Move the fst_stream_into_hashset method in the helper methods

This commit is contained in:
Clément Renault
2022-01-19 14:30:03 +01:00
committed by Kerollmops
parent c90fa95f93
commit 5404bc02dd
3 changed files with 24 additions and 21 deletions

View File

@ -2,9 +2,11 @@ mod clonable_mmap;
mod grenad_helpers;
mod merge_functions;
use std::collections::HashSet;
use std::convert::{TryFrom, TryInto};
pub use clonable_mmap::{ClonableMmap, CursorClonableMmap};
use fst::{IntoStreamer, Streamer};
pub use grenad_helpers::{
create_sorter, create_writer, grenad_obkv_into_chunks, into_clonable_grenad, merge_readers,
sorter_into_lmdb_database, sorter_into_reader, write_into_lmdb_database, writer_into_reader,
@ -43,3 +45,17 @@ where
pub fn read_u32_ne_bytes(bytes: &[u8]) -> impl Iterator<Item = u32> + '_ {
bytes.chunks_exact(4).flat_map(TryInto::try_into).map(u32::from_ne_bytes)
}
/// Converts an fst Stream into an HashSet.
pub fn fst_stream_into_hashset<'f, I, S>(stream: I) -> HashSet<Vec<u8>>
where
I: for<'a> IntoStreamer<'a, Into = S, Item = &'a [u8]>,
S: 'f + for<'a> Streamer<'a, Item = &'a [u8]>,
{
let mut hashset = HashSet::new();
let mut stream = stream.into_stream();
while let Some(value) = stream.next() {
hashset.insert(value.to_owned());
}
hashset
}

View File

@ -15,8 +15,9 @@ use serde::{Deserialize, Serialize};
use typed_chunk::{write_typed_chunk_into_index, TypedChunk};
pub use self::helpers::{
create_sorter, create_writer, merge_cbo_roaring_bitmaps, merge_roaring_bitmaps,
sorter_into_lmdb_database, write_into_lmdb_database, writer_into_reader, ClonableMmap, MergeFn,
create_sorter, create_writer, fst_stream_into_hashset, merge_cbo_roaring_bitmaps,
merge_roaring_bitmaps, sorter_into_lmdb_database, write_into_lmdb_database, writer_into_reader,
ClonableMmap, MergeFn,
};
use self::helpers::{grenad_obkv_into_chunks, GrenadParameters};
pub use self::transform::{Transform, TransformOutput};