remove superfluous impls

Co-authored-by: Paul Masurel <paul@quickwit.io>

.github/workflows/test.yml

@@ -48,7 +48,7 @@ jobs:
     strategy:
       matrix:
         features: [
-            { label: "all", flags: "mmap,brotli-compression,lz4-compression,snappy-compression,zstd-compression,failpoints" },
+            { label: "all", flags: "mmap,stopwords,brotli-compression,lz4-compression,snappy-compression,zstd-compression,failpoints" },
             { label: "quickwit", flags: "mmap,quickwit,failpoints" }
         ]
 

CHANGELOG.md

@@ -1,6 +1,7 @@
 Tantivy 0.19
 ================================
 
+- Skip score calculation, when no scoring is required [#1646](https://github.com/quickwit-oss/tantivy/pull/1646) (@PSeitz)
 - Limit fast fields to u32 (`get_val(u32)`) [#1644](https://github.com/quickwit-oss/tantivy/pull/1644) (@PSeitz)
 - Major bugfix: Fix missing fieldnorms for u64, i64, f64, bool, bytes and date [#1620](https://github.com/quickwit-oss/tantivy/pull/1620) (@PSeitz)
 - Updated [Date Field Type](https://github.com/quickwit-oss/tantivy/pull/1396)
@@ -23,10 +24,11 @@ Tantivy 0.19
   - Add support for keyed parameter in range and histgram aggregations [#1424](https://github.com/quickwit-oss/tantivy/pull/1424) (@k-yomo)
   - Add aggregation bucket limit [#1363](https://github.com/quickwit-oss/tantivy/pull/1363) (@PSeitz)
 - Faster indexing
-  - [#1610](https://github.com/quickwit-oss/tantivy/pull/1610 (@PSeitz)
-  - [#1594](https://github.com/quickwit-oss/tantivy/pull/1594 (@PSeitz)
-  - [#1582](https://github.com/quickwit-oss/tantivy/pull/1582 (@PSeitz)
-  - [#1611](https://github.com/quickwit-oss/tantivy/pull/1611 (@PSeitz)
+  - [#1610](https://github.com/quickwit-oss/tantivy/pull/1610) (@PSeitz)
+  - [#1594](https://github.com/quickwit-oss/tantivy/pull/1594) (@PSeitz)
+  - [#1582](https://github.com/quickwit-oss/tantivy/pull/1582) (@PSeitz)
+  - [#1611](https://github.com/quickwit-oss/tantivy/pull/1611) (@PSeitz)
+  - Added a pre-configured stop word filter for various language [#1666](https://github.com/quickwit-oss/tantivy/pull/1666) (@adamreichold)
 
 Tantivy 0.18
 ================================

Cargo.toml

@@ -71,10 +71,10 @@ maplit = "1.0.2"
 matches = "0.1.9"
 pretty_assertions = "1.2.1"
 proptest = "1.0.0"
-criterion = "0.3.5"
+criterion = "0.4"
 test-log = "0.2.10"
 env_logger = "0.9.0"
-pprof = { version = "0.10.0", features = ["flamegraph", "criterion"] }
+pprof = { version = "0.11.0", features = ["flamegraph", "criterion"] }
 futures = "0.3.21"
 
 [dev-dependencies.fail]
@@ -91,8 +91,9 @@ debug-assertions = true
 overflow-checks = true
 
 [features]
-default = ["mmap", "lz4-compression" ]
+default = ["mmap", "stopwords", "lz4-compression"]
 mmap = ["fs2", "tempfile", "memmap2"]
+stopwords = []
 
 brotli-compression = ["brotli"]
 lz4-compression = ["lz4_flex"]

common/src/vint.rs

@@ -157,7 +157,7 @@ fn vint_len(data: &[u8]) -> usize {
 /// If the buffer does not start by a valid
 /// vint payload
 pub fn read_u32_vint(data: &mut &[u8]) -> u32 {
-    let (result, vlen) = read_u32_vint_no_advance(*data);
+    let (result, vlen) = read_u32_vint_no_advance(data);
     *data = &data[vlen..];
     result
 }

examples/custom_collector.rs

@@ -9,7 +9,7 @@
 
 use std::sync::Arc;
 
-use fastfield_codecs::Column;
+use fastfield_codecs::OptionalColumn;
 // ---
 // Importing tantivy...
 use tantivy::collector::{Collector, SegmentCollector};
@@ -97,7 +97,7 @@ impl Collector for StatsCollector {
 }
 
 struct StatsSegmentCollector {
-    fast_field_reader: Arc<dyn Column<u64>>,
+    fast_field_reader: Arc<dyn OptionalColumn<u64>>,
     stats: Stats,
 }
 
@@ -105,10 +105,12 @@ impl SegmentCollector for StatsSegmentCollector {
     type Fruit = Option<Stats>;
 
     fn collect(&mut self, doc: u32, _score: Score) {
-        let value = self.fast_field_reader.get_val(doc) as f64;
-        self.stats.count += 1;
-        self.stats.sum += value;
-        self.stats.squared_sum += value * value;
+        if let Some(value) = self.fast_field_reader.get_val(doc) {
+            let value = value as f64;
+            self.stats.count += 1;
+            self.stats.sum += value;
+            self.stats.squared_sum += value * value;
+        }
     }
 
     fn harvest(self) -> <Self as SegmentCollector>::Fruit {

examples/warmer.rs

@@ -51,7 +51,7 @@ impl Warmer for DynamicPriceColumn {
             let product_id_reader = segment.fast_fields().u64(self.field)?;
             let product_ids: Vec<ProductId> = segment
                 .doc_ids_alive()
-                .map(|doc| product_id_reader.get_val(doc))
+                .flat_map(|doc| product_id_reader.get_val(doc))
                 .collect();
             let mut prices_it = self.price_fetcher.fetch_prices(&product_ids).into_iter();
             let mut price_vals: Vec<Price> = Vec::new();

fastfield_codecs/benches/bench.rs

@@ -41,7 +41,7 @@ mod tests {
     ) -> Arc<dyn Column<T>> {
         let mut buffer = Vec::new();
         serialize(VecColumn::from(&column), &mut buffer, &ALL_CODEC_TYPES).unwrap();
-        open(OwnedBytes::new(buffer)).unwrap()
+        open(OwnedBytes::new(buffer)).unwrap().to_full().unwrap()
     }
 
     #[bench]
@@ -103,7 +103,7 @@ mod tests {
         let iter_gen = || data.iter().cloned();
         serialize_u128(iter_gen, data.len() as u32, &mut out).unwrap();
         let out = OwnedBytes::new(out);
-        open_u128::<u128>(out).unwrap()
+        open_u128::<u128>(out).unwrap().to_full().unwrap()
     }
 
     #[bench]
@@ -113,7 +113,7 @@ mod tests {
 
         b.iter(|| {
             let mut positions = Vec::new();
-            column.get_positions_for_value_range(
+            column.get_docids_for_value_range(
                 major_item..=major_item,
                 0..data.len() as u32,
                 &mut positions,
@@ -129,7 +129,7 @@ mod tests {
 
         b.iter(|| {
             let mut positions = Vec::new();
-            column.get_positions_for_value_range(
+            column.get_docids_for_value_range(
                 minor_item..=minor_item,
                 0..data.len() as u32,
                 &mut positions,
@@ -145,11 +145,7 @@ mod tests {
 
         b.iter(|| {
             let mut positions = Vec::new();
-            column.get_positions_for_value_range(
-                0..=u128::MAX,
-                0..data.len() as u32,
-                &mut positions,
-            );
+            column.get_docids_for_value_range(0..=u128::MAX, 0..data.len() as u32, &mut positions);
             positions
         });
     }

fastfield_codecs/src/column.rs

@@ -35,7 +35,7 @@ pub trait Column<T: PartialOrd = u64>: Send + Sync {
     ///
     /// Note that position == docid for single value fast fields
     #[inline]
-    fn get_positions_for_value_range(
+    fn get_docids_for_value_range(
         &self,
         value_range: RangeInclusive<T>,
         doc_id_range: Range<u32>,
@@ -222,13 +222,13 @@ where
         )
     }
 
-    fn get_positions_for_value_range(
+    fn get_docids_for_value_range(
         &self,
         range: RangeInclusive<Output>,
         doc_id_range: Range<u32>,
         positions: &mut Vec<u32>,
     ) {
-        self.from_column.get_positions_for_value_range(
+        self.from_column.get_docids_for_value_range(
             self.monotonic_mapping.inverse(range.start().clone())
                 ..=self.monotonic_mapping.inverse(range.end().clone()),
             doc_id_range,
@@ -240,6 +240,7 @@ where
     // and we do not have any specialized implementation anyway.
 }
 
+/// Wraps an iterator into a `Column`.
 pub struct IterColumn<T>(T);
 
 impl<T> From<T> for IterColumn<T>

fastfield_codecs/src/compact_space/mod.rs

@@ -306,13 +306,13 @@ impl Column<u128> for CompactSpaceDecompressor {
     }
 
     #[inline]
-    fn get_positions_for_value_range(
+    fn get_docids_for_value_range(
         &self,
         value_range: RangeInclusive<u128>,
-        doc_id_range: Range<u32>,
+        positions_range: Range<u32>,
         positions: &mut Vec<u32>,
     ) {
-        self.get_positions_for_value_range(value_range, doc_id_range, positions)
+        self.get_positions_for_value_range(value_range, positions_range, positions)
     }
 }
 
@@ -351,13 +351,13 @@ impl CompactSpaceDecompressor {
     pub fn get_positions_for_value_range(
         &self,
         value_range: RangeInclusive<u128>,
-        doc_id_range: Range<u32>,
+        position_range: Range<u32>,
         positions: &mut Vec<u32>,
     ) {
         if value_range.start() > value_range.end() {
             return;
         }
-        let doc_id_range = doc_id_range.start..doc_id_range.end.min(self.num_vals());
+        let position_range = position_range.start..position_range.end.min(self.num_vals());
         let from_value = *value_range.start();
         let to_value = *value_range.end();
         assert!(to_value >= from_value);
@@ -390,10 +390,10 @@ impl CompactSpaceDecompressor {
 
         let range = compact_from..=compact_to;
 
-        let scan_num_docs = doc_id_range.end - doc_id_range.start;
+        let scan_num_docs = position_range.end - position_range.start;
 
         let step_size = 4;
-        let cutoff = doc_id_range.start + scan_num_docs - scan_num_docs % step_size;
+        let cutoff = position_range.start + scan_num_docs - scan_num_docs % step_size;
 
         let mut push_if_in_range = |idx, val| {
             if range.contains(&val) {
@@ -402,7 +402,7 @@ impl CompactSpaceDecompressor {
         };
         let get_val = |idx| self.params.bit_unpacker.get(idx, &self.data);
         // unrolled loop
-        for idx in (doc_id_range.start..cutoff).step_by(step_size as usize) {
+        for idx in (position_range.start..cutoff).step_by(step_size as usize) {
             let idx1 = idx;
             let idx2 = idx + 1;
             let idx3 = idx + 2;
@@ -418,7 +418,7 @@ impl CompactSpaceDecompressor {
         }
 
         // handle rest
-        for idx in cutoff..doc_id_range.end {
+        for idx in cutoff..position_range.end {
             push_if_in_range(idx, get_val(idx as u32));
         }
     }
@@ -456,6 +456,7 @@ impl CompactSpaceDecompressor {
 mod tests {
 
     use super::*;
+    use crate::serialize::U128Header;
     use crate::{open_u128, serialize_u128};
 
     #[test]
@@ -501,7 +502,8 @@ mod tests {
         assert_eq!(amplitude, 2);
     }
 
-    fn test_all(data: OwnedBytes, expected: &[u128]) {
+    fn test_all(mut data: OwnedBytes, expected: &[u128]) {
+        let _header = U128Header::deserialize(&mut data);
         let decompressor = CompactSpaceDecompressor::open(data).unwrap();
         for (idx, expected_val) in expected.iter().cloned().enumerate() {
             let val = decompressor.get(idx as u32);
@@ -556,7 +558,8 @@ mod tests {
             4_000_211_222u128,
             333u128,
         ];
-        let data = test_aux_vals(vals);
+        let mut data = test_aux_vals(vals);
+        let _header = U128Header::deserialize(&mut data);
         let decomp = CompactSpaceDecompressor::open(data).unwrap();
         let complete_range = 0..vals.len() as u32;
         for (pos, val) in vals.iter().enumerate() {
@@ -681,7 +684,8 @@ mod tests {
             4_000_211_222u128,
             333u128,
         ];
-        let data = test_aux_vals(vals);
+        let mut data = test_aux_vals(vals);
+        let _header = U128Header::deserialize(&mut data);
         let decomp = CompactSpaceDecompressor::open(data).unwrap();
         let complete_range = 0..vals.len() as u32;
         assert_eq!(
@@ -704,7 +708,7 @@ mod tests {
         doc_id_range: Range<u32>,
     ) -> Vec<u32> {
         let mut positions = Vec::new();
-        column.get_positions_for_value_range(value_range, doc_id_range, &mut positions);
+        column.get_docids_for_value_range(value_range, doc_id_range, &mut positions);
         positions
     }
 
@@ -727,7 +731,10 @@ mod tests {
         ];
         let mut out = Vec::new();
         serialize_u128(|| vals.iter().cloned(), vals.len() as u32, &mut out).unwrap();
-        let decomp = open_u128::<u128>(OwnedBytes::new(out)).unwrap();
+        let decomp = open_u128::<u128>(OwnedBytes::new(out))
+            .unwrap()
+            .to_full()
+            .unwrap();
         let complete_range = 0..vals.len() as u32;
 
         assert_eq!(

fastfield_codecs/src/gcd.rs

@@ -59,11 +59,11 @@ mod tests {
         crate::serialize(VecColumn::from(&vals), &mut buffer, &[codec_type])?;
         let buffer = OwnedBytes::new(buffer);
         let column = crate::open::<i64>(buffer.clone())?;
-        assert_eq!(column.get_val(0), -4000i64);
-        assert_eq!(column.get_val(1), -3000i64);
-        assert_eq!(column.get_val(2), -2000i64);
-        assert_eq!(column.max_value(), (num_vals as i64 - 5) * 1000);
-        assert_eq!(column.min_value(), -4000i64);
+        assert_eq!(column.get_val(0), Some(-4000i64));
+        assert_eq!(column.get_val(1), Some(-3000i64));
+        assert_eq!(column.get_val(2), Some(-2000i64));
+        assert_eq!(column.max_value(), Some((num_vals as i64 - 5) * 1000));
+        assert_eq!(column.min_value(), Some(-4000i64));
 
         // Can't apply gcd
         let mut buffer_without_gcd = Vec::new();
@@ -101,11 +101,11 @@ mod tests {
         crate::serialize(VecColumn::from(&vals), &mut buffer, &[codec_type])?;
         let buffer = OwnedBytes::new(buffer);
         let column = crate::open::<u64>(buffer.clone())?;
-        assert_eq!(column.get_val(0), 1000u64);
-        assert_eq!(column.get_val(1), 2000u64);
-        assert_eq!(column.get_val(2), 3000u64);
-        assert_eq!(column.max_value(), num_vals as u64 * 1000);
-        assert_eq!(column.min_value(), 1000u64);
+        assert_eq!(column.get_val(0), Some(1000u64));
+        assert_eq!(column.get_val(1), Some(2000u64));
+        assert_eq!(column.get_val(2), Some(3000u64));
+        assert_eq!(column.max_value(), Some(num_vals as u64 * 1000));
+        assert_eq!(column.min_value(), Some(1000u64));
 
         // Can't apply gcd
         let mut buffer_without_gcd = Vec::new();

fastfield_codecs/src/lib.rs

@@ -24,8 +24,10 @@ use monotonic_mapping::{
     StrictlyMonotonicMappingInverter, StrictlyMonotonicMappingToInternal,
     StrictlyMonotonicMappingToInternalBaseval, StrictlyMonotonicMappingToInternalGCDBaseval,
 };
+pub use optional_column::OptionalColumn;
+use optional_column::ToOptionalColumn;
 use ownedbytes::OwnedBytes;
-use serialize::Header;
+use serialize::{Header, U128Header};
 
 mod bitpacked;
 mod blockwise_linear;
@@ -34,6 +36,7 @@ mod line;
 mod linear;
 mod monotonic_mapping;
 mod monotonic_mapping_u128;
+mod optional_column;
 
 mod column;
 mod gcd;
@@ -41,7 +44,7 @@ mod serialize;
 
 use self::bitpacked::BitpackedCodec;
 use self::blockwise_linear::BlockwiseLinearCodec;
-pub use self::column::{monotonic_map_column, Column, VecColumn};
+pub use self::column::{monotonic_map_column, Column, IterColumn, VecColumn};
 use self::linear::LinearCodec;
 pub use self::monotonic_mapping::{MonotonicallyMappableToU64, StrictlyMonotonicFn};
 pub use self::monotonic_mapping_u128::MonotonicallyMappableToU128;
@@ -92,20 +95,60 @@ impl FastFieldCodecType {
     }
 }
 
+#[derive(PartialEq, Eq, PartialOrd, Ord, Debug, Clone, Copy)]
+#[repr(u8)]
+/// Available codecs to use to encode the u128 (via [`MonotonicallyMappableToU128`]) converted data.
+pub enum U128FastFieldCodecType {
+    /// This codec takes a large number space (u128) and reduces it to a compact number space, by
+    /// removing the holes.
+    CompactSpace = 1,
+}
+
+impl BinarySerializable for U128FastFieldCodecType {
+    fn serialize<W: Write>(&self, wrt: &mut W) -> io::Result<()> {
+        self.to_code().serialize(wrt)
+    }
+
+    fn deserialize<R: io::Read>(reader: &mut R) -> io::Result<Self> {
+        let code = u8::deserialize(reader)?;
+        let codec_type: Self = Self::from_code(code)
+            .ok_or_else(|| io::Error::new(io::ErrorKind::InvalidData, "Unknown code `{code}.`"))?;
+        Ok(codec_type)
+    }
+}
+
+impl U128FastFieldCodecType {
+    pub(crate) fn to_code(self) -> u8 {
+        self as u8
+    }
+
+    pub(crate) fn from_code(code: u8) -> Option<Self> {
+        match code {
+            1 => Some(Self::CompactSpace),
+            _ => None,
+        }
+    }
+}
+
 /// Returns the correct codec reader wrapped in the `Arc` for the data.
 pub fn open_u128<Item: MonotonicallyMappableToU128>(
-    bytes: OwnedBytes,
-) -> io::Result<Arc<dyn Column<Item>>> {
+    mut bytes: OwnedBytes,
+) -> io::Result<Arc<dyn OptionalColumn<Item>>> {
+    let header = U128Header::deserialize(&mut bytes)?;
+    assert_eq!(header.codec_type, U128FastFieldCodecType::CompactSpace);
     let reader = CompactSpaceDecompressor::open(bytes)?;
     let inverted: StrictlyMonotonicMappingInverter<StrictlyMonotonicMappingToInternal<Item>> =
         StrictlyMonotonicMappingToInternal::<Item>::new().into();
-    Ok(Arc::new(monotonic_map_column(reader, inverted)))
+
+    Ok(Arc::new(ToOptionalColumn::new(Arc::new(
+        monotonic_map_column(reader, inverted),
+    ))))
 }
 
 /// Returns the correct codec reader wrapped in the `Arc` for the data.
 pub fn open<T: MonotonicallyMappableToU64>(
     mut bytes: OwnedBytes,
-) -> io::Result<Arc<dyn Column<T>>> {
+) -> io::Result<Arc<dyn OptionalColumn<T>>> {
     let header = Header::deserialize(&mut bytes)?;
     match header.codec_type {
         FastFieldCodecType::Bitpacked => open_specific_codec::<BitpackedCodec, _>(bytes, &header),
@@ -119,7 +162,7 @@ pub fn open<T: MonotonicallyMappableToU64>(
 fn open_specific_codec<C: FastFieldCodec, Item: MonotonicallyMappableToU64>(
     bytes: OwnedBytes,
     header: &Header,
-) -> io::Result<Arc<dyn Column<Item>>> {
+) -> io::Result<Arc<dyn OptionalColumn<Item>>> {
     let normalized_header = header.normalized();
     let reader = C::open_from_bytes(bytes, normalized_header)?;
     let min_value = header.min_value;
@@ -127,12 +170,16 @@ fn open_specific_codec<C: FastFieldCodec, Item: MonotonicallyMappableToU64>(
         let mapping = StrictlyMonotonicMappingInverter::from(
             StrictlyMonotonicMappingToInternalGCDBaseval::new(gcd.get(), min_value),
         );
-        Ok(Arc::new(monotonic_map_column(reader, mapping)))
+        Ok(Arc::new(ToOptionalColumn::new(Arc::new(
+            monotonic_map_column(reader, mapping),
+        ))))
     } else {
         let mapping = StrictlyMonotonicMappingInverter::from(
             StrictlyMonotonicMappingToInternalBaseval::new(min_value),
         );
-        Ok(Arc::new(monotonic_map_column(reader, mapping)))
+        Ok(Arc::new(ToOptionalColumn::new(Arc::new(
+            monotonic_map_column(reader, mapping),
+        ))))
     }
 }
 
@@ -203,8 +250,9 @@ mod tests {
         for (doc, orig_val) in data.iter().copied().enumerate() {
             let val = reader.get_val(doc as u32);
             assert_eq!(
-                val, orig_val,
-                "val `{val}` does not match orig_val {orig_val:?}, in data set {name}, data \
+                val,
+                Some(orig_val),
+                "val `{val:?}` does not match orig_val {orig_val:?}, in data set {name}, data \
                  `{data:?}`",
             );
         }
@@ -218,7 +266,7 @@ mod tests {
                 .map(|(pos, _)| pos as u32)
                 .collect();
             let mut positions = Vec::new();
-            reader.get_positions_for_value_range(
+            reader.get_docids_for_value_range(
                 data[test_rand_idx]..=data[test_rand_idx],
                 0..data.len() as u32,
                 &mut positions,

fastfield_codecs/src/main.rs

@@ -113,13 +113,16 @@ fn bench_ip() {
         (data.len() * 8) as f32 / dataset.len() as f32
     );
 
-    let decompressor = open_u128::<u128>(OwnedBytes::new(data)).unwrap();
+    let decompressor = open_u128::<u128>(OwnedBytes::new(data))
+        .unwrap()
+        .to_full()
+        .unwrap();
     // Sample some ranges
     let mut doc_values = Vec::new();
     for value in dataset.iter().take(1110).skip(1100).cloned() {
         doc_values.clear();
         print_time!("get range");
-        decompressor.get_positions_for_value_range(
+        decompressor.get_docids_for_value_range(
             value..=value,
             0..decompressor.num_vals(),
             &mut doc_values,

fastfield_codecs/src/optional_column.rs

@@ -0,0 +1,119 @@
+use std::ops::{Range, RangeInclusive};
+use std::sync::Arc;
+
+use crate::Column;
+
+/// `OptionalColumn` provides columnar access on a field.
+pub trait OptionalColumn<T: PartialOrd = u64>: Send + Sync {
+    /// Return the value associated with the given idx.
+    ///
+    /// This accessor should return as fast as possible.
+    ///
+    /// # Panics
+    ///
+    /// May panic if `idx` is greater than the column length.
+    fn get_val(&self, idx: u32) -> Option<T>;
+
+    /// Fills an output buffer with the fast field values
+    /// associated with the `DocId` going from
+    /// `start` to `start + output.len()`.
+    ///
+    /// # Panics
+    ///
+    /// Must panic if `start + output.len()` is greater than
+    /// the segment's `maxdoc`.
+    fn get_range(&self, start: u64, output: &mut [Option<T>]) {
+        for (out, idx) in output.iter_mut().zip(start..) {
+            *out = self.get_val(idx as u32);
+        }
+    }
+
+    /// Return the positions of values which are in the provided range.
+    fn get_docids_for_value_range(
+        &self,
+        value_range: RangeInclusive<T>,
+        doc_id_range: Range<u32>,
+        positions: &mut Vec<u32>,
+    ) {
+        let doc_id_range = doc_id_range.start..doc_id_range.end.min(self.num_vals());
+
+        for idx in doc_id_range {
+            let val = self.get_val(idx);
+            if let Some(val) = val {
+                if value_range.contains(&val) {
+                    positions.push(idx);
+                }
+            }
+        }
+    }
+
+    /// Returns the minimum value for this fast field.
+    ///
+    /// This min_value may not be exact.
+    /// For instance, the min value does not take in account of possible
+    /// deleted document. All values are however guaranteed to be higher than
+    /// `.min_value()`.
+    fn min_value(&self) -> Option<T>;
+
+    /// Returns the maximum value for this fast field.
+    ///
+    /// This max_value may not be exact.
+    /// For instance, the max value does not take in account of possible
+    /// deleted document. All values are however guaranteed to be higher than
+    /// `.max_value()`.
+    fn max_value(&self) -> Option<T>;
+
+    /// The number of values including `None` in the column.
+    fn num_vals(&self) -> u32;
+
+    /// Returns a iterator over the data
+    fn iter<'a>(&'a self) -> Box<dyn Iterator<Item = Option<T>> + 'a> {
+        Box::new((0..self.num_vals()).map(|idx| self.get_val(idx)))
+    }
+
+    /// return full column if all values are set and is not empty
+    fn to_full(&self) -> Option<Arc<dyn Column<T>>> {
+        None
+    }
+}
+
+/// Temporary wrapper to migrate to optional column
+pub(crate) struct ToOptionalColumn<T> {
+    column: Arc<dyn Column<T>>,
+}
+
+impl<T: PartialOrd> ToOptionalColumn<T> {
+    pub(crate) fn new(column: Arc<dyn Column<T>>) -> Self {
+        Self { column }
+    }
+}
+
+impl<T: PartialOrd> OptionalColumn<T> for ToOptionalColumn<T> {
+    #[inline]
+    fn get_val(&self, idx: u32) -> Option<T> {
+        let val = self.column.get_val(idx);
+        Some(val)
+    }
+
+    fn min_value(&self) -> Option<T> {
+        let min_value = self.column.min_value();
+        Some(min_value)
+    }
+
+    fn max_value(&self) -> Option<T> {
+        let max_value = self.column.max_value();
+        Some(max_value)
+    }
+
+    fn num_vals(&self) -> u32 {
+        self.column.num_vals()
+    }
+
+    fn iter(&self) -> Box<dyn Iterator<Item = Option<T>> + '_> {
+        Box::new(self.column.iter().map(|el| Some(el)))
+    }
+    /// return full column if all values are set and is not empty
+    fn to_full(&self) -> Option<Arc<dyn Column<T>>> {
+        Some(self.column.clone())
+    }
+}

fastfield_codecs/src/serialize.rs

@@ -35,7 +35,7 @@ use crate::monotonic_mapping::{
 };
 use crate::{
     monotonic_map_column, Column, FastFieldCodec, FastFieldCodecType, MonotonicallyMappableToU64,
-    VecColumn, ALL_CODEC_TYPES,
+    U128FastFieldCodecType, VecColumn, ALL_CODEC_TYPES,
 };
 
 /// The normalized header gives some parameters after applying the following
@@ -98,6 +98,29 @@ impl Header {
     }
 }
 
+#[derive(Debug, Copy, Clone, PartialEq, Eq)]
+pub(crate) struct U128Header {
+    pub num_vals: u32,
+    pub codec_type: U128FastFieldCodecType,
+}
+
+impl BinarySerializable for U128Header {
+    fn serialize<W: io::Write>(&self, writer: &mut W) -> io::Result<()> {
+        VInt(self.num_vals as u64).serialize(writer)?;
+        self.codec_type.serialize(writer)?;
+        Ok(())
+    }
+
+    fn deserialize<R: io::Read>(reader: &mut R) -> io::Result<Self> {
+        let num_vals = VInt::deserialize(reader)?.0 as u32;
+        let codec_type = U128FastFieldCodecType::deserialize(reader)?;
+        Ok(U128Header {
+            num_vals,
+            codec_type,
+        })
+    }
+}
+
 pub fn normalize_column<C: Column>(
     from_column: C,
     min_value: u64,
@@ -167,7 +190,11 @@ pub fn serialize_u128<F: Fn() -> I, I: Iterator<Item = u128>>(
     num_vals: u32,
     output: &mut impl io::Write,
 ) -> io::Result<()> {
-    // TODO write header, to later support more codecs
+    let header = U128Header {
+        num_vals,
+        codec_type: U128FastFieldCodecType::CompactSpace,
+    };
+    header.serialize(output)?;
     let compressor = CompactSpaceCompressor::train_from(iter_gen(), num_vals);
     compressor.compress_into(iter_gen(), output).unwrap();
 
@@ -251,13 +278,28 @@ pub fn serialize_and_load<T: MonotonicallyMappableToU64 + Ord + Default>(
 ) -> Arc<dyn Column<T>> {
     let mut buffer = Vec::new();
     super::serialize(VecColumn::from(&column), &mut buffer, &ALL_CODEC_TYPES).unwrap();
-    super::open(OwnedBytes::new(buffer)).unwrap()
+    super::open(OwnedBytes::new(buffer))
+        .unwrap()
+        .to_full()
+        .unwrap()
 }
 
 #[cfg(test)]
 mod tests {
     use super::*;
 
+    #[test]
+    fn test_serialize_deserialize_u128_header() {
+        let original = U128Header {
+            num_vals: 11,
+            codec_type: U128FastFieldCodecType::CompactSpace,
+        };
+        let mut out = Vec::new();
+        original.serialize(&mut out).unwrap();
+        let restored = U128Header::deserialize(&mut &out[..]).unwrap();
+        assert_eq!(restored, original);
+    }
+
     #[test]
     fn test_serialize_deserialize() {
         let original = [1u64, 5u64, 10u64];

query-grammar/src/query_grammar.rs

@@ -5,7 +5,8 @@ use combine::parser::range::{take_while, take_while1};
 use combine::parser::repeat::escaped;
 use combine::parser::Parser;
 use combine::{
-    attempt, choice, eof, many, many1, one_of, optional, parser, satisfy, skip_many1, value,
+    attempt, between, choice, eof, many, many1, one_of, optional, parser, satisfy, sep_by,
+    skip_many1, value,
 };
 use once_cell::sync::Lazy;
 use regex::Regex;
@@ -264,6 +265,17 @@ fn range<'a>() -> impl Parser<&'a str, Output = UserInputLeaf> {
     })
 }
 
+/// Function that parses a set out of a Stream
+/// Supports ranges like: `IN [val1 val2 val3]`
+fn set<'a>() -> impl Parser<&'a str, Output = UserInputLeaf> {
+    let term_list = between(char('['), char(']'), sep_by(term_val(), spaces()));
+
+    let set_content = ((string("IN"), spaces()), term_list).map(|(_, elements)| elements);
+
+    (optional(attempt(field_name().skip(spaces()))), set_content)
+        .map(|(field, elements)| UserInputLeaf::Set { field, elements })
+}
+
 fn negate(expr: UserInputAst) -> UserInputAst {
     expr.unary(Occur::MustNot)
 }
@@ -278,6 +290,7 @@ fn leaf<'a>() -> impl Parser<&'a str, Output = UserInputAst> {
                 string("NOT").skip(spaces1()).with(leaf()).map(negate),
             ))
             .or(attempt(range().map(UserInputAst::from)))
+            .or(attempt(set().map(UserInputAst::from)))
             .or(literal().map(UserInputAst::from))
             .parse_stream(input)
             .into_result()
@@ -747,6 +760,14 @@ mod test {
         test_parse_query_to_ast_helper("+(a b) +d", "(+(*\"a\" *\"b\") +\"d\")");
     }
 
+    #[test]
+    fn test_parse_test_query_set() {
+        test_parse_query_to_ast_helper("abc: IN [a b c]", r#""abc": IN ["a" "b" "c"]"#);
+        test_parse_query_to_ast_helper("abc: IN [1]", r#""abc": IN ["1"]"#);
+        test_parse_query_to_ast_helper("abc: IN []", r#""abc": IN []"#);
+        test_parse_query_to_ast_helper("IN [1 2]", r#"IN ["1" "2"]"#);
+    }
+
     #[test]
     fn test_parse_test_query_other() {
         test_parse_query_to_ast_helper("(+a +b) d", "(*(+\"a\" +\"b\") *\"d\")");

query-grammar/src/user_input_ast.rs

@@ -12,6 +12,10 @@ pub enum UserInputLeaf {
         lower: UserInputBound,
         upper: UserInputBound,
     },
+    Set {
+        field: Option<String>,
+        elements: Vec<String>,
+    },
 }
 
 impl Debug for UserInputLeaf {
@@ -31,6 +35,19 @@ impl Debug for UserInputLeaf {
                 upper.display_upper(formatter)?;
                 Ok(())
             }
+            UserInputLeaf::Set { field, elements } => {
+                if let Some(ref field) = field {
+                    write!(formatter, "\"{}\": ", field)?;
+                }
+                write!(formatter, "IN [")?;
+                for (i, element) in elements.iter().enumerate() {
+                    if i != 0 {
+                        write!(formatter, " ")?;
+                    }
+                    write!(formatter, "\"{}\"", element)?;
+                }
+                write!(formatter, "]")
+            }
             UserInputLeaf::All => write!(formatter, "*"),
         }
     }

src/aggregation/agg_req_with_accessor.rs

@@ -4,7 +4,7 @@ use std::rc::Rc;
 use std::sync::atomic::AtomicU32;
 use std::sync::Arc;
 
-use fastfield_codecs::Column;
+use fastfield_codecs::OptionalColumn;
 
 use super::agg_req::{Aggregation, Aggregations, BucketAggregationType, MetricAggregation};
 use super::bucket::{HistogramAggregation, RangeAggregation, TermsAggregation};
@@ -37,16 +37,16 @@ impl AggregationsWithAccessor {
 #[derive(Clone)]
 pub(crate) enum FastFieldAccessor {
     Multi(MultiValuedFastFieldReader<u64>),
-    Single(Arc<dyn Column<u64>>),
+    Single(Arc<dyn OptionalColumn<u64>>),
 }
 impl FastFieldAccessor {
-    pub fn as_single(&self) -> Option<&dyn Column<u64>> {
+    pub fn as_single(&self) -> Option<&dyn OptionalColumn<u64>> {
         match self {
             FastFieldAccessor::Multi(_) => None,
             FastFieldAccessor::Single(reader) => Some(&**reader),
         }
     }
-    pub fn into_single(self) -> Option<Arc<dyn Column<u64>>> {
+    pub fn into_single(self) -> Option<Arc<dyn OptionalColumn<u64>>> {
         match self {
             FastFieldAccessor::Multi(_) => None,
             FastFieldAccessor::Single(reader) => Some(reader),
@@ -124,7 +124,7 @@ impl BucketAggregationWithAccessor {
 pub struct MetricAggregationWithAccessor {
     pub metric: MetricAggregation,
     pub field_type: Type,
-    pub accessor: Arc<dyn Column>,
+    pub accessor: Arc<dyn OptionalColumn>,
 }
 
 impl MetricAggregationWithAccessor {

src/aggregation/agg_result.rs

@@ -4,8 +4,6 @@
 //! intermediate average results, which is the sum and the number of values. The actual average is
 //! calculated on the step from intermediate to final aggregation result tree.
 
-use std::collections::HashMap;
-
 use rustc_hash::FxHashMap;
 use serde::{Deserialize, Serialize};
 
@@ -18,7 +16,7 @@ use crate::TantivyError;
 
 #[derive(Clone, Default, Debug, PartialEq, Serialize, Deserialize)]
 /// The final aggegation result.
-pub struct AggregationResults(pub HashMap<String, AggregationResult>);
+pub struct AggregationResults(pub FxHashMap<String, AggregationResult>);
 
 impl AggregationResults {
     pub(crate) fn get_value_from_aggregation(

src/aggregation/bucket/histogram/histogram.rs

@@ -1,7 +1,7 @@
 use std::cmp::Ordering;
 use std::fmt::Display;
 
-use fastfield_codecs::Column;
+use fastfield_codecs::OptionalColumn;
 use itertools::Itertools;
 use serde::{Deserialize, Serialize};
 
@@ -263,13 +263,17 @@ impl SegmentHistogramCollector {
         req: &HistogramAggregation,
         sub_aggregation: &AggregationsWithAccessor,
         field_type: Type,
-        accessor: &dyn Column<u64>,
+        accessor: &dyn OptionalColumn<u64>,
     ) -> crate::Result<Self> {
         req.validate()?;
-        let min = f64_from_fastfield_u64(accessor.min_value(), &field_type);
-        let max = f64_from_fastfield_u64(accessor.max_value(), &field_type);
+        let min_max_u64 = accessor.min_value().zip(accessor.max_value());
+        let min_max_f64 = min_max_u64.map(|(min, max)| {
+            let min = f64_from_fastfield_u64(min, &field_type);
+            let max = f64_from_fastfield_u64(max, &field_type);
+            (min, max)
+        });
 
-        let (min, max) = get_req_min_max(req, Some((min, max)));
+        let (min, max) = get_req_min_max(req, min_max_f64);
 
         // We compute and generate the buckets range (min, max) based on the request and the min
         // max in the fast field, but this is likely not ideal when this is a subbucket, where many
@@ -331,47 +335,58 @@ impl SegmentHistogramCollector {
             .expect("unexpected fast field cardinatility");
         let mut iter = doc.chunks_exact(4);
         for docs in iter.by_ref() {
-            let val0 = self.f64_from_fastfield_u64(accessor.get_val(docs[0]));
-            let val1 = self.f64_from_fastfield_u64(accessor.get_val(docs[1]));
-            let val2 = self.f64_from_fastfield_u64(accessor.get_val(docs[2]));
-            let val3 = self.f64_from_fastfield_u64(accessor.get_val(docs[3]));
+            if let Some(val) = accessor.get_val(docs[0]) {
+                let val = self.f64_from_fastfield_u64(val);
+                let bucket_pos = get_bucket_num(val);
+                self.increment_bucket_if_in_bounds(
+                    val,
+                    &bounds,
+                    bucket_pos,
+                    docs[0],
+                    &bucket_with_accessor.sub_aggregation,
+                )?;
+            }
 
-            let bucket_pos0 = get_bucket_num(val0);
-            let bucket_pos1 = get_bucket_num(val1);
-            let bucket_pos2 = get_bucket_num(val2);
-            let bucket_pos3 = get_bucket_num(val3);
+            if let Some(val) = accessor.get_val(docs[1]) {
+                let val = self.f64_from_fastfield_u64(val);
+                let bucket_pos = get_bucket_num(val);
+                self.increment_bucket_if_in_bounds(
+                    val,
+                    &bounds,
+                    bucket_pos,
+                    docs[1],
+                    &bucket_with_accessor.sub_aggregation,
+                )?;
+            }
 
-            self.increment_bucket_if_in_bounds(
-                val0,
-                &bounds,
-                bucket_pos0,
-                docs[0],
-                &bucket_with_accessor.sub_aggregation,
-            )?;
-            self.increment_bucket_if_in_bounds(
-                val1,
-                &bounds,
-                bucket_pos1,
-                docs[1],
-                &bucket_with_accessor.sub_aggregation,
-            )?;
-            self.increment_bucket_if_in_bounds(
-                val2,
-                &bounds,
-                bucket_pos2,
-                docs[2],
-                &bucket_with_accessor.sub_aggregation,
-            )?;
-            self.increment_bucket_if_in_bounds(
-                val3,
-                &bounds,
-                bucket_pos3,
-                docs[3],
-                &bucket_with_accessor.sub_aggregation,
-            )?;
+            if let Some(val) = accessor.get_val(docs[2]) {
+                let val = self.f64_from_fastfield_u64(val);
+                let bucket_pos = get_bucket_num(val);
+                self.increment_bucket_if_in_bounds(
+                    val,
+                    &bounds,
+                    bucket_pos,
+                    docs[2],
+                    &bucket_with_accessor.sub_aggregation,
+                )?;
+            }
+
+            if let Some(val) = accessor.get_val(docs[3]) {
+                let val = self.f64_from_fastfield_u64(val);
+                let bucket_pos = get_bucket_num(val);
+                self.increment_bucket_if_in_bounds(
+                    val,
+                    &bounds,
+                    bucket_pos,
+                    docs[3],
+                    &bucket_with_accessor.sub_aggregation,
+                )?;
+            }
         }
         for &doc in iter.remainder() {
-            let val = f64_from_fastfield_u64(accessor.get_val(doc), &self.field_type);
+            let Some(val) = accessor.get_val(doc).map(|val|f64_from_fastfield_u64(val, &self.field_type)) else{
+                continue;
+            };
             if !bounds.contains(val) {
                 continue;
             }

src/aggregation/bucket/range.rs

@@ -267,20 +267,29 @@ impl SegmentRangeCollector {
             let val2 = accessor.get_val(docs[1]);
             let val3 = accessor.get_val(docs[2]);
             let val4 = accessor.get_val(docs[3]);
-            let bucket_pos1 = self.get_bucket_pos(val1);
-            let bucket_pos2 = self.get_bucket_pos(val2);
-            let bucket_pos3 = self.get_bucket_pos(val3);
-            let bucket_pos4 = self.get_bucket_pos(val4);
-
-            self.increment_bucket(bucket_pos1, docs[0], &bucket_with_accessor.sub_aggregation)?;
-            self.increment_bucket(bucket_pos2, docs[1], &bucket_with_accessor.sub_aggregation)?;
-            self.increment_bucket(bucket_pos3, docs[2], &bucket_with_accessor.sub_aggregation)?;
-            self.increment_bucket(bucket_pos4, docs[3], &bucket_with_accessor.sub_aggregation)?;
+            if let Some(val) = val1 {
+                let bucket_pos = self.get_bucket_pos(val);
+                self.increment_bucket(bucket_pos, docs[0], &bucket_with_accessor.sub_aggregation)?;
+            }
+            if let Some(val) = val2 {
+                let bucket_pos = self.get_bucket_pos(val);
+                self.increment_bucket(bucket_pos, docs[1], &bucket_with_accessor.sub_aggregation)?;
+            }
+            if let Some(val) = val3 {
+                let bucket_pos = self.get_bucket_pos(val);
+                self.increment_bucket(bucket_pos, docs[2], &bucket_with_accessor.sub_aggregation)?;
+            }
+            if let Some(val) = val4 {
+                let bucket_pos = self.get_bucket_pos(val);
+                self.increment_bucket(bucket_pos, docs[3], &bucket_with_accessor.sub_aggregation)?;
+            }
         }
         for &doc in iter.remainder() {
             let val = accessor.get_val(doc);
-            let bucket_pos = self.get_bucket_pos(val);
-            self.increment_bucket(bucket_pos, doc, &bucket_with_accessor.sub_aggregation)?;
+            if let Some(val) = val {
+                let bucket_pos = self.get_bucket_pos(val);
+                self.increment_bucket(bucket_pos, doc, &bucket_with_accessor.sub_aggregation)?;
+            }
         }
         if force_flush {
             for bucket in &mut self.buckets {

src/aggregation/intermediate_agg_result.rs

@@ -3,7 +3,6 @@
 //! indices.
 
 use std::cmp::Ordering;
-use std::collections::HashMap;
 
 use itertools::Itertools;
 use rustc_hash::FxHashMap;
@@ -51,7 +50,7 @@ impl IntermediateAggregationResults {
         // Important assumption:
         // When the tree contains buckets/metric, we expect it to have all buckets/metrics from the
         // request
-        let mut results: HashMap<String, AggregationResult> = HashMap::new();
+        let mut results: FxHashMap<String, AggregationResult> = FxHashMap::default();
 
         if let Some(buckets) = self.buckets {
             convert_and_add_final_buckets_to_result(&mut results, buckets, &req.buckets)?
@@ -132,7 +131,7 @@ impl IntermediateAggregationResults {
 }
 
 fn convert_and_add_final_metrics_to_result(
-    results: &mut HashMap<String, AggregationResult>,
+    results: &mut FxHashMap<String, AggregationResult>,
     metrics: VecWithNames<IntermediateMetricResult>,
 ) {
     results.extend(
@@ -143,7 +142,7 @@ fn convert_and_add_final_metrics_to_result(
 }
 
 fn add_empty_final_metrics_to_result(
-    results: &mut HashMap<String, AggregationResult>,
+    results: &mut FxHashMap<String, AggregationResult>,
     req_metrics: &VecWithNames<MetricAggregation>,
 ) -> crate::Result<()> {
     results.extend(req_metrics.iter().map(|(key, req)| {
@@ -157,7 +156,7 @@ fn add_empty_final_metrics_to_result(
 }
 
 fn add_empty_final_buckets_to_result(
-    results: &mut HashMap<String, AggregationResult>,
+    results: &mut FxHashMap<String, AggregationResult>,
     req_buckets: &VecWithNames<BucketAggregationInternal>,
 ) -> crate::Result<()> {
     let requested_buckets = req_buckets.iter();
@@ -169,7 +168,7 @@ fn add_empty_final_buckets_to_result(
 }
 
 fn convert_and_add_final_buckets_to_result(
-    results: &mut HashMap<String, AggregationResult>,
+    results: &mut FxHashMap<String, AggregationResult>,
     buckets: VecWithNames<IntermediateBucketResult>,
     req_buckets: &VecWithNames<BucketAggregationInternal>,
 ) -> crate::Result<()> {

src/aggregation/metric/average.rs

@@ -1,6 +1,6 @@
 use std::fmt::Debug;
 
-use fastfield_codecs::Column;
+use fastfield_codecs::OptionalColumn;
 use serde::{Deserialize, Serialize};
 
 use crate::aggregation::f64_from_fastfield_u64;
@@ -57,26 +57,33 @@ impl SegmentAverageCollector {
             data: Default::default(),
         }
     }
-    pub(crate) fn collect_block(&mut self, doc: &[DocId], field: &dyn Column<u64>) {
+    pub(crate) fn collect_block(&mut self, doc: &[DocId], field: &dyn OptionalColumn<u64>) {
         let mut iter = doc.chunks_exact(4);
         for docs in iter.by_ref() {
-            let val1 = field.get_val(docs[0]);
-            let val2 = field.get_val(docs[1]);
-            let val3 = field.get_val(docs[2]);
-            let val4 = field.get_val(docs[3]);
-            let val1 = f64_from_fastfield_u64(val1, &self.field_type);
-            let val2 = f64_from_fastfield_u64(val2, &self.field_type);
-            let val3 = f64_from_fastfield_u64(val3, &self.field_type);
-            let val4 = f64_from_fastfield_u64(val4, &self.field_type);
-            self.data.collect(val1);
-            self.data.collect(val2);
-            self.data.collect(val3);
-            self.data.collect(val4);
+            if let Some(val) = field.get_val(docs[0]) {
+                let val = f64_from_fastfield_u64(val, &self.field_type);
+                self.data.collect(val);
+            }
+            if let Some(val) = field.get_val(docs[1]) {
+                let val = f64_from_fastfield_u64(val, &self.field_type);
+                self.data.collect(val);
+            }
+
+            if let Some(val) = field.get_val(docs[2]) {
+                let val = f64_from_fastfield_u64(val, &self.field_type);
+                self.data.collect(val);
+            }
+
+            if let Some(val) = field.get_val(docs[3]) {
+                let val = f64_from_fastfield_u64(val, &self.field_type);
+                self.data.collect(val);
+            }
         }
         for &doc in iter.remainder() {
-            let val = field.get_val(doc);
-            let val = f64_from_fastfield_u64(val, &self.field_type);
-            self.data.collect(val);
+            if let Some(val) = field.get_val(doc) {
+                let val = f64_from_fastfield_u64(val, &self.field_type);
+                self.data.collect(val);
+            }
         }
     }
 }

src/aggregation/metric/stats.rs

@@ -1,4 +1,4 @@
-use fastfield_codecs::Column;
+use fastfield_codecs::OptionalColumn;
 use serde::{Deserialize, Serialize};
 
 use crate::aggregation::f64_from_fastfield_u64;
@@ -163,26 +163,31 @@ impl SegmentStatsCollector {
             stats: IntermediateStats::default(),
         }
     }
-    pub(crate) fn collect_block(&mut self, doc: &[DocId], field: &dyn Column<u64>) {
+    pub(crate) fn collect_block(&mut self, doc: &[DocId], field: &dyn OptionalColumn<u64>) {
         let mut iter = doc.chunks_exact(4);
         for docs in iter.by_ref() {
-            let val1 = field.get_val(docs[0]);
-            let val2 = field.get_val(docs[1]);
-            let val3 = field.get_val(docs[2]);
-            let val4 = field.get_val(docs[3]);
-            let val1 = f64_from_fastfield_u64(val1, &self.field_type);
-            let val2 = f64_from_fastfield_u64(val2, &self.field_type);
-            let val3 = f64_from_fastfield_u64(val3, &self.field_type);
-            let val4 = f64_from_fastfield_u64(val4, &self.field_type);
-            self.stats.collect(val1);
-            self.stats.collect(val2);
-            self.stats.collect(val3);
-            self.stats.collect(val4);
+            if let Some(val) = field.get_val(docs[0]) {
+                let val = f64_from_fastfield_u64(val, &self.field_type);
+                self.stats.collect(val);
+            }
+            if let Some(val) = field.get_val(docs[1]) {
+                let val = f64_from_fastfield_u64(val, &self.field_type);
+                self.stats.collect(val);
+            }
+            if let Some(val) = field.get_val(docs[2]) {
+                let val = f64_from_fastfield_u64(val, &self.field_type);
+                self.stats.collect(val);
+            }
+            if let Some(val) = field.get_val(docs[3]) {
+                let val = f64_from_fastfield_u64(val, &self.field_type);
+                self.stats.collect(val);
+            }
         }
         for &doc in iter.remainder() {
-            let val = field.get_val(doc);
-            let val = f64_from_fastfield_u64(val, &self.field_type);
-            self.stats.collect(val);
+            if let Some(val) = field.get_val(doc) {
+                let val = f64_from_fastfield_u64(val, &self.field_type);
+                self.stats.collect(val);
+            }
         }
     }
 }

src/collector/facet_collector.rs

@@ -616,7 +616,7 @@ mod tests {
             .map(|mut doc| {
                 doc.add_facet(
                     facet_field,
-                    &format!("/facet/{}", thread_rng().sample(&uniform)),
+                    &format!("/facet/{}", thread_rng().sample(uniform)),
                 );
                 doc
             })

src/collector/filter_collector_wrapper.rs

@@ -130,7 +130,9 @@ where
 
         let fast_field_reader = segment_reader
             .fast_fields()
-            .typed_fast_field_reader(self.field)?;
+            .typed_fast_field_reader(self.field)?
+            .to_full()
+            .expect("temp migration solution");
 
         let segment_collector = self
             .collector

src/collector/histogram_collector.rs

@@ -112,7 +112,11 @@ impl Collector for HistogramCollector {
         _segment_local_id: crate::SegmentOrdinal,
         segment: &crate::SegmentReader,
     ) -> crate::Result<Self::Child> {
-        let ff_reader = segment.fast_fields().u64_lenient(self.field)?;
+        let ff_reader = segment
+            .fast_fields()
+            .u64_lenient(self.field)?
+            .to_full()
+            .expect("temp migration solution");
         Ok(SegmentHistogramCollector {
             histogram_computer: HistogramComputer {
                 counts: vec![0; self.num_buckets],

src/collector/mod.rs

@@ -172,17 +172,33 @@ pub trait Collector: Sync + Send {
     ) -> crate::Result<<Self::Child as SegmentCollector>::Fruit> {
         let mut segment_collector = self.for_segment(segment_ord as u32, reader)?;
 
-        if let Some(alive_bitset) = reader.alive_bitset() {
-            weight.for_each(reader, &mut |doc, score| {
-                if alive_bitset.is_alive(doc) {
+        match (reader.alive_bitset(), self.requires_scoring()) {
+            (Some(alive_bitset), true) => {
+                weight.for_each(reader, &mut |doc, score| {
+                    if alive_bitset.is_alive(doc) {
+                        segment_collector.collect(doc, score);
+                    }
+                })?;
+            }
+            (Some(alive_bitset), false) => {
+                weight.for_each_no_score(reader, &mut |doc| {
+                    if alive_bitset.is_alive(doc) {
+                        segment_collector.collect(doc, 0.0);
+                    }
+                })?;
+            }
+            (None, true) => {
+                weight.for_each(reader, &mut |doc, score| {
                     segment_collector.collect(doc, score);
-                }
-            })?;
-        } else {
-            weight.for_each(reader, &mut |doc, score| {
-                segment_collector.collect(doc, score);
-            })?;
+                })?;
+            }
+            (None, false) => {
+                weight.for_each_no_score(reader, &mut |doc| {
+                    segment_collector.collect(doc, 0.0);
+                })?;
+            }
         }
+
         Ok(segment_collector.harvest())
     }
 }

src/collector/tests.rs

@@ -1,6 +1,6 @@
 use std::sync::Arc;
 
-use fastfield_codecs::Column;
+use fastfield_codecs::OptionalColumn;
 
 use super::*;
 use crate::collector::{Count, FilterCollector, TopDocs};
@@ -160,7 +160,7 @@ pub struct FastFieldTestCollector {
 
 pub struct FastFieldSegmentCollector {
     vals: Vec<u64>,
-    reader: Arc<dyn Column<u64>>,
+    reader: Arc<dyn OptionalColumn<u64>>,
 }
 
 impl FastFieldTestCollector {
@@ -202,7 +202,9 @@ impl SegmentCollector for FastFieldSegmentCollector {
 
     fn collect(&mut self, doc: DocId, _score: Score) {
         let val = self.reader.get_val(doc);
-        self.vals.push(val);
+        if let Some(val) = val {
+            self.vals.push(val);
+        }
     }
 
     fn harvest(self) -> Vec<u64> {

src/collector/top_score_collector.rs

@@ -156,7 +156,9 @@ impl CustomScorer<u64> for ScorerByField {
         // The conversion will then happen only on the top-K docs.
         let ff_reader = segment_reader
             .fast_fields()
-            .typed_fast_field_reader(self.field)?;
+            .typed_fast_field_reader(self.field)?
+            .to_full()
+            .expect("temp migration solution");
         Ok(ScorerByFastFieldReader { ff_reader })
     }
 }
@@ -458,7 +460,7 @@ impl TopDocs {
     ///
     ///             // We can now define our actual scoring function
     ///             move |doc: DocId, original_score: Score| {
-    ///                 let popularity: u64 = popularity_reader.get_val(doc);
+    ///                 let popularity: u64 = popularity_reader.get_val(doc).unwrap();
     ///                 // Well.. For the sake of the example we use a simple logarithm
     ///                 // function.
     ///                 let popularity_boost_score = ((2u64 + popularity) as Score).log2();
@@ -567,8 +569,8 @@ impl TopDocs {
     ///
     ///             // We can now define our actual scoring function
     ///             move |doc: DocId| {
-    ///                 let popularity: u64 = popularity_reader.get_val(doc);
-    ///                 let boosted: u64 = boosted_reader.get_val(doc);
+    ///                 let popularity: u64 = popularity_reader.get_val(doc).unwrap();
+    ///                 let boosted: u64 = boosted_reader.get_val(doc).unwrap();
     ///                 // Score do not have to be `f64` in tantivy.
     ///                 // Here we return a couple to get lexicographical order
     ///                 // for free.

src/core/index_meta.rs

@@ -133,7 +133,7 @@ impl SegmentMeta {
     /// associated with a segment component.
     pub fn relative_path(&self, component: SegmentComponent) -> PathBuf {
         let mut path = self.id().uuid_string();
-        path.push_str(&*match component {
+        path.push_str(&match component {
             SegmentComponent::Postings => ".idx".to_string(),
             SegmentComponent::Positions => ".pos".to_string(),
             SegmentComponent::Terms => ".term".to_string(),

src/core/inverted_index_reader.rs

@@ -230,6 +230,18 @@ impl InvertedIndexReader {
         Ok(())
     }
 
+    /// Read the block postings for all terms.
+    /// This method is for an advanced usage only.
+    ///
+    /// If you know which terms to pre-load, prefer using [`Self::warm_postings`] instead.
+    pub async fn warm_postings_full(&self, with_positions: bool) -> crate::AsyncIoResult<()> {
+        self.postings_file_slice.read_bytes_async().await?;
+        if with_positions {
+            self.positions_file_slice.read_bytes_async().await?;
+        }
+        Ok(())
+    }
+
     /// Returns the number of documents containing the term asynchronously.
     pub async fn doc_freq_async(&self, term: &Term) -> crate::AsyncIoResult<u32> {
         Ok(self

src/core/searcher.rs

@@ -4,7 +4,7 @@ use std::{fmt, io};
 
 use crate::collector::Collector;
 use crate::core::{Executor, SegmentReader};
-use crate::query::Query;
+use crate::query::{EnableScoring, Query};
 use crate::schema::{Document, Schema, Term};
 use crate::space_usage::SearcherSpaceUsage;
 use crate::store::{CacheStats, StoreReader};
@@ -199,7 +199,12 @@ impl Searcher {
         executor: &Executor,
     ) -> crate::Result<C::Fruit> {
         let scoring_enabled = collector.requires_scoring();
-        let weight = query.weight(self, scoring_enabled)?;
+        let enabled_scoring = if scoring_enabled {
+            EnableScoring::Enabled(self)
+        } else {
+            EnableScoring::Disabled(self.schema())
+        };
+        let weight = query.weight(enabled_scoring)?;
         let segment_readers = self.segment_readers();
         let fruits = executor.map(
             |(segment_ord, segment_reader)| {

src/directory/directory.rs

@@ -55,7 +55,7 @@ impl<T: Send + Sync + 'static> From<Box<T>> for DirectoryLock {
 
 impl Drop for DirectoryLockGuard {
     fn drop(&mut self) {
-        if let Err(e) = self.directory.delete(&*self.path) {
+        if let Err(e) = self.directory.delete(&self.path) {
             error!("Failed to remove the lock file. {:?}", e);
         }
     }

src/fastfield/bytes/mod.rs

@@ -6,7 +6,7 @@ pub use self::writer::BytesFastFieldWriter;
 
 #[cfg(test)]
 mod tests {
-    use crate::query::TermQuery;
+    use crate::query::{EnableScoring, TermQuery};
     use crate::schema::{BytesOptions, IndexRecordOption, Schema, Value, FAST, INDEXED, STORED};
     use crate::{DocAddress, DocSet, Index, Searcher, Term};
 
@@ -82,7 +82,7 @@ mod tests {
         let field = searcher.schema().get_field("string_bytes").unwrap();
         let term = Term::from_field_bytes(field, b"lucene".as_ref());
         let term_query = TermQuery::new(term, IndexRecordOption::Basic);
-        let term_weight = term_query.specialized_weight(&searcher, true)?;
+        let term_weight = term_query.specialized_weight(EnableScoring::Enabled(&searcher))?;
         let term_scorer = term_weight.specialized_scorer(searcher.segment_reader(0), 1.0)?;
         assert_eq!(term_scorer.doc(), 0u32);
         Ok(())
@@ -95,7 +95,8 @@ mod tests {
         let field = searcher.schema().get_field("string_bytes").unwrap();
         let term = Term::from_field_bytes(field, b"lucene".as_ref());
         let term_query = TermQuery::new(term, IndexRecordOption::Basic);
-        let term_weight_err = term_query.specialized_weight(&searcher, false);
+        let term_weight_err =
+            term_query.specialized_weight(EnableScoring::Disabled(searcher.schema()));
         assert!(matches!(
             term_weight_err,
             Err(crate::TantivyError::SchemaError(_))

src/fastfield/bytes/reader.rs

@@ -1,10 +1,9 @@
-use std::ops::Range;
 use std::sync::Arc;
 
 use fastfield_codecs::Column;
 
 use crate::directory::{FileSlice, OwnedBytes};
-use crate::fastfield::MultiValueLength;
+use crate::fastfield::MultiValueIndex;
 use crate::DocId;
 
 /// Reader for byte array fast fields
@@ -19,7 +18,7 @@ use crate::DocId;
 /// and the start index for the next document, and keeping the bytes in between.
 #[derive(Clone)]
 pub struct BytesFastFieldReader {
-    idx_reader: Arc<dyn Column<u64>>,
+    idx_reader: MultiValueIndex,
     values: OwnedBytes,
 }
 
@@ -29,41 +28,31 @@ impl BytesFastFieldReader {
         values_file: FileSlice,
     ) -> crate::Result<BytesFastFieldReader> {
         let values = values_file.read_bytes()?;
-        Ok(BytesFastFieldReader { idx_reader, values })
+        Ok(BytesFastFieldReader {
+            idx_reader: MultiValueIndex::new(idx_reader),
+            values,
+        })
     }
 
-    fn range(&self, doc: DocId) -> Range<u32> {
-        let start = self.idx_reader.get_val(doc) as u32;
-        let end = self.idx_reader.get_val(doc + 1) as u32;
-        start..end
+    /// returns the multivalue index
+    pub fn get_index_reader(&self) -> &MultiValueIndex {
+        &self.idx_reader
     }
 
     /// Returns the bytes associated with the given `doc`
     pub fn get_bytes(&self, doc: DocId) -> &[u8] {
-        let range = self.range(doc);
+        let range = self.idx_reader.range(doc);
         &self.values.as_slice()[range.start as usize..range.end as usize]
     }
 
     /// Returns the length of the bytes associated with the given `doc`
     pub fn num_bytes(&self, doc: DocId) -> u64 {
-        let range = self.range(doc);
+        let range = self.idx_reader.range(doc);
         (range.end - range.start) as u64
     }
 
     /// Returns the overall number of bytes in this bytes fast field.
-    pub fn total_num_bytes(&self) -> u64 {
-        self.values.len() as u64
-    }
-}
-
-impl MultiValueLength for BytesFastFieldReader {
-    fn get_range(&self, doc_id: DocId) -> std::ops::Range<u32> {
-        self.range(doc_id)
-    }
-    fn get_len(&self, doc_id: DocId) -> u64 {
-        self.num_bytes(doc_id)
-    }
-    fn get_total_len(&self) -> u64 {
-        self.total_num_bytes()
+    pub fn total_num_bytes(&self) -> u32 {
+        self.values.len() as u32
     }
 }

src/fastfield/mod.rs

@@ -27,8 +27,8 @@ pub use self::error::{FastFieldNotAvailableError, Result};
 pub use self::facet_reader::FacetReader;
 pub(crate) use self::multivalued::{get_fastfield_codecs_for_multivalue, MultivalueStartIndex};
 pub use self::multivalued::{
-    MultiValueU128FastFieldWriter, MultiValuedFastFieldReader, MultiValuedFastFieldWriter,
-    MultiValuedU128FastFieldReader,
+    MultiValueIndex, MultiValueU128FastFieldWriter, MultiValuedFastFieldReader,
+    MultiValuedFastFieldWriter, MultiValuedU128FastFieldReader,
 };
 pub use self::readers::FastFieldReaders;
 pub(crate) use self::readers::{type_and_cardinality, FastType};
@@ -36,7 +36,7 @@ pub use self::serializer::{Column, CompositeFastFieldSerializer};
 use self::writer::unexpected_value;
 pub use self::writer::{FastFieldsWriter, IntFastFieldWriter};
 use crate::schema::{Type, Value};
-use crate::{DateTime, DocId};
+use crate::DateTime;
 
 mod alive_bitset;
 mod bytes;
@@ -47,17 +47,6 @@ mod readers;
 mod serializer;
 mod writer;
 
-/// Trait for `BytesFastFieldReader` and `MultiValuedFastFieldReader` to return the length of data
-/// for a doc_id
-pub trait MultiValueLength {
-    /// returns the positions for a docid
-    fn get_range(&self, doc_id: DocId) -> std::ops::Range<u32>;
-    /// returns the num of values associated with a doc_id
-    fn get_len(&self, doc_id: DocId) -> u64;
-    /// returns the sum of num values for all doc_ids
-    fn get_total_len(&self) -> u64;
-}
-
 /// Trait for types that are allowed for fast fields:
 /// (u64, i64 and f64, bool, DateTime).
 pub trait FastValue:
@@ -221,7 +210,7 @@ mod tests {
         assert_eq!(file.len(), 25);
         let composite_file = CompositeFile::open(&file)?;
         let fast_field_bytes = composite_file.open_read(*FIELD).unwrap().read_bytes()?;
-        let fast_field_reader = open::<u64>(fast_field_bytes)?;
+        let fast_field_reader = open::<u64>(fast_field_bytes)?.to_full().unwrap();
         assert_eq!(fast_field_reader.get_val(0), 13u64);
         assert_eq!(fast_field_reader.get_val(1), 14u64);
         assert_eq!(fast_field_reader.get_val(2), 2u64);
@@ -274,7 +263,7 @@ mod tests {
                 .open_read(*FIELD)
                 .unwrap()
                 .read_bytes()?;
-            let fast_field_reader = open::<u64>(data)?;
+            let fast_field_reader = open::<u64>(data)?.to_full().unwrap();
             assert_eq!(fast_field_reader.get_val(0), 4u64);
             assert_eq!(fast_field_reader.get_val(1), 14_082_001u64);
             assert_eq!(fast_field_reader.get_val(2), 3_052u64);
@@ -315,7 +304,7 @@ mod tests {
                 .open_read(*FIELD)
                 .unwrap()
                 .read_bytes()?;
-            let fast_field_reader = open::<u64>(data)?;
+            let fast_field_reader = open::<u64>(data)?.to_full().unwrap();
             for doc in 0..10_000 {
                 assert_eq!(fast_field_reader.get_val(doc), 100_000u64);
             }
@@ -354,7 +343,7 @@ mod tests {
                 .open_read(*FIELD)
                 .unwrap()
                 .read_bytes()?;
-            let fast_field_reader = open::<u64>(data)?;
+            let fast_field_reader = open::<u64>(data)?.to_full().unwrap();
             assert_eq!(fast_field_reader.get_val(0), 0u64);
             for doc in 1..10_001 {
                 assert_eq!(
@@ -397,7 +386,7 @@ mod tests {
                 .open_read(i64_field)
                 .unwrap()
                 .read_bytes()?;
-            let fast_field_reader = open::<i64>(data)?;
+            let fast_field_reader = open::<i64>(data)?.to_full().unwrap();
 
             assert_eq!(fast_field_reader.min_value(), -100i64);
             assert_eq!(fast_field_reader.max_value(), 9_999i64);
@@ -440,7 +429,7 @@ mod tests {
                 .open_read(i64_field)
                 .unwrap()
                 .read_bytes()?;
-            let fast_field_reader = open::<i64>(data)?;
+            let fast_field_reader = open::<i64>(data)?.to_full().unwrap();
             assert_eq!(fast_field_reader.get_val(0), 0i64);
         }
         Ok(())
@@ -481,7 +470,7 @@ mod tests {
                 .open_read(*FIELD)
                 .unwrap()
                 .read_bytes()?;
-            let fast_field_reader = open::<u64>(data)?;
+            let fast_field_reader = open::<u64>(data)?.to_full().unwrap();
 
             for a in 0..n {
                 assert_eq!(fast_field_reader.get_val(a as u32), permutation[a as usize]);
@@ -774,19 +763,28 @@ mod tests {
         let dates_fast_field = fast_fields.dates(multi_date_field).unwrap();
         let mut dates = vec![];
         {
-            assert_eq!(date_fast_field.get_val(0).into_timestamp_micros(), 1i64);
+            assert_eq!(
+                date_fast_field.get_val(0).unwrap().into_timestamp_micros(),
+                1i64
+            );
             dates_fast_field.get_vals(0u32, &mut dates);
             assert_eq!(dates.len(), 2);
             assert_eq!(dates[0].into_timestamp_micros(), 2i64);
             assert_eq!(dates[1].into_timestamp_micros(), 3i64);
         }
         {
-            assert_eq!(date_fast_field.get_val(1).into_timestamp_micros(), 4i64);
+            assert_eq!(
+                date_fast_field.get_val(1).unwrap().into_timestamp_micros(),
+                4i64
+            );
             dates_fast_field.get_vals(1u32, &mut dates);
             assert!(dates.is_empty());
         }
         {
-            assert_eq!(date_fast_field.get_val(2).into_timestamp_micros(), 0i64);
+            assert_eq!(
+                date_fast_field.get_val(2).unwrap().into_timestamp_micros(),
+                0i64
+            );
             dates_fast_field.get_vals(2u32, &mut dates);
             assert_eq!(dates.len(), 2);
             assert_eq!(dates[0].into_timestamp_micros(), 5i64);
@@ -836,7 +834,7 @@ mod tests {
         assert_eq!(file.len(), 24);
         let composite_file = CompositeFile::open(&file)?;
         let data = composite_file.open_read(field).unwrap().read_bytes()?;
-        let fast_field_reader = open::<bool>(data)?;
+        let fast_field_reader = open::<bool>(data)?.to_full().unwrap();
         assert_eq!(fast_field_reader.get_val(0), true);
         assert_eq!(fast_field_reader.get_val(1), false);
         assert_eq!(fast_field_reader.get_val(2), true);
@@ -874,7 +872,7 @@ mod tests {
         assert_eq!(file.len(), 36);
         let composite_file = CompositeFile::open(&file)?;
         let data = composite_file.open_read(field).unwrap().read_bytes()?;
-        let fast_field_reader = open::<bool>(data)?;
+        let fast_field_reader = open::<bool>(data)?.to_full().unwrap();
         for i in 0..25 {
             assert_eq!(fast_field_reader.get_val(i * 2), true);
             assert_eq!(fast_field_reader.get_val(i * 2 + 1), false);
@@ -905,7 +903,7 @@ mod tests {
         let composite_file = CompositeFile::open(&file)?;
         assert_eq!(file.len(), 23);
         let data = composite_file.open_read(field).unwrap().read_bytes()?;
-        let fast_field_reader = open::<bool>(data)?;
+        let fast_field_reader = open::<bool>(data)?.to_full().unwrap();
         assert_eq!(fast_field_reader.get_val(0), false);
 
         Ok(())
@@ -973,7 +971,9 @@ mod tests {
         let composite_file = CompositeFile::open(&file)?;
         let file = composite_file.open_read(*FIELD).unwrap();
         let len = file.len();
-        let test_fastfield = open::<DateTime>(file.read_bytes()?)?;
+        let test_fastfield = open::<DateTime>(file.read_bytes()?)?
+            .to_full()
+            .expect("temp migration solution");
 
         for (i, time) in times.iter().enumerate() {
             assert_eq!(test_fastfield.get_val(i as u32), time.truncate(precision));

src/fastfield/multivalued/index.rs

@@ -0,0 +1,148 @@
+use std::ops::Range;
+use std::sync::Arc;
+
+use fastfield_codecs::Column;
+
+use crate::DocId;
+
+#[derive(Clone)]
+/// Index to resolve value range for given doc_id.
+/// Starts at 0.
+pub struct MultiValueIndex {
+    idx: Arc<dyn Column<u64>>,
+}
+
+impl MultiValueIndex {
+    pub(crate) fn new(idx: Arc<dyn Column<u64>>) -> Self {
+        Self { idx }
+    }
+
+    /// Returns `[start, end)`, such that the values associated with
+    /// the given document are `start..end`.
+    #[inline]
+    pub(crate) fn range(&self, doc: DocId) -> Range<u32> {
+        let start = self.idx.get_val(doc) as u32;
+        let end = self.idx.get_val(doc + 1) as u32;
+        start..end
+    }
+
+    /// Given a range of documents, returns the Range of value offsets fo
+    /// these documents.
+    ///
+    /// For instance, `given start_doc..end_doc`,
+    /// if we assume Document #start_doc end #end_doc both
+    /// have values, this function returns `start..end`
+    /// such that `value_column.get(start_doc)` is the first value of
+    /// `start_doc` (well, if there is one), and `value_column.get(end_doc - 1)`
+    /// is the last value of `end_doc`.
+    ///
+    /// The passed end range is allowed to be out of bounds, in which case
+    /// it will be clipped to make it valid.
+    #[inline]
+    pub(crate) fn docid_range_to_position_range(&self, range: Range<DocId>) -> Range<u32> {
+        let end_docid = range.end.min(self.num_docs() - 1) + 1;
+        let start_docid = range.start.min(end_docid);
+
+        let start = self.idx.get_val(start_docid) as u32;
+        let end = self.idx.get_val(end_docid) as u32;
+        assert!(start <= end);
+
+        start..end
+    }
+
+    /// returns the num of values associated with a doc_id
+    pub(crate) fn num_vals_for_doc(&self, doc: DocId) -> u32 {
+        let range = self.range(doc);
+        range.end - range.start
+    }
+
+    /// Returns the overall number of values in this field.
+    #[inline]
+    pub fn total_num_vals(&self) -> u32 {
+        self.idx.max_value() as u32
+    }
+
+    /// Returns the number of documents in the index.
+    #[inline]
+    pub fn num_docs(&self) -> u32 {
+        self.idx.num_vals() - 1
+    }
+
+    /// Converts a list of positions of values in a 1:n index to the corresponding list of DocIds.
+    /// Positions are converted inplace to docids.
+    ///
+    /// Since there is no index for value pos -> docid, but docid -> value pos range, we scan the
+    /// index.
+    ///
+    /// Correctness: positions needs to be sorted. idx_reader needs to contain monotonically
+    /// increasing positions.
+    ///
+    ///
+    /// TODO: Instead of a linear scan we can employ a exponential search into binary search to
+    /// match a docid to its value position.
+    pub(crate) fn positions_to_docids(&self, doc_id_range: Range<u32>, positions: &mut Vec<u32>) {
+        if positions.is_empty() {
+            return;
+        }
+        let mut cur_doc = doc_id_range.start;
+        let mut last_doc = None;
+
+        assert!(self.idx.get_val(doc_id_range.start) as u32 <= positions[0]);
+
+        let mut write_doc_pos = 0;
+        for i in 0..positions.len() {
+            let pos = positions[i];
+            loop {
+                let end = self.idx.get_val(cur_doc + 1) as u32;
+                if end > pos {
+                    positions[write_doc_pos] = cur_doc;
+                    write_doc_pos += if last_doc == Some(cur_doc) { 0 } else { 1 };
+                    last_doc = Some(cur_doc);
+                    break;
+                }
+                cur_doc += 1;
+            }
+        }
+        positions.truncate(write_doc_pos);
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use std::ops::Range;
+    use std::sync::Arc;
+
+    use fastfield_codecs::IterColumn;
+
+    use crate::fastfield::MultiValueIndex;
+
+    fn index_to_pos_helper(
+        index: &MultiValueIndex,
+        doc_id_range: Range<u32>,
+        positions: &[u32],
+    ) -> Vec<u32> {
+        let mut positions = positions.to_vec();
+        index.positions_to_docids(doc_id_range, &mut positions);
+        positions
+    }
+
+    #[test]
+    fn test_positions_to_docid() {
+        let offsets = vec![0, 10, 12, 15, 22, 23]; // docid values are [0..10, 10..12, 12..15, etc.]
+        let column = IterColumn::from(offsets.into_iter());
+        let index = MultiValueIndex::new(Arc::new(column));
+        assert_eq!(index.num_docs(), 5);
+        {
+            let positions = vec![10u32, 11, 15, 20, 21, 22];
+
+            assert_eq!(index_to_pos_helper(&index, 0..5, &positions), vec![1, 3, 4]);
+            assert_eq!(index_to_pos_helper(&index, 1..5, &positions), vec![1, 3, 4]);
+            assert_eq!(index_to_pos_helper(&index, 0..5, &[9]), vec![0]);
+            assert_eq!(index_to_pos_helper(&index, 1..5, &[10]), vec![1]);
+            assert_eq!(index_to_pos_helper(&index, 1..5, &[11]), vec![1]);
+            assert_eq!(index_to_pos_helper(&index, 2..5, &[12]), vec![2]);
+            assert_eq!(index_to_pos_helper(&index, 2..5, &[12, 14]), vec![2]);
+            assert_eq!(index_to_pos_helper(&index, 2..5, &[12, 14, 15]), vec![2, 3]);
+        }
+    }
+}

src/fastfield/multivalued/mod.rs

@@ -1,7 +1,9 @@
+mod index;
 mod reader;
 mod writer;
 
 use fastfield_codecs::FastFieldCodecType;
+pub use index::MultiValueIndex;
 
 pub use self::reader::{MultiValuedFastFieldReader, MultiValuedU128FastFieldReader};
 pub(crate) use self::writer::MultivalueStartIndex;
@@ -531,14 +533,17 @@ mod bench {
                 .unwrap()
                 .read_bytes()
                 .unwrap();
-            let idx_reader = fastfield_codecs::open(data_idx).unwrap();
+            let idx_reader = fastfield_codecs::open(data_idx).unwrap().to_full().unwrap();
 
             let data_vals = fast_fields_composite
                 .open_read_with_idx(field, 1)
                 .unwrap()
                 .read_bytes()
                 .unwrap();
-            let vals_reader = fastfield_codecs::open(data_vals).unwrap();
+            let vals_reader = fastfield_codecs::open(data_vals)
+                .unwrap()
+                .to_full()
+                .unwrap();
             let fast_field_reader = MultiValuedFastFieldReader::open(idx_reader, vals_reader);
             b.iter(|| {
                 let mut sum = 0u64;

src/fastfield/multivalued/reader.rs

@@ -3,7 +3,8 @@ use std::sync::Arc;
 
 use fastfield_codecs::{Column, MonotonicallyMappableToU128};
 
-use crate::fastfield::{FastValue, MultiValueLength};
+use super::MultiValueIndex;
+use crate::fastfield::FastValue;
 use crate::DocId;
 
 /// Reader for a multivalued `u64` fast field.
@@ -13,9 +14,10 @@ use crate::DocId;
 /// The `vals_reader` will access the concatenated list of all
 /// values for all reader.
 /// The `idx_reader` associated, for each document, the index of its first value.
+/// Stores the start position for each document.
 #[derive(Clone)]
 pub struct MultiValuedFastFieldReader<Item: FastValue> {
-    idx_reader: Arc<dyn Column<u64>>,
+    idx_reader: MultiValueIndex,
     vals_reader: Arc<dyn Column<Item>>,
 }
 
@@ -25,20 +27,11 @@ impl<Item: FastValue> MultiValuedFastFieldReader<Item> {
         vals_reader: Arc<dyn Column<Item>>,
     ) -> MultiValuedFastFieldReader<Item> {
         MultiValuedFastFieldReader {
-            idx_reader,
+            idx_reader: MultiValueIndex::new(idx_reader),
             vals_reader,
         }
     }
 
-    /// Returns `[start, end)`, such that the values associated with
-    /// the given document are `start..end`.
-    #[inline]
-    fn range(&self, doc: DocId) -> Range<u32> {
-        let start = self.idx_reader.get_val(doc) as u32;
-        let end = self.idx_reader.get_val(doc + 1) as u32;
-        start..end
-    }
-
     /// Returns the array of values associated with the given `doc`.
     #[inline]
     fn get_vals_for_range(&self, range: Range<u32>, vals: &mut Vec<Item>) {
@@ -51,10 +44,15 @@ impl<Item: FastValue> MultiValuedFastFieldReader<Item> {
     /// Returns the array of values associated with the given `doc`.
     #[inline]
     pub fn get_vals(&self, doc: DocId, vals: &mut Vec<Item>) {
-        let range = self.range(doc);
+        let range = self.idx_reader.range(doc);
         self.get_vals_for_range(range, vals);
     }
 
+    /// returns the multivalue index
+    pub fn get_index_reader(&self) -> &MultiValueIndex {
+        &self.idx_reader
+    }
+
     /// Returns the minimum value for this fast field.
     ///
     /// The min value does not take in account of possible
@@ -75,28 +73,14 @@ impl<Item: FastValue> MultiValuedFastFieldReader<Item> {
 
     /// Returns the number of values associated with the document `DocId`.
     #[inline]
-    pub fn num_vals(&self, doc: DocId) -> usize {
-        let range = self.range(doc);
-        (range.end - range.start) as usize
+    pub fn num_vals(&self, doc: DocId) -> u32 {
+        self.idx_reader.num_vals_for_doc(doc)
     }
 
-    /// Returns the overall number of values in this field  .
+    /// Returns the overall number of values in this field.
     #[inline]
-    pub fn total_num_vals(&self) -> u64 {
-        self.idx_reader.max_value()
-    }
-}
-
-impl<Item: FastValue> MultiValueLength for MultiValuedFastFieldReader<Item> {
-    fn get_range(&self, doc_id: DocId) -> Range<u32> {
-        self.range(doc_id)
-    }
-    fn get_len(&self, doc_id: DocId) -> u64 {
-        self.num_vals(doc_id) as u64
-    }
-
-    fn get_total_len(&self) -> u64 {
-        self.total_num_vals() as u64
+    pub fn total_num_vals(&self) -> u32 {
+        self.idx_reader.total_num_vals()
     }
 }
 
@@ -109,7 +93,7 @@ impl<Item: FastValue> MultiValueLength for MultiValuedFastFieldReader<Item> {
 /// The `idx_reader` associated, for each document, the index of its first value.
 #[derive(Clone)]
 pub struct MultiValuedU128FastFieldReader<T: MonotonicallyMappableToU128> {
-    idx_reader: Arc<dyn Column<u64>>,
+    idx_reader: MultiValueIndex,
     vals_reader: Arc<dyn Column<T>>,
 }
 
@@ -119,24 +103,15 @@ impl<T: MonotonicallyMappableToU128> MultiValuedU128FastFieldReader<T> {
         vals_reader: Arc<dyn Column<T>>,
     ) -> MultiValuedU128FastFieldReader<T> {
         Self {
-            idx_reader,
+            idx_reader: MultiValueIndex::new(idx_reader),
             vals_reader,
         }
     }
 
-    /// Returns `[start, end)`, such that the values associated
-    /// to the given document are `start..end`.
-    #[inline]
-    fn range(&self, doc: DocId) -> Range<u32> {
-        let start = self.idx_reader.get_val(doc) as u32;
-        let end = self.idx_reader.get_val(doc + 1) as u32;
-        start..end
-    }
-
     /// Returns the array of values associated to the given `doc`.
     #[inline]
     pub fn get_first_val(&self, doc: DocId) -> Option<T> {
-        let range = self.range(doc);
+        let range = self.idx_reader.range(doc);
         if range.is_empty() {
             return None;
         }
@@ -152,26 +127,18 @@ impl<T: MonotonicallyMappableToU128> MultiValuedU128FastFieldReader<T> {
             .get_range(range.start as u64, &mut vals[..]);
     }
 
+    /// Returns the index reader
+    pub fn get_index_reader(&self) -> &MultiValueIndex {
+        &self.idx_reader
+    }
+
     /// Returns the array of values associated to the given `doc`.
     #[inline]
     pub fn get_vals(&self, doc: DocId, vals: &mut Vec<T>) {
-        let range = self.range(doc);
+        let range = self.idx_reader.range(doc);
         self.get_vals_for_range(range, vals);
     }
 
-    /// Returns all docids which are in the provided value range
-    pub fn get_positions_for_value_range(
-        &self,
-        value_range: RangeInclusive<T>,
-        doc_id_range: Range<u32>,
-    ) -> Vec<DocId> {
-        let mut positions = Vec::new(); // TODO replace
-        self.vals_reader
-            .get_positions_for_value_range(value_range, doc_id_range, &mut positions);
-
-        positions_to_docids(&positions, self.idx_reader.as_ref())
-    }
-
     /// Iterates over all elements in the fast field
     pub fn iter(&self) -> impl Iterator<Item = T> + '_ {
         self.vals_reader.iter()
@@ -197,85 +164,44 @@ impl<T: MonotonicallyMappableToU128> MultiValuedU128FastFieldReader<T> {
 
     /// Returns the number of values associated with the document `DocId`.
     #[inline]
-    pub fn num_vals(&self, doc: DocId) -> usize {
-        let range = self.range(doc);
-        (range.end - range.start) as usize
+    pub fn num_vals(&self, doc: DocId) -> u32 {
+        self.idx_reader.num_vals_for_doc(doc)
     }
 
-    /// Returns the overall number of values in this field.
+    /// Returns the overall number of values in this field. It does not include deletes.
     #[inline]
-    pub fn total_num_vals(&self) -> u64 {
-        self.idx_reader.max_value()
-    }
-}
-
-impl<T: MonotonicallyMappableToU128> MultiValueLength for MultiValuedU128FastFieldReader<T> {
-    fn get_range(&self, doc_id: DocId) -> std::ops::Range<u32> {
-        self.range(doc_id)
-    }
-    fn get_len(&self, doc_id: DocId) -> u64 {
-        self.num_vals(doc_id) as u64
-    }
-    fn get_total_len(&self) -> u64 {
-        self.total_num_vals() as u64
-    }
-}
-
-/// Converts a list of positions of values in a 1:n index to the corresponding list of DocIds.
-///
-/// Since there is no index for value pos -> docid, but docid -> value pos range, we scan the index.
-///
-/// Correctness: positions needs to be sorted. idx_reader needs to contain monotonically increasing
-/// positions.
-///
-/// TODO: Instead of a linear scan we can employ a expotential search into binary search to match a
-/// docid to its value position.
-fn positions_to_docids<C: Column + ?Sized>(positions: &[u32], idx_reader: &C) -> Vec<DocId> {
-    let mut docs = vec![];
-    let mut cur_doc = 0u32;
-    let mut last_doc = None;
-
-    for pos in positions {
-        loop {
-            let end = idx_reader.get_val(cur_doc + 1) as u32;
-            if end > *pos {
-                // avoid duplicates
-                if Some(cur_doc) == last_doc {
-                    break;
-                }
-                docs.push(cur_doc);
-                last_doc = Some(cur_doc);
-                break;
-            }
-            cur_doc += 1;
-        }
+    pub fn total_num_vals(&self) -> u32 {
+        assert_eq!(
+            self.vals_reader.num_vals(),
+            self.get_index_reader().total_num_vals()
+        );
+        self.idx_reader.total_num_vals()
     }
 
-    docs
+    /// Returns the docids matching given doc_id_range and value_range.
+    #[inline]
+    pub fn get_docids_for_value_range(
+        &self,
+        value_range: RangeInclusive<T>,
+        doc_id_range: Range<u32>,
+        positions: &mut Vec<u32>,
+    ) {
+        let position_range = self
+            .get_index_reader()
+            .docid_range_to_position_range(doc_id_range.clone());
+        self.vals_reader
+            .get_docids_for_value_range(value_range, position_range, positions);
+
+        self.idx_reader.positions_to_docids(doc_id_range, positions);
+    }
 }
 
 #[cfg(test)]
 mod tests {
 
-    use fastfield_codecs::VecColumn;
-
     use crate::core::Index;
-    use crate::fastfield::multivalued::reader::positions_to_docids;
     use crate::schema::{Cardinality, Facet, FacetOptions, NumericOptions, Schema};
 
-    #[test]
-    fn test_positions_to_docid() {
-        let positions = vec![10u32, 11, 15, 20, 21, 22];
-
-        let offsets = vec![0, 10, 12, 15, 22, 23];
-        {
-            let column = VecColumn::from(&offsets);
-
-            let docids = positions_to_docids(&positions, &column);
-            assert_eq!(docids, vec![1, 3, 4]);
-        }
-    }
-
     #[test]
     fn test_multifastfield_reader() -> crate::Result<()> {
         let mut schema_builder = Schema::builder();

src/fastfield/readers.rs

@@ -1,7 +1,7 @@
 use std::net::Ipv6Addr;
 use std::sync::Arc;
 
-use fastfield_codecs::{open, open_u128, Column};
+use fastfield_codecs::{open, open_u128, Column, OptionalColumn};
 
 use super::multivalued::MultiValuedU128FastFieldReader;
 use crate::directory::{CompositeFile, FileSlice};
@@ -118,7 +118,7 @@ impl FastFieldReaders {
         &self,
         field: Field,
         index: usize,
-    ) -> crate::Result<Arc<dyn Column<TFastValue>>> {
+    ) -> crate::Result<Arc<dyn OptionalColumn<TFastValue>>> {
         let fast_field_slice = self.fast_field_data(field, index)?;
         let bytes = fast_field_slice.read_bytes()?;
         let column = fastfield_codecs::open(bytes)?;
@@ -128,7 +128,7 @@ impl FastFieldReaders {
     pub(crate) fn typed_fast_field_reader<TFastValue: FastValue>(
         &self,
         field: Field,
-    ) -> crate::Result<Arc<dyn Column<TFastValue>>> {
+    ) -> crate::Result<Arc<dyn OptionalColumn<TFastValue>>> {
         self.typed_fast_field_reader_with_idx(field, 0)
     }
 
@@ -138,13 +138,20 @@ impl FastFieldReaders {
     ) -> crate::Result<MultiValuedFastFieldReader<TFastValue>> {
         let idx_reader = self.typed_fast_field_reader(field)?;
         let vals_reader = self.typed_fast_field_reader_with_idx(field, 1)?;
-        Ok(MultiValuedFastFieldReader::open(idx_reader, vals_reader))
+        Ok(MultiValuedFastFieldReader::open(
+            idx_reader
+                .to_full()
+                .expect("multivalue fast field are always full"),
+            vals_reader
+                .to_full()
+                .expect("multivalue fast field are always full"),
+        ))
     }
 
     /// Returns the `u64` fast field reader reader associated with `field`.
     ///
     /// If `field` is not a u64 fast field, this method returns an Error.
-    pub fn u64(&self, field: Field) -> crate::Result<Arc<dyn Column<u64>>> {
+    pub fn u64(&self, field: Field) -> crate::Result<Arc<dyn OptionalColumn<u64>>> {
         self.check_type(field, FastType::U64, Cardinality::SingleValue)?;
         self.typed_fast_field_reader(field)
     }
@@ -152,7 +159,7 @@ impl FastFieldReaders {
     /// Returns the `ip` fast field reader reader associated to `field`.
     ///
     /// If `field` is not a u128 fast field, this method returns an Error.
-    pub fn ip_addr(&self, field: Field) -> crate::Result<Arc<dyn Column<Ipv6Addr>>> {
+    pub fn ip_addr(&self, field: Field) -> crate::Result<Arc<dyn OptionalColumn<Ipv6Addr>>> {
         self.check_type(field, FastType::U128, Cardinality::SingleValue)?;
         let bytes = self.fast_field_data(field, 0)?.read_bytes()?;
         Ok(open_u128::<Ipv6Addr>(bytes)?)
@@ -166,10 +173,15 @@ impl FastFieldReaders {
         field: Field,
     ) -> crate::Result<MultiValuedU128FastFieldReader<Ipv6Addr>> {
         self.check_type(field, FastType::U128, Cardinality::MultiValues)?;
-        let idx_reader: Arc<dyn Column<u64>> = self.typed_fast_field_reader(field)?;
+        let idx_reader: Arc<dyn Column<u64>> = self
+            .typed_fast_field_reader(field)?
+            .to_full()
+            .expect("multivalue fast fields are always full");
 
         let bytes = self.fast_field_data(field, 1)?.read_bytes()?;
-        let vals_reader = open_u128::<Ipv6Addr>(bytes)?;
+        let vals_reader = open_u128::<Ipv6Addr>(bytes)?
+            .to_full()
+            .expect("multivalue fields are always full");
 
         Ok(MultiValuedU128FastFieldReader::open(
             idx_reader,
@@ -179,8 +191,9 @@ impl FastFieldReaders {
 
     /// Returns the `u128` fast field reader reader associated to `field`.
     ///
-    /// If `field` is not a u128 fast field, this method returns an Error.
-    pub(crate) fn u128(&self, field: Field) -> crate::Result<Arc<dyn Column<u128>>> {
+    /// If `field` is not a u128 base type fast field, this method returns an Error.
+    /// Ip addresses use u128 as base type.
+    pub(crate) fn u128(&self, field: Field) -> crate::Result<Arc<dyn OptionalColumn<u128>>> {
         self.check_type(field, FastType::U128, Cardinality::SingleValue)?;
         let bytes = self.fast_field_data(field, 0)?.read_bytes()?;
         Ok(open_u128::<u128>(bytes)?)
@@ -191,10 +204,15 @@ impl FastFieldReaders {
     /// If `field` is not a u128 multi-valued fast field, this method returns an Error.
     pub fn u128s(&self, field: Field) -> crate::Result<MultiValuedU128FastFieldReader<u128>> {
         self.check_type(field, FastType::U128, Cardinality::MultiValues)?;
-        let idx_reader: Arc<dyn Column<u64>> = self.typed_fast_field_reader(field)?;
+        let idx_reader: Arc<dyn Column<u64>> = self
+            .typed_fast_field_reader(field)?
+            .to_full()
+            .expect("multivalue fast fields are always full");
 
         let bytes = self.fast_field_data(field, 1)?.read_bytes()?;
-        let vals_reader = open_u128::<u128>(bytes)?;
+        let vals_reader = open_u128::<u128>(bytes)?
+            .to_full()
+            .expect("multivalue fast fields are always full");
 
         Ok(MultiValuedU128FastFieldReader::open(
             idx_reader,
@@ -207,14 +225,14 @@ impl FastFieldReaders {
     ///
     /// If not, the fastfield reader will returns the u64-value associated with the original
     /// FastValue.
-    pub fn u64_lenient(&self, field: Field) -> crate::Result<Arc<dyn Column<u64>>> {
+    pub fn u64_lenient(&self, field: Field) -> crate::Result<Arc<dyn OptionalColumn<u64>>> {
         self.typed_fast_field_reader(field)
     }
 
     /// Returns the `i64` fast field reader reader associated with `field`.
     ///
     /// If `field` is not a i64 fast field, this method returns an Error.
-    pub fn i64(&self, field: Field) -> crate::Result<Arc<dyn Column<i64>>> {
+    pub fn i64(&self, field: Field) -> crate::Result<Arc<dyn OptionalColumn<i64>>> {
         self.check_type(field, FastType::I64, Cardinality::SingleValue)?;
         self.typed_fast_field_reader(field)
     }
@@ -222,7 +240,7 @@ impl FastFieldReaders {
     /// Returns the `date` fast field reader reader associated with `field`.
     ///
     /// If `field` is not a date fast field, this method returns an Error.
-    pub fn date(&self, field: Field) -> crate::Result<Arc<dyn Column<DateTime>>> {
+    pub fn date(&self, field: Field) -> crate::Result<Arc<dyn OptionalColumn<DateTime>>> {
         self.check_type(field, FastType::Date, Cardinality::SingleValue)?;
         self.typed_fast_field_reader(field)
     }
@@ -230,7 +248,7 @@ impl FastFieldReaders {
     /// Returns the `f64` fast field reader reader associated with `field`.
     ///
     /// If `field` is not a f64 fast field, this method returns an Error.
-    pub fn f64(&self, field: Field) -> crate::Result<Arc<dyn Column<f64>>> {
+    pub fn f64(&self, field: Field) -> crate::Result<Arc<dyn OptionalColumn<f64>>> {
         self.check_type(field, FastType::F64, Cardinality::SingleValue)?;
         self.typed_fast_field_reader(field)
     }
@@ -238,7 +256,7 @@ impl FastFieldReaders {
     /// Returns the `bool` fast field reader reader associated with `field`.
     ///
     /// If `field` is not a bool fast field, this method returns an Error.
-    pub fn bool(&self, field: Field) -> crate::Result<Arc<dyn Column<bool>>> {
+    pub fn bool(&self, field: Field) -> crate::Result<Arc<dyn OptionalColumn<bool>>> {
         self.check_type(field, FastType::Bool, Cardinality::SingleValue)?;
         self.typed_fast_field_reader(field)
     }
@@ -309,7 +327,12 @@ impl FastFieldReaders {
             let fast_field_idx_bytes = fast_field_idx_file.read_bytes()?;
             let idx_reader = open(fast_field_idx_bytes)?;
             let data = self.fast_field_data(field, 1)?;
-            BytesFastFieldReader::open(idx_reader, data)
+            BytesFastFieldReader::open(
+                idx_reader
+                    .to_full()
+                    .expect("multivalue fields are always full"),
+                data,
+            )
         } else {
             Err(FastFieldNotAvailableError::new(field_entry).into())
         }

src/fieldnorm/mod.rs

@@ -34,7 +34,7 @@ mod tests {
 
     use crate::directory::{CompositeFile, Directory, RamDirectory, WritePtr};
     use crate::fieldnorm::{FieldNormReader, FieldNormsSerializer, FieldNormsWriter};
-    use crate::query::{Query, TermQuery};
+    use crate::query::{EnableScoring, Query, TermQuery};
     use crate::schema::{
         Field, IndexRecordOption, Schema, TextFieldIndexing, TextOptions, STORED, TEXT,
     };
@@ -112,7 +112,7 @@ mod tests {
             Term::from_field_text(text, "hello"),
             IndexRecordOption::WithFreqs,
         );
-        let weight = query.weight(&searcher, true)?;
+        let weight = query.weight(EnableScoring::Enabled(&searcher))?;
         let mut scorer = weight.scorer(searcher.segment_reader(0), 1.0f32)?;
         assert_eq!(scorer.doc(), 0);
         assert!((scorer.score() - 0.22920431).abs() < 0.001f32);
@@ -141,7 +141,7 @@ mod tests {
             Term::from_field_text(text, "hello"),
             IndexRecordOption::WithFreqs,
         );
-        let weight = query.weight(&searcher, true)?;
+        let weight = query.weight(EnableScoring::Enabled(&searcher))?;
         let mut scorer = weight.scorer(searcher.segment_reader(0), 1.0f32)?;
         assert_eq!(scorer.doc(), 0);
         assert!((scorer.score() - 0.22920431).abs() < 0.001f32);

src/fieldnorm/writer.rs

@@ -9,7 +9,7 @@ use crate::DocId;
 /// The `FieldNormsWriter` is in charge of tracking the fieldnorm byte
 /// of each document for each field with field norms.
 ///
-/// `FieldNormsWriter` stores a Vec<u8> for each tracked field, using a
+/// `FieldNormsWriter` stores a `Vec<u8>` for each tracked field, using a
 /// byte per document per field.
 pub struct FieldNormsWriter {
     fieldnorms_buffers: Vec<Option<Vec<u8>>>,

src/indexer/doc_id_mapping.rs

@@ -465,9 +465,9 @@ mod tests_indexsorting {
         let my_number = index.schema().get_field("my_number").unwrap();
 
         let fast_field = fast_fields.u64(my_number).unwrap();
-        assert_eq!(fast_field.get_val(0), 10u64);
-        assert_eq!(fast_field.get_val(1), 20u64);
-        assert_eq!(fast_field.get_val(2), 30u64);
+        assert_eq!(fast_field.get_val(0), Some(10u64));
+        assert_eq!(fast_field.get_val(1), Some(20u64));
+        assert_eq!(fast_field.get_val(2), Some(30u64));
 
         let multi_numbers = index.schema().get_field("multi_numbers").unwrap();
         let multifield = fast_fields.u64s(multi_numbers).unwrap();

src/indexer/index_writer.rs

@@ -19,9 +19,9 @@ use crate::indexer::index_writer_status::IndexWriterStatus;
 use crate::indexer::operation::DeleteOperation;
 use crate::indexer::stamper::Stamper;
 use crate::indexer::{MergePolicy, SegmentEntry, SegmentWriter};
-use crate::query::{Query, TermQuery};
+use crate::query::{EnableScoring, Query, TermQuery};
 use crate::schema::{Document, IndexRecordOption, Term};
-use crate::{FutureResult, IndexReader, Opstamp};
+use crate::{FutureResult, Opstamp};
 
 // Size of the margin for the `memory_arena`. A segment is closed when the remaining memory
 // in the `memory_arena` goes below MARGIN_IN_BYTES.
@@ -57,7 +57,6 @@ pub struct IndexWriter {
     _directory_lock: Option<DirectoryLock>,
 
     index: Index,
-    index_reader: IndexReader,
 
     memory_arena_in_bytes_per_thread: usize,
 
@@ -95,7 +94,7 @@ fn compute_deleted_bitset(
         // document that were inserted before it.
         delete_op
             .target
-            .for_each(segment_reader, &mut |doc_matching_delete_query, _| {
+            .for_each_no_score(segment_reader, &mut |doc_matching_delete_query| {
                 if doc_opstamps.is_deleted(doc_matching_delete_query, delete_op.opstamp) {
                     alive_bitset.remove(doc_matching_delete_query);
                     might_have_changed = true;
@@ -298,8 +297,6 @@ impl IndexWriter {
 
             memory_arena_in_bytes_per_thread,
             index: index.clone(),
-            index_reader: index.reader()?,
-
             index_writer_status: IndexWriterStatus::from(document_receiver),
             operation_sender: document_sender,
 
@@ -681,8 +678,7 @@ impl IndexWriter {
     /// only after calling `commit()`.
     #[doc(hidden)]
     pub fn delete_query(&self, query: Box<dyn Query>) -> crate::Result<Opstamp> {
-        let weight = query.weight(&self.index_reader.searcher(), false)?;
-
+        let weight = query.weight(EnableScoring::Disabled(&self.index.schema()))?;
         let opstamp = self.stamper.stamp();
         let delete_operation = DeleteOperation {
             opstamp,
@@ -763,8 +759,7 @@ impl IndexWriter {
             match user_op {
                 UserOperation::Delete(term) => {
                     let query = TermQuery::new(term, IndexRecordOption::Basic);
-                    let weight = query.weight(&self.index_reader.searcher(), false)?;
-
+                    let weight = query.weight(EnableScoring::Disabled(&self.index.schema()))?;
                     let delete_operation = DeleteOperation {
                         opstamp,
                         target: weight,
@@ -1472,7 +1467,7 @@ mod tests {
         let fast_field_reader = segment_reader.fast_fields().u64(id_field)?;
         let in_order_alive_ids: Vec<u64> = segment_reader
             .doc_ids_alive()
-            .map(|doc| fast_field_reader.get_val(doc))
+            .map(|doc| fast_field_reader.get_val(doc).unwrap())
             .collect();
         assert_eq!(&in_order_alive_ids[..], &[9, 8, 7, 6, 5, 4, 1, 0]);
         Ok(())
@@ -1533,7 +1528,7 @@ mod tests {
         let fast_field_reader = segment_reader.fast_fields().u64(id_field)?;
         let in_order_alive_ids: Vec<u64> = segment_reader
             .doc_ids_alive()
-            .map(|doc| fast_field_reader.get_val(doc))
+            .map(|doc| fast_field_reader.get_val(doc).unwrap())
             .collect();
         assert_eq!(&in_order_alive_ids[..], &[9, 8, 7, 6, 5, 4, 2, 0]);
         Ok(())
@@ -1591,6 +1586,25 @@ mod tests {
         (existing_ids, deleted_ids)
     }
 
+    fn get_id_list(ops: &[IndexingOp]) -> Vec<u64> {
+        let mut id_list = Vec::new();
+        for &op in ops {
+            match op {
+                IndexingOp::AddDoc { id } => {
+                    id_list.push(id);
+                }
+                IndexingOp::DeleteDoc { id } => {
+                    id_list.retain(|el| *el != id);
+                }
+                IndexingOp::DeleteDocQuery { id } => {
+                    id_list.retain(|el| *el != id);
+                }
+                _ => {}
+            }
+        }
+        id_list
+    }
+
     fn test_operation_strategy(
         ops: &[IndexingOp],
         sort_index: bool,
@@ -1600,7 +1614,9 @@ mod tests {
         let ip_field = schema_builder.add_ip_addr_field("ip", FAST | INDEXED | STORED);
         let ips_field = schema_builder.add_ip_addr_field(
             "ips",
-            IpAddrOptions::default().set_fast(Cardinality::MultiValues),
+            IpAddrOptions::default()
+                .set_fast(Cardinality::MultiValues)
+                .set_indexed(),
         );
         let id_field = schema_builder.add_u64_field("id", FAST | INDEXED | STORED);
         let i64_field = schema_builder.add_i64_field("i64", INDEXED);
@@ -1665,11 +1681,13 @@ mod tests {
         // rotate right
         let multi_text_field_text3 = "test3 test1 test2 test3 test1 test2";
 
+        let ip_from_id = |id| Ipv6Addr::from_u128(id as u128);
+
         for &op in ops {
             match op {
                 IndexingOp::AddDoc { id } => {
                     let facet = Facet::from(&("/cola/".to_string() + &id.to_string()));
-                    let ip_from_id = Ipv6Addr::from_u128(id as u128);
+                    let ip = ip_from_id(id);
 
                     if !ip_exists(id) {
                         // every 3rd doc has no ip field
@@ -1693,9 +1711,9 @@ mod tests {
                     } else {
                         index_writer.add_document(doc!(id_field=>id,
                                 bytes_field => id.to_le_bytes().as_slice(),
-                                ip_field => ip_from_id,
-                                ips_field => ip_from_id,
-                                ips_field => ip_from_id,
+                                ip_field => ip,
+                                ips_field => ip,
+                                ips_field => ip,
                                 multi_numbers=> id,
                                 multi_numbers => id,
                                 bool_field => (id % 2u64) != 0,
@@ -1738,6 +1756,7 @@ mod tests {
         index_writer.commit()?;
 
         let searcher = index.reader()?.searcher();
+        let num_segments_before_merge = searcher.segment_readers().len();
         if force_end_merge {
             index_writer.wait_merging_threads()?;
             let mut index_writer = index.writer_for_tests()?;
@@ -1749,6 +1768,7 @@ mod tests {
                 assert!(index_writer.wait_merging_threads().is_ok());
             }
         }
+        let num_segments_after_merge = searcher.segment_readers().len();
 
         old_reader.reload()?;
         let old_searcher = old_reader.searcher();
@@ -1757,7 +1777,12 @@ mod tests {
             .segment_readers()
             .iter()
             .flat_map(|segment_reader| {
-                let ff_reader = segment_reader.fast_fields().u64(id_field).unwrap();
+                let ff_reader = segment_reader
+                    .fast_fields()
+                    .u64(id_field)
+                    .unwrap()
+                    .to_full()
+                    .unwrap();
                 segment_reader
                     .doc_ids_alive()
                     .map(move |doc| ff_reader.get_val(doc))
@@ -1768,7 +1793,12 @@ mod tests {
             .segment_readers()
             .iter()
             .flat_map(|segment_reader| {
-                let ff_reader = segment_reader.fast_fields().u64(id_field).unwrap();
+                let ff_reader = segment_reader
+                    .fast_fields()
+                    .u64(id_field)
+                    .unwrap()
+                    .to_full()
+                    .unwrap();
                 segment_reader
                     .doc_ids_alive()
                     .map(move |doc| ff_reader.get_val(doc))
@@ -1776,6 +1806,22 @@ mod tests {
             .collect();
 
         let (expected_ids_and_num_occurrences, deleted_ids) = expected_ids(ops);
+
+        let id_list = get_id_list(ops);
+
+        // multivalue fast field content
+        let mut all_ips = Vec::new();
+        let mut num_ips = 0;
+        for segment_reader in searcher.segment_readers().iter() {
+            let ip_reader = segment_reader.fast_fields().ip_addrs(ips_field).unwrap();
+            for doc in segment_reader.doc_ids_alive() {
+                let mut vals = vec![];
+                ip_reader.get_vals(doc, &mut vals);
+                all_ips.extend_from_slice(&vals);
+            }
+            num_ips += ip_reader.total_num_vals();
+        }
+
         let num_docs_expected = expected_ids_and_num_occurrences
             .iter()
             .map(|(_, id_occurrences)| *id_occurrences as usize)
@@ -1797,6 +1843,30 @@ mod tests {
                 .collect::<HashSet<_>>()
         );
 
+        if force_end_merge && num_segments_before_merge > 1 && num_segments_after_merge == 1 {
+            let mut expected_multi_ips: Vec<_> = id_list
+                .iter()
+                .filter(|id| ip_exists(**id))
+                .flat_map(|id| vec![ip_from_id(*id), ip_from_id(*id)])
+                .collect();
+            assert_eq!(num_ips, expected_multi_ips.len() as u32);
+
+            expected_multi_ips.sort();
+            all_ips.sort();
+            assert_eq!(expected_multi_ips, all_ips);
+
+            // Test fastfield num_docs
+            let num_docs: usize = searcher
+                .segment_readers()
+                .iter()
+                .map(|segment_reader| {
+                    let ff_reader = segment_reader.fast_fields().ip_addrs(ips_field).unwrap();
+                    ff_reader.get_index_reader().num_docs() as usize
+                })
+                .sum();
+            assert_eq!(num_docs, num_docs_expected);
+        }
+
         // Load all ips addr
         let ips: HashSet<Ipv6Addr> = searcher
             .segment_readers()
@@ -1804,7 +1874,7 @@ mod tests {
             .flat_map(|segment_reader| {
                 let ff_reader = segment_reader.fast_fields().ip_addr(ip_field).unwrap();
                 segment_reader.doc_ids_alive().flat_map(move |doc| {
-                    let val = ff_reader.get_val(doc);
+                    let val = ff_reader.get_val(doc).unwrap(); // TODO handle null
                     if val == Ipv6Addr::from_u128(0) {
                         // TODO Fix null handling
                         None
@@ -1861,7 +1931,7 @@ mod tests {
                 ff_reader.get_vals(doc, &mut vals);
                 assert_eq!(vals.len(), 2);
                 assert_eq!(vals[0], vals[1]);
-                assert_eq!(id_reader.get_val(doc), vals[0]);
+                assert_eq!(id_reader.get_val(doc).unwrap(), vals[0]);
 
                 let mut bool_vals = vec![];
                 bool_ff_reader.get_vals(doc, &mut bool_vals);
@@ -2000,6 +2070,51 @@ mod tests {
                 assert_eq!(do_search_ip_field(&format!("\"{}\"", ip_addr)), count);
             }
         }
+
+        // assert data is like expected
+        //
+        for (existing_id, count) in expected_ids_and_num_occurrences.iter().take(10) {
+            let (existing_id, count) = (*existing_id, *count);
+            if !ip_exists(existing_id) {
+                continue;
+            }
+            let gen_query_inclusive = |field: &str, from: Ipv6Addr, to: Ipv6Addr| {
+                format!("{}:[{} TO {}]", field, &from.to_string(), &to.to_string())
+            };
+            let ip = ip_from_id(existing_id);
+
+            let do_search_ip_field = |term: &str| do_search(term, ip_field).len() as u64;
+            // Range query on single value field
+            // let query = gen_query_inclusive("ip", ip, ip);
+            // assert_eq!(do_search_ip_field(&query), count);
+
+            // Range query on multi value field
+            let query = gen_query_inclusive("ips", ip, ip);
+            assert_eq!(do_search_ip_field(&query), count);
+        }
+
+        // ip range query on fast field
+        //
+        for (existing_id, count) in expected_ids_and_num_occurrences.iter().take(10) {
+            let (existing_id, count) = (*existing_id, *count);
+            if !ip_exists(existing_id) {
+                continue;
+            }
+            let gen_query_inclusive = |field: &str, from: Ipv6Addr, to: Ipv6Addr| {
+                format!("{}:[{} TO {}]", field, &from.to_string(), &to.to_string())
+            };
+            let ip = ip_from_id(existing_id);
+
+            let do_search_ip_field = |term: &str| do_search(term, ip_field).len() as u64;
+            // Range query on single value field
+            // let query = gen_query_inclusive("ip", ip, ip);
+            // assert_eq!(do_search_ip_field(&query), count);
+
+            // Range query on multi value field
+            let query = gen_query_inclusive("ips", ip, ip);
+            assert_eq!(do_search_ip_field(&query), count);
+        }
+
         // test facets
         for segment_reader in searcher.segment_readers().iter() {
             let mut facet_reader = segment_reader.facet_reader(facet_field).unwrap();
@@ -2012,7 +2127,7 @@ mod tests {
                 facet_reader
                     .facet_from_ord(facet_ords[0], &mut facet)
                     .unwrap();
-                let id = ff_reader.get_val(doc_id);
+                let id = ff_reader.get_val(doc_id).unwrap();
                 let facet_expected = Facet::from(&("/cola/".to_string() + &id.to_string()));
 
                 assert_eq!(facet, facet_expected);
@@ -2021,6 +2136,40 @@ mod tests {
         Ok(())
     }
 
+    #[test]
+    fn test_ip_range_query_multivalue_bug() {
+        assert!(test_operation_strategy(
+            &[
+                IndexingOp::AddDoc { id: 2 },
+                IndexingOp::Commit,
+                IndexingOp::AddDoc { id: 1 },
+                IndexingOp::AddDoc { id: 1 },
+                IndexingOp::Commit,
+                IndexingOp::Merge
+            ],
+            true,
+            false
+        )
+        .is_ok());
+    }
+
+    #[test]
+    fn test_ff_num_ips_regression() {
+        assert!(test_operation_strategy(
+            &[
+                IndexingOp::AddDoc { id: 13 },
+                IndexingOp::AddDoc { id: 1 },
+                IndexingOp::Commit,
+                IndexingOp::DeleteDocQuery { id: 13 },
+                IndexingOp::AddDoc { id: 1 },
+                IndexingOp::Commit,
+            ],
+            false,
+            true
+        )
+        .is_ok());
+    }
+
     #[test]
     fn test_minimal() {
         assert!(test_operation_strategy(
@@ -2030,7 +2179,7 @@ mod tests {
                 IndexingOp::DeleteDoc { id: 13 }
             ],
             true,
-            false
+            true
         )
         .is_ok());
 

src/indexer/json_term_writer.rs

@@ -261,6 +261,39 @@ pub struct JsonTermWriter<'a> {
     path_stack: Vec<usize>,
 }
 
+/// Splits a json path supplied to the query parser in such a way that
+/// `.` can be escaped.
+///
+/// In other words,
+/// - `k8s.node` ends up as `["k8s", "node"]`.
+/// - `k8s\.node` ends up as `["k8s.node"]`.
+fn split_json_path(json_path: &str) -> Vec<String> {
+    let mut escaped_state: bool = false;
+    let mut json_path_segments = Vec::new();
+    let mut buffer = String::new();
+    for ch in json_path.chars() {
+        if escaped_state {
+            buffer.push(ch);
+            escaped_state = false;
+            continue;
+        }
+        match ch {
+            '\\' => {
+                escaped_state = true;
+            }
+            '.' => {
+                let new_segment = std::mem::take(&mut buffer);
+                json_path_segments.push(new_segment);
+            }
+            _ => {
+                buffer.push(ch);
+            }
+        }
+    }
+    json_path_segments.push(buffer);
+    json_path_segments
+}
+
 impl<'a> JsonTermWriter<'a> {
     pub fn from_field_and_json_path(
         field: Field,
@@ -269,8 +302,8 @@ impl<'a> JsonTermWriter<'a> {
     ) -> Self {
         term_buffer.set_field_and_type(field, Type::Json);
         let mut json_term_writer = Self::wrap(term_buffer);
-        for segment in json_path.split('.') {
-            json_term_writer.push_path_segment(segment);
+        for segment in split_json_path(json_path) {
+            json_term_writer.push_path_segment(&segment);
         }
         json_term_writer
     }
@@ -350,7 +383,7 @@ impl<'a> JsonTermWriter<'a> {
 
 #[cfg(test)]
 mod tests {
-    use super::JsonTermWriter;
+    use super::{split_json_path, JsonTermWriter};
     use crate::schema::{Field, Type};
     use crate::Term;
 
@@ -495,4 +528,48 @@ mod tests {
         json_writer.set_str("pink");
         assert_eq!(json_writer.path(), b"color\x01hue");
     }
+
+    #[test]
+    fn test_split_json_path_simple() {
+        let json_path = split_json_path("titi.toto");
+        assert_eq!(&json_path, &["titi", "toto"]);
+    }
+
+    #[test]
+    fn test_split_json_path_single_segment() {
+        let json_path = split_json_path("toto");
+        assert_eq!(&json_path, &["toto"]);
+    }
+
+    #[test]
+    fn test_split_json_path_trailing_dot() {
+        let json_path = split_json_path("toto.");
+        assert_eq!(&json_path, &["toto", ""]);
+    }
+
+    #[test]
+    fn test_split_json_path_heading_dot() {
+        let json_path = split_json_path(".toto");
+        assert_eq!(&json_path, &["", "toto"]);
+    }
+
+    #[test]
+    fn test_split_json_path_escaped_dot() {
+        let json_path = split_json_path(r#"toto\.titi"#);
+        assert_eq!(&json_path, &["toto.titi"]);
+        let json_path_2 = split_json_path(r#"k8s\.container\.name"#);
+        assert_eq!(&json_path_2, &["k8s.container.name"]);
+    }
+
+    #[test]
+    fn test_split_json_path_escaped_backslash() {
+        let json_path = split_json_path(r#"toto\\titi"#);
+        assert_eq!(&json_path, &[r#"toto\titi"#]);
+    }
+
+    #[test]
+    fn test_split_json_path_escaped_normal_letter() {
+        let json_path = split_json_path(r#"toto\titi"#);
+        assert_eq!(&json_path, &[r#"tototiti"#]);
+    }
 }

src/indexer/merger.rs

@@ -13,7 +13,7 @@ use crate::docset::{DocSet, TERMINATED};
 use crate::error::DataCorruption;
 use crate::fastfield::{
     get_fastfield_codecs_for_multivalue, AliveBitSet, Column, CompositeFastFieldSerializer,
-    MultiValueLength, MultiValuedFastFieldReader, MultiValuedU128FastFieldReader,
+    MultiValueIndex, MultiValuedFastFieldReader, MultiValuedU128FastFieldReader,
 };
 use crate::fieldnorm::{FieldNormReader, FieldNormReaders, FieldNormsSerializer, FieldNormsWriter};
 use crate::indexer::doc_id_mapping::{expect_field_id_for_sort_field, SegmentDocIdMapping};
@@ -348,9 +348,29 @@ impl IndexMerger {
             field,
             fast_field_serializer,
             doc_id_mapping,
-            &segment_and_ff_readers,
+            &segment_and_ff_readers
+                .iter()
+                .map(|(segment_reader, u64s_reader)| {
+                    (*segment_reader, u64s_reader.get_index_reader())
+                })
+                .collect::<Vec<_>>(),
         )?;
 
+        let num_vals = segment_and_ff_readers
+            .iter()
+            .map(|(segment_reader, reader)| {
+                // TODO implement generic version, implement reverse scan, all - deletes
+                if let Some(alive_bitset) = segment_reader.alive_bitset() {
+                    alive_bitset
+                        .iter_alive()
+                        .map(|doc| reader.num_vals(doc))
+                        .sum()
+                } else {
+                    reader.total_num_vals() as u32
+                }
+            })
+            .sum();
+
         let fast_field_readers = segment_and_ff_readers
             .into_iter()
             .map(|(_, ff_reader)| ff_reader)
@@ -365,12 +385,7 @@ impl IndexMerger {
                 })
         };
 
-        fast_field_serializer.create_u128_fast_field_with_idx(
-            field,
-            iter_gen,
-            doc_id_mapping.len() as u32,
-            1,
-        )?;
+        fast_field_serializer.create_u128_fast_field_with_idx(field, iter_gen, num_vals, 1)?;
 
         Ok(())
     }
@@ -386,10 +401,15 @@ impl IndexMerger {
             .readers
             .iter()
             .map(|reader| {
-                let u128_reader: Arc<dyn Column<u128>> = reader.fast_fields().u128(field).expect(
-                    "Failed to find a reader for single fast field. This is a tantivy bug and it \
-                     should never happen.",
-                );
+                let u128_reader: Arc<dyn Column<u128>> = reader
+                    .fast_fields()
+                    .u128(field)
+                    .expect(
+                        "Failed to find a reader for single fast field. This is a tantivy bug and \
+                         it should never happen.",
+                    )
+                    .to_full()
+                    .expect("temp migration solution");
                 u128_reader
             })
             .collect::<Vec<_>>();
@@ -450,7 +470,11 @@ impl IndexMerger {
         sort_by_field: &IndexSortByField,
     ) -> crate::Result<Arc<dyn Column>> {
         let field_id = expect_field_id_for_sort_field(reader.schema(), sort_by_field)?; // for now expect fastfield, but not strictly required
-        let value_accessor = reader.fast_fields().u64_lenient(field_id)?;
+        let value_accessor = reader
+            .fast_fields()
+            .u64_lenient(field_id)?
+            .to_full()
+            .expect("temp migration solution");
         Ok(value_accessor)
     }
     /// Collecting value_accessors into a vec to bind the lifetime.
@@ -529,11 +553,11 @@ impl IndexMerger {
     // Creating the index file to point into the data, generic over `BytesFastFieldReader` and
     // `MultiValuedFastFieldReader`
     //
-    fn write_1_n_fast_field_idx_generic<T: MultiValueLength + Send + Sync>(
+    fn write_1_n_fast_field_idx_generic(
         field: Field,
         fast_field_serializer: &mut CompositeFastFieldSerializer,
         doc_id_mapping: &SegmentDocIdMapping,
-        segment_and_ff_readers: &[(&SegmentReader, T)],
+        segment_and_ff_readers: &[(&SegmentReader, &MultiValueIndex)],
     ) -> crate::Result<()> {
         let column =
             RemappedDocIdMultiValueIndexColumn::new(segment_and_ff_readers, doc_id_mapping);
@@ -567,7 +591,12 @@ impl IndexMerger {
             field,
             fast_field_serializer,
             doc_id_mapping,
-            &segment_and_ff_readers,
+            &segment_and_ff_readers
+                .iter()
+                .map(|(segment_reader, u64s_reader)| {
+                    (*segment_reader, u64s_reader.get_index_reader())
+                })
+                .collect::<Vec<_>>(),
         )
     }
 
@@ -697,7 +726,12 @@ impl IndexMerger {
             field,
             fast_field_serializer,
             doc_id_mapping,
-            &segment_and_ff_readers,
+            &segment_and_ff_readers
+                .iter()
+                .map(|(segment_reader, u64s_reader)| {
+                    (*segment_reader, u64s_reader.get_index_reader())
+                })
+                .collect::<Vec<_>>(),
         )?;
 
         let mut serialize_vals = fast_field_serializer.new_bytes_fast_field(field);
@@ -804,7 +838,7 @@ impl IndexMerger {
             // Let's compute the list of non-empty posting lists
             for (segment_ord, term_info) in merged_terms.current_segment_ords_and_term_infos() {
                 let segment_reader = &self.readers[segment_ord];
-                let inverted_index: &InvertedIndexReader = &*field_readers[segment_ord];
+                let inverted_index: &InvertedIndexReader = &field_readers[segment_ord];
                 let segment_postings = inverted_index
                     .read_postings_from_terminfo(&term_info, segment_postings_option)?;
                 let alive_bitset_opt = segment_reader.alive_bitset();
@@ -1039,7 +1073,7 @@ mod tests {
     };
     use crate::collector::{Count, FacetCollector};
     use crate::core::Index;
-    use crate::query::{AllQuery, BooleanQuery, Scorer, TermQuery};
+    use crate::query::{AllQuery, BooleanQuery, EnableScoring, Scorer, TermQuery};
     use crate::schema::{
         Cardinality, Document, Facet, FacetOptions, IndexRecordOption, NumericOptions, Term,
         TextFieldIndexing, INDEXED, TEXT,
@@ -1343,16 +1377,16 @@ mod tests {
                 .fast_fields()
                 .u64(score_field)
                 .unwrap();
-            assert_eq!(score_field_reader.min_value(), 4000);
-            assert_eq!(score_field_reader.max_value(), 7000);
+            assert_eq!(score_field_reader.min_value(), Some(4000));
+            assert_eq!(score_field_reader.max_value(), Some(7000));
 
             let score_field_reader = searcher
                 .segment_reader(1)
                 .fast_fields()
                 .u64(score_field)
                 .unwrap();
-            assert_eq!(score_field_reader.min_value(), 1);
-            assert_eq!(score_field_reader.max_value(), 3);
+            assert_eq!(score_field_reader.min_value(), Some(1));
+            assert_eq!(score_field_reader.max_value(), Some(3));
         }
         {
             // merging the segments
@@ -1397,8 +1431,8 @@ mod tests {
                 .fast_fields()
                 .u64(score_field)
                 .unwrap();
-            assert_eq!(score_field_reader.min_value(), 3);
-            assert_eq!(score_field_reader.max_value(), 7000);
+            assert_eq!(score_field_reader.min_value(), Some(3));
+            assert_eq!(score_field_reader.max_value(), Some(7000));
         }
         {
             // test a commit with only deletes
@@ -1444,8 +1478,8 @@ mod tests {
                 .fast_fields()
                 .u64(score_field)
                 .unwrap();
-            assert_eq!(score_field_reader.min_value(), 3);
-            assert_eq!(score_field_reader.max_value(), 7000);
+            assert_eq!(score_field_reader.min_value(), Some(3));
+            assert_eq!(score_field_reader.max_value(), Some(7000));
         }
         {
             // Test merging a single segment in order to remove deletes.
@@ -1491,8 +1525,8 @@ mod tests {
                 .fast_fields()
                 .u64(score_field)
                 .unwrap();
-            assert_eq!(score_field_reader.min_value(), 6000);
-            assert_eq!(score_field_reader.max_value(), 7000);
+            assert_eq!(score_field_reader.min_value(), Some(6000));
+            assert_eq!(score_field_reader.max_value(), Some(7000));
         }
 
         {
@@ -1952,7 +1986,7 @@ mod tests {
         let reader = index.reader()?;
         let searcher = reader.searcher();
         let mut term_scorer = term_query
-            .specialized_weight(&searcher, true)?
+            .specialized_weight(EnableScoring::Enabled(&searcher))?
             .specialized_scorer(searcher.segment_reader(0u32), 1.0)?;
         assert_eq!(term_scorer.doc(), 0);
         assert_nearly_equals!(term_scorer.block_max_score(), 0.0079681855);
@@ -1967,7 +2001,7 @@ mod tests {
         assert_eq!(searcher.segment_readers().len(), 2);
         for segment_reader in searcher.segment_readers() {
             let mut term_scorer = term_query
-                .specialized_weight(&searcher, true)?
+                .specialized_weight(EnableScoring::Enabled(&searcher))?
                 .specialized_scorer(segment_reader, 1.0)?;
             // the difference compared to before is intrinsic to the bm25 formula. no worries
             // there.
@@ -1992,7 +2026,7 @@ mod tests {
 
         let segment_reader = searcher.segment_reader(0u32);
         let mut term_scorer = term_query
-            .specialized_weight(&searcher, true)?
+            .specialized_weight(EnableScoring::Enabled(&searcher))?
             .specialized_scorer(segment_reader, 1.0)?;
         // the difference compared to before is intrinsic to the bm25 formula. no worries there.
         for doc in segment_reader.doc_ids_alive() {

src/indexer/merger_sorted_index_test.rs

@@ -186,17 +186,17 @@ mod tests {
 
         let fast_fields = segment_reader.fast_fields();
         let fast_field = fast_fields.u64(int_field).unwrap();
-        assert_eq!(fast_field.get_val(5), 1u64);
-        assert_eq!(fast_field.get_val(4), 2u64);
-        assert_eq!(fast_field.get_val(3), 3u64);
+        assert_eq!(fast_field.get_val(5), Some(1u64));
+        assert_eq!(fast_field.get_val(4), Some(2u64));
+        assert_eq!(fast_field.get_val(3), Some(3u64));
         if force_disjunct_segment_sort_values {
-            assert_eq!(fast_field.get_val(2), 20u64);
-            assert_eq!(fast_field.get_val(1), 100u64);
+            assert_eq!(fast_field.get_val(2), Some(20u64));
+            assert_eq!(fast_field.get_val(1), Some(100u64));
         } else {
-            assert_eq!(fast_field.get_val(2), 10u64);
-            assert_eq!(fast_field.get_val(1), 20u64);
+            assert_eq!(fast_field.get_val(2), Some(10u64));
+            assert_eq!(fast_field.get_val(1), Some(20u64));
         }
-        assert_eq!(fast_field.get_val(0), 1_000u64);
+        assert_eq!(fast_field.get_val(0), Some(1_000u64));
 
         // test new field norm mapping
         {
@@ -373,12 +373,12 @@ mod tests {
 
         let fast_fields = segment_reader.fast_fields();
         let fast_field = fast_fields.u64(int_field).unwrap();
-        assert_eq!(fast_field.get_val(0), 1u64);
-        assert_eq!(fast_field.get_val(1), 2u64);
-        assert_eq!(fast_field.get_val(2), 3u64);
-        assert_eq!(fast_field.get_val(3), 10u64);
-        assert_eq!(fast_field.get_val(4), 20u64);
-        assert_eq!(fast_field.get_val(5), 1_000u64);
+        assert_eq!(fast_field.get_val(0), Some(1u64));
+        assert_eq!(fast_field.get_val(1), Some(2u64));
+        assert_eq!(fast_field.get_val(2), Some(3u64));
+        assert_eq!(fast_field.get_val(3), Some(10u64));
+        assert_eq!(fast_field.get_val(4), Some(20u64));
+        assert_eq!(fast_field.get_val(5), Some(1_000u64));
 
         let get_vals = |fast_field: &MultiValuedFastFieldReader<u64>, doc_id: u32| -> Vec<u64> {
             let mut vals = vec![];
@@ -535,11 +535,15 @@ mod bench_sorted_index_merge {
         b.iter(|| {
             let sorted_doc_ids = doc_id_mapping.iter_old_doc_addrs().map(|doc_addr| {
                 let reader = &merger.readers[doc_addr.segment_ord as usize];
-                let u64_reader: Arc<dyn Column<u64>> =
-                    reader.fast_fields().typed_fast_field_reader(field).expect(
+                let u64_reader: Arc<dyn Column<u64>> = reader
+                    .fast_fields()
+                    .typed_fast_field_reader(field)
+                    .expect(
                         "Failed to find a reader for single fast field. This is a tantivy bug and \
                          it should never happen.",
-                    );
+                    )
+                    .to_full()
+                    .unwrap();
                 (doc_addr.doc_id, reader, u64_reader)
             });
             // add values in order of the new doc_ids

src/indexer/mod.rs

@@ -58,13 +58,15 @@ type AddBatchReceiver = channel::Receiver<AddBatch>;
 #[cfg(feature = "mmap")]
 #[cfg(test)]
 mod tests_mmap {
-    use crate::schema::{self, Schema};
+    use crate::collector::Count;
+    use crate::query::QueryParser;
+    use crate::schema::{Schema, STORED, TEXT};
     use crate::{Index, Term};
 
     #[test]
     fn test_advance_delete_bug() -> crate::Result<()> {
         let mut schema_builder = Schema::builder();
-        let text_field = schema_builder.add_text_field("text", schema::TEXT);
+        let text_field = schema_builder.add_text_field("text", TEXT);
         let index = Index::create_from_tempdir(schema_builder.build())?;
         let mut index_writer = index.writer_for_tests()?;
         // there must be one deleted document in the segment
@@ -75,7 +77,26 @@ mod tests_mmap {
             index_writer.add_document(doc!(text_field=>"c"))?;
         }
         index_writer.commit()?;
-        index_writer.commit()?;
         Ok(())
     }
+
+    #[test]
+    fn test_json_field_espace() {
+        let mut schema_builder = Schema::builder();
+        let json_field = schema_builder.add_json_field("json", TEXT | STORED);
+        let index = Index::create_in_ram(schema_builder.build());
+        let mut index_writer = index.writer_for_tests().unwrap();
+        let json = serde_json::json!({"k8s.container.name": "prometheus", "val": "hello"});
+        index_writer.add_document(doc!(json_field=>json)).unwrap();
+        index_writer.commit().unwrap();
+        let reader = index.reader().unwrap();
+        let searcher = reader.searcher();
+        assert_eq!(searcher.num_docs(), 1);
+        let parse_query = QueryParser::for_index(&index, Vec::new());
+        let query = parse_query
+            .parse_query(r#"json.k8s\.container\.name:prometheus"#)
+            .unwrap();
+        let num_docs = searcher.search(&query, &Count).unwrap();
+        assert_eq!(num_docs, 1);
+    }
 }

src/indexer/sorted_doc_id_column.rs

@@ -46,11 +46,15 @@ impl<'a> RemappedDocIdColumn<'a> {
         let (min_value, max_value) = readers
             .iter()
             .filter_map(|reader| {
-                let u64_reader: Arc<dyn Column<u64>> =
-                    reader.fast_fields().typed_fast_field_reader(field).expect(
+                let u64_reader: Arc<dyn Column<u64>> = reader
+                    .fast_fields()
+                    .typed_fast_field_reader(field)
+                    .expect(
                         "Failed to find a reader for single fast field. This is a tantivy bug and \
                          it should never happen.",
-                    );
+                    )
+                    .to_full()
+                    .expect("temp migration solution");
                 compute_min_max_val(&*u64_reader, reader)
             })
             .reduce(|a, b| (a.0.min(b.0), a.1.max(b.1)))
@@ -59,11 +63,15 @@ impl<'a> RemappedDocIdColumn<'a> {
         let fast_field_readers = readers
             .iter()
             .map(|reader| {
-                let u64_reader: Arc<dyn Column<u64>> =
-                    reader.fast_fields().typed_fast_field_reader(field).expect(
+                let u64_reader: Arc<dyn Column<u64>> = reader
+                    .fast_fields()
+                    .typed_fast_field_reader(field)
+                    .expect(
                         "Failed to find a reader for single fast field. This is a tantivy bug and \
                          it should never happen.",
-                    );
+                    )
+                    .to_full()
+                    .expect("temp migration solution");
                 u64_reader
             })
             .collect::<Vec<_>>();

src/indexer/sorted_doc_id_multivalue_column.rs

@@ -3,7 +3,7 @@ use std::cmp;
 use fastfield_codecs::Column;
 
 use super::flat_map_with_buffer::FlatMapWithBufferIter;
-use crate::fastfield::{MultiValueLength, MultiValuedFastFieldReader};
+use crate::fastfield::{MultiValueIndex, MultiValuedFastFieldReader};
 use crate::indexer::doc_id_mapping::SegmentDocIdMapping;
 use crate::schema::Field;
 use crate::{DocAddress, SegmentReader};
@@ -94,17 +94,17 @@ impl<'a> Column for RemappedDocIdMultiValueColumn<'a> {
     }
 }
 
-pub(crate) struct RemappedDocIdMultiValueIndexColumn<'a, T: MultiValueLength> {
+pub(crate) struct RemappedDocIdMultiValueIndexColumn<'a> {
     doc_id_mapping: &'a SegmentDocIdMapping,
-    multi_value_length_readers: Vec<&'a T>,
+    multi_value_length_readers: Vec<&'a MultiValueIndex>,
     min_value: u64,
     max_value: u64,
     num_vals: u32,
 }
 
-impl<'a, T: MultiValueLength> RemappedDocIdMultiValueIndexColumn<'a, T> {
+impl<'a> RemappedDocIdMultiValueIndexColumn<'a> {
     pub(crate) fn new(
-        segment_and_ff_readers: &'a [(&'a SegmentReader, T)],
+        segment_and_ff_readers: &'a [(&'a SegmentReader, &'a MultiValueIndex)],
         doc_id_mapping: &'a SegmentDocIdMapping,
     ) -> Self {
         // We go through a complete first pass to compute the minimum and the
@@ -115,17 +115,19 @@ impl<'a, T: MultiValueLength> RemappedDocIdMultiValueIndexColumn<'a, T> {
         let mut multi_value_length_readers = Vec::with_capacity(segment_and_ff_readers.len());
         for segment_and_ff_reader in segment_and_ff_readers {
             let segment_reader = segment_and_ff_reader.0;
-            let multi_value_length_reader = &segment_and_ff_reader.1;
+            let multi_value_length_reader = segment_and_ff_reader.1;
             if !segment_reader.has_deletes() {
-                max_value += multi_value_length_reader.get_total_len();
+                max_value += multi_value_length_reader.total_num_vals() as u64;
             } else {
                 for doc in segment_reader.doc_ids_alive() {
-                    max_value += multi_value_length_reader.get_len(doc);
+                    max_value += multi_value_length_reader.num_vals_for_doc(doc) as u64;
                 }
             }
             num_vals += segment_reader.num_docs();
             multi_value_length_readers.push(multi_value_length_reader);
         }
+        // The value range is always get_val(doc)..get_val(doc + 1)
+        num_vals += 1;
         Self {
             doc_id_mapping,
             multi_value_length_readers,
@@ -136,7 +138,7 @@ impl<'a, T: MultiValueLength> RemappedDocIdMultiValueIndexColumn<'a, T> {
     }
 }
 
-impl<'a, T: MultiValueLength + Send + Sync> Column for RemappedDocIdMultiValueIndexColumn<'a, T> {
+impl<'a> Column for RemappedDocIdMultiValueIndexColumn<'a> {
     fn get_val(&self, _pos: u32) -> u64 {
         unimplemented!()
     }
@@ -148,8 +150,8 @@ impl<'a, T: MultiValueLength + Send + Sync> Column for RemappedDocIdMultiValueIn
                 move |old_doc_addr| {
                     let ff_reader =
                         &self.multi_value_length_readers[old_doc_addr.segment_ord as usize];
-                    offset += ff_reader.get_len(old_doc_addr.doc_id);
-                    offset
+                    offset += ff_reader.num_vals_for_doc(old_doc_addr.doc_id);
+                    offset as u64
                 },
             )),
         )

src/lib.rs

@@ -277,6 +277,8 @@ pub mod fastfield;
 pub mod fieldnorm;
 pub mod positions;
 pub mod postings;
+
+/// Module containing the different query implementations.
 pub mod query;
 pub mod schema;
 pub mod space_usage;
@@ -1035,21 +1037,21 @@ pub mod tests {
             let fast_field_reader_opt = segment_reader.fast_fields().u64(fast_field_unsigned);
             assert!(fast_field_reader_opt.is_ok());
             let fast_field_reader = fast_field_reader_opt.unwrap();
-            assert_eq!(fast_field_reader.get_val(0), 4u64)
+            assert_eq!(fast_field_reader.get_val(0), Some(4u64))
         }
 
         {
             let fast_field_reader_res = segment_reader.fast_fields().i64(fast_field_signed);
             assert!(fast_field_reader_res.is_ok());
             let fast_field_reader = fast_field_reader_res.unwrap();
-            assert_eq!(fast_field_reader.get_val(0), 4i64)
+            assert_eq!(fast_field_reader.get_val(0), Some(4i64))
         }
 
         {
             let fast_field_reader_res = segment_reader.fast_fields().f64(fast_field_float);
             assert!(fast_field_reader_res.is_ok());
             let fast_field_reader = fast_field_reader_res.unwrap();
-            assert_eq!(fast_field_reader.get_val(0), 4f64)
+            assert_eq!(fast_field_reader.get_val(0), Some(4f64))
         }
         Ok(())
     }

src/query/all_query.rs

@@ -1,8 +1,8 @@
-use crate::core::{Searcher, SegmentReader};
+use crate::core::SegmentReader;
 use crate::docset::{DocSet, TERMINATED};
 use crate::query::boost_query::BoostScorer;
 use crate::query::explanation::does_not_match;
-use crate::query::{Explanation, Query, Scorer, Weight};
+use crate::query::{EnableScoring, Explanation, Query, Scorer, Weight};
 use crate::{DocId, Score};
 
 /// Query that matches all of the documents.
@@ -12,7 +12,7 @@ use crate::{DocId, Score};
 pub struct AllQuery;
 
 impl Query for AllQuery {
-    fn weight(&self, _: &Searcher, _: bool) -> crate::Result<Box<dyn Weight>> {
+    fn weight(&self, _: EnableScoring<'_>) -> crate::Result<Box<dyn Weight>> {
         Ok(Box::new(AllWeight))
     }
 }
@@ -72,7 +72,7 @@ impl Scorer for AllScorer {
 mod tests {
     use super::AllQuery;
     use crate::docset::TERMINATED;
-    use crate::query::Query;
+    use crate::query::{EnableScoring, Query};
     use crate::schema::{Schema, TEXT};
     use crate::Index;
 
@@ -95,7 +95,7 @@ mod tests {
         let index = create_test_index()?;
         let reader = index.reader()?;
         let searcher = reader.searcher();
-        let weight = AllQuery.weight(&searcher, false)?;
+        let weight = AllQuery.weight(EnableScoring::Disabled(&index.schema()))?;
         {
             let reader = searcher.segment_reader(0);
             let mut scorer = weight.scorer(reader, 1.0)?;
@@ -118,7 +118,7 @@ mod tests {
         let index = create_test_index()?;
         let reader = index.reader()?;
         let searcher = reader.searcher();
-        let weight = AllQuery.weight(&searcher, false)?;
+        let weight = AllQuery.weight(EnableScoring::Disabled(searcher.schema()))?;
         let reader = searcher.segment_reader(0);
         {
             let mut scorer = weight.scorer(reader, 2.0)?;

src/query/automaton_weight.rs

@@ -33,7 +33,7 @@ where
         &'a self,
         term_dict: &'a TermDictionary,
     ) -> io::Result<TermStreamer<'a, &'a A>> {
-        let automaton: &A = &*self.automaton;
+        let automaton: &A = &self.automaton;
         let term_stream_builder = term_dict.search(automaton);
         term_stream_builder.into_stream()
     }

src/query/boolean_query/boolean_query.rs

@@ -1,7 +1,6 @@
 use super::boolean_weight::BooleanWeight;
-use crate::query::{Occur, Query, SumWithCoordsCombiner, TermQuery, Weight};
+use crate::query::{EnableScoring, Occur, Query, SumWithCoordsCombiner, TermQuery, Weight};
 use crate::schema::{IndexRecordOption, Term};
-use crate::Searcher;
 
 /// The boolean query returns a set of documents
 /// that matches the Boolean combination of constituent subqueries.
@@ -143,17 +142,15 @@ impl From<Vec<(Occur, Box<dyn Query>)>> for BooleanQuery {
 }
 
 impl Query for BooleanQuery {
-    fn weight(&self, searcher: &Searcher, scoring_enabled: bool) -> crate::Result<Box<dyn Weight>> {
+    fn weight(&self, enable_scoring: EnableScoring<'_>) -> crate::Result<Box<dyn Weight>> {
         let sub_weights = self
             .subqueries
             .iter()
-            .map(|&(ref occur, ref subquery)| {
-                Ok((*occur, subquery.weight(searcher, scoring_enabled)?))
-            })
+            .map(|&(ref occur, ref subquery)| Ok((*occur, subquery.weight(enable_scoring)?)))
             .collect::<crate::Result<_>>()?;
         Ok(Box::new(BooleanWeight::new(
             sub_weights,
-            scoring_enabled,
+            enable_scoring.is_scoring_enabled(),
             Box::new(SumWithCoordsCombiner::default),
         )))
     }

src/query/boolean_query/boolean_weight.rs

@@ -5,7 +5,7 @@ use crate::postings::FreqReadingOption;
 use crate::query::explanation::does_not_match;
 use crate::query::score_combiner::{DoNothingCombiner, ScoreCombiner};
 use crate::query::term_query::TermScorer;
-use crate::query::weight::{for_each_pruning_scorer, for_each_scorer};
+use crate::query::weight::{for_each_docset, for_each_pruning_scorer, for_each_scorer};
 use crate::query::{
     intersect_scorers, EmptyScorer, Exclude, Explanation, Occur, RequiredOptionalScorer, Scorer,
     Union, Weight,
@@ -219,6 +219,24 @@ impl<TScoreCombiner: ScoreCombiner + Sync> Weight for BooleanWeight<TScoreCombin
         Ok(())
     }
 
+    fn for_each_no_score(
+        &self,
+        reader: &SegmentReader,
+        callback: &mut dyn FnMut(DocId),
+    ) -> crate::Result<()> {
+        let scorer = self.complex_scorer(reader, 1.0, || DoNothingCombiner)?;
+        match scorer {
+            SpecializedScorer::TermUnion(term_scorers) => {
+                let mut union_scorer = Union::build(term_scorers, &self.score_combiner_fn);
+                for_each_docset(&mut union_scorer, callback);
+            }
+            SpecializedScorer::Other(mut scorer) => {
+                for_each_docset(scorer.as_mut(), callback);
+            }
+        }
+        Ok(())
+    }
+
     /// Calls `callback` with all of the `(doc, score)` for which score
     /// is exceeding a given threshold.
     ///

src/query/boolean_query/mod.rs

@@ -15,7 +15,8 @@ mod tests {
     use crate::query::score_combiner::SumWithCoordsCombiner;
     use crate::query::term_query::TermScorer;
     use crate::query::{
-        Intersection, Occur, Query, QueryParser, RequiredOptionalScorer, Scorer, TermQuery,
+        EnableScoring, Intersection, Occur, Query, QueryParser, RequiredOptionalScorer, Scorer,
+        TermQuery,
     };
     use crate::schema::*;
     use crate::{assert_nearly_equals, DocAddress, DocId, Index, Score};
@@ -54,7 +55,7 @@ mod tests {
         let query_parser = QueryParser::for_index(&index, vec![text_field]);
         let query = query_parser.parse_query("+a")?;
         let searcher = index.reader()?.searcher();
-        let weight = query.weight(&searcher, true)?;
+        let weight = query.weight(EnableScoring::Enabled(&searcher))?;
         let scorer = weight.scorer(searcher.segment_reader(0u32), 1.0)?;
         assert!(scorer.is::<TermScorer>());
         Ok(())
@@ -67,13 +68,13 @@ mod tests {
         let searcher = index.reader()?.searcher();
         {
             let query = query_parser.parse_query("+a +b +c")?;
-            let weight = query.weight(&searcher, true)?;
+            let weight = query.weight(EnableScoring::Enabled(&searcher))?;
             let scorer = weight.scorer(searcher.segment_reader(0u32), 1.0)?;
             assert!(scorer.is::<Intersection<TermScorer>>());
         }
         {
             let query = query_parser.parse_query("+a +(b c)")?;
-            let weight = query.weight(&searcher, true)?;
+            let weight = query.weight(EnableScoring::Enabled(&searcher))?;
             let scorer = weight.scorer(searcher.segment_reader(0u32), 1.0)?;
             assert!(scorer.is::<Intersection<Box<dyn Scorer>>>());
         }
@@ -87,7 +88,7 @@ mod tests {
         let searcher = index.reader()?.searcher();
         {
             let query = query_parser.parse_query("+a b")?;
-            let weight = query.weight(&searcher, true)?;
+            let weight = query.weight(EnableScoring::Enabled(&searcher))?;
             let scorer = weight.scorer(searcher.segment_reader(0u32), 1.0)?;
             assert!(scorer.is::<RequiredOptionalScorer<
                 Box<dyn Scorer>,
@@ -97,7 +98,7 @@ mod tests {
         }
         {
             let query = query_parser.parse_query("+a b")?;
-            let weight = query.weight(&searcher, false)?;
+            let weight = query.weight(EnableScoring::Disabled(searcher.schema()))?;
             let scorer = weight.scorer(searcher.segment_reader(0u32), 1.0)?;
             assert!(scorer.is::<TermScorer>());
         }
@@ -241,7 +242,9 @@ mod tests {
         let searcher = reader.searcher();
         let boolean_query =
             BooleanQuery::new(vec![(Occur::Should, term_a), (Occur::Should, term_b)]);
-        let boolean_weight = boolean_query.weight(&searcher, true).unwrap();
+        let boolean_weight = boolean_query
+            .weight(EnableScoring::Enabled(&searcher))
+            .unwrap();
         {
             let mut boolean_scorer = boolean_weight.scorer(searcher.segment_reader(0u32), 1.0)?;
             assert_eq!(boolean_scorer.doc(), 0u32);

src/query/boost_query.rs

@@ -2,8 +2,8 @@ use std::fmt;
 
 use crate::fastfield::AliveBitSet;
 use crate::query::explanation::does_not_match;
-use crate::query::{Explanation, Query, Scorer, Weight};
-use crate::{DocId, DocSet, Score, Searcher, SegmentReader, Term};
+use crate::query::{EnableScoring, Explanation, Query, Scorer, Weight};
+use crate::{DocId, DocSet, Score, SegmentReader, Term};
 
 /// `BoostQuery` is a wrapper over a query used to boost its score.
 ///
@@ -38,9 +38,9 @@ impl fmt::Debug for BoostQuery {
 }
 
 impl Query for BoostQuery {
-    fn weight(&self, searcher: &Searcher, scoring_enabled: bool) -> crate::Result<Box<dyn Weight>> {
-        let weight_without_boost = self.query.weight(searcher, scoring_enabled)?;
-        let boosted_weight = if scoring_enabled {
+    fn weight(&self, enable_scoring: EnableScoring<'_>) -> crate::Result<Box<dyn Weight>> {
+        let weight_without_boost = self.query.weight(enable_scoring)?;
+        let boosted_weight = if enable_scoring.is_scoring_enabled() {
             Box::new(BoostWeight::new(weight_without_boost, self.boost))
         } else {
             weight_without_boost

src/query/const_score_query.rs

@@ -1,7 +1,7 @@
 use std::fmt;
 
-use crate::query::{Explanation, Query, Scorer, Weight};
-use crate::{DocId, DocSet, Score, Searcher, SegmentReader, TantivyError, Term};
+use crate::query::{EnableScoring, Explanation, Query, Scorer, Weight};
+use crate::{DocId, DocSet, Score, SegmentReader, TantivyError, Term};
 
 /// `ConstScoreQuery` is a wrapper over a query to provide a constant score.
 /// It can avoid unnecessary score computation on the wrapped query.
@@ -36,9 +36,9 @@ impl fmt::Debug for ConstScoreQuery {
 }
 
 impl Query for ConstScoreQuery {
-    fn weight(&self, searcher: &Searcher, scoring_enabled: bool) -> crate::Result<Box<dyn Weight>> {
-        let inner_weight = self.query.weight(searcher, scoring_enabled)?;
-        Ok(if scoring_enabled {
+    fn weight(&self, enable_scoring: EnableScoring<'_>) -> crate::Result<Box<dyn Weight>> {
+        let inner_weight = self.query.weight(enable_scoring)?;
+        Ok(if enable_scoring.is_scoring_enabled() {
             Box::new(ConstWeight::new(inner_weight, self.score))
         } else {
             inner_weight

src/query/disjunction_max_query.rs

@@ -1,7 +1,7 @@
 use tantivy_query_grammar::Occur;
 
-use crate::query::{BooleanWeight, DisjunctionMaxCombiner, Query, Weight};
-use crate::{Score, Searcher, Term};
+use crate::query::{BooleanWeight, DisjunctionMaxCombiner, EnableScoring, Query, Weight};
+use crate::{Score, Term};
 
 /// The disjunction max query кeturns documents matching one or more wrapped queries,
 /// called query clauses or clauses.
@@ -91,16 +91,16 @@ impl Clone for DisjunctionMaxQuery {
 }
 
 impl Query for DisjunctionMaxQuery {
-    fn weight(&self, searcher: &Searcher, scoring_enabled: bool) -> crate::Result<Box<dyn Weight>> {
+    fn weight(&self, enable_scoring: EnableScoring<'_>) -> crate::Result<Box<dyn Weight>> {
         let disjuncts = self
             .disjuncts
             .iter()
-            .map(|disjunct| Ok((Occur::Should, disjunct.weight(searcher, scoring_enabled)?)))
+            .map(|disjunct| Ok((Occur::Should, disjunct.weight(enable_scoring)?)))
             .collect::<crate::Result<_>>()?;
         let tie_breaker = self.tie_breaker;
         Ok(Box::new(BooleanWeight::new(
             disjuncts,
-            scoring_enabled,
+            enable_scoring.is_scoring_enabled(),
             Box::new(move || DisjunctionMaxCombiner::with_tie_breaker(tie_breaker)),
         )))
     }

src/query/empty_query.rs

@@ -1,7 +1,7 @@
 use super::Scorer;
 use crate::docset::TERMINATED;
 use crate::query::explanation::does_not_match;
-use crate::query::{Explanation, Query, Weight};
+use crate::query::{EnableScoring, Explanation, Query, Weight};
 use crate::{DocId, DocSet, Score, Searcher, SegmentReader};
 
 /// `EmptyQuery` is a dummy `Query` in which no document matches.
@@ -11,11 +11,7 @@ use crate::{DocId, DocSet, Score, Searcher, SegmentReader};
 pub struct EmptyQuery;
 
 impl Query for EmptyQuery {
-    fn weight(
-        &self,
-        _searcher: &Searcher,
-        _scoring_enabled: bool,
-    ) -> crate::Result<Box<dyn Weight>> {
+    fn weight(&self, _enable_scoring: EnableScoring<'_>) -> crate::Result<Box<dyn Weight>> {
         Ok(Box::new(EmptyWeight))
     }
 

src/query/fuzzy_query.rs

@@ -5,9 +5,8 @@ use levenshtein_automata::{Distance, LevenshteinAutomatonBuilder, DFA};
 use once_cell::sync::Lazy;
 use tantivy_fst::Automaton;
 
-use crate::query::{AutomatonWeight, Query, Weight};
+use crate::query::{AutomatonWeight, EnableScoring, Query, Weight};
 use crate::schema::Term;
-use crate::Searcher;
 use crate::TantivyError::InvalidArgument;
 
 pub(crate) struct DfaWrapper(pub DFA);
@@ -158,11 +157,7 @@ impl FuzzyTermQuery {
 }
 
 impl Query for FuzzyTermQuery {
-    fn weight(
-        &self,
-        _searcher: &Searcher,
-        _scoring_enabled: bool,
-    ) -> crate::Result<Box<dyn Weight>> {
+    fn weight(&self, _enable_scoring: EnableScoring<'_>) -> crate::Result<Box<dyn Weight>> {
         Ok(Box::new(self.specialized_weight()?))
     }
 }

src/query/mod.rs

@@ -1,5 +1,3 @@
-//! Query Module
-
 mod all_query;
 mod automaton_weight;
 mod bitset;
@@ -51,7 +49,7 @@ pub use self::fuzzy_query::FuzzyTermQuery;
 pub use self::intersection::{intersect_scorers, Intersection};
 pub use self::more_like_this::{MoreLikeThisQuery, MoreLikeThisQueryBuilder};
 pub use self::phrase_query::PhraseQuery;
-pub use self::query::{Query, QueryClone};
+pub use self::query::{EnableScoring, Query, QueryClone};
 pub use self::query_parser::{QueryParser, QueryParserError};
 pub use self::range_query::RangeQuery;
 pub use self::regex_query::RegexQuery;

src/query/more_like_this/mod.rs

@@ -1,4 +1,6 @@
 mod more_like_this;
+
+/// Module containing the different query implementations.
 mod query;
 
 pub use self::more_like_this::MoreLikeThis;

src/query/more_like_this/query.rs

@@ -1,7 +1,7 @@
 use super::MoreLikeThis;
-use crate::query::{Query, Weight};
+use crate::query::{EnableScoring, Query, Weight};
 use crate::schema::{Field, Value};
-use crate::{DocAddress, Result, Searcher};
+use crate::DocAddress;
 
 /// A query that matches all of the documents similar to a document
 /// or a set of field values provided.
@@ -42,16 +42,23 @@ impl MoreLikeThisQuery {
 }
 
 impl Query for MoreLikeThisQuery {
-    fn weight(&self, searcher: &Searcher, scoring_enabled: bool) -> Result<Box<dyn Weight>> {
+    fn weight(&self, enable_scoring: EnableScoring<'_>) -> crate::Result<Box<dyn Weight>> {
+        let searcher = match enable_scoring {
+            EnableScoring::Enabled(searcher) => searcher,
+            EnableScoring::Disabled(_) => {
+                let err = "MoreLikeThisQuery requires to enable scoring.".to_string();
+                return Err(crate::TantivyError::InvalidArgument(err));
+            }
+        };
         match &self.target {
             TargetDocument::DocumentAdress(doc_address) => self
                 .mlt
                 .query_with_document(searcher, *doc_address)?
-                .weight(searcher, scoring_enabled),
+                .weight(enable_scoring),
             TargetDocument::DocumentFields(doc_fields) => self
                 .mlt
                 .query_with_document_fields(searcher, doc_fields)?
-                .weight(searcher, scoring_enabled),
+                .weight(enable_scoring),
         }
     }
 }

src/query/phrase_query/mod.rs

@@ -14,7 +14,7 @@ pub mod tests {
     use super::*;
     use crate::collector::tests::{TEST_COLLECTOR_WITHOUT_SCORE, TEST_COLLECTOR_WITH_SCORE};
     use crate::core::Index;
-    use crate::query::{QueryParser, Weight};
+    use crate::query::{EnableScoring, QueryParser, Weight};
     use crate::schema::{Schema, Term, TEXT};
     use crate::{assert_nearly_equals, DocAddress, DocId, TERMINATED};
 
@@ -79,7 +79,8 @@ pub mod tests {
             .map(|text| Term::from_field_text(text_field, text))
             .collect();
         let phrase_query = PhraseQuery::new(terms);
-        let phrase_weight = phrase_query.phrase_weight(&searcher, false)?;
+        let phrase_weight =
+            phrase_query.phrase_weight(EnableScoring::Disabled(searcher.schema()))?;
         let mut phrase_scorer = phrase_weight.scorer(searcher.segment_reader(0), 1.0)?;
         assert_eq!(phrase_scorer.doc(), 1);
         assert_eq!(phrase_scorer.advance(), TERMINATED);
@@ -359,7 +360,9 @@ pub mod tests {
         let matching_docs = |query: &str| {
             let query_parser = QueryParser::for_index(&index, vec![json_field]);
             let phrase_query = query_parser.parse_query(query).unwrap();
-            let phrase_weight = phrase_query.weight(&searcher, false).unwrap();
+            let phrase_weight = phrase_query
+                .weight(EnableScoring::Disabled(searcher.schema()))
+                .unwrap();
             let mut phrase_scorer = phrase_weight
                 .scorer(searcher.segment_reader(0), 1.0f32)
                 .unwrap();

src/query/phrase_query/phrase_query.rs

@@ -1,7 +1,6 @@
 use super::PhraseWeight;
-use crate::core::searcher::Searcher;
 use crate::query::bm25::Bm25Weight;
-use crate::query::{Query, Weight};
+use crate::query::{EnableScoring, Query, Weight};
 use crate::schema::{Field, IndexRecordOption, Term};
 
 /// `PhraseQuery` matches a specific sequence of words.
@@ -67,7 +66,7 @@ impl PhraseQuery {
     /// 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.  
+    /// By default the slop is 0 meaning query terms need to be adjacent.
     pub fn set_slop(&mut self, value: u32) {
         self.slop = value;
     }
@@ -91,10 +90,9 @@ impl PhraseQuery {
     /// a specialized type [`PhraseWeight`] instead of a Boxed trait.
     pub(crate) fn phrase_weight(
         &self,
-        searcher: &Searcher,
-        scoring_enabled: bool,
+        enable_scoring: EnableScoring<'_>,
     ) -> crate::Result<PhraseWeight> {
-        let schema = searcher.schema();
+        let schema = enable_scoring.schema();
         let field_entry = schema.get_field_entry(self.field);
         let has_positions = field_entry
             .field_type()
@@ -109,8 +107,11 @@ impl PhraseQuery {
             )));
         }
         let terms = self.phrase_terms();
-        let bm25_weight = Bm25Weight::for_terms(searcher, &terms)?;
-        let mut weight = PhraseWeight::new(self.phrase_terms.clone(), bm25_weight, scoring_enabled);
+        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);
         }
@@ -122,8 +123,8 @@ impl Query for PhraseQuery {
     /// Create the weight associated with a query.
     ///
     /// See [`Weight`].
-    fn weight(&self, searcher: &Searcher, scoring_enabled: bool) -> crate::Result<Box<dyn Weight>> {
-        let phrase_weight = self.phrase_weight(searcher, scoring_enabled)?;
+    fn weight(&self, enable_scoring: EnableScoring<'_>) -> crate::Result<Box<dyn Weight>> {
+        let phrase_weight = self.phrase_weight(enable_scoring)?;
         Ok(Box::new(phrase_weight))
     }
 

src/query/phrase_query/phrase_scorer.rs

@@ -50,8 +50,7 @@ pub struct PhraseScorer<TPostings: Postings> {
     right: Vec<u32>,
     phrase_count: u32,
     fieldnorm_reader: FieldNormReader,
-    similarity_weight: Bm25Weight,
-    scoring_enabled: bool,
+    similarity_weight_opt: Option<Bm25Weight>,
     slop: u32,
 }
 
@@ -245,11 +244,11 @@ fn intersection_exists_with_slop(left: &[u32], right: &[u32], slop: u32) -> bool
 }
 
 impl<TPostings: Postings> PhraseScorer<TPostings> {
+    // If similarity_weight is None, then scoring is disabled.
     pub fn new(
         term_postings: Vec<(usize, TPostings)>,
-        similarity_weight: Bm25Weight,
+        similarity_weight_opt: Option<Bm25Weight>,
         fieldnorm_reader: FieldNormReader,
-        scoring_enabled: bool,
         slop: u32,
     ) -> PhraseScorer<TPostings> {
         let max_offset = term_postings
@@ -270,9 +269,8 @@ impl<TPostings: Postings> PhraseScorer<TPostings> {
             left: Vec::with_capacity(100),
             right: Vec::with_capacity(100),
             phrase_count: 0u32,
-            similarity_weight,
+            similarity_weight_opt,
             fieldnorm_reader,
-            scoring_enabled,
             slop,
         };
         if scorer.doc() != TERMINATED && !scorer.phrase_match() {
@@ -286,7 +284,7 @@ impl<TPostings: Postings> PhraseScorer<TPostings> {
     }
 
     fn phrase_match(&mut self) -> bool {
-        if self.scoring_enabled {
+        if self.similarity_weight_opt.is_some() {
             let count = self.compute_phrase_count();
             self.phrase_count = count;
             count > 0u32
@@ -388,8 +386,11 @@ impl<TPostings: Postings> Scorer for PhraseScorer<TPostings> {
     fn score(&mut self) -> Score {
         let doc = self.doc();
         let fieldnorm_id = self.fieldnorm_reader.fieldnorm_id(doc);
-        self.similarity_weight
-            .score(fieldnorm_id, self.phrase_count)
+        if let Some(similarity_weight) = self.similarity_weight_opt.as_ref() {
+            similarity_weight.score(fieldnorm_id, self.phrase_count)
+        } else {
+            1.0f32
+        }
     }
 }
 

src/query/phrase_query/phrase_weight.rs

@@ -10,30 +10,28 @@ use crate::{DocId, DocSet, Score};
 
 pub struct PhraseWeight {
     phrase_terms: Vec<(usize, Term)>,
-    similarity_weight: Bm25Weight,
-    scoring_enabled: bool,
+    similarity_weight_opt: Option<Bm25Weight>,
     slop: u32,
 }
 
 impl PhraseWeight {
     /// Creates a new phrase weight.
+    /// If `similarity_weight_opt` is None, then scoring is disabled
     pub fn new(
         phrase_terms: Vec<(usize, Term)>,
-        similarity_weight: Bm25Weight,
-        scoring_enabled: bool,
+        similarity_weight_opt: Option<Bm25Weight>,
     ) -> PhraseWeight {
         let slop = 0;
         PhraseWeight {
             phrase_terms,
-            similarity_weight,
-            scoring_enabled,
+            similarity_weight_opt,
             slop,
         }
     }
 
     fn fieldnorm_reader(&self, reader: &SegmentReader) -> crate::Result<FieldNormReader> {
         let field = self.phrase_terms[0].1.field();
-        if self.scoring_enabled {
+        if self.similarity_weight_opt.is_some() {
             if let Some(fieldnorm_reader) = reader.fieldnorms_readers().get_field(field)? {
                 return Ok(fieldnorm_reader);
             }
@@ -46,7 +44,10 @@ impl PhraseWeight {
         reader: &SegmentReader,
         boost: Score,
     ) -> crate::Result<Option<PhraseScorer<SegmentPostings>>> {
-        let similarity_weight = self.similarity_weight.boost_by(boost);
+        let similarity_weight_opt = self
+            .similarity_weight_opt
+            .as_ref()
+            .map(|similarity_weight| similarity_weight.boost_by(boost));
         let fieldnorm_reader = self.fieldnorm_reader(reader)?;
         let mut term_postings_list = Vec::new();
         if reader.has_deletes() {
@@ -74,9 +75,8 @@ impl PhraseWeight {
         }
         Ok(Some(PhraseScorer::new(
             term_postings_list,
-            similarity_weight,
+            similarity_weight_opt,
             fieldnorm_reader,
-            self.scoring_enabled,
             self.slop,
         )))
     }
@@ -108,7 +108,9 @@ impl Weight for PhraseWeight {
         let fieldnorm_id = fieldnorm_reader.fieldnorm_id(doc);
         let phrase_count = scorer.phrase_count();
         let mut explanation = Explanation::new("Phrase Scorer", scorer.score());
-        explanation.add_detail(self.similarity_weight.explain(fieldnorm_id, phrase_count));
+        if let Some(similarity_weight) = self.similarity_weight_opt.as_ref() {
+            explanation.add_detail(similarity_weight.explain(fieldnorm_id, phrase_count));
+        }
         Ok(explanation)
     }
 }
@@ -117,7 +119,7 @@ impl Weight for PhraseWeight {
 mod tests {
     use super::super::tests::create_index;
     use crate::docset::TERMINATED;
-    use crate::query::PhraseQuery;
+    use crate::query::{EnableScoring, PhraseQuery};
     use crate::{DocSet, Term};
 
     #[test]
@@ -130,7 +132,8 @@ mod tests {
             Term::from_field_text(text_field, "a"),
             Term::from_field_text(text_field, "b"),
         ]);
-        let phrase_weight = phrase_query.phrase_weight(&searcher, true).unwrap();
+        let enable_scoring = EnableScoring::Enabled(&searcher);
+        let phrase_weight = phrase_query.phrase_weight(enable_scoring).unwrap();
         let mut phrase_scorer = phrase_weight
             .phrase_scorer(searcher.segment_reader(0u32), 1.0)?
             .unwrap();

src/query/query.rs

@@ -5,8 +5,37 @@ use downcast_rs::impl_downcast;
 use super::Weight;
 use crate::core::searcher::Searcher;
 use crate::query::Explanation;
+use crate::schema::Schema;
 use crate::{DocAddress, Term};
 
+/// Argument used in `Query::weight(..)`
+#[derive(Copy, Clone)]
+pub enum EnableScoring<'a> {
+    /// Pass this to enable scoring.
+    Enabled(&'a Searcher),
+    /// Pass this to disable scoring.
+    /// This can improve performance.
+    Disabled(&'a Schema),
+}
+
+impl<'a> EnableScoring<'a> {
+    /// Returns the schema.
+    pub fn schema(&self) -> &Schema {
+        match self {
+            EnableScoring::Enabled(searcher) => searcher.schema(),
+            EnableScoring::Disabled(schema) => schema,
+        }
+    }
+
+    /// Returns true if the scoring is enabled.
+    pub fn is_scoring_enabled(&self) -> bool {
+        match self {
+            EnableScoring::Enabled(_) => true,
+            EnableScoring::Disabled(_) => false,
+        }
+    }
+}
+
 /// The `Query` trait defines a set of documents and a scoring method
 /// for those documents.
 ///
@@ -48,18 +77,18 @@ pub trait Query: QueryClone + Send + Sync + downcast_rs::Downcast + fmt::Debug {
     /// can increase performances.
     ///
     /// See [`Weight`].
-    fn weight(&self, searcher: &Searcher, scoring_enabled: bool) -> crate::Result<Box<dyn Weight>>;
+    fn weight(&self, enable_scoring: EnableScoring<'_>) -> crate::Result<Box<dyn Weight>>;
 
     /// Returns an `Explanation` for the score of the document.
     fn explain(&self, searcher: &Searcher, doc_address: DocAddress) -> crate::Result<Explanation> {
+        let weight = self.weight(EnableScoring::Enabled(searcher))?;
         let reader = searcher.segment_reader(doc_address.segment_ord);
-        let weight = self.weight(searcher, true)?;
         weight.explain(reader, doc_address.doc_id)
     }
 
     /// Returns the number of documents matching the query.
     fn count(&self, searcher: &Searcher) -> crate::Result<usize> {
-        let weight = self.weight(searcher, false)?;
+        let weight = self.weight(EnableScoring::Disabled(searcher.schema()))?;
         let mut result = 0;
         for reader in searcher.segment_readers() {
             result += weight.count(reader)? as usize;
@@ -93,8 +122,8 @@ where T: 'static + Query + Clone
 }
 
 impl Query for Box<dyn Query> {
-    fn weight(&self, searcher: &Searcher, scoring_enabled: bool) -> crate::Result<Box<dyn Weight>> {
-        self.as_ref().weight(searcher, scoring_enabled)
+    fn weight(&self, enabled_scoring: EnableScoring) -> crate::Result<Box<dyn Weight>> {
+        self.as_ref().weight(enabled_scoring)
     }
 
     fn count(&self, searcher: &Searcher) -> crate::Result<usize> {

src/query/query_parser/logical_ast.rs

@@ -15,6 +15,11 @@ pub enum LogicalLiteral {
         lower: Bound<Term>,
         upper: Bound<Term>,
     },
+    Set {
+        field: Field,
+        value_type: Type,
+        elements: Vec<Term>,
+    },
     All,
 }
 
@@ -87,6 +92,27 @@ impl fmt::Debug for LogicalLiteral {
                 ref upper,
                 ..
             } => write!(formatter, "({:?} TO {:?})", lower, upper),
+            LogicalLiteral::Set { ref elements, .. } => {
+                const MAX_DISPLAYED: usize = 10;
+
+                write!(formatter, "IN [")?;
+                for (i, element) in elements.iter().enumerate() {
+                    if i == 0 {
+                        write!(formatter, "{:?}", element)?;
+                    } else if i == MAX_DISPLAYED - 1 {
+                        write!(
+                            formatter,
+                            ", {:?}, ... ({} more)",
+                            element,
+                            elements.len() - i - 1
+                        )?;
+                        break;
+                    } else {
+                        write!(formatter, ", {:?}", element)?;
+                    }
+                }
+                write!(formatter, "]")
+            }
             LogicalLiteral::All => write!(formatter, "*"),
         }
     }

src/query/query_parser/query_parser.rs

@@ -13,7 +13,7 @@ use crate::indexer::{
 };
 use crate::query::{
     AllQuery, BooleanQuery, BoostQuery, EmptyQuery, Occur, PhraseQuery, Query, RangeQuery,
-    TermQuery,
+    TermQuery, TermSetQuery,
 };
 use crate::schema::{
     Facet, FacetParseError, Field, FieldType, IndexRecordOption, IntoIpv6Addr, Schema, Term, Type,
@@ -333,7 +333,9 @@ impl QueryParser {
     ) -> Result<Term, QueryParserError> {
         let field_entry = self.schema.get_field_entry(field);
         let field_type = field_entry.field_type();
-        if !field_type.is_indexed() {
+
+        let is_ip_and_fast = field_type.is_ip_addr() && field_type.is_fast();
+        if !field_type.is_indexed() && !is_ip_and_fast {
             return Err(QueryParserError::FieldNotIndexed(
                 field_entry.name().to_string(),
             ));
@@ -683,6 +685,31 @@ impl QueryParser {
                 }));
                 Ok(logical_ast)
             }
+            UserInputLeaf::Set {
+                field: full_field_opt,
+                elements,
+            } => {
+                let full_path = full_field_opt.ok_or_else(|| {
+                    QueryParserError::UnsupportedQuery(
+                        "Set query need to target a specific field.".to_string(),
+                    )
+                })?;
+                let (field, json_path) = self
+                    .split_full_path(&full_path)
+                    .ok_or_else(|| QueryParserError::FieldDoesNotExist(full_path.clone()))?;
+                let field_entry = self.schema.get_field_entry(field);
+                let value_type = field_entry.field_type().value_type();
+                let logical_ast = LogicalAst::Leaf(Box::new(LogicalLiteral::Set {
+                    elements: elements
+                        .into_iter()
+                        .map(|element| self.compute_boundary_term(field, json_path, &element))
+                        .collect::<Result<Vec<_>, _>>()?,
+
+                    field,
+                    value_type,
+                }));
+                Ok(logical_ast)
+            }
         }
     }
 }
@@ -701,6 +728,7 @@ fn convert_literal_to_query(logical_literal: LogicalLiteral) -> Box<dyn Query> {
         } => Box::new(RangeQuery::new_term_bounds(
             field, value_type, &lower, &upper,
         )),
+        LogicalLiteral::Set { elements, .. } => Box::new(TermSetQuery::new(elements)),
         LogicalLiteral::All => Box::new(AllQuery),
     }
 }
@@ -1060,6 +1088,28 @@ mod test {
         );
     }
 
+    fn extract_query_term_json_path(query: &str) -> String {
+        let LogicalAst::Leaf(literal) = parse_query_to_logical_ast(query, false).unwrap() else {
+            panic!();
+        };
+        let LogicalLiteral::Term(term) = *literal else {
+            panic!();
+        };
+        std::str::from_utf8(term.value_bytes()).unwrap().to_string()
+    }
+
+    #[test]
+    fn test_json_field_query_with_espaced_dot() {
+        assert_eq!(
+            extract_query_term_json_path(r#"json.k8s.node.name:hello"#),
+            "k8s\u{1}node\u{1}name\0shello"
+        );
+        assert_eq!(
+            extract_query_term_json_path(r#"json.k8s\.node\.name:hello"#),
+            "k8s.node.name\0shello"
+        );
+    }
+
     #[test]
     fn test_json_field_possibly_a_number() {
         test_parse_query_to_logical_ast_helper(
@@ -1539,4 +1589,29 @@ mod test {
             false,
         );
     }
+
+    #[test]
+    pub fn test_term_set_query() {
+        test_parse_query_to_logical_ast_helper(
+            "title: IN [a b cd]",
+            r#"IN [Term(type=Str, field=0, "a"), Term(type=Str, field=0, "b"), Term(type=Str, field=0, "cd")]"#,
+            false,
+        );
+        test_parse_query_to_logical_ast_helper(
+            "bytes: IN [AA== ABA= ABCD]",
+            r#"IN [Term(type=Bytes, field=12, [0]), Term(type=Bytes, field=12, [0, 16]), Term(type=Bytes, field=12, [0, 16, 131])]"#,
+            false,
+        );
+        test_parse_query_to_logical_ast_helper(
+            "signed: IN [1 2 -3]",
+            r#"IN [Term(type=I64, field=2, 1), Term(type=I64, field=2, 2), Term(type=I64, field=2, -3)]"#,
+            false,
+        );
+
+        test_parse_query_to_logical_ast_helper(
+            "float: IN [1.1 2.2 -3.3]",
+            r#"IN [Term(type=F64, field=10, 1.1), Term(type=F64, field=10, 2.2), Term(type=F64, field=10, -3.3)]"#,
+            false,
+        );
+    }
 }

src/query/range_query.rs

@@ -3,11 +3,11 @@ use std::ops::{Bound, Range};
 
 use common::BitSet;
 
-use crate::core::{Searcher, SegmentReader};
+use crate::core::SegmentReader;
 use crate::error::TantivyError;
 use crate::query::explanation::does_not_match;
 use crate::query::range_query_ip_fastfield::IPFastFieldRangeWeight;
-use crate::query::{BitSetDocSet, ConstScorer, Explanation, Query, Scorer, Weight};
+use crate::query::{BitSetDocSet, ConstScorer, EnableScoring, Explanation, Query, Scorer, Weight};
 use crate::schema::{Field, IndexRecordOption, Term, Type};
 use crate::termdict::{TermDictionary, TermStreamer};
 use crate::{DocId, Score};
@@ -253,12 +253,8 @@ impl RangeQuery {
 }
 
 impl Query for RangeQuery {
-    fn weight(
-        &self,
-        searcher: &Searcher,
-        _scoring_enabled: bool,
-    ) -> crate::Result<Box<dyn Weight>> {
-        let schema = searcher.schema();
+    fn weight(&self, enable_scoring: EnableScoring<'_>) -> crate::Result<Box<dyn Weight>> {
+        let schema = enable_scoring.schema();
         let field_type = schema.get_field_entry(self.field).field_type();
         let value_type = field_type.value_type();
         if value_type != self.value_type {

src/query/range_query_ip_fastfield.rs

@@ -7,10 +7,11 @@ use std::ops::{Bound, RangeInclusive};
 use std::sync::Arc;
 
 use common::BinarySerializable;
-use fastfield_codecs::{Column, MonotonicallyMappableToU128};
+use fastfield_codecs::{MonotonicallyMappableToU128, OptionalColumn};
 
 use super::range_query::map_bound;
 use super::{ConstScorer, Explanation, Scorer, Weight};
+use crate::fastfield::MultiValuedU128FastFieldReader;
 use crate::schema::{Cardinality, Field};
 use crate::{DocId, DocSet, Score, SegmentReader, TantivyError, TERMINATED};
 
@@ -44,15 +45,32 @@ impl Weight for IPFastFieldRangeWeight {
         match field_type.fastfield_cardinality().unwrap() {
             Cardinality::SingleValue => {
                 let ip_addr_fast_field = reader.fast_fields().ip_addr(self.field)?;
+                let minmax = ip_addr_fast_field
+                    .min_value()
+                    .zip(ip_addr_fast_field.max_value());
+                let value_range = bound_to_value_range(&self.left_bound, &self.right_bound, minmax);
+                let docset = IpRangeDocSet::new(
+                    value_range,
+                    IpFastFieldCardinality::SingleValue(ip_addr_fast_field),
+                );
+                Ok(Box::new(ConstScorer::new(docset, boost)))
+            }
+            Cardinality::MultiValues => {
+                let ip_addr_fast_field = reader.fast_fields().ip_addrs(self.field)?;
                 let value_range = bound_to_value_range(
                     &self.left_bound,
                     &self.right_bound,
-                    ip_addr_fast_field.as_ref(),
+                    Some((
+                        ip_addr_fast_field.min_value(),
+                        ip_addr_fast_field.max_value(),
+                    )),
+                );
+                let docset = IpRangeDocSet::new(
+                    value_range,
+                    IpFastFieldCardinality::MultiValue(ip_addr_fast_field),
                 );
-                let docset = IpRangeDocSet::new(value_range, ip_addr_fast_field);
                 Ok(Box::new(ConstScorer::new(docset, boost)))
             }
-            Cardinality::MultiValues => unimplemented!(),
         }
     }
 
@@ -73,18 +91,20 @@ impl Weight for IPFastFieldRangeWeight {
 fn bound_to_value_range(
     left_bound: &Bound<Ipv6Addr>,
     right_bound: &Bound<Ipv6Addr>,
-    column: &dyn Column<Ipv6Addr>,
+    min_max: Option<(Ipv6Addr, Ipv6Addr)>,
 ) -> RangeInclusive<Ipv6Addr> {
+    let (min_value, max_value) =
+        min_max.unwrap_or((Ipv6Addr::from(u128::MIN), Ipv6Addr::from(u128::MAX)));
     let start_value = match left_bound {
         Bound::Included(ip_addr) => *ip_addr,
         Bound::Excluded(ip_addr) => Ipv6Addr::from(ip_addr.to_u128() + 1),
-        Bound::Unbounded => column.min_value(),
+        Bound::Unbounded => min_value,
     };
 
     let end_value = match right_bound {
         Bound::Included(ip_addr) => *ip_addr,
         Bound::Excluded(ip_addr) => Ipv6Addr::from(ip_addr.to_u128() - 1),
-        Bound::Unbounded => column.max_value(),
+        Bound::Unbounded => max_value,
     };
     start_value..=end_value
 }
@@ -109,22 +129,39 @@ impl VecCursor {
     fn current(&self) -> Option<u32> {
         self.docs.get(self.current_pos).map(|el| *el as u32)
     }
-
     fn get_cleared_data(&mut self) -> &mut Vec<u32> {
         self.docs.clear();
         self.current_pos = 0;
         &mut self.docs
     }
-
+    fn last_value(&self) -> Option<u32> {
+        self.docs.iter().last().cloned()
+    }
     fn is_empty(&self) -> bool {
         self.current_pos >= self.docs.len()
     }
 }
 
+pub(crate) enum IpFastFieldCardinality {
+    SingleValue(Arc<dyn OptionalColumn<Ipv6Addr>>),
+    MultiValue(MultiValuedU128FastFieldReader<Ipv6Addr>),
+}
+
+impl IpFastFieldCardinality {
+    fn num_docs(&self) -> u32 {
+        match self {
+            IpFastFieldCardinality::SingleValue(single_value) => single_value.num_vals(),
+            IpFastFieldCardinality::MultiValue(multi_value) => {
+                multi_value.get_index_reader().num_docs()
+            }
+        }
+    }
+}
+
 struct IpRangeDocSet {
     /// The range filter on the values.
     value_range: RangeInclusive<Ipv6Addr>,
-    ip_addr_fast_field: Arc<dyn Column<Ipv6Addr>>,
+    ip_addr_fast_field: IpFastFieldCardinality,
     /// The next docid start range to fetch (inclusive).
     next_fetch_start: u32,
     /// Number of docs range checked in a batch.
@@ -141,18 +178,18 @@ struct IpRangeDocSet {
     last_seek_pos_opt: Option<u32>,
 }
 
-const DEFALT_FETCH_HORIZON: u32 = 128;
+const DEFAULT_FETCH_HORIZON: u32 = 128;
 impl IpRangeDocSet {
     fn new(
         value_range: RangeInclusive<Ipv6Addr>,
-        ip_addr_fast_field: Arc<dyn Column<Ipv6Addr>>,
+        ip_addr_fast_field: IpFastFieldCardinality,
     ) -> Self {
         let mut ip_range_docset = Self {
             value_range,
             ip_addr_fast_field,
             loaded_docs: VecCursor::new(),
             next_fetch_start: 0,
-            fetch_horizon: DEFALT_FETCH_HORIZON,
+            fetch_horizon: DEFAULT_FETCH_HORIZON,
             last_seek_pos_opt: None,
         };
         ip_range_docset.reset_fetch_range();
@@ -161,7 +198,7 @@ impl IpRangeDocSet {
     }
 
     fn reset_fetch_range(&mut self) {
-        self.fetch_horizon = DEFALT_FETCH_HORIZON;
+        self.fetch_horizon = DEFAULT_FETCH_HORIZON;
     }
 
     /// Returns true if more data could be fetched
@@ -190,20 +227,40 @@ impl IpRangeDocSet {
     fn fetch_horizon(&mut self, horizon: u32) -> bool {
         let mut finished_to_end = false;
 
-        let limit = self.ip_addr_fast_field.num_vals();
+        let limit = self.ip_addr_fast_field.num_docs();
         let mut end = self.next_fetch_start + horizon;
         if end >= limit {
             end = limit;
             finished_to_end = true;
         }
 
-        let data = self.loaded_docs.get_cleared_data();
-        self.ip_addr_fast_field.get_positions_for_value_range(
-            self.value_range.clone(),
-            self.next_fetch_start..end,
-            data,
-        );
+        match &self.ip_addr_fast_field {
+            IpFastFieldCardinality::MultiValue(multi) => {
+                let last_value = self.loaded_docs.last_value();
+
+                multi.get_docids_for_value_range(
+                    self.value_range.clone(),
+                    self.next_fetch_start..end,
+                    self.loaded_docs.get_cleared_data(),
+                );
+                // In case of multivalues, we may have an overlap of the same docid between fetching
+                // blocks
+                if let Some(last_value) = last_value {
+                    while self.loaded_docs.current() == Some(last_value) {
+                        self.loaded_docs.next();
+                    }
+                }
+            }
+            IpFastFieldCardinality::SingleValue(single) => {
+                single.get_docids_for_value_range(
+                    self.value_range.clone(),
+                    self.next_fetch_start..end,
+                    self.loaded_docs.get_cleared_data(),
+                );
+            }
+        }
         self.next_fetch_start = end;
+
         finished_to_end
     }
 }
@@ -214,7 +271,7 @@ impl DocSet for IpRangeDocSet {
         if let Some(docid) = self.loaded_docs.next() {
             docid as u32
         } else {
-            if self.next_fetch_start >= self.ip_addr_fast_field.num_vals() as u32 {
+            if self.next_fetch_start >= self.ip_addr_fast_field.num_docs() as u32 {
                 return TERMINATED;
             }
             self.fetch_block();
@@ -269,7 +326,7 @@ mod tests {
     use super::*;
     use crate::collector::Count;
     use crate::query::QueryParser;
-    use crate::schema::{Schema, FAST, INDEXED, STORED, STRING};
+    use crate::schema::{IpAddrOptions, Schema, FAST, STORED, STRING};
     use crate::Index;
 
     #[derive(Clone, Debug)]
@@ -280,12 +337,13 @@ mod tests {
 
     fn operation_strategy() -> impl Strategy<Value = Doc> {
         prop_oneof![
-            (0u64..100u64).prop_map(doc_from_id_1),
-            (1u64..100u64).prop_map(doc_from_id_2),
+            (0u64..10_000u64).prop_map(doc_from_id_1),
+            (1u64..10_000u64).prop_map(doc_from_id_2),
         ]
     }
 
     pub fn doc_from_id_1(id: u64) -> Doc {
+        let id = id * 1000;
         Doc {
             // ip != id
             id: id.to_string(),
@@ -293,6 +351,7 @@ mod tests {
         }
     }
     fn doc_from_id_2(id: u64) -> Doc {
+        let id = id * 1000;
         Doc {
             // ip != id
             id: (id - 1).to_string(),
@@ -310,6 +369,12 @@ mod tests {
 
     #[test]
     fn ip_range_regression1_test() {
+        let ops = vec![doc_from_id_1(0)];
+        assert!(test_ip_range_for_docs(ops).is_ok());
+    }
+
+    #[test]
+    fn ip_range_regression2_test() {
         let ops = vec![
             doc_from_id_1(52),
             doc_from_id_1(63),
@@ -321,14 +386,20 @@ mod tests {
     }
 
     #[test]
-    fn ip_range_regression2_test() {
-        let ops = vec![doc_from_id_1(0)];
+    fn ip_range_regression3_test() {
+        let ops = vec![doc_from_id_1(1), doc_from_id_1(2), doc_from_id_1(3)];
         assert!(test_ip_range_for_docs(ops).is_ok());
     }
 
     pub fn create_index_from_docs(docs: &[Doc]) -> Index {
         let mut schema_builder = Schema::builder();
-        let ip_field = schema_builder.add_ip_addr_field("ip", INDEXED | STORED | FAST);
+        let ip_field = schema_builder.add_ip_addr_field("ip", STORED | FAST);
+        let ips_field = schema_builder.add_ip_addr_field(
+            "ips",
+            IpAddrOptions::default()
+                .set_fast(Cardinality::MultiValues)
+                .set_indexed(),
+        );
         let text_field = schema_builder.add_text_field("id", STRING | STORED);
         let schema = schema_builder.build();
         let index = Index::create_in_ram(schema);
@@ -338,6 +409,8 @@ mod tests {
             for doc in docs.iter() {
                 index_writer
                     .add_document(doc!(
+                        ips_field => doc.ip,
+                        ips_field => doc.ip,
                         ip_field => doc.ip,
                         text_field => doc.id.to_string(),
                     ))
@@ -361,8 +434,8 @@ mod tests {
                 .unwrap()
         };
 
-        let gen_query_inclusive = |from: Ipv6Addr, to: Ipv6Addr| {
-            format!("ip:[{} TO {}]", &from.to_string(), &to.to_string())
+        let gen_query_inclusive = |field: &str, from: Ipv6Addr, to: Ipv6Addr| {
+            format!("{}:[{} TO {}]", field, &from.to_string(), &to.to_string())
         };
 
         let test_sample = |sample_docs: Vec<Doc>| {
@@ -373,7 +446,10 @@ mod tests {
                 .filter(|doc| (ips[0]..=ips[1]).contains(&doc.ip))
                 .count();
 
-            let query = gen_query_inclusive(ips[0], ips[1]);
+            let query = gen_query_inclusive("ip", ips[0], ips[1]);
+            assert_eq!(get_num_hits(query_from_text(&query)), expected_num_hits);
+
+            let query = gen_query_inclusive("ips", ips[0], ips[1]);
             assert_eq!(get_num_hits(query_from_text(&query)), expected_num_hits);
 
             // Intersection search
@@ -382,7 +458,20 @@ mod tests {
                 .iter()
                 .filter(|doc| (ips[0]..=ips[1]).contains(&doc.ip) && doc.id == id_filter)
                 .count();
-            let query = format!("{} AND id:{}", query, &id_filter);
+            let query = format!(
+                "{} AND id:{}",
+                gen_query_inclusive("ip", ips[0], ips[1]),
+                &id_filter
+            );
+            assert_eq!(get_num_hits(query_from_text(&query)), expected_num_hits);
+
+            // Intersection search on multivalue ip field
+            let id_filter = sample_docs[0].id.to_string();
+            let query = format!(
+                "{} AND id:{}",
+                gen_query_inclusive("ips", ips[0], ips[1]),
+                &id_filter
+            );
             assert_eq!(get_num_hits(query_from_text(&query)), expected_num_hits);
         };
 
@@ -402,7 +491,8 @@ mod tests {
 #[cfg(all(test, feature = "unstable"))]
 mod bench {
 
-    use rand::{thread_rng, Rng};
+    use rand::rngs::StdRng;
+    use rand::{Rng, SeedableRng};
     use test::Bencher;
 
     use super::tests::*;
@@ -412,7 +502,7 @@ mod bench {
     use crate::Index;
 
     fn get_index_0_to_100() -> Index {
-        let mut rng = thread_rng();
+        let mut rng = StdRng::from_seed([1u8; 32]);
         let num_vals = 100_000;
         let docs: Vec<_> = (0..num_vals)
             .map(|_i| {
@@ -424,8 +514,10 @@ mod bench {
                     "many".to_string() // 90%
                 };
                 Doc {
-                    id: id,
+                    id,
                     // Multiply by 1000, so that we create many buckets in the compact space
+                    // The benches depend on this range to select n-percent of elements with the
+                    // methods below.
                     ip: Ipv6Addr::from_u128(rng.gen_range(0..100) * 1000),
                 }
             })
@@ -434,22 +526,42 @@ mod bench {
         let index = create_index_from_docs(&docs);
         index
     }
+
+    fn get_90_percent() -> RangeInclusive<Ipv6Addr> {
+        let start = Ipv6Addr::from_u128(0);
+        let end = Ipv6Addr::from_u128(90 * 1000);
+        start..=end
+    }
+
+    fn get_10_percent() -> RangeInclusive<Ipv6Addr> {
+        let start = Ipv6Addr::from_u128(0);
+        let end = Ipv6Addr::from_u128(10 * 1000);
+        start..=end
+    }
+
+    fn get_1_percent() -> RangeInclusive<Ipv6Addr> {
+        let start = Ipv6Addr::from_u128(10 * 1000);
+        let end = Ipv6Addr::from_u128(10 * 1000);
+        start..=end
+    }
+
     fn excute_query(
-        start_inclusive: Ipv6Addr,
-        end_inclusive: Ipv6Addr,
+        field: &str,
+        ip_range: RangeInclusive<Ipv6Addr>,
         suffix: &str,
         index: &Index,
     ) -> usize {
-        let gen_query_inclusive = |from: Ipv6Addr, to: Ipv6Addr| {
+        let gen_query_inclusive = |from: &Ipv6Addr, to: &Ipv6Addr| {
             format!(
-                "ip:[{} TO {}] {}",
+                "{}:[{} TO {}] {}",
+                field,
                 &from.to_string(),
                 &to.to_string(),
                 suffix
             )
         };
 
-        let query = gen_query_inclusive(start_inclusive, end_inclusive);
+        let query = gen_query_inclusive(ip_range.start(), ip_range.end());
         let query_from_text = |text: &str| {
             QueryParser::for_index(&index, vec![])
                 .parse_query(text)
@@ -465,131 +577,153 @@ mod bench {
     fn bench_ip_range_hit_90_percent(bench: &mut Bencher) {
         let index = get_index_0_to_100();
 
-        bench.iter(|| {
-            let start = Ipv6Addr::from_u128(0);
-            let end = Ipv6Addr::from_u128(90 * 1000);
-
-            excute_query(start, end, "", &index)
-        });
+        bench.iter(|| excute_query("ip", get_90_percent(), "", &index));
     }
 
     #[bench]
     fn bench_ip_range_hit_10_percent(bench: &mut Bencher) {
         let index = get_index_0_to_100();
 
-        bench.iter(|| {
-            let start = Ipv6Addr::from_u128(0);
-            let end = Ipv6Addr::from_u128(10 * 1000);
-
-            excute_query(start, end, "", &index)
-        });
+        bench.iter(|| excute_query("ip", get_10_percent(), "", &index));
     }
 
     #[bench]
     fn bench_ip_range_hit_1_percent(bench: &mut Bencher) {
         let index = get_index_0_to_100();
 
-        bench.iter(|| {
-            let start = Ipv6Addr::from_u128(10 * 1000);
-            let end = Ipv6Addr::from_u128(10 * 1000);
-
-            excute_query(start, end, "", &index)
-        });
+        bench.iter(|| excute_query("ip", get_1_percent(), "", &index));
     }
 
     #[bench]
     fn bench_ip_range_hit_10_percent_intersect_with_10_percent(bench: &mut Bencher) {
         let index = get_index_0_to_100();
 
-        bench.iter(|| {
-            let start = Ipv6Addr::from_u128(0);
-            let end = Ipv6Addr::from_u128(10 * 1000);
-
-            excute_query(start, end, "AND id:few", &index)
-        });
+        bench.iter(|| excute_query("ip", get_10_percent(), "AND id:few", &index));
     }
 
     #[bench]
     fn bench_ip_range_hit_1_percent_intersect_with_10_percent(bench: &mut Bencher) {
         let index = get_index_0_to_100();
 
-        bench.iter(|| {
-            let start = Ipv6Addr::from_u128(10 * 1000);
-            let end = Ipv6Addr::from_u128(10 * 1000);
-
-            excute_query(start, end, "AND id:few", &index)
-        });
+        bench.iter(|| excute_query("ip", get_1_percent(), "AND id:few", &index));
     }
 
     #[bench]
     fn bench_ip_range_hit_1_percent_intersect_with_90_percent(bench: &mut Bencher) {
         let index = get_index_0_to_100();
 
-        bench.iter(|| {
-            let start = Ipv6Addr::from_u128(10 * 1000);
-            let end = Ipv6Addr::from_u128(10 * 1000);
-
-            excute_query(start, end, "AND id:many", &index)
-        });
+        bench.iter(|| excute_query("ip", get_1_percent(), "AND id:many", &index));
     }
 
     #[bench]
     fn bench_ip_range_hit_1_percent_intersect_with_1_percent(bench: &mut Bencher) {
         let index = get_index_0_to_100();
 
-        bench.iter(|| {
-            let start = Ipv6Addr::from_u128(10 * 1000);
-            let end = Ipv6Addr::from_u128(10 * 1000);
-
-            excute_query(start, end, "AND id:veryfew", &index)
-        });
+        bench.iter(|| excute_query("ip", get_1_percent(), "AND id:veryfew", &index));
     }
 
     #[bench]
     fn bench_ip_range_hit_10_percent_intersect_with_90_percent(bench: &mut Bencher) {
         let index = get_index_0_to_100();
 
-        bench.iter(|| {
-            let start = Ipv6Addr::from_u128(0);
-            let end = Ipv6Addr::from_u128(10 * 1000);
-
-            excute_query(start, end, "AND id:many", &index)
-        });
+        bench.iter(|| excute_query("ip", get_10_percent(), "AND id:many", &index));
     }
 
     #[bench]
     fn bench_ip_range_hit_90_percent_intersect_with_90_percent(bench: &mut Bencher) {
         let index = get_index_0_to_100();
 
-        bench.iter(|| {
-            let start = Ipv6Addr::from_u128(0);
-            let end = Ipv6Addr::from_u128(90 * 1000);
-
-            excute_query(start, end, "AND id:many", &index)
-        });
+        bench.iter(|| excute_query("ip", get_90_percent(), "AND id:many", &index));
     }
 
     #[bench]
     fn bench_ip_range_hit_90_percent_intersect_with_10_percent(bench: &mut Bencher) {
         let index = get_index_0_to_100();
 
-        bench.iter(|| {
-            let start = Ipv6Addr::from_u128(0);
-            let end = Ipv6Addr::from_u128(90 * 1000);
-
-            excute_query(start, end, "AND id:few", &index)
-        });
+        bench.iter(|| excute_query("ip", get_90_percent(), "AND id:few", &index));
     }
 
     #[bench]
     fn bench_ip_range_hit_90_percent_intersect_with_1_percent(bench: &mut Bencher) {
         let index = get_index_0_to_100();
 
-        bench.iter(|| {
-            let start = Ipv6Addr::from_u128(0);
-            let end = Ipv6Addr::from_u128(90 * 1000);
+        bench.iter(|| excute_query("ip", get_90_percent(), "AND id:veryfew", &index));
+    }
 
-            excute_query(start, end, "AND id:veryfew", &index)
-        });
+    #[bench]
+    fn bench_ip_range_hit_90_percent_multi(bench: &mut Bencher) {
+        let index = get_index_0_to_100();
+
+        bench.iter(|| excute_query("ips", get_90_percent(), "", &index));
+    }
+
+    #[bench]
+    fn bench_ip_range_hit_10_percent_multi(bench: &mut Bencher) {
+        let index = get_index_0_to_100();
+
+        bench.iter(|| excute_query("ips", get_10_percent(), "", &index));
+    }
+
+    #[bench]
+    fn bench_ip_range_hit_1_percent_multi(bench: &mut Bencher) {
+        let index = get_index_0_to_100();
+
+        bench.iter(|| excute_query("ips", get_1_percent(), "", &index));
+    }
+
+    #[bench]
+    fn bench_ip_range_hit_10_percent_intersect_with_10_percent_multi(bench: &mut Bencher) {
+        let index = get_index_0_to_100();
+
+        bench.iter(|| excute_query("ips", get_10_percent(), "AND id:few", &index));
+    }
+
+    #[bench]
+    fn bench_ip_range_hit_1_percent_intersect_with_10_percent_multi(bench: &mut Bencher) {
+        let index = get_index_0_to_100();
+
+        bench.iter(|| excute_query("ips", get_1_percent(), "AND id:few", &index));
+    }
+
+    #[bench]
+    fn bench_ip_range_hit_1_percent_intersect_with_90_percent_multi(bench: &mut Bencher) {
+        let index = get_index_0_to_100();
+
+        bench.iter(|| excute_query("ips", get_1_percent(), "AND id:many", &index));
+    }
+
+    #[bench]
+    fn bench_ip_range_hit_1_percent_intersect_with_1_percent_multi(bench: &mut Bencher) {
+        let index = get_index_0_to_100();
+
+        bench.iter(|| excute_query("ips", get_1_percent(), "AND id:veryfew", &index));
+    }
+
+    #[bench]
+    fn bench_ip_range_hit_10_percent_intersect_with_90_percent_multi(bench: &mut Bencher) {
+        let index = get_index_0_to_100();
+
+        bench.iter(|| excute_query("ips", get_10_percent(), "AND id:many", &index));
+    }
+
+    #[bench]
+    fn bench_ip_range_hit_90_percent_intersect_with_90_percent_multi(bench: &mut Bencher) {
+        let index = get_index_0_to_100();
+
+        bench.iter(|| excute_query("ips", get_90_percent(), "AND id:many", &index));
+    }
+
+    #[bench]
+    fn bench_ip_range_hit_90_percent_intersect_with_10_percent_multi(bench: &mut Bencher) {
+        let index = get_index_0_to_100();
+
+        bench.iter(|| excute_query("ips", get_90_percent(), "AND id:few", &index));
+    }
+
+    #[bench]
+    fn bench_ip_range_hit_90_percent_intersect_with_1_percent_multi(bench: &mut Bencher) {
+        let index = get_index_0_to_100();
+
+        bench.iter(|| excute_query("ips", get_90_percent(), "AND id:veryfew", &index));
     }
 }

src/query/regex_query.rs

@@ -4,9 +4,8 @@ use std::sync::Arc;
 use tantivy_fst::Regex;
 
 use crate::error::TantivyError;
-use crate::query::{AutomatonWeight, Query, Weight};
+use crate::query::{AutomatonWeight, EnableScoring, Query, Weight};
 use crate::schema::Field;
-use crate::Searcher;
 
 /// A Regex Query matches all of the documents
 /// containing a specific term that matches
@@ -82,11 +81,7 @@ impl RegexQuery {
 }
 
 impl Query for RegexQuery {
-    fn weight(
-        &self,
-        _searcher: &Searcher,
-        _scoring_enabled: bool,
-    ) -> crate::Result<Box<dyn Weight>> {
+    fn weight(&self, _enabled_scoring: EnableScoring<'_>) -> crate::Result<Box<dyn Weight>> {
         Ok(Box::new(self.specialized_weight()))
     }
 }

src/query/set_query.rs

@@ -4,9 +4,9 @@ use tantivy_fst::raw::CompiledAddr;
 use tantivy_fst::{Automaton, Map};
 
 use crate::query::score_combiner::DoNothingCombiner;
-use crate::query::{AutomatonWeight, BooleanWeight, Occur, Query, Weight};
-use crate::schema::Field;
-use crate::{Searcher, Term};
+use crate::query::{AutomatonWeight, BooleanWeight, EnableScoring, Occur, Query, Weight};
+use crate::schema::{Field, Schema};
+use crate::Term;
 
 /// A Term Set Query matches all of the documents containing any of the Term provided
 #[derive(Debug, Clone)]
@@ -32,12 +32,12 @@ impl TermSetQuery {
 
     fn specialized_weight(
         &self,
-        searcher: &Searcher,
+        schema: &Schema,
     ) -> crate::Result<BooleanWeight<DoNothingCombiner>> {
         let mut sub_queries: Vec<(_, Box<dyn Weight>)> = Vec::with_capacity(self.terms_map.len());
 
         for (&field, sorted_terms) in self.terms_map.iter() {
-            let field_entry = searcher.schema().get_field_entry(field);
+            let field_entry = schema.get_field_entry(field);
             let field_type = field_entry.field_type();
             if !field_type.is_indexed() {
                 let error_msg = format!("Field {:?} is not indexed.", field_entry.name());
@@ -65,12 +65,8 @@ impl TermSetQuery {
 }
 
 impl Query for TermSetQuery {
-    fn weight(
-        &self,
-        searcher: &Searcher,
-        _scoring_enabled: bool,
-    ) -> crate::Result<Box<dyn Weight>> {
-        Ok(Box::new(self.specialized_weight(searcher)?))
+    fn weight(&self, enable_scoring: EnableScoring<'_>) -> crate::Result<Box<dyn Weight>> {
+        Ok(Box::new(self.specialized_weight(enable_scoring.schema())?))
     }
 }
 
@@ -105,9 +101,8 @@ impl Automaton for SetDfaWrapper {
 
 #[cfg(test)]
 mod tests {
-
     use crate::collector::TopDocs;
-    use crate::query::TermSetQuery;
+    use crate::query::{QueryParser, TermSetQuery};
     use crate::schema::{Schema, TEXT};
     use crate::{assert_nearly_equals, Index, Term};
 
@@ -219,4 +214,31 @@ mod tests {
 
         Ok(())
     }
+
+    #[test]
+    fn test_term_set_query_parser() -> crate::Result<()> {
+        let mut schema_builder = Schema::builder();
+        schema_builder.add_text_field("field", TEXT);
+        let schema = schema_builder.build();
+        let index = Index::create_in_ram(schema.clone());
+        let mut index_writer = index.writer_for_tests()?;
+        let field = schema.get_field("field").unwrap();
+        index_writer.add_document(doc!(
+          field => "val1",
+        ))?;
+        index_writer.add_document(doc!(
+          field => "val2",
+        ))?;
+        index_writer.add_document(doc!(
+          field => "val3",
+        ))?;
+        index_writer.commit()?;
+        let reader = index.reader()?;
+        let searcher = reader.searcher();
+        let query_parser = QueryParser::for_index(&index, vec![]);
+        let query = query_parser.parse_query("field: IN [val1 val2]")?;
+        let top_docs = searcher.search(&query, &TopDocs::with_limit(3))?;
+        assert_eq!(top_docs.len(), 2);
+        Ok(())
+    }
 }

src/query/term_query/mod.rs

@@ -12,7 +12,7 @@ mod tests {
     use crate::collector::TopDocs;
     use crate::docset::DocSet;
     use crate::postings::compression::COMPRESSION_BLOCK_SIZE;
-    use crate::query::{Query, QueryParser, Scorer, TermQuery};
+    use crate::query::{EnableScoring, Query, QueryParser, Scorer, TermQuery};
     use crate::schema::{Field, IndexRecordOption, Schema, STRING, TEXT};
     use crate::{assert_nearly_equals, DocAddress, Index, Term, TERMINATED};
 
@@ -34,7 +34,7 @@ mod tests {
             Term::from_field_text(text_field, "a"),
             IndexRecordOption::Basic,
         );
-        let term_weight = term_query.weight(&searcher, true)?;
+        let term_weight = term_query.weight(EnableScoring::Enabled(&searcher))?;
         let segment_reader = searcher.segment_reader(0);
         let mut term_scorer = term_weight.scorer(segment_reader, 1.0)?;
         assert_eq!(term_scorer.doc(), 0);
@@ -62,7 +62,7 @@ mod tests {
             Term::from_field_text(text_field, "a"),
             IndexRecordOption::Basic,
         );
-        let term_weight = term_query.weight(&searcher, true)?;
+        let term_weight = term_query.weight(EnableScoring::Enabled(&searcher))?;
         let segment_reader = searcher.segment_reader(0);
         let mut term_scorer = term_weight.scorer(segment_reader, 1.0)?;
         for i in 0u32..COMPRESSION_BLOCK_SIZE as u32 {
@@ -158,7 +158,7 @@ mod tests {
         let term_a = Term::from_field_text(text_field, "a");
         let term_query = TermQuery::new(term_a, IndexRecordOption::Basic);
         let searcher = index.reader()?.searcher();
-        let term_weight = term_query.weight(&searcher, false)?;
+        let term_weight = term_query.weight(EnableScoring::Disabled(searcher.schema()))?;
         let mut term_scorer = term_weight.scorer(searcher.segment_reader(0u32), 1.0)?;
         assert_eq!(term_scorer.doc(), 0u32);
         term_scorer.seek(1u32);

src/query/term_query/term_query.rs

@@ -2,9 +2,9 @@ use std::fmt;
 
 use super::term_weight::TermWeight;
 use crate::query::bm25::Bm25Weight;
-use crate::query::{Explanation, Query, Weight};
+use crate::query::{EnableScoring, Explanation, Query, Weight};
 use crate::schema::IndexRecordOption;
-use crate::{Searcher, Term};
+use crate::Term;
 
 /// A Term query matches all of the documents
 /// containing a specific term.
@@ -87,19 +87,23 @@ impl TermQuery {
     /// This is useful for optimization purpose.
     pub fn specialized_weight(
         &self,
-        searcher: &Searcher,
-        scoring_enabled: bool,
+        enable_scoring: EnableScoring<'_>,
     ) -> crate::Result<TermWeight> {
-        let field_entry = searcher.schema().get_field_entry(self.term.field());
+        let schema = enable_scoring.schema();
+        let field_entry = schema.get_field_entry(self.term.field());
         if !field_entry.is_indexed() {
             let error_msg = format!("Field {:?} is not indexed.", field_entry.name());
             return Err(crate::TantivyError::SchemaError(error_msg));
         }
-        let bm25_weight = if scoring_enabled {
-            Bm25Weight::for_terms(searcher, &[self.term.clone()])?
-        } else {
-            Bm25Weight::new(Explanation::new("<no score>".to_string(), 1.0f32), 1.0f32)
+        let bm25_weight = match enable_scoring {
+            EnableScoring::Enabled(searcher) => {
+                Bm25Weight::for_terms(searcher, &[self.term.clone()])?
+            }
+            EnableScoring::Disabled(_schema) => {
+                Bm25Weight::new(Explanation::new("<no score>".to_string(), 1.0f32), 1.0f32)
+            }
         };
+        let scoring_enabled = enable_scoring.is_scoring_enabled();
         let index_record_option = if scoring_enabled {
             self.index_record_option
         } else {
@@ -115,10 +119,8 @@ impl TermQuery {
 }
 
 impl Query for TermQuery {
-    fn weight(&self, searcher: &Searcher, scoring_enabled: bool) -> crate::Result<Box<dyn Weight>> {
-        Ok(Box::new(
-            self.specialized_weight(searcher, scoring_enabled)?,
-        ))
+    fn weight(&self, enable_scoring: EnableScoring<'_>) -> crate::Result<Box<dyn Weight>> {
+        Ok(Box::new(self.specialized_weight(enable_scoring)?))
     }
     fn query_terms<'a>(&'a self, visitor: &mut dyn FnMut(&'a Term, bool)) {
         visitor(&self.term, false);

src/query/term_query/term_scorer.rs

@@ -130,7 +130,7 @@ mod tests {
     use crate::merge_policy::NoMergePolicy;
     use crate::postings::compression::COMPRESSION_BLOCK_SIZE;
     use crate::query::term_query::TermScorer;
-    use crate::query::{Bm25Weight, Scorer, TermQuery};
+    use crate::query::{Bm25Weight, EnableScoring, Scorer, TermQuery};
     use crate::schema::{IndexRecordOption, Schema, TEXT};
     use crate::{
         assert_nearly_equals, DocId, DocSet, Index, Score, Searcher, SegmentId, Term, TERMINATED,
@@ -250,7 +250,7 @@ mod tests {
     }
 
     fn test_block_wand_aux(term_query: &TermQuery, searcher: &Searcher) -> crate::Result<()> {
-        let term_weight = term_query.specialized_weight(searcher, true)?;
+        let term_weight = term_query.specialized_weight(EnableScoring::Enabled(searcher))?;
         for reader in searcher.segment_readers() {
             let mut block_max_scores = vec![];
             let mut block_max_scores_b = vec![];

src/query/term_query/term_weight.rs

@@ -5,7 +5,7 @@ use crate::fieldnorm::FieldNormReader;
 use crate::postings::SegmentPostings;
 use crate::query::bm25::Bm25Weight;
 use crate::query::explanation::does_not_match;
-use crate::query::weight::for_each_scorer;
+use crate::query::weight::{for_each_docset, for_each_scorer};
 use crate::query::{Explanation, Scorer, Weight};
 use crate::schema::IndexRecordOption;
 use crate::{DocId, Score, Term};
@@ -56,6 +56,18 @@ impl Weight for TermWeight {
         Ok(())
     }
 
+    /// Iterates through all of the document matched by the DocSet
+    /// `DocSet` and push the scored documents to the collector.
+    fn for_each_no_score(
+        &self,
+        reader: &SegmentReader,
+        callback: &mut dyn FnMut(DocId),
+    ) -> crate::Result<()> {
+        let mut scorer = self.specialized_scorer(reader, 1.0)?;
+        for_each_docset(&mut scorer, callback);
+        Ok(())
+    }
+
     /// Calls `callback` with all of the `(doc, score)` for which score
     /// is exceeding a given threshold.
     ///

src/query/union.rs

@@ -94,8 +94,8 @@ impl<TScorer: Scorer, TScoreCombiner: ScoreCombiner> Union<TScorer, TScoreCombin
             self.doc = min_doc;
             refill(
                 &mut self.docsets,
-                &mut *self.bitsets,
-                &mut *self.scores,
+                &mut self.bitsets,
+                &mut self.scores,
                 min_doc,
             );
             true

src/query/weight.rs

@@ -1,10 +1,10 @@
 use super::Scorer;
 use crate::core::SegmentReader;
 use crate::query::Explanation;
-use crate::{DocId, Score, TERMINATED};
+use crate::{DocId, DocSet, Score, TERMINATED};
 
-/// Iterates through all of the document matched by the DocSet
-/// `DocSet` and push the scored documents to the collector.
+/// Iterates through all of the documents and scores matched by the DocSet
+/// `DocSet`.
 pub(crate) fn for_each_scorer<TScorer: Scorer + ?Sized>(
     scorer: &mut TScorer,
     callback: &mut dyn FnMut(DocId, Score),
@@ -16,6 +16,16 @@ pub(crate) fn for_each_scorer<TScorer: Scorer + ?Sized>(
     }
 }
 
+/// Iterates through all of the documents matched by the DocSet
+/// `DocSet`.
+pub(crate) fn for_each_docset<T: DocSet + ?Sized>(docset: &mut T, callback: &mut dyn FnMut(DocId)) {
+    let mut doc = docset.doc();
+    while doc != TERMINATED {
+        callback(doc);
+        doc = docset.advance();
+    }
+}
+
 /// Calls `callback` with all of the `(doc, score)` for which score
 /// is exceeding a given threshold.
 ///
@@ -78,6 +88,18 @@ pub trait Weight: Send + Sync + 'static {
         Ok(())
     }
 
+    /// Iterates through all of the document matched by the DocSet
+    /// `DocSet` and push the scored documents to the collector.
+    fn for_each_no_score(
+        &self,
+        reader: &SegmentReader,
+        callback: &mut dyn FnMut(DocId),
+    ) -> crate::Result<()> {
+        let mut docset = self.scorer(reader, 1.0)?;
+        for_each_docset(docset.as_mut(), callback);
+        Ok(())
+    }
+
     /// Calls `callback` with all of the `(doc, score)` for which score
     /// is exceeding a given threshold.
     ///

src/store/footer.rs

@@ -2,7 +2,7 @@ use std::io;
 
 use common::{BinarySerializable, FixedSize, HasLen};
 
-use super::Decompressor;
+use super::{Decompressor, DOC_STORE_VERSION};
 use crate::directory::FileSlice;
 
 #[derive(Debug, Clone, PartialEq)]
@@ -17,6 +17,7 @@ pub struct DocStoreFooter {
 /// - reserved for future use: 15 bytes
 impl BinarySerializable for DocStoreFooter {
     fn serialize<W: io::Write>(&self, writer: &mut W) -> io::Result<()> {
+        BinarySerializable::serialize(&DOC_STORE_VERSION, writer)?;
         BinarySerializable::serialize(&self.offset, writer)?;
         BinarySerializable::serialize(&self.decompressor.get_id(), writer)?;
         writer.write_all(&[0; 15])?;
@@ -24,6 +25,13 @@ impl BinarySerializable for DocStoreFooter {
     }
 
     fn deserialize<R: io::Read>(reader: &mut R) -> io::Result<Self> {
+        let doc_store_version = u32::deserialize(reader)?;
+        if doc_store_version != DOC_STORE_VERSION {
+            panic!(
+                "actual doc store version: {}, expected: {}",
+                doc_store_version, DOC_STORE_VERSION
+            );
+        }
         let offset = u64::deserialize(reader)?;
         let compressor_id = u8::deserialize(reader)?;
         let mut skip_buf = [0; 15];
@@ -36,7 +44,7 @@ impl BinarySerializable for DocStoreFooter {
 }
 
 impl FixedSize for DocStoreFooter {
-    const SIZE_IN_BYTES: usize = 24;
+    const SIZE_IN_BYTES: usize = 28;
 }
 
 impl DocStoreFooter {

src/store/mod.rs

@@ -44,6 +44,9 @@ pub use self::reader::{CacheStats, StoreReader};
 pub use self::writer::StoreWriter;
 mod store_compressor;
 
+/// Doc store version in footer to handle format changes.
+pub(crate) const DOC_STORE_VERSION: u32 = 1;
+
 #[cfg(feature = "lz4-compression")]
 mod compression_lz4_block;
 

src/termdict/tests.rs

@@ -229,10 +229,10 @@ fn test_empty_string() -> crate::Result<()> {
     let buffer: Vec<u8> = {
         let mut term_dictionary_builder = TermDictionaryBuilder::create(vec![]).unwrap();
         term_dictionary_builder
-            .insert(&[], &make_term_info(1_u64))
+            .insert([], &make_term_info(1_u64))
             .unwrap();
         term_dictionary_builder
-            .insert(&[1u8], &make_term_info(2_u64))
+            .insert([1u8], &make_term_info(2_u64))
             .unwrap();
         term_dictionary_builder.finish()?
     };
@@ -252,7 +252,7 @@ fn stream_range_test_dict() -> crate::Result<TermDictionary> {
         let mut term_dictionary_builder = TermDictionaryBuilder::create(Vec::new())?;
         for i in 0u8..10u8 {
             let number_arr = [i; 1];
-            term_dictionary_builder.insert(&number_arr, &make_term_info(i as u64))?;
+            term_dictionary_builder.insert(number_arr, &make_term_info(i as u64))?;
         }
         term_dictionary_builder.finish()?
     };

src/tokenizer/stop_word_filter/gen_stopwords.py

@@ -0,0 +1,42 @@
+import requests
+
+LANGUAGES = [
+    "danish",
+    "dutch",
+    "finnish",
+    "french",
+    "german",
+    "italian",
+    "norwegian",
+    "portuguese",
+    "russian",
+    "spanish",
+    "swedish",
+]
+
+with requests.Session() as sess, open("stopwords.rs", "w") as mod:
+    mod.write("/*\n")
+    mod.write(
+        "These stop word lists are from the Snowball project (https://snowballstem.org/)\nwhich carries the following copyright and license:\n\n"
+    )
+
+    resp = sess.get(
+        "https://raw.githubusercontent.com/snowballstem/snowball/master/COPYING"
+    )
+    resp.raise_for_status()
+    mod.write(resp.text)
+    mod.write("*/\n\n")
+
+    for lang in LANGUAGES:
+        resp = sess.get(f"https://snowballstem.org/algorithms/{lang}/stop.txt")
+        resp.raise_for_status()
+
+        mod.write(f"pub const {lang.upper()}: &[&str] = &[\n")
+
+        for line in resp.text.splitlines():
+            line, _, _ = line.partition("|")
+
+            for word in line.split():
+                mod.write(f'    "{word}",\n')
+
+        mod.write("];\n\n")

src/tokenizer/stop_word_filter/mod.rs

@@ -10,42 +10,65 @@
 //! assert_eq!(stream.next().unwrap().text, "crafty");
 //! assert!(stream.next().is_none());
 //! ```
+#[cfg(feature = "stopwords")]
+#[rustfmt::skip]
+mod stopwords;
+
+use std::sync::Arc;
+
 use rustc_hash::FxHashSet;
 
-use super::{Token, TokenFilter, TokenStream};
-use crate::tokenizer::BoxTokenStream;
+use super::{BoxTokenStream, Language, Token, TokenFilter, TokenStream};
 
 /// `TokenFilter` that removes stop words from a token stream
 #[derive(Clone)]
 pub struct StopWordFilter {
-    words: FxHashSet<String>,
+    words: Arc<FxHashSet<String>>,
 }
 
 impl StopWordFilter {
-    /// Creates a `StopWordFilter` given a list of words to remove
-    pub fn remove(words: Vec<String>) -> StopWordFilter {
-        let mut set = FxHashSet::default();
+    /// Creates a new [`StopWordFilter`] for the given [`Language`]
+    ///
+    /// Returns `Some` if a list of stop words is available and `None` otherwise.
+    #[cfg(feature = "stopwords")]
+    pub fn new(language: Language) -> Option<Self> {
+        let words = match language {
+            Language::Danish => stopwords::DANISH,
+            Language::Dutch => stopwords::DUTCH,
+            Language::English => {
+                // This is the same list of words used by the Apache-licensed Lucene project,
+                // c.f. https://github.com/apache/lucene/blob/d5d6dc079395c47cd6d12dcce3bcfdd2c7d9dc63/lucene/analysis/common/src/java/org/apache/lucene/analysis/en/EnglishAnalyzer.java#L46
+                &[
+                    "a", "an", "and", "are", "as", "at", "be", "but", "by", "for", "if", "in",
+                    "into", "is", "it", "no", "not", "of", "on", "or", "such", "that", "the",
+                    "their", "then", "there", "these", "they", "this", "to", "was", "will", "with",
+                ]
+            }
+            Language::Finnish => stopwords::FINNISH,
+            Language::French => stopwords::FRENCH,
+            Language::German => stopwords::GERMAN,
+            Language::Italian => stopwords::ITALIAN,
+            Language::Norwegian => stopwords::NORWEGIAN,
+            Language::Portuguese => stopwords::PORTUGUESE,
+            Language::Russian => stopwords::RUSSIAN,
+            Language::Spanish => stopwords::SPANISH,
+            Language::Swedish => stopwords::SWEDISH,
+            _ => return None,
+        };
 
-        for word in words {
-            set.insert(word);
-        }
-
-        StopWordFilter { words: set }
+        Some(Self::remove(words.iter().map(|&word| word.to_owned())))
     }
 
-    fn english() -> StopWordFilter {
-        let words: [&'static str; 33] = [
-            "a", "an", "and", "are", "as", "at", "be", "but", "by", "for", "if", "in", "into",
-            "is", "it", "no", "not", "of", "on", "or", "such", "that", "the", "their", "then",
-            "there", "these", "they", "this", "to", "was", "will", "with",
-        ];
-
-        StopWordFilter::remove(words.iter().map(|&s| s.to_string()).collect())
+    /// Creates a `StopWordFilter` given a list of words to remove
+    pub fn remove<W: IntoIterator<Item = String>>(words: W) -> StopWordFilter {
+        StopWordFilter {
+            words: Arc::new(words.into_iter().collect()),
+        }
     }
 }
 
 pub struct StopWordFilterStream<'a> {
-    words: FxHashSet<String>,
+    words: Arc<FxHashSet<String>>,
     tail: BoxTokenStream<'a>,
 }
 
@@ -83,12 +106,6 @@ impl<'a> TokenStream for StopWordFilterStream<'a> {
     }
 }
 
-impl Default for StopWordFilter {
-    fn default() -> StopWordFilter {
-        StopWordFilter::english()
-    }
-}
-
 #[cfg(test)]
 mod tests {
     use crate::tokenizer::tests::assert_token;