searchengines-web/MeiliSearch
searchengines-web/MeiliSearch
1
0
Fork 0
mirror of https://github.com/meilisearch/MeiliSearch synced 2024-12-25 22:20:06 +01:00
Code Issues Releases Wiki Activity

Move to the v0.20.0-alpha.9 of heed

Browse Source
This commit is contained in:
Clément Renault 2023-11-27 11:52:22 +01:00
parent 58dac8af42
commit d32eb11329
No known key found for this signature in database
GPG Key ID: F250A4C4E3AE5F5F
28 changed files with 230 additions and 255 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

38
Cargo.lock generated
View File

@ -1260,9 +1260,9 @@ dependencies = [
[[package]]
name = "doxygen-rs"
version = "0.4.2"
version = "0.2.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "415b6ec780d34dcf624666747194393603d0373b7141eef01d12ee58881507d9"
checksum = "bff670ea0c9bbb8414e7efa6e23ebde2b8f520a7eef78273a3918cf1903e7505"
dependencies = [
"phf",
]
@ -1823,8 +1823,9 @@ checksum = "95505c38b4572b2d910cecb0281560f54b440a19336cbbcb27bf6ce6adc6f5a8"
[[package]]
name = "heed"
version = "0.20.0-alpha.6"
source = "git+https://github.com/meilisearch/heed?branch=main#321b0d4f2d26ceb504e70b4cb4eac0e0bdb3d796"
version = "0.20.0-alpha.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9648a50991c86df7d00c56c268c27754fcf4c80be2ba57fc4a00dc928c6fe934"
dependencies = [
"bitflags 2.3.3",
"bytemuck",
@ -1834,23 +1835,24 @@ dependencies = [
"libc",
"lmdb-master-sys",
"once_cell",
"page_size",
"page_size 0.6.0",
"synchronoise",
"url",
]
[[package]]
name = "heed-traits"
version = "0.20.0-alpha.6"
source = "git+https://github.com/meilisearch/heed?branch=main#321b0d4f2d26ceb504e70b4cb4eac0e0bdb3d796"
version = "0.20.0-alpha.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "5ab0b7d9cde969ad36dde692e487dc89d97f7168bf6a7bd3b894ad4bf7278298"
[[package]]
name = "heed-types"
version = "0.20.0-alpha.6"
source = "git+https://github.com/meilisearch/heed?branch=main#321b0d4f2d26ceb504e70b4cb4eac0e0bdb3d796"
version = "0.20.0-alpha.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f0cb3567a7363f28b597bf6e9897b9466397951dd0e52df2c8196dd8a71af44a"
dependencies = [
"bincode",
"bytemuck",
"byteorder",
"heed-traits",
"serde",
@ -2472,7 +2474,7 @@ dependencies = [
"meilisearch-auth",
"meilisearch-types",
"nelson",
"page_size",
"page_size 0.5.0",
"puffin",
"roaring",
"serde",
@ -2984,11 +2986,13 @@ dependencies = [
[[package]]
name = "lmdb-master-sys"
version = "0.1.0"
source = "git+https://github.com/meilisearch/heed?branch=main#321b0d4f2d26ceb504e70b4cb4eac0e0bdb3d796"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "629c123f5321b48fa4f8f4d3b868165b748d9ba79c7103fb58e3a94f736bcedd"
dependencies = [
"cc",
"doxygen-rs",
"libc",
"pkg-config",
]
[[package]]
@ -3494,6 +3498,16 @@ dependencies = [
"winapi",
]
[[package]]
name = "page_size"
version = "0.6.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "30d5b2194ed13191c1999ae0704b7839fb18384fa22e49b57eeaa97d79ce40da"
dependencies = [
"libc",
"winapi",
]
[[package]]
name = "parking_lot"
version = "0.12.1"

6
meilisearch-auth/src/store.rs
View File

@ -14,7 +14,7 @@ use meilisearch_types::heed::BoxedError;
use meilisearch_types::index_uid_pattern::IndexUidPattern;
use meilisearch_types::keys::KeyId;
use meilisearch_types::milli;
use meilisearch_types::milli::heed::types::{ByteSlice, DecodeIgnore, SerdeJson};
use meilisearch_types::milli::heed::types::{Bytes, DecodeIgnore, SerdeJson};
use meilisearch_types::milli::heed::{Database, Env, EnvOpenOptions, RwTxn};
use sha2::Sha256;
use time::OffsetDateTime;
@ -32,7 +32,7 @@ const KEY_ID_ACTION_INDEX_EXPIRATION_DB_NAME: &str = "keyid-action-index-expirat
#[derive(Clone)]
pub struct HeedAuthStore {
env: Arc<Env>,
keys: Database<ByteSlice, SerdeJson<Key>>,
keys: Database<Bytes, SerdeJson<Key>>,
action_keyid_index_expiration: Database<KeyIdActionCodec, SerdeJson<Option<OffsetDateTime>>>,
should_close_on_drop: bool,
}
@ -278,7 +278,7 @@ impl HeedAuthStore {
fn delete_key_from_inverted_db(&self, wtxn: &mut RwTxn, key: &KeyId) -> Result<()> {
let mut iter = self
.action_keyid_index_expiration
.remap_types::<ByteSlice, DecodeIgnore>()
.remap_types::<Bytes, DecodeIgnore>()
.prefix_iter_mut(wtxn, key.as_bytes())?;
while iter.next().transpose()?.is_some() {
// safety: we don't keep references from inside the LMDB database.

10
meilisearch-types/src/settings.rs
View File

@ -532,7 +532,10 @@ pub fn settings(
let faceting = FacetingSettings {
max_values_per_facet: Setting::Set(
index.max_values_per_facet(rtxn)?.unwrap_or(DEFAULT_VALUES_PER_FACET),
index
.max_values_per_facet(rtxn)?
.map(|x| x as usize)
.unwrap_or(DEFAULT_VALUES_PER_FACET),
),
sort_facet_values_by: Setting::Set(
index
@ -545,7 +548,10 @@ pub fn settings(
let pagination = PaginationSettings {
max_total_hits: Setting::Set(
index.pagination_max_total_hits(rtxn)?.unwrap_or(DEFAULT_PAGINATION_MAX_TOTAL_HITS),
index
.pagination_max_total_hits(rtxn)?
.map(|x| x as usize)
.unwrap_or(DEFAULT_PAGINATION_MAX_TOTAL_HITS),
),
};

2
meilisearch/src/routes/indexes/documents.rs
View File

@ -3,7 +3,7 @@ use std::io::ErrorKind;
use actix_web::http::header::CONTENT_TYPE;
use actix_web::web::Data;
use actix_web::{web, HttpMessage, HttpRequest, HttpResponse};
use bstr::ByteSlice;
use bstr::ByteSlice as _;
use deserr::actix_web::{AwebJson, AwebQueryParameter};
use deserr::Deserr;
use futures::StreamExt;

2
meilisearch/src/search.rs
View File

@ -360,6 +360,7 @@ fn prepare_search<'t>(
let max_total_hits = index
.pagination_max_total_hits(rtxn)
.map_err(milli::Error::from)?
.map(|x| x as usize)
.unwrap_or(DEFAULT_PAGINATION_MAX_TOTAL_HITS);
search.exhaustive_number_hits(is_finite_pagination);
@ -586,6 +587,7 @@ pub fn perform_search(
let max_values_by_facet = index
.max_values_per_facet(&rtxn)
.map_err(milli::Error::from)?
.map(|x| x as usize)
.unwrap_or(DEFAULT_VALUES_PER_FACET);
facet_distribution.max_values_per_facet(max_values_by_facet);

2
milli/Cargo.toml
View File

@ -29,7 +29,7 @@ geoutils = "0.5.1"
grenad = { version = "0.4.5", default-features = false, features = [
"rayon", "tempfile"
] }
heed = { git = "https://github.com/meilisearch/heed", branch = "main", default-features = false, features = [
heed = { version = "0.20.0-alpha.9", default-features = false, features = [
"serde-json", "serde-bincode", "read-txn-no-tls"
] }
indexmap = { version = "2.0.0", features = ["serde"] }

8
milli/src/heed_codec/byte_slice_ref.rs
View File

@ -2,11 +2,11 @@ use std::borrow::Cow;
use heed::{BoxedError, BytesDecode, BytesEncode};
/// A codec for values of type `&[u8]`. Unlike `ByteSlice`, its `EItem` and `DItem` associated
/// A codec for values of type `&[u8]`. Unlike `Bytes`, its `EItem` and `DItem` associated
/// types are equivalent (= `&'a [u8]`) and these values can reside within another structure.
pub struct ByteSliceRefCodec;
pub struct BytesRefCodec;
impl<'a> BytesEncode<'a> for ByteSliceRefCodec {
impl<'a> BytesEncode<'a> for BytesRefCodec {
type EItem = &'a [u8];
fn bytes_encode(item: &'a Self::EItem) -> Result<Cow<'a, [u8]>, BoxedError> {
@ -14,7 +14,7 @@ impl<'a> BytesEncode<'a> for ByteSliceRefCodec {
}
}
impl<'a> BytesDecode<'a> for ByteSliceRefCodec {
impl<'a> BytesDecode<'a> for BytesRefCodec {
type DItem = &'a [u8];
fn bytes_decode(bytes: &'a [u8]) -> Result<Self::DItem, BoxedError> {

2
milli/src/heed_codec/mod.rs
View File

@ -12,7 +12,7 @@ mod str_beu32_codec;
mod str_ref;
mod str_str_u8_codec;
pub use byte_slice_ref::ByteSliceRefCodec;
pub use byte_slice_ref::BytesRefCodec;
use heed::BoxedError;
pub use str_ref::StrRefCodec;

78
milli/src/index.rs
View File

@ -26,7 +26,7 @@ use crate::{
default_criteria, CboRoaringBitmapCodec, Criterion, DocumentId, ExternalDocumentsIds,
FacetDistribution, FieldDistribution, FieldId, FieldIdWordCountCodec, GeoPoint, ObkvCodec,
OrderBy, Result, RoaringBitmapCodec, RoaringBitmapLenCodec, Search, U8StrStrCodec, BEU16,
BEU32,
BEU32, BEU64,
};
/// The HNSW data-structure that we serialize, fill and search in.
@ -498,7 +498,7 @@ impl Index {
let i = i as u32;
let mut key = main_key::VECTOR_HNSW_KEY_PREFIX.as_bytes().to_vec();
key.extend_from_slice(&i.to_be_bytes());
self.main.remap_types::<ByteSlice, ByteSlice>().put(wtxn, &key, chunk)?;
self.main.remap_types::<Bytes, Bytes>().put(wtxn, &key, chunk)?;
}
Ok(())
}
@ -507,7 +507,7 @@ impl Index {
pub(crate) fn delete_vector_hnsw(&self, wtxn: &mut RwTxn) -> heed::Result<bool> {
let mut iter = self
.main
.remap_types::<ByteSlice, DecodeIgnore>()
.remap_types::<Bytes, DecodeIgnore>()
.prefix_iter_mut(wtxn, main_key::VECTOR_HNSW_KEY_PREFIX.as_bytes())?;
let mut deleted = false;
while iter.next().transpose()?.is_some() {
@ -522,7 +522,7 @@ impl Index {
let mut slices = Vec::new();
for result in self
.main
.remap_types::<Str, ByteSlice>()
.remap_types::<Str, Bytes>()
.prefix_iter(rtxn, main_key::VECTOR_HNSW_KEY_PREFIX)?
{
let (_, slice) = result?;
@ -994,7 +994,7 @@ impl Index {
wtxn: &mut RwTxn,
fst: &fst::Set<A>,
) -> heed::Result<()> {
self.main.remap_types::<Str, ByteSlice>().put(
self.main.remap_types::<Str, Bytes>().put(
wtxn,
main_key::WORDS_FST_KEY,
fst.as_fst().as_bytes(),
@ -1003,7 +1003,7 @@ impl Index {
/// Returns the FST which is the words dictionary of the engine.
pub fn words_fst<'t>(&self, rtxn: &'t RoTxn) -> Result<fst::Set<Cow<'t, [u8]>>> {
match self.main.remap_types::<Str, ByteSlice>().get(rtxn, main_key::WORDS_FST_KEY)? {
match self.main.remap_types::<Str, Bytes>().get(rtxn, main_key::WORDS_FST_KEY)? {
Some(bytes) => Ok(fst::Set::new(bytes)?.map_data(Cow::Borrowed)?),
None => Ok(fst::Set::default().map_data(Cow::Owned)?),
}
@ -1016,7 +1016,7 @@ impl Index {
wtxn: &mut RwTxn,
fst: &fst::Set<A>,
) -> heed::Result<()> {
self.main.remap_types::<Str, ByteSlice>().put(
self.main.remap_types::<Str, Bytes>().put(
wtxn,
main_key::STOP_WORDS_KEY,
fst.as_fst().as_bytes(),
@ -1028,7 +1028,7 @@ impl Index {
}
pub fn stop_words<'t>(&self, rtxn: &'t RoTxn) -> Result<Option<fst::Set<&'t [u8]>>> {
match self.main.remap_types::<Str, ByteSlice>().get(rtxn, main_key::STOP_WORDS_KEY)? {
match self.main.remap_types::<Str, Bytes>().get(rtxn, main_key::STOP_WORDS_KEY)? {
Some(bytes) => Ok(Some(fst::Set::new(bytes)?)),
None => Ok(None),
}
@ -1186,7 +1186,7 @@ impl Index {
wtxn: &mut RwTxn,
fst: &fst::Set<A>,
) -> heed::Result<()> {
self.main.remap_types::<Str, ByteSlice>().put(
self.main.remap_types::<Str, Bytes>().put(
wtxn,
main_key::WORDS_PREFIXES_FST_KEY,
fst.as_fst().as_bytes(),
@ -1195,11 +1195,7 @@ impl Index {
/// Returns the FST which is the words prefixes dictionnary of the engine.
pub fn words_prefixes_fst<'t>(&self, rtxn: &'t RoTxn) -> Result<fst::Set<Cow<'t, [u8]>>> {
match self
.main
.remap_types::<Str, ByteSlice>()
.get(rtxn, main_key::WORDS_PREFIXES_FST_KEY)?
{
match self.main.remap_types::<Str, Bytes>().get(rtxn, main_key::WORDS_PREFIXES_FST_KEY)? {
Some(bytes) => Ok(fst::Set::new(bytes)?.map_data(Cow::Borrowed)?),
None => Ok(fst::Set::default().map_data(Cow::Owned)?),
}
@ -1325,7 +1321,7 @@ impl Index {
// It is not possible to put a bool in heed with OwnedType, so we put a u8 instead. We
// identify 0 as being false, and anything else as true. The absence of a value is true,
// because by default, we authorize typos.
match self.main.remap_types::<Str, OwnedType<u8>>().get(txn, main_key::AUTHORIZE_TYPOS)? {
match self.main.remap_types::<Str, U8>().get(txn, main_key::AUTHORIZE_TYPOS)? {
Some(0) => Ok(false),
_ => Ok(true),
}
@ -1335,11 +1331,7 @@ impl Index {
// It is not possible to put a bool in heed with OwnedType, so we put a u8 instead. We
// identify 0 as being false, and anything else as true. The absence of a value is true,
// because by default, we authorize typos.
self.main.remap_types::<Str, OwnedType<u8>>().put(
txn,
main_key::AUTHORIZE_TYPOS,
&(flag as u8),
)?;
self.main.remap_types::<Str, U8>().put(txn, main_key::AUTHORIZE_TYPOS, &(flag as u8))?;
Ok(())
}
@ -1350,7 +1342,7 @@ impl Index {
// because by default, we authorize typos.
Ok(self
.main
.remap_types::<Str, OwnedType<u8>>()
.remap_types::<Str, U8>()
.get(txn, main_key::ONE_TYPO_WORD_LEN)?
.unwrap_or(DEFAULT_MIN_WORD_LEN_ONE_TYPO))
}
@ -1359,11 +1351,7 @@ impl Index {
// It is not possible to put a bool in heed with OwnedType, so we put a u8 instead. We
// identify 0 as being false, and anything else as true. The absence of a value is true,
// because by default, we authorize typos.
self.main.remap_types::<Str, OwnedType<u8>>().put(
txn,
main_key::ONE_TYPO_WORD_LEN,
&val,
)?;
self.main.remap_types::<Str, U8>().put(txn, main_key::ONE_TYPO_WORD_LEN, &val)?;
Ok(())
}
@ -1373,7 +1361,7 @@ impl Index {
// because by default, we authorize typos.
Ok(self
.main
.remap_types::<Str, OwnedType<u8>>()
.remap_types::<Str, U8>()
.get(txn, main_key::TWO_TYPOS_WORD_LEN)?
.unwrap_or(DEFAULT_MIN_WORD_LEN_TWO_TYPOS))
}
@ -1382,17 +1370,13 @@ impl Index {
// It is not possible to put a bool in heed with OwnedType, so we put a u8 instead. We
// identify 0 as being false, and anything else as true. The absence of a value is true,
// because by default, we authorize typos.
self.main.remap_types::<Str, OwnedType<u8>>().put(
txn,
main_key::TWO_TYPOS_WORD_LEN,
&val,
)?;
self.main.remap_types::<Str, U8>().put(txn, main_key::TWO_TYPOS_WORD_LEN, &val)?;
Ok(())
}
/// List the words on which typo are not allowed
pub fn exact_words<'t>(&self, txn: &'t RoTxn) -> Result<Option<fst::Set<Cow<'t, [u8]>>>> {
match self.main.remap_types::<Str, ByteSlice>().get(txn, main_key::EXACT_WORDS)? {
match self.main.remap_types::<Str, Bytes>().get(txn, main_key::EXACT_WORDS)? {
Some(bytes) => Ok(Some(fst::Set::new(bytes)?.map_data(Cow::Borrowed)?)),
None => Ok(None),
}
@ -1403,7 +1387,7 @@ impl Index {
txn: &mut RwTxn,
words: &fst::Set<A>,
) -> Result<()> {
self.main.remap_types::<Str, ByteSlice>().put(
self.main.remap_types::<Str, Bytes>().put(
txn,
main_key::EXACT_WORDS,
words.as_fst().as_bytes(),
@ -1442,16 +1426,12 @@ impl Index {
self.main.remap_key_type::<Str>().delete(txn, main_key::EXACT_ATTRIBUTES)
}
pub fn max_values_per_facet(&self, txn: &RoTxn) -> heed::Result<Option<usize>> {
self.main.remap_types::<Str, OwnedType<usize>>().get(txn, main_key::MAX_VALUES_PER_FACET)
pub fn max_values_per_facet(&self, txn: &RoTxn) -> heed::Result<Option<u64>> {
self.main.remap_types::<Str, BEU64>().get(txn, main_key::MAX_VALUES_PER_FACET)
}
pub(crate) fn put_max_values_per_facet(&self, txn: &mut RwTxn, val: usize) -> heed::Result<()> {
self.main.remap_types::<Str, OwnedType<usize>>().put(
txn,
main_key::MAX_VALUES_PER_FACET,
&val,
)
pub(crate) fn put_max_values_per_facet(&self, txn: &mut RwTxn, val: u64) -> heed::Result<()> {
self.main.remap_types::<Str, BEU64>().put(txn, main_key::MAX_VALUES_PER_FACET, &val)
}
pub(crate) fn delete_max_values_per_facet(&self, txn: &mut RwTxn) -> heed::Result<bool> {
@ -1481,22 +1461,16 @@ impl Index {
self.main.remap_key_type::<Str>().delete(txn, main_key::SORT_FACET_VALUES_BY)
}
pub fn pagination_max_total_hits(&self, txn: &RoTxn) -> heed::Result<Option<usize>> {
self.main
.remap_types::<Str, OwnedType<usize>>()
.get(txn, main_key::PAGINATION_MAX_TOTAL_HITS)
pub fn pagination_max_total_hits(&self, txn: &RoTxn) -> heed::Result<Option<u64>> {
self.main.remap_types::<Str, BEU64>().get(txn, main_key::PAGINATION_MAX_TOTAL_HITS)
}
pub(crate) fn put_pagination_max_total_hits(
&self,
txn: &mut RwTxn,
val: usize,
val: u64,
) -> heed::Result<()> {
self.main.remap_types::<Str, OwnedType<usize>>().put(
txn,
main_key::PAGINATION_MAX_TOTAL_HITS,
&val,
)
self.main.remap_types::<Str, BEU64>().put(txn, main_key::PAGINATION_MAX_TOTAL_HITS, &val)
}
pub(crate) fn delete_pagination_max_total_hits(&self, txn: &mut RwTxn) -> heed::Result<bool> {

16
milli/src/search/facet/facet_distribution.rs
View File

@ -2,7 +2,7 @@ use std::collections::{BTreeMap, HashMap, HashSet};
use std::ops::ControlFlow;
use std::{fmt, mem};
use heed::types::ByteSlice;
use heed::types::Bytes;
use heed::BytesDecode;
use indexmap::IndexMap;
use roaring::RoaringBitmap;
@ -13,7 +13,7 @@ use crate::facet::FacetType;
use crate::heed_codec::facet::{
FacetGroupKeyCodec, FieldDocIdFacetF64Codec, FieldDocIdFacetStringCodec, OrderedF64Codec,
};
use crate::heed_codec::{ByteSliceRefCodec, StrRefCodec};
use crate::heed_codec::{BytesRefCodec, StrRefCodec};
use crate::search::facet::facet_distribution_iter::{
count_iterate_over_facet_distribution, lexicographically_iterate_over_facet_distribution,
};
@ -105,7 +105,7 @@ impl<'a> FacetDistribution<'a> {
key_buffer.truncate(mem::size_of::<FieldId>());
key_buffer.extend_from_slice(&docid.to_be_bytes());
let iter = db
.remap_key_type::<ByteSlice>()
.remap_key_type::<Bytes>()
.prefix_iter(self.rtxn, &key_buffer)?
.remap_key_type::<FieldDocIdFacetF64Codec>();
@ -129,7 +129,7 @@ impl<'a> FacetDistribution<'a> {
key_buffer.truncate(mem::size_of::<FieldId>());
key_buffer.extend_from_slice(&docid.to_be_bytes());
let iter = db
.remap_key_type::<ByteSlice>()
.remap_key_type::<Bytes>()
.prefix_iter(self.rtxn, &key_buffer)?
.remap_key_type::<FieldDocIdFacetStringCodec>();
@ -172,9 +172,7 @@ impl<'a> FacetDistribution<'a> {
search_function(
self.rtxn,
self.index
.facet_id_f64_docids
.remap_key_type::<FacetGroupKeyCodec<ByteSliceRefCodec>>(),
self.index.facet_id_f64_docids.remap_key_type::<FacetGroupKeyCodec<BytesRefCodec>>(),
field_id,
candidates,
|facet_key, nbr_docids, _| {
@ -203,9 +201,7 @@ impl<'a> FacetDistribution<'a> {
search_function(
self.rtxn,
self.index
.facet_id_string_docids
.remap_key_type::<FacetGroupKeyCodec<ByteSliceRefCodec>>(),
self.index.facet_id_string_docids.remap_key_type::<FacetGroupKeyCodec<BytesRefCodec>>(),
field_id,
candidates,
|facet_key, nbr_docids, any_docid| {

16
milli/src/search/facet/facet_distribution_iter.rs
View File

@ -7,7 +7,7 @@ use roaring::RoaringBitmap;
use super::{get_first_facet_value, get_highest_level};
use crate::heed_codec::facet::{FacetGroupKey, FacetGroupKeyCodec, FacetGroupValueCodec};
use crate::heed_codec::ByteSliceRefCodec;
use crate::heed_codec::BytesRefCodec;
use crate::DocumentId;
/// Call the given closure on the facet distribution of the candidate documents.
@ -23,7 +23,7 @@ use crate::DocumentId;
/// keep iterating over the different facet values or stop.
pub fn lexicographically_iterate_over_facet_distribution<'t, CB>(
rtxn: &'t heed::RoTxn<'t>,
db: heed::Database<FacetGroupKeyCodec<ByteSliceRefCodec>, FacetGroupValueCodec>,
db: heed::Database<FacetGroupKeyCodec<BytesRefCodec>, FacetGroupValueCodec>,
field_id: u16,
candidates: &RoaringBitmap,
callback: CB,
@ -34,11 +34,11 @@ where
let mut fd = LexicographicFacetDistribution { rtxn, db, field_id, callback };
let highest_level = get_highest_level(
rtxn,
db.remap_key_type::<FacetGroupKeyCodec<ByteSliceRefCodec>>(),
db.remap_key_type::<FacetGroupKeyCodec<BytesRefCodec>>(),
field_id,
)?;
if let Some(first_bound) = get_first_facet_value::<ByteSliceRefCodec>(rtxn, db, field_id)? {
if let Some(first_bound) = get_first_facet_value::<BytesRefCodec>(rtxn, db, field_id)? {
fd.iterate(candidates, highest_level, first_bound, usize::MAX)?;
Ok(())
} else {
@ -48,7 +48,7 @@ where
pub fn count_iterate_over_facet_distribution<'t, CB>(
rtxn: &'t heed::RoTxn<'t>,
db: heed::Database<FacetGroupKeyCodec<ByteSliceRefCodec>, FacetGroupValueCodec>,
db: heed::Database<FacetGroupKeyCodec<BytesRefCodec>, FacetGroupValueCodec>,
field_id: u16,
candidates: &RoaringBitmap,
mut callback: CB,
@ -77,11 +77,11 @@ where
let mut heap = BinaryHeap::new();
let highest_level = get_highest_level(
rtxn,
db.remap_key_type::<FacetGroupKeyCodec<ByteSliceRefCodec>>(),
db.remap_key_type::<FacetGroupKeyCodec<BytesRefCodec>>(),
field_id,
)?;
if let Some(first_bound) = get_first_facet_value::<ByteSliceRefCodec>(rtxn, db, field_id)? {
if let Some(first_bound) = get_first_facet_value::<BytesRefCodec>(rtxn, db, field_id)? {
// We first fill the heap with values from the highest level
let starting_key =
FacetGroupKey { field_id, level: highest_level, left_bound: first_bound };
@ -146,7 +146,7 @@ where
CB: FnMut(&'t [u8], u64, DocumentId) -> Result<ControlFlow<()>>,
{
rtxn: &'t heed::RoTxn<'t>,
db: heed::Database<FacetGroupKeyCodec<ByteSliceRefCodec>, FacetGroupValueCodec>,
db: heed::Database<FacetGroupKeyCodec<BytesRefCodec>, FacetGroupValueCodec>,
field_id: u16,
callback: CB,
}

15
milli/src/search/facet/facet_range_search.rs
View File

@ -5,7 +5,7 @@ use roaring::RoaringBitmap;
use super::{get_first_facet_value, get_highest_level, get_last_facet_value};
use crate::heed_codec::facet::{FacetGroupKey, FacetGroupKeyCodec, FacetGroupValueCodec};
use crate::heed_codec::ByteSliceRefCodec;
use crate::heed_codec::BytesRefCodec;
use crate::Result;
/// Find all the document ids for which the given field contains a value contained within
@ -46,16 +46,13 @@ where
}
Bound::Unbounded => Bound::Unbounded,
};
let db = db.remap_key_type::<FacetGroupKeyCodec<ByteSliceRefCodec>>();
let db = db.remap_key_type::<FacetGroupKeyCodec<BytesRefCodec>>();
let mut f = FacetRangeSearch { rtxn, db, field_id, left, right, docids };
let highest_level = get_highest_level(rtxn, db, field_id)?;
if let Some(starting_left_bound) =
get_first_facet_value::<ByteSliceRefCodec>(rtxn, db, field_id)?
{
let rightmost_bound = Bound::Included(
get_last_facet_value::<ByteSliceRefCodec>(rtxn, db, field_id)?.unwrap(),
); // will not fail because get_first_facet_value succeeded
if let Some(starting_left_bound) = get_first_facet_value::<BytesRefCodec>(rtxn, db, field_id)? {
let rightmost_bound =
Bound::Included(get_last_facet_value::<BytesRefCodec>(rtxn, db, field_id)?.unwrap()); // will not fail because get_first_facet_value succeeded
let group_size = usize::MAX;
f.run(highest_level, starting_left_bound, rightmost_bound, group_size)?;
Ok(())
@ -67,7 +64,7 @@ where
/// Fetch the document ids that have a facet with a value between the two given bounds
struct FacetRangeSearch<'t, 'b, 'bitmap> {
rtxn: &'t heed::RoTxn<'t>,
db: heed::Database<FacetGroupKeyCodec<ByteSliceRefCodec>, FacetGroupValueCodec>,
db: heed::Database<FacetGroupKeyCodec<BytesRefCodec>, FacetGroupValueCodec>,
field_id: u16,
left: Bound<&'b [u8]>,
right: Bound<&'b [u8]>,

12
milli/src/search/facet/facet_sort_ascending.rs
View File

@ -5,7 +5,7 @@ use super::{get_first_facet_value, get_highest_level};
use crate::heed_codec::facet::{
FacetGroupKey, FacetGroupKeyCodec, FacetGroupValue, FacetGroupValueCodec,
};
use crate::heed_codec::ByteSliceRefCodec;
use crate::heed_codec::BytesRefCodec;
/// Return an iterator which iterates over the given candidate documents in
/// ascending order of their facet value for the given field id.
@ -31,12 +31,12 @@ use crate::heed_codec::ByteSliceRefCodec;
/// Note that once a document id is returned by the iterator, it is never returned again.
pub fn ascending_facet_sort<'t>(
rtxn: &'t heed::RoTxn<'t>,
db: heed::Database<FacetGroupKeyCodec<ByteSliceRefCodec>, FacetGroupValueCodec>,
db: heed::Database<FacetGroupKeyCodec<BytesRefCodec>, FacetGroupValueCodec>,
field_id: u16,
candidates: RoaringBitmap,
) -> Result<impl Iterator<Item = Result<(RoaringBitmap, &'t [u8])>> + 't> {
let highest_level = get_highest_level(rtxn, db, field_id)?;
if let Some(first_bound) = get_first_facet_value::<ByteSliceRefCodec>(rtxn, db, field_id)? {
if let Some(first_bound) = get_first_facet_value::<BytesRefCodec>(rtxn, db, field_id)? {
let first_key = FacetGroupKey { field_id, level: highest_level, left_bound: first_bound };
let iter = db.range(rtxn, &(first_key..)).unwrap().take(usize::MAX);
@ -53,14 +53,12 @@ pub fn ascending_facet_sort<'t>(
struct AscendingFacetSort<'t, 'e> {
rtxn: &'t heed::RoTxn<'e>,
db: heed::Database<FacetGroupKeyCodec<ByteSliceRefCodec>, FacetGroupValueCodec>,
db: heed::Database<FacetGroupKeyCodec<BytesRefCodec>, FacetGroupValueCodec>,
field_id: u16,
#[allow(clippy::type_complexity)]
stack: Vec<(
RoaringBitmap,
std::iter::Take<
heed::RoRange<'t, FacetGroupKeyCodec<ByteSliceRefCodec>, FacetGroupValueCodec>,
>,
std::iter::Take<heed::RoRange<'t, FacetGroupKeyCodec<BytesRefCodec>, FacetGroupValueCodec>>,
)>,
}

20
milli/src/search/facet/facet_sort_descending.rs
View File

@ -7,21 +7,21 @@ use super::{get_first_facet_value, get_highest_level, get_last_facet_value};
use crate::heed_codec::facet::{
FacetGroupKey, FacetGroupKeyCodec, FacetGroupValue, FacetGroupValueCodec,
};
use crate::heed_codec::ByteSliceRefCodec;
use crate::heed_codec::BytesRefCodec;
/// See documentationg for [`ascending_facet_sort`](super::ascending_facet_sort).
///
/// This function does the same thing, but in the opposite order.
pub fn descending_facet_sort<'t>(
rtxn: &'t heed::RoTxn<'t>,
db: heed::Database<FacetGroupKeyCodec<ByteSliceRefCodec>, FacetGroupValueCodec>,
db: heed::Database<FacetGroupKeyCodec<BytesRefCodec>, FacetGroupValueCodec>,
field_id: u16,
candidates: RoaringBitmap,
) -> Result<impl Iterator<Item = Result<(RoaringBitmap, &'t [u8])>> + 't> {
let highest_level = get_highest_level(rtxn, db, field_id)?;
if let Some(first_bound) = get_first_facet_value::<ByteSliceRefCodec>(rtxn, db, field_id)? {
if let Some(first_bound) = get_first_facet_value::<BytesRefCodec>(rtxn, db, field_id)? {
let first_key = FacetGroupKey { field_id, level: highest_level, left_bound: first_bound };
let last_bound = get_last_facet_value::<ByteSliceRefCodec>(rtxn, db, field_id)?.unwrap();
let last_bound = get_last_facet_value::<BytesRefCodec>(rtxn, db, field_id)?.unwrap();
let last_key = FacetGroupKey { field_id, level: highest_level, left_bound: last_bound };
let iter = db.rev_range(rtxn, &(first_key..=last_key))?.take(usize::MAX);
Ok(itertools::Either::Left(DescendingFacetSort {
@ -37,13 +37,13 @@ pub fn descending_facet_sort<'t>(
struct DescendingFacetSort<'t> {
rtxn: &'t heed::RoTxn<'t>,
db: heed::Database<FacetGroupKeyCodec<ByteSliceRefCodec>, FacetGroupValueCodec>,
db: heed::Database<FacetGroupKeyCodec<BytesRefCodec>, FacetGroupValueCodec>,
field_id: u16,
#[allow(clippy::type_complexity)]
stack: Vec<(
RoaringBitmap,
std::iter::Take<
heed::RoRevRange<'t, FacetGroupKeyCodec<ByteSliceRefCodec>, FacetGroupValueCodec>,
heed::RoRevRange<'t, FacetGroupKeyCodec<BytesRefCodec>, FacetGroupValueCodec>,
>,
Bound<&'t [u8]>,
)>,
@ -100,7 +100,7 @@ impl<'t> Iterator for DescendingFacetSort<'t> {
*right_bound = Bound::Excluded(left_bound);
let iter = match self
.db
.remap_key_type::<FacetGroupKeyCodec<ByteSliceRefCodec>>()
.remap_key_type::<FacetGroupKeyCodec<BytesRefCodec>>()
.rev_range(self.rtxn, &(Bound::Included(starting_key_below), end_key_kelow))
{
Ok(iter) => iter,
@ -123,7 +123,7 @@ mod tests {
use roaring::RoaringBitmap;
use crate::heed_codec::facet::FacetGroupKeyCodec;
use crate::heed_codec::ByteSliceRefCodec;
use crate::heed_codec::BytesRefCodec;
use crate::milli_snap;
use crate::search::facet::facet_sort_descending::descending_facet_sort;
use crate::search::facet::tests::{
@ -144,7 +144,7 @@ mod tests {
let txn = index.env.read_txn().unwrap();
let candidates = (200..=300).collect::<RoaringBitmap>();
let mut results = String::new();
let db = index.content.remap_key_type::<FacetGroupKeyCodec<ByteSliceRefCodec>>();
let db = index.content.remap_key_type::<FacetGroupKeyCodec<BytesRefCodec>>();
let iter = descending_facet_sort(&txn, db, 0, candidates).unwrap();
for el in iter {
let (docids, _) = el.unwrap();
@ -167,7 +167,7 @@ mod tests {
let txn = index.env.read_txn().unwrap();
let candidates = (200..=300).collect::<RoaringBitmap>();
let mut results = String::new();
let db = index.content.remap_key_type::<FacetGroupKeyCodec<ByteSliceRefCodec>>();
let db = index.content.remap_key_type::<FacetGroupKeyCodec<BytesRefCodec>>();
let iter = descending_facet_sort(&txn, db, 0, candidates.clone()).unwrap();
for el in iter {
let (docids, _) = el.unwrap();

25
milli/src/search/facet/mod.rs
View File

@ -1,13 +1,13 @@
pub use facet_sort_ascending::ascending_facet_sort;
pub use facet_sort_descending::descending_facet_sort;
use heed::types::{ByteSlice, DecodeIgnore};
use heed::types::{Bytes, DecodeIgnore};
use heed::{BytesDecode, RoTxn};
use roaring::RoaringBitmap;
pub use self::facet_distribution::{FacetDistribution, OrderBy, DEFAULT_VALUES_PER_FACET};
pub use self::filter::{BadGeoError, Filter};
use crate::heed_codec::facet::{FacetGroupKeyCodec, FacetGroupValueCodec, OrderedF64Codec};
use crate::heed_codec::ByteSliceRefCodec;
use crate::heed_codec::BytesRefCodec;
use crate::{Index, Result};
mod facet_distribution;
mod facet_distribution_iter;
@ -34,7 +34,7 @@ pub fn facet_min_value<'t>(
field_id: u16,
candidates: RoaringBitmap,
) -> Result<Option<f64>> {
let db = index.facet_id_f64_docids.remap_key_type::<FacetGroupKeyCodec<ByteSliceRefCodec>>();
let db = index.facet_id_f64_docids.remap_key_type::<FacetGroupKeyCodec<BytesRefCodec>>();
let it = ascending_facet_sort(rtxn, db, field_id, candidates)?;
facet_extreme_value(it)
}
@ -45,7 +45,7 @@ pub fn facet_max_value<'t>(
field_id: u16,
candidates: RoaringBitmap,
) -> Result<Option<f64>> {
let db = index.facet_id_f64_docids.remap_key_type::<FacetGroupKeyCodec<ByteSliceRefCodec>>();
let db = index.facet_id_f64_docids.remap_key_type::<FacetGroupKeyCodec<BytesRefCodec>>();
let it = descending_facet_sort(rtxn, db, field_id, candidates)?;
facet_extreme_value(it)
}
@ -53,7 +53,7 @@ pub fn facet_max_value<'t>(
/// Get the first facet value in the facet database
pub(crate) fn get_first_facet_value<'t, BoundCodec>(
txn: &'t RoTxn,
db: heed::Database<FacetGroupKeyCodec<ByteSliceRefCodec>, FacetGroupValueCodec>,
db: heed::Database<FacetGroupKeyCodec<BytesRefCodec>, FacetGroupValueCodec>,
field_id: u16,
) -> heed::Result<Option<BoundCodec::DItem>>
where
@ -63,7 +63,7 @@ where
level0prefix.extend_from_slice(&field_id.to_be_bytes());
level0prefix.push(0);
let mut level0_iter_forward =
db.remap_types::<ByteSlice, DecodeIgnore>().prefix_iter(txn, level0prefix.as_slice())?;
db.remap_types::<Bytes, DecodeIgnore>().prefix_iter(txn, level0prefix.as_slice())?;
if let Some(first) = level0_iter_forward.next() {
let (first_key, _) = first?;
let first_key = FacetGroupKeyCodec::<BoundCodec>::bytes_decode(first_key)
@ -77,7 +77,7 @@ where
/// Get the last facet value in the facet database
pub(crate) fn get_last_facet_value<'t, BoundCodec>(
txn: &'t RoTxn,
db: heed::Database<FacetGroupKeyCodec<ByteSliceRefCodec>, FacetGroupValueCodec>,
db: heed::Database<FacetGroupKeyCodec<BytesRefCodec>, FacetGroupValueCodec>,
field_id: u16,
) -> heed::Result<Option<BoundCodec::DItem>>
where
@ -86,9 +86,8 @@ where
let mut level0prefix = vec![];
level0prefix.extend_from_slice(&field_id.to_be_bytes());
level0prefix.push(0);
let mut level0_iter_backward = db
.remap_types::<ByteSlice, DecodeIgnore>()
.rev_prefix_iter(txn, level0prefix.as_slice())?;
let mut level0_iter_backward =
db.remap_types::<Bytes, DecodeIgnore>().rev_prefix_iter(txn, level0prefix.as_slice())?;
if let Some(last) = level0_iter_backward.next() {
let (last_key, _) = last?;
let last_key = FacetGroupKeyCodec::<BoundCodec>::bytes_decode(last_key)
@ -102,17 +101,17 @@ where
/// Get the height of the highest level in the facet database
pub(crate) fn get_highest_level<'t>(
txn: &'t RoTxn<'t>,
db: heed::Database<FacetGroupKeyCodec<ByteSliceRefCodec>, FacetGroupValueCodec>,
db: heed::Database<FacetGroupKeyCodec<BytesRefCodec>, FacetGroupValueCodec>,
field_id: u16,
) -> heed::Result<u8> {
let field_id_prefix = &field_id.to_be_bytes();
Ok(db
.remap_types::<ByteSlice, DecodeIgnore>()
.remap_types::<Bytes, DecodeIgnore>()
.rev_prefix_iter(txn, field_id_prefix)?
.next()
.map(|el| {
let (key, _) = el.unwrap();
let key = FacetGroupKeyCodec::<ByteSliceRefCodec>::bytes_decode(key).unwrap();
let key = FacetGroupKeyCodec::<BytesRefCodec>::bytes_decode(key).unwrap();
key.level
})
.unwrap_or(0))

40
milli/src/search/new/db_cache.rs
View File

@ -3,7 +3,7 @@ use std::collections::hash_map::Entry;
use std::hash::Hash;
use fxhash::FxHashMap;
use heed::types::ByteSlice;
use heed::types::Bytes;
use heed::{BytesEncode, Database, RoTxn};
use roaring::RoaringBitmap;
@ -50,7 +50,7 @@ impl<'ctx> DatabaseCache<'ctx> {
cache_key: K1,
db_key: &'v KC::EItem,
cache: &mut FxHashMap<K1, Option<Cow<'ctx, [u8]>>>,
db: Database<KC, ByteSlice>,
db: Database<KC, Bytes>,
) -> Result<Option<DC::DItem>>
where
K1: Copy + Eq + Hash,
@ -80,7 +80,7 @@ impl<'ctx> DatabaseCache<'ctx> {
cache_key: K1,
db_keys: &'v [KC::EItem],
cache: &mut FxHashMap<K1, Option<Cow<'ctx, [u8]>>>,
db: Database<KC, ByteSlice>,
db: Database<KC, Bytes>,
merger: MergeFn,
) -> Result<Option<DC::DItem>>
where
@ -168,7 +168,7 @@ impl<'ctx> SearchContext<'ctx> {
word,
&keys[..],
&mut self.db_cache.word_docids,
self.index.word_fid_docids.remap_data_type::<ByteSlice>(),
self.index.word_fid_docids.remap_data_type::<Bytes>(),
merge_cbo_roaring_bitmaps,
)
}
@ -177,7 +177,7 @@ impl<'ctx> SearchContext<'ctx> {
word,
self.word_interner.get(word).as_str(),
&mut self.db_cache.word_docids,
self.index.word_docids.remap_data_type::<ByteSlice>(),
self.index.word_docids.remap_data_type::<Bytes>(),
),
}
}
@ -191,7 +191,7 @@ impl<'ctx> SearchContext<'ctx> {
word,
self.word_interner.get(word).as_str(),
&mut self.db_cache.exact_word_docids,
self.index.exact_word_docids.remap_data_type::<ByteSlice>(),
self.index.exact_word_docids.remap_data_type::<Bytes>(),
)
}
@ -230,7 +230,7 @@ impl<'ctx> SearchContext<'ctx> {
prefix,
&keys[..],
&mut self.db_cache.word_prefix_docids,
self.index.word_prefix_fid_docids.remap_data_type::<ByteSlice>(),
self.index.word_prefix_fid_docids.remap_data_type::<Bytes>(),
merge_cbo_roaring_bitmaps,
)
}
@ -239,7 +239,7 @@ impl<'ctx> SearchContext<'ctx> {
prefix,
self.word_interner.get(prefix).as_str(),
&mut self.db_cache.word_prefix_docids,
self.index.word_prefix_docids.remap_data_type::<ByteSlice>(),
self.index.word_prefix_docids.remap_data_type::<Bytes>(),
),
}
}
@ -253,7 +253,7 @@ impl<'ctx> SearchContext<'ctx> {
prefix,
self.word_interner.get(prefix).as_str(),
&mut self.db_cache.exact_word_prefix_docids,
self.index.exact_word_prefix_docids.remap_data_type::<ByteSlice>(),
self.index.exact_word_prefix_docids.remap_data_type::<Bytes>(),
)
}
@ -272,7 +272,7 @@ impl<'ctx> SearchContext<'ctx> {
self.word_interner.get(word2).as_str(),
),
&mut self.db_cache.word_pair_proximity_docids,
self.index.word_pair_proximity_docids.remap_data_type::<ByteSlice>(),
self.index.word_pair_proximity_docids.remap_data_type::<Bytes>(),
)
}
@ -291,7 +291,7 @@ impl<'ctx> SearchContext<'ctx> {
self.word_interner.get(word2).as_str(),
),
&mut self.db_cache.word_pair_proximity_docids,
self.index.word_pair_proximity_docids.remap_data_type::<ByteSlice>(),
self.index.word_pair_proximity_docids.remap_data_type::<Bytes>(),
)
}
@ -320,7 +320,7 @@ impl<'ctx> SearchContext<'ctx> {
let remap_key_type = self
.index
.word_pair_proximity_docids
.remap_key_type::<ByteSlice>()
.remap_key_type::<Bytes>()
.prefix_iter(self.txn, &key)?;
for result in remap_key_type {
let (_, docids) = result?;
@ -359,7 +359,7 @@ impl<'ctx> SearchContext<'ctx> {
(word, fid),
&(self.word_interner.get(word).as_str(), fid),
&mut self.db_cache.word_fid_docids,
self.index.word_fid_docids.remap_data_type::<ByteSlice>(),
self.index.word_fid_docids.remap_data_type::<Bytes>(),
)
}
@ -378,7 +378,7 @@ impl<'ctx> SearchContext<'ctx> {
(word_prefix, fid),
&(self.word_interner.get(word_prefix).as_str(), fid),
&mut self.db_cache.word_prefix_fid_docids,
self.index.word_prefix_fid_docids.remap_data_type::<ByteSlice>(),
self.index.word_prefix_fid_docids.remap_data_type::<Bytes>(),
)
}
@ -392,7 +392,7 @@ impl<'ctx> SearchContext<'ctx> {
let remap_key_type = self
.index
.word_fid_docids
.remap_types::<ByteSlice, ByteSlice>()
.remap_types::<Bytes, Bytes>()
.prefix_iter(self.txn, &key)?
.remap_key_type::<StrBEU16Codec>();
for result in remap_key_type {
@ -418,7 +418,7 @@ impl<'ctx> SearchContext<'ctx> {
let remap_key_type = self
.index
.word_prefix_fid_docids
.remap_types::<ByteSlice, ByteSlice>()
.remap_types::<Bytes, Bytes>()
.prefix_iter(self.txn, &key)?
.remap_key_type::<StrBEU16Codec>();
for result in remap_key_type {
@ -446,7 +446,7 @@ impl<'ctx> SearchContext<'ctx> {
(word, position),
&(self.word_interner.get(word).as_str(), position),
&mut self.db_cache.word_position_docids,
self.index.word_position_docids.remap_data_type::<ByteSlice>(),
self.index.word_position_docids.remap_data_type::<Bytes>(),
)
}
@ -460,7 +460,7 @@ impl<'ctx> SearchContext<'ctx> {
(word_prefix, position),
&(self.word_interner.get(word_prefix).as_str(), position),
&mut self.db_cache.word_prefix_position_docids,
self.index.word_prefix_position_docids.remap_data_type::<ByteSlice>(),
self.index.word_prefix_position_docids.remap_data_type::<Bytes>(),
)
}
@ -474,7 +474,7 @@ impl<'ctx> SearchContext<'ctx> {
let remap_key_type = self
.index
.word_position_docids
.remap_types::<ByteSlice, ByteSlice>()
.remap_types::<Bytes, Bytes>()
.prefix_iter(self.txn, &key)?
.remap_key_type::<StrBEU16Codec>();
for result in remap_key_type {
@ -505,7 +505,7 @@ impl<'ctx> SearchContext<'ctx> {
let remap_key_type = self
.index
.word_prefix_position_docids
.remap_types::<ByteSlice, ByteSlice>()
.remap_types::<Bytes, Bytes>()
.prefix_iter(self.txn, &key)?
.remap_key_type::<StrBEU16Codec>();
for result in remap_key_type {

14
milli/src/search/new/distinct.rs
View File

@ -1,4 +1,4 @@
use heed::types::{ByteSlice, Str, Unit};
use heed::types::{Bytes, Str, Unit};
use heed::{Database, RoPrefix, RoTxn};
use roaring::RoaringBitmap;
@ -8,7 +8,7 @@ const DOCID_SIZE: usize = 4;
use crate::heed_codec::facet::{
FacetGroupKey, FacetGroupKeyCodec, FacetGroupValueCodec, FieldDocIdFacetCodec,
};
use crate::heed_codec::ByteSliceRefCodec;
use crate::heed_codec::BytesRefCodec;
use crate::{Index, Result, SearchContext};
pub struct DistinctOutput {
@ -71,7 +71,7 @@ pub fn distinct_single_docid(
/// Return all the docids containing the given value in the given field
fn facet_value_docids(
database: Database<FacetGroupKeyCodec<ByteSliceRefCodec>, FacetGroupValueCodec>,
database: Database<FacetGroupKeyCodec<BytesRefCodec>, FacetGroupValueCodec>,
txn: &RoTxn,
field_id: u16,
facet_value: &[u8],
@ -87,12 +87,12 @@ fn facet_number_values<'a>(
field_id: u16,
index: &Index,
txn: &'a RoTxn,
) -> Result<RoPrefix<'a, FieldDocIdFacetCodec<ByteSliceRefCodec>, Unit>> {
) -> Result<RoPrefix<'a, FieldDocIdFacetCodec<BytesRefCodec>, Unit>> {
let key = facet_values_prefix_key(field_id, docid);
let iter = index
.field_id_docid_facet_f64s
.remap_key_type::<ByteSlice>()
.remap_key_type::<Bytes>()
.prefix_iter(txn, &key)?
.remap_key_type();
@ -105,12 +105,12 @@ pub fn facet_string_values<'a>(
field_id: u16,
index: &Index,
txn: &'a RoTxn,
) -> Result<RoPrefix<'a, FieldDocIdFacetCodec<ByteSliceRefCodec>, Str>> {
) -> Result<RoPrefix<'a, FieldDocIdFacetCodec<BytesRefCodec>, Str>> {
let key = facet_values_prefix_key(field_id, docid);
let iter = index
.field_id_docid_facet_strings
.remap_key_type::<ByteSlice>()
.remap_key_type::<Bytes>()
.prefix_iter(txn, &key)?
.remap_types();

4
milli/src/search/new/geo_sort.rs
View File

@ -1,7 +1,7 @@
use std::collections::VecDeque;
use std::iter::FromIterator;
use heed::types::{ByteSlice, Unit};
use heed::types::{Bytes, Unit};
use heed::{RoPrefix, RoTxn};
use roaring::RoaringBitmap;
use rstar::RTree;
@ -34,7 +34,7 @@ fn facet_number_values<'a>(
let iter = index
.field_id_docid_facet_f64s
.remap_key_type::<ByteSlice>()
.remap_key_type::<Bytes>()
.prefix_iter(txn, &key)?
.remap_key_type();

6
milli/src/search/new/sort.rs
View File

@ -4,7 +4,7 @@ use roaring::RoaringBitmap;
use super::logger::SearchLogger;
use super::{RankingRule, RankingRuleOutput, RankingRuleQueryTrait, SearchContext};
use crate::heed_codec::facet::{FacetGroupKeyCodec, OrderedF64Codec};
use crate::heed_codec::{ByteSliceRefCodec, StrRefCodec};
use crate::heed_codec::{BytesRefCodec, StrRefCodec};
use crate::score_details::{self, ScoreDetails};
use crate::search::facet::{ascending_facet_sort, descending_facet_sort};
use crate::{FieldId, Index, Result};
@ -100,11 +100,11 @@ impl<'ctx, Query: RankingRuleQueryTrait> RankingRule<'ctx, Query> for Sort<'ctx,
let number_db = ctx
.index
.facet_id_f64_docids
.remap_key_type::<FacetGroupKeyCodec<ByteSliceRefCodec>>();
.remap_key_type::<FacetGroupKeyCodec<BytesRefCodec>>();
let string_db = ctx
.index
.facet_id_string_docids
.remap_key_type::<FacetGroupKeyCodec<ByteSliceRefCodec>>();
.remap_key_type::<FacetGroupKeyCodec<BytesRefCodec>>();
let (number_iter, string_iter) = if self.is_ascending {
let number_iter = ascending_facet_sort(

34
milli/src/update/facet/bulk.rs
View File

@ -2,7 +2,7 @@ use std::fs::File;
use std::io::BufReader;
use grenad::CompressionType;
use heed::types::ByteSlice;
use heed::types::Bytes;
use heed::{BytesDecode, BytesEncode, Error, PutFlags, RoTxn, RwTxn};
use roaring::RoaringBitmap;
@ -11,7 +11,7 @@ use crate::facet::FacetType;
use crate::heed_codec::facet::{
FacetGroupKey, FacetGroupKeyCodec, FacetGroupValue, FacetGroupValueCodec,
};
use crate::heed_codec::ByteSliceRefCodec;
use crate::heed_codec::BytesRefCodec;
use crate::update::del_add::{DelAdd, KvReaderDelAdd};
use crate::update::index_documents::{create_writer, valid_lmdb_key, writer_into_reader};
use crate::{CboRoaringBitmapCodec, CboRoaringBitmapLenCodec, FieldId, Index, Result};
@ -70,11 +70,11 @@ impl<'i> FacetsUpdateBulk<'i> {
let Self { index, field_ids, group_size, min_level_size, facet_type, delta_data } = self;
let db = match facet_type {
FacetType::String => index
.facet_id_string_docids
.remap_key_type::<FacetGroupKeyCodec<ByteSliceRefCodec>>(),
FacetType::String => {
index.facet_id_string_docids.remap_key_type::<FacetGroupKeyCodec<BytesRefCodec>>()
}
FacetType::Number => {
index.facet_id_f64_docids.remap_key_type::<FacetGroupKeyCodec<ByteSliceRefCodec>>()
index.facet_id_f64_docids.remap_key_type::<FacetGroupKeyCodec<BytesRefCodec>>()
}
};
@ -88,7 +88,7 @@ impl<'i> FacetsUpdateBulk<'i> {
/// Implementation of `FacetsUpdateBulk` that is independent of milli's `Index` type
pub(crate) struct FacetsUpdateBulkInner<R: std::io::Read + std::io::Seek> {
pub db: heed::Database<FacetGroupKeyCodec<ByteSliceRefCodec>, FacetGroupValueCodec>,
pub db: heed::Database<FacetGroupKeyCodec<BytesRefCodec>, FacetGroupValueCodec>,
pub delta_data: Option<grenad::Reader<R>>,
pub group_size: u8,
pub min_level_size: u8,
@ -106,7 +106,7 @@ impl<R: std::io::Read + std::io::Seek> FacetsUpdateBulkInner<R> {
for level_reader in level_readers {
let mut cursor = level_reader.into_cursor()?;
while let Some((k, v)) = cursor.move_on_next()? {
self.db.remap_types::<ByteSlice, ByteSlice>().put(wtxn, k, v)?;
self.db.remap_types::<Bytes, Bytes>().put(wtxn, k, v)?;
}
}
}
@ -128,7 +128,7 @@ impl<R: std::io::Read + std::io::Seek> FacetsUpdateBulkInner<R> {
};
if self.db.is_empty(wtxn)? {
let mut buffer = Vec::new();
let mut database = self.db.iter_mut(wtxn)?.remap_types::<ByteSlice, ByteSlice>();
let mut database = self.db.iter_mut(wtxn)?.remap_types::<Bytes, Bytes>();
let mut cursor = delta_data.into_cursor()?;
while let Some((key, value)) = cursor.move_on_next()? {
if !valid_lmdb_key(key) {
@ -147,16 +147,12 @@ impl<R: std::io::Read + std::io::Seek> FacetsUpdateBulkInner<R> {
// then we extend the buffer with the docids bitmap
buffer.extend_from_slice(value);
unsafe {
database.put_current_with_options::<ByteSlice>(
PutFlags::APPEND,
key,
&buffer,
)?
database.put_current_with_options::<Bytes>(PutFlags::APPEND, key, &buffer)?
};
}
} else {
let mut buffer = Vec::new();
let database = self.db.remap_types::<ByteSlice, ByteSlice>();
let database = self.db.remap_types::<Bytes, Bytes>();
let mut cursor = delta_data.into_cursor()?;
while let Some((key, value)) = cursor.move_on_next()? {
@ -225,9 +221,9 @@ impl<R: std::io::Read + std::io::Seek> FacetsUpdateBulkInner<R> {
let level_0_iter = self
.db
.remap_types::<ByteSlice, ByteSlice>()
.remap_types::<Bytes, Bytes>()
.prefix_iter(rtxn, level_0_prefix.as_slice())?
.remap_types::<FacetGroupKeyCodec<ByteSliceRefCodec>, FacetGroupValueCodec>();
.remap_types::<FacetGroupKeyCodec<BytesRefCodec>, FacetGroupValueCodec>();
let mut left_bound: &[u8] = &[];
let mut first_iteration_for_new_group = true;
@ -313,7 +309,7 @@ impl<R: std::io::Read + std::io::Seek> FacetsUpdateBulkInner<R> {
bitmaps.drain(..).zip(left_bounds.drain(..)).zip(group_sizes.drain(..))
{
let key = FacetGroupKey { field_id, level, left_bound };
let key = FacetGroupKeyCodec::<ByteSliceRefCodec>::bytes_encode(&key)
let key = FacetGroupKeyCodec::<BytesRefCodec>::bytes_encode(&key)
.map_err(Error::Encoding)?;
let value = FacetGroupValue { size: group_size, bitmap };
let value =
@ -342,7 +338,7 @@ impl<R: std::io::Read + std::io::Seek> FacetsUpdateBulkInner<R> {
bitmaps.drain(..).zip(left_bounds.drain(..)).zip(group_sizes.drain(..))
{
let key = FacetGroupKey { field_id, level, left_bound };
let key = FacetGroupKeyCodec::<ByteSliceRefCodec>::bytes_encode(&key)
let key = FacetGroupKeyCodec::<BytesRefCodec>::bytes_encode(&key)
.map_err(Error::Encoding)?;
let value = FacetGroupValue { size: group_size, bitmap };
let value = FacetGroupValueCodec::bytes_encode(&value).map_err(Error::Encoding)?;

54
milli/src/update/facet/incremental.rs
View File

@ -1,7 +1,7 @@
use std::fs::File;
use std::io::BufReader;
use heed::types::{ByteSlice, DecodeIgnore};
use heed::types::{Bytes, DecodeIgnore};
use heed::{BytesDecode, Error, RoTxn, RwTxn};
use obkv::KvReader;
use roaring::RoaringBitmap;
@ -10,7 +10,7 @@ use crate::facet::FacetType;
use crate::heed_codec::facet::{
FacetGroupKey, FacetGroupKeyCodec, FacetGroupValue, FacetGroupValueCodec,
};
use crate::heed_codec::ByteSliceRefCodec;
use crate::heed_codec::BytesRefCodec;
use crate::search::facet::get_highest_level;
use crate::update::del_add::DelAdd;
use crate::update::index_documents::valid_lmdb_key;
@ -48,10 +48,10 @@ impl FacetsUpdateIncremental {
db: match facet_type {
FacetType::String => index
.facet_id_string_docids
.remap_key_type::<FacetGroupKeyCodec<ByteSliceRefCodec>>(),
.remap_key_type::<FacetGroupKeyCodec<BytesRefCodec>>(),
FacetType::Number => index
.facet_id_f64_docids
.remap_key_type::<FacetGroupKeyCodec<ByteSliceRefCodec>>(),
.remap_key_type::<FacetGroupKeyCodec<BytesRefCodec>>(),
},
group_size,
max_group_size,
@ -67,7 +67,7 @@ impl FacetsUpdateIncremental {
if !valid_lmdb_key(key) {
continue;
}
let key = FacetGroupKeyCodec::<ByteSliceRefCodec>::bytes_decode(key)
let key = FacetGroupKeyCodec::<BytesRefCodec>::bytes_decode(key)
.map_err(heed::Error::Encoding)?;
let value = KvReader::new(value);
@ -98,7 +98,7 @@ impl FacetsUpdateIncremental {
/// Implementation of `FacetsUpdateIncremental` that is independent of milli's `Index` type
pub struct FacetsUpdateIncrementalInner {
pub db: heed::Database<FacetGroupKeyCodec<ByteSliceRefCodec>, FacetGroupValueCodec>,
pub db: heed::Database<FacetGroupKeyCodec<BytesRefCodec>, FacetGroupValueCodec>,
pub group_size: u8,
pub min_level_size: u8,
pub max_group_size: u8,
@ -136,11 +136,11 @@ impl FacetsUpdateIncrementalInner {
let mut iter = self
.db
.remap_types::<ByteSlice, FacetGroupValueCodec>()
.remap_types::<Bytes, FacetGroupValueCodec>()
.prefix_iter(txn, prefix.as_slice())?;
let (key_bytes, value) = iter.next().unwrap()?;
Ok((
FacetGroupKeyCodec::<ByteSliceRefCodec>::bytes_decode(key_bytes)
FacetGroupKeyCodec::<BytesRefCodec>::bytes_decode(key_bytes)
.map_err(Error::Encoding)?
.into_owned(),
value,
@ -177,7 +177,7 @@ impl FacetsUpdateIncrementalInner {
level0_prefix.push(0);
let mut iter =
self.db.remap_types::<ByteSlice, DecodeIgnore>().prefix_iter(txn, &level0_prefix)?;
self.db.remap_types::<Bytes, DecodeIgnore>().prefix_iter(txn, &level0_prefix)?;
if iter.next().is_none() {
drop(iter);
@ -379,11 +379,8 @@ impl FacetsUpdateIncrementalInner {
highest_level_prefix.extend_from_slice(&field_id.to_be_bytes());
highest_level_prefix.push(highest_level);
let size_highest_level = self
.db
.remap_types::<ByteSlice, ByteSlice>()
.prefix_iter(txn, &highest_level_prefix)?
.count();
let size_highest_level =
self.db.remap_types::<Bytes, Bytes>().prefix_iter(txn, &highest_level_prefix)?.count();
if size_highest_level < self.group_size as usize * self.min_level_size as usize {
return Ok(());
@ -391,7 +388,7 @@ impl FacetsUpdateIncrementalInner {
let mut groups_iter = self
.db
.remap_types::<ByteSlice, FacetGroupValueCodec>()
.remap_types::<Bytes, FacetGroupValueCodec>()
.prefix_iter(txn, &highest_level_prefix)?;
let nbr_new_groups = size_highest_level / self.group_size as usize;
@ -403,7 +400,7 @@ impl FacetsUpdateIncrementalInner {
let mut values = RoaringBitmap::new();
for _ in 0..group_size {
let (key_bytes, value_i) = groups_iter.next().unwrap()?;
let key_i = FacetGroupKeyCodec::<ByteSliceRefCodec>::bytes_decode(key_bytes)
let key_i = FacetGroupKeyCodec::<BytesRefCodec>::bytes_decode(key_bytes)
.map_err(Error::Encoding)?;
if first_key.is_none() {
@ -426,7 +423,7 @@ impl FacetsUpdateIncrementalInner {
let mut values = RoaringBitmap::new();
for _ in 0..nbr_leftover_elements {
let (key_bytes, value_i) = groups_iter.next().unwrap()?;
let key_i = FacetGroupKeyCodec::<ByteSliceRefCodec>::bytes_decode(key_bytes)
let key_i = FacetGroupKeyCodec::<BytesRefCodec>::bytes_decode(key_bytes)
.map_err(Error::Encoding)?;
if first_key.is_none() {
@ -594,7 +591,7 @@ impl FacetsUpdateIncrementalInner {
if highest_level == 0
|| self
.db
.remap_types::<ByteSlice, ByteSlice>()
.remap_types::<Bytes, Bytes>()
.prefix_iter(txn, &highest_level_prefix)?
.count()
>= self.min_level_size as usize
@ -602,14 +599,12 @@ impl FacetsUpdateIncrementalInner {
return Ok(());
}
let mut to_delete = vec![];
let mut iter = self
.db
.remap_types::<ByteSlice, ByteSlice>()
.prefix_iter(txn, &highest_level_prefix)?;
let mut iter =
self.db.remap_types::<Bytes, Bytes>().prefix_iter(txn, &highest_level_prefix)?;
for el in iter.by_ref() {
let (k, _) = el?;
to_delete.push(
FacetGroupKeyCodec::<ByteSliceRefCodec>::bytes_decode(k)
FacetGroupKeyCodec::<BytesRefCodec>::bytes_decode(k)
.map_err(Error::Encoding)?
.into_owned(),
);
@ -1118,7 +1113,7 @@ mod fuzz {
#[no_coverage]
fn compare_with_trivial_database(tempdir: Rc<TempDir>, operations: &[Operation]) {
let index = FacetIndex::<ByteSliceRefCodec>::open_from_tempdir(tempdir, 4, 8, 5); // dummy params, they'll be overwritten
let index = FacetIndex::<BytesRefCodec>::open_from_tempdir(tempdir, 4, 8, 5); // dummy params, they'll be overwritten
let mut txn = index.env.write_txn().unwrap();
let mut trivial_db = TrivialDatabase::<Vec<u8>>::default();
@ -1164,16 +1159,13 @@ mod fuzz {
let level0iter = index
.content
.as_polymorph()
.prefix_iter::<_, ByteSlice, FacetGroupValueCodec>(
&mut txn,
&field_id.to_be_bytes(),
)
.prefix_iter::<_, Bytes, FacetGroupValueCodec>(&mut txn, &field_id.to_be_bytes())
.unwrap();
for ((key, values), group) in values_field_id.iter().zip(level0iter) {
let (group_key, group_values) = group.unwrap();
let group_key =
FacetGroupKeyCodec::<ByteSliceRefCodec>::bytes_decode(group_key).unwrap();
FacetGroupKeyCodec::<BytesRefCodec>::bytes_decode(group_key).unwrap();
assert_eq!(key, &group_key.left_bound);
assert_eq!(values, &group_values.bitmap);
}
@ -1183,13 +1175,13 @@ mod fuzz {
let level0iter = index
.content
.as_polymorph()
.prefix_iter::<_, ByteSlice, FacetGroupValueCodec>(&txn, &field_id.to_be_bytes())
.prefix_iter::<_, Bytes, FacetGroupValueCodec>(&txn, &field_id.to_be_bytes())
.unwrap();
for ((key, values), group) in values_field_id.iter().zip(level0iter) {
let (group_key, group_values) = group.unwrap();
let group_key =
FacetGroupKeyCodec::<ByteSliceRefCodec>::bytes_decode(group_key).unwrap();
FacetGroupKeyCodec::<BytesRefCodec>::bytes_decode(group_key).unwrap();
assert_eq!(key, &group_key.left_bound);
assert_eq!(values, &group_values.bitmap);
}

39
milli/src/update/facet/mod.rs
View File

@ -83,7 +83,7 @@ use std::iter::FromIterator;
use charabia::normalizer::{Normalize, NormalizerOption};
use grenad::{CompressionType, SortAlgorithm};
use heed::types::{ByteSlice, DecodeIgnore, SerdeJson};
use heed::types::{Bytes, DecodeIgnore, SerdeJson};
use heed::BytesEncode;
use log::debug;
use time::OffsetDateTime;
@ -92,7 +92,7 @@ use self::incremental::FacetsUpdateIncremental;
use super::FacetsUpdateBulk;
use crate::facet::FacetType;
use crate::heed_codec::facet::{FacetGroupKey, FacetGroupKeyCodec, FacetGroupValueCodec};
use crate::heed_codec::ByteSliceRefCodec;
use crate::heed_codec::BytesRefCodec;
use crate::update::index_documents::create_sorter;
use crate::update::merge_btreeset_string;
use crate::{BEU16StrCodec, Index, Result, MAX_FACET_VALUE_LENGTH};
@ -106,7 +106,7 @@ pub mod incremental;
/// a bulk update method or an incremental update method.
pub struct FacetsUpdate<'i> {
index: &'i Index,
database: heed::Database<FacetGroupKeyCodec<ByteSliceRefCodec>, FacetGroupValueCodec>,
database: heed::Database<FacetGroupKeyCodec<BytesRefCodec>, FacetGroupValueCodec>,
facet_type: FacetType,
delta_data: grenad::Reader<BufReader<File>>,
group_size: u8,
@ -120,11 +120,11 @@ impl<'i> FacetsUpdate<'i> {
delta_data: grenad::Reader<BufReader<File>>,
) -> Self {
let database = match facet_type {
FacetType::String => index
.facet_id_string_docids
.remap_key_type::<FacetGroupKeyCodec<ByteSliceRefCodec>>(),
FacetType::String => {
index.facet_id_string_docids.remap_key_type::<FacetGroupKeyCodec<BytesRefCodec>>()
}
FacetType::Number => {
index.facet_id_f64_docids.remap_key_type::<FacetGroupKeyCodec<ByteSliceRefCodec>>()
index.facet_id_f64_docids.remap_key_type::<FacetGroupKeyCodec<BytesRefCodec>>()
}
};
Self {
@ -217,10 +217,11 @@ impl<'i> FacetsUpdate<'i> {
// as the grenad sorter already merged them for us.
let mut merger_iter = sorter.into_stream_merger_iter()?;
while let Some((key_bytes, btreeset_bytes)) = merger_iter.next()? {
self.index
.facet_id_normalized_string_strings
.remap_types::<ByteSlice, ByteSlice>()
.put(wtxn, key_bytes, btreeset_bytes)?;
self.index.facet_id_normalized_string_strings.remap_types::<Bytes, Bytes>().put(
wtxn,
key_bytes,
btreeset_bytes,
)?;
}
// We compute one FST by string facet
@ -267,7 +268,7 @@ pub(crate) mod test_helpers {
use std::marker::PhantomData;
use std::rc::Rc;
use heed::types::ByteSlice;
use heed::types::Bytes;
use heed::{BytesDecode, BytesEncode, Env, RoTxn, RwTxn};
use roaring::RoaringBitmap;
@ -275,7 +276,7 @@ pub(crate) mod test_helpers {
use crate::heed_codec::facet::{
FacetGroupKey, FacetGroupKeyCodec, FacetGroupValue, FacetGroupValueCodec,
};
use crate::heed_codec::ByteSliceRefCodec;
use crate::heed_codec::BytesRefCodec;
use crate::search::facet::get_highest_level;
use crate::snapshot_tests::display_bitmap;
use crate::update::del_add::{DelAdd, KvWriterDelAdd};
@ -306,7 +307,7 @@ pub(crate) mod test_helpers {
BytesEncode<'a> + BytesDecode<'a, DItem = <BoundCodec as BytesEncode<'a>>::EItem>,
{
pub env: Env,
pub content: heed::Database<FacetGroupKeyCodec<ByteSliceRefCodec>, FacetGroupValueCodec>,
pub content: heed::Database<FacetGroupKeyCodec<BytesRefCodec>, FacetGroupValueCodec>,
pub group_size: Cell<u8>,
pub min_level_size: Cell<u8>,
pub max_group_size: Cell<u8>,
@ -454,7 +455,7 @@ pub(crate) mod test_helpers {
let left_bound_bytes = BoundCodec::bytes_encode(left_bound).unwrap().into_owned();
let key: FacetGroupKey<&[u8]> =
FacetGroupKey { field_id: *field_id, level: 0, left_bound: &left_bound_bytes };
let key = FacetGroupKeyCodec::<ByteSliceRefCodec>::bytes_encode(&key).unwrap();
let key = FacetGroupKeyCodec::<BytesRefCodec>::bytes_encode(&key).unwrap();
let mut inner_writer = KvWriterDelAdd::memory();
let value = CboRoaringBitmapCodec::bytes_encode(docids).unwrap();
inner_writer.insert(DelAdd::Addition, value).unwrap();
@ -486,12 +487,12 @@ pub(crate) mod test_helpers {
let iter = self
.content
.remap_types::<ByteSlice, FacetGroupValueCodec>()
.remap_types::<Bytes, FacetGroupValueCodec>()
.prefix_iter(txn, &level_no_prefix)
.unwrap();
for el in iter {
let (key, value) = el.unwrap();
let key = FacetGroupKeyCodec::<ByteSliceRefCodec>::bytes_decode(key).unwrap();
let key = FacetGroupKeyCodec::<BytesRefCodec>::bytes_decode(key).unwrap();
let mut prefix_start_below = vec![];
prefix_start_below.extend_from_slice(&field_id.to_be_bytes());
@ -501,11 +502,11 @@ pub(crate) mod test_helpers {
let start_below = {
let mut start_below_iter = self
.content
.remap_types::<ByteSlice, FacetGroupValueCodec>()
.remap_types::<Bytes, FacetGroupValueCodec>()
.prefix_iter(txn, &prefix_start_below)
.unwrap();
let (key_bytes, _) = start_below_iter.next().unwrap().unwrap();
FacetGroupKeyCodec::<ByteSliceRefCodec>::bytes_decode(key_bytes).unwrap()
FacetGroupKeyCodec::<BytesRefCodec>::bytes_decode(key_bytes).unwrap()
};
assert!(value.size > 0);

4
milli/src/update/index_documents/helpers/grenad_helpers.rs
View File

@ -3,7 +3,7 @@ use std::fs::File;
use std::io::{self, BufReader, BufWriter, Seek};
use grenad::{CompressionType, Sorter};
use heed::types::ByteSlice;
use heed::types::Bytes;
use super::{ClonableMmap, MergeFn};
use crate::update::index_documents::valid_lmdb_key;
@ -255,7 +255,7 @@ where
puffin::profile_function!();
let mut buffer = Vec::new();
let database = database.remap_types::<ByteSlice, ByteSlice>();
let database = database.remap_types::<Bytes, Bytes>();
let mut merger_iter = sorter.into_stream_merger_iter()?;
while let Some((key, value)) = merger_iter.next()? {

4
milli/src/update/index_documents/transform.rs
View File

@ -247,7 +247,7 @@ impl<'a, 'i> Transform<'a, 'i> {
let base_obkv = self
.index
.documents
.remap_data_type::<heed::types::ByteSlice>()
.remap_data_type::<heed::types::Bytes>()
.get(wtxn, &original_key)?
.ok_or(InternalError::DatabaseMissingEntry {
db_name: db_name::DOCUMENTS,
@ -501,7 +501,7 @@ impl<'a, 'i> Transform<'a, 'i> {
let base_obkv = self
.index
.documents
.remap_data_type::<heed::types::ByteSlice>()
.remap_data_type::<heed::types::Bytes>()
.get(txn, &original_key)?
.ok_or(InternalError::DatabaseMissingEntry {
db_name: db_name::DOCUMENTS,

14
milli/src/update/index_documents/typed_chunk.rs
View File

@ -6,7 +6,7 @@ use std::io::{self, BufReader};
use bytemuck::allocation::pod_collect_to_vec;
use charabia::{Language, Script};
use grenad::MergerBuilder;
use heed::types::ByteSlice;
use heed::types::Bytes;
use heed::{PutFlags, RwTxn};
use log::error;
use obkv::{KvReader, KvWriter};
@ -144,7 +144,7 @@ pub(crate) fn write_typed_chunk_into_index(
}
}
let db = index.documents.remap_data_type::<ByteSlice>();
let db = index.documents.remap_data_type::<Bytes>();
if !writer.is_empty() {
db.put(wtxn, &docid, &writer.into_inner().unwrap())?;
@ -293,7 +293,7 @@ pub(crate) fn write_typed_chunk_into_index(
}
TypedChunk::FieldIdDocidFacetNumbers(fid_docid_facet_number) => {
let index_fid_docid_facet_numbers =
index.field_id_docid_facet_f64s.remap_types::<ByteSlice, ByteSlice>();
index.field_id_docid_facet_f64s.remap_types::<Bytes, Bytes>();
let mut cursor = fid_docid_facet_number.into_cursor()?;
while let Some((key, value)) = cursor.move_on_next()? {
let reader = KvReaderDelAdd::new(value);
@ -313,7 +313,7 @@ pub(crate) fn write_typed_chunk_into_index(
}
TypedChunk::FieldIdDocidFacetStrings(fid_docid_facet_string) => {
let index_fid_docid_facet_strings =
index.field_id_docid_facet_strings.remap_types::<ByteSlice, ByteSlice>();
index.field_id_docid_facet_strings.remap_types::<Bytes, Bytes>();
let mut cursor = fid_docid_facet_string.into_cursor()?;
while let Some((key, value)) = cursor.move_on_next()? {
let reader = KvReaderDelAdd::new(value);
@ -498,7 +498,7 @@ where
puffin::profile_function!(format!("number of entries: {}", data.len()));
let mut buffer = Vec::new();
let database = database.remap_types::<ByteSlice, ByteSlice>();
let database = database.remap_types::<Bytes, Bytes>();
let mut cursor = data.into_cursor()?;
while let Some((key, value)) = cursor.move_on_next()? {
@ -556,7 +556,7 @@ where
}
let mut buffer = Vec::new();
let mut database = database.iter_mut(wtxn)?.remap_types::<ByteSlice, ByteSlice>();
let mut database = database.iter_mut(wtxn)?.remap_types::<Bytes, Bytes>();
let mut cursor = data.into_cursor()?;
while let Some((key, value)) = cursor.move_on_next()? {
@ -571,7 +571,7 @@ where
let value = serialize_value(value, &mut buffer)?;
unsafe {
// safety: We do not keep a reference to anything that lives inside the database
database.put_current_with_options::<ByteSlice>(PutFlags::APPEND, key, value)?
database.put_current_with_options::<Bytes>(PutFlags::APPEND, key, value)?
};
}
}

10
milli/src/update/settings.rs
View File

@ -822,7 +822,7 @@ impl<'a, 't, 'i> Settings<'a, 't, 'i> {
fn update_max_values_per_facet(&mut self) -> Result<()> {
match self.max_values_per_facet {
Setting::Set(max) => {
self.index.put_max_values_per_facet(self.wtxn, max)?;
self.index.put_max_values_per_facet(self.wtxn, max as u64)?;
}
Setting::Reset => {
self.index.delete_max_values_per_facet(self.wtxn)?;
@ -850,7 +850,7 @@ impl<'a, 't, 'i> Settings<'a, 't, 'i> {
fn update_pagination_max_total_hits(&mut self) -> Result<()> {
match self.pagination_max_total_hits {
Setting::Set(max) => {
self.index.put_pagination_max_total_hits(self.wtxn, max)?;
self.index.put_pagination_max_total_hits(self.wtxn, max as u64)?;
}
Setting::Reset => {
self.index.delete_pagination_max_total_hits(self.wtxn)?;
@ -917,7 +917,7 @@ impl<'a, 't, 'i> Settings<'a, 't, 'i> {
#[cfg(test)]
mod tests {
use big_s::S;
use heed::types::ByteSlice;
use heed::types::Bytes;
use maplit::{btreemap, btreeset, hashset};
use super::*;
@ -1130,7 +1130,7 @@ mod tests {
}
let count = index
.facet_id_f64_docids
.remap_key_type::<ByteSlice>()
.remap_key_type::<Bytes>()
// The faceted field id is 1u16
.prefix_iter(&rtxn, &[0, 1, 0])
.unwrap()
@ -1151,7 +1151,7 @@ mod tests {
// Only count the field_id 0 and level 0 facet values.
let count = index
.facet_id_f64_docids
.remap_key_type::<ByteSlice>()
.remap_key_type::<Bytes>()
.prefix_iter(&rtxn, &[0, 1, 0])
.unwrap()
.count();

4
milli/src/update/word_prefix_docids.rs
View File

@ -1,7 +1,7 @@
use std::collections::{HashMap, HashSet};
use grenad::CompressionType;
use heed::types::{ByteSlice, Str};
use heed::types::{Bytes, Str};
use heed::Database;
use crate::update::del_add::{deladd_serialize_add_side, DelAdd, KvWriterDelAdd};
@ -93,7 +93,7 @@ impl<'t, 'i> WordPrefixDocids<'t, 'i> {
}
// We fetch the docids associated to the newly added word prefix fst only.
let db = self.word_docids.remap_data_type::<ByteSlice>();
let db = self.word_docids.remap_data_type::<Bytes>();
let mut buffer = Vec::new();
for prefix in new_prefix_fst_words {
let prefix = std::str::from_utf8(prefix.as_bytes())?;

6
milli/src/update/words_prefix_integer_docids.rs
View File

@ -2,7 +2,7 @@ use std::collections::{HashMap, HashSet};
use std::str;
use grenad::CompressionType;
use heed::types::ByteSlice;
use heed::types::Bytes;
use heed::{BytesDecode, BytesEncode, Database};
use log::debug;
@ -110,7 +110,7 @@ impl<'t, 'i> WordPrefixIntegerDocids<'t, 'i> {
}
// We fetch the docids associated to the newly added word prefix fst only.
let db = self.word_database.remap_data_type::<ByteSlice>();
let db = self.word_database.remap_data_type::<Bytes>();
let mut buffer = Vec::new();
for prefix_bytes in new_prefix_fst_words {
let prefix = str::from_utf8(prefix_bytes.as_bytes()).map_err(|_| {
@ -119,7 +119,7 @@ impl<'t, 'i> WordPrefixIntegerDocids<'t, 'i> {
// iter over all lines of the DB where the key is prefixed by the current prefix.
let iter = db
.remap_key_type::<ByteSlice>()
.remap_key_type::<Bytes>()
.prefix_iter(self.wtxn, prefix_bytes.as_bytes())?
.remap_key_type::<StrBEU16Codec>();
for result in iter {