaringBitmapCodec>,
|
2021-08-16 13:36:30 +02:00
|
|
|
strings_db: heed::Database<FacetStringLevelZeroCodec, FacetStringLevelZeroValueCodec>,
|
2020-11-26 20:42:54 +01:00
|
|
|
field_id: FieldId,
|
2021-05-03 11:45:45 +02:00
|
|
|
operator: &Operator,
|
2021-06-16 18:33:33 +02:00
|
|
|
) -> Result<RoaringBitmap> {
|
2020-11-26 20:42:54 +01:00
|
|
|
// Make sure we always bound the ranges with the field id and the level,
|
|
|
|
// as the facets values are all in the same database and prefixed by the
|
|
|
|
// field id and the level.
|
|
|
|
let (left, right) = match operator {
|
2021-09-28 11:50:15 +02:00
|
|
|
Operator::GreaterThan(val) => (Excluded(*val), Included(f64::MAX)),
|
|
|
|
Operator::GreaterThanOrEqual(val) => (Included(*val), Included(f64::MAX)),
|
|
|
|
Operator::Equal(number, string) => {
|
2021-07-17 12:50:01 +02:00
|
|
|
let (_original_value, string_docids) =
|
|
|
|
strings_db.get(rtxn, &(field_id, &string))?.unwrap_or_default();
|
2021-05-03 11:45:45 +02:00
|
|
|
let number_docids = match number {
|
|
|
|
Some(n) => {
|
|
|
|
let n = Included(*n);
|
|
|
|
let mut output = RoaringBitmap::new();
|
2021-06-16 18:33:33 +02:00
|
|
|
Self::explore_facet_number_levels(
|
|
|
|
rtxn,
|
|
|
|
numbers_db,
|
|
|
|
field_id,
|
|
|
|
0,
|
|
|
|
n,
|
|
|
|
n,
|
|
|
|
&mut output,
|
|
|
|
)?;
|
2021-05-03 11:45:45 +02:00
|
|
|
output
|
2021-06-16 18:33:33 +02:00
|
|
|
}
|
2021-05-03 11:45:45 +02:00
|
|
|
None => RoaringBitmap::new(),
|
|
|
|
};
|
|
|
|
return Ok(string_docids | number_docids);
|
2021-06-16 18:33:33 +02:00
|
|
|
}
|
2021-09-28 11:50:15 +02:00
|
|
|
Operator::NotEqual(number, string) => {
|
2021-05-03 11:45:45 +02:00
|
|
|
let all_numbers_ids = if number.is_some() {
|
|
|
|
index.number_faceted_documents_ids(rtxn, field_id)?
|
|
|
|
} else {
|
|
|
|
RoaringBitmap::new()
|
|
|
|
};
|
|
|
|
let all_strings_ids = index.string_faceted_documents_ids(rtxn, field_id)?;
|
2021-09-28 11:50:15 +02:00
|
|
|
let operator = Operator::Equal(*number, string.clone());
|
2021-06-16 18:33:33 +02:00
|
|
|
let docids = Self::evaluate_operator(
|
|
|
|
rtxn, index, numbers_db, strings_db, field_id, &operator,
|
|
|
|
)?;
|
2021-05-03 11:45:45 +02:00
|
|
|
return Ok((all_numbers_ids | all_strings_ids) - docids);
|
2021-06-16 18:33:33 +02:00
|
|
|
}
|
2021-09-28 11:50:15 +02:00
|
|
|
Operator::LowerThan(val) => (Included(f64::MIN), Excluded(*val)),
|
|
|
|
Operator::LowerThanOrEqual(val) => (Included(f64::MIN), Included(*val)),
|
|
|
|
Operator::Between(left, right) => (Included(*left), Included(*right)),
|
|
|
|
Operator::GeoLowerThan(base_point, distance) => {
|
2021-08-26 16:38:29 +02:00
|
|
|
let rtree = match index.geo_rtree(rtxn)? {
|
|
|
|
Some(rtree) => rtree,
|
2021-09-09 12:20:08 +02:00
|
|
|
None => return Ok(RoaringBitmap::new()),
|
2021-08-26 16:38:29 +02:00
|
|
|
};
|
|
|
|
|
2021-09-09 12:20:08 +02:00
|
|
|
let result = rtree
|
2021-09-07 12:11:03 +02:00
|
|
|
.nearest_neighbor_iter(base_point)
|
|
|
|
.take_while(|point| {
|
2021-09-09 12:20:08 +02:00
|
|
|
distance_between_two_points(base_point, point.geom()) < *distance
|
2021-09-07 12:11:03 +02:00
|
|
|
})
|
2021-09-09 12:20:08 +02:00
|
|
|
.map(|point| point.data)
|
|
|
|
.collect();
|
2021-08-26 16:38:29 +02:00
|
|
|
|
|
|
|
return Ok(result);
|
|
|
|
}
|
2021-09-28 11:50:15 +02:00
|
|
|
Operator::GeoGreaterThan(point, distance) => {
|
2021-08-26 16:38:29 +02:00
|
|
|
let result = Self::evaluate_operator(
|
|
|
|
rtxn,
|
|
|
|
index,
|
|
|
|
numbers_db,
|
|
|
|
strings_db,
|
|
|
|
field_id,
|
2021-10-10 08:50:59 +02:00
|
|
|
&Operator::GeoLowerThan(point.clone(), *distance),
|
2021-08-26 16:38:29 +02:00
|
|
|
)?;
|
|
|
|
let geo_faceted_doc_ids = index.geo_faceted_documents_ids(rtxn)?;
|
|
|
|
return Ok(geo_faceted_doc_ids - result);
|
|
|
|
}
|
2020-11-26 20:42:54 +01:00
|
|
|
};
|
|
|
|
|
|
|
|
// Ask for the biggest value that can exist for this specific field, if it exists
|
|
|
|
// that's fine if it don't, the value just before will be returned instead.
|
2021-05-03 11:45:45 +02:00
|
|
|
let biggest_level = numbers_db
|
2021-04-07 11:57:16 +02:00
|
|
|
.remap_data_type::<DecodeIgnore>()
|
|
|
|
.get_lower_than_or_equal_to(rtxn, &(field_id, u8::MAX, f64::MAX, f64::MAX))?
|
2020-11-26 20:42:54 +01:00
|
|
|
.and_then(|((id, level, _, _), _)| if id == field_id { Some(level) } else { None });
|
|
|
|
|
|
|
|
match biggest_level {
|
|
|
|
Some(level) => {
|
|
|
|
let mut output = RoaringBitmap::new();
|
2021-06-16 18:33:33 +02:00
|
|
|
Self::explore_facet_number_levels(
|
|
|
|
rtxn,
|
|
|
|
numbers_db,
|
|
|
|
field_id,
|
|
|
|
level,
|
|
|
|
left,
|
|
|
|
right,
|
|
|
|
&mut output,
|
|
|
|
)?;
|
2020-11-26 20:42:54 +01:00
|
|
|
Ok(output)
|
2021-06-16 18:33:33 +02:00
|
|
|
}
|
2020-11-26 20:42:54 +01:00
|
|
|
None => Ok(RoaringBitmap::new()),
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2021-06-16 18:33:33 +02:00
|
|
|
pub fn evaluate(&self, rtxn: &heed::RoTxn, index: &Index) -> Result<RoaringBitmap> {
|
2021-05-03 11:45:45 +02:00
|
|
|
let numbers_db = index.facet_id_f64_docids;
|
|
|
|
let strings_db = index.facet_id_string_docids;
|
|
|
|
|
2020-11-26 20:42:54 +01:00
|
|
|
match self {
|
2021-05-03 11:45:45 +02:00
|
|
|
Operator(fid, op) => {
|
|
|
|
Self::evaluate_operator(rtxn, index, numbers_db, strings_db, *fid, op)
|
2021-06-16 18:33:33 +02:00
|
|
|
}
|
2020-11-26 20:42:54 +01:00
|
|
|
Or(lhs, rhs) => {
|
|
|
|
let lhs = lhs.evaluate(rtxn, index)?;
|
|
|
|
let rhs = rhs.evaluate(rtxn, index)?;
|
|
|
|
Ok(lhs | rhs)
|
2021-06-16 18:33:33 +02:00
|
|
|
}
|
2020-11-26 20:42:54 +01:00
|
|
|
And(lhs, rhs) => {
|
|
|
|
let lhs = lhs.evaluate(rtxn, index)?;
|
|
|
|
let rhs = rhs.evaluate(rtxn, index)?;
|
|
|
|
Ok(lhs & rhs)
|
2021-06-16 18:33:33 +02:00
|
|
|
}
|
2021-07-27 16:24:21 +02:00
|
|
|
Empty => Ok(RoaringBitmap::new()),
|
2020-11-26 20:42:54 +01:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|