Compare commits

...

2 Commits

Author SHA1 Message Date
lancedb-gatefixer[bot] 0d19a6c546 fix: support nested-list FTS indexing (#4059)
## Summary

- validate native FTS fields against the recursively resolved terminal
text leaf
- preserve canonical public paths and list depth for Lance
document-boundary handling
- cover async nested-list index creation and deepest-list `_doc_index`
search coordinates

## Root cause

The FTS resolver recursively found the terminal text field but returned
the outer list field. Native validation therefore rejected
`List(List(Utf8))` before Lance could create a list-element index.

## Validation

- `cargo test --quiet --features remote -p lancedb
test_nested_list_fts_uses_deepest_document_coordinates -- --nocapture`
- `cargo test --quiet --features remote -p lancedb
test_execute_async_validates_fts_input_before_starting_job`
- `cargo test --quiet --features remote -p lancedb
test_public_fts_field_path_prefers_exact_case`
- `cargo check --quiet --features remote --tests --examples`
- `cargo clippy --quiet --features remote --tests --examples`
- `cargo fmt --all -- --check`

Fixes #4058

<!-- lance-gatekeeper-fix:v1 agent=a4e48f49d12b2b6de7f10adb1c5bf5f5
generation=1 -->

Co-authored-by: Gatefixer <313497061+lancedb-gatefixer[bot]@users.noreply.github.com>
2026-09-02 17:51:21 +08:00
Lance Release c0f33f8627 Bump version: 0.39.0-beta.0 → 0.39.0-beta.1 2026-09-02 05:28:44 +00:00
18 changed files with 101 additions and 24 deletions
+1 -1
View File
@@ -1,5 +1,5 @@
[tool.bumpversion]
current_version = "0.39.0-beta.0"
current_version = "0.39.0-beta.1"
parse = """(?x)
(?P<major>0|[1-9]\\d*)\\.
(?P<minor>0|[1-9]\\d*)\\.
Generated
+3 -3
View File
@@ -5403,7 +5403,7 @@ dependencies = [
[[package]]
name = "lancedb"
version = "0.39.0-beta.0"
version = "0.39.0-beta.1"
dependencies = [
"ahash",
"anyhow",
@@ -5491,7 +5491,7 @@ dependencies = [
[[package]]
name = "lancedb-nodejs"
version = "0.39.0-beta.0"
version = "0.39.0-beta.1"
dependencies = [
"arrow-array",
"arrow-buffer",
@@ -5516,7 +5516,7 @@ dependencies = [
[[package]]
name = "lancedb-python"
version = "0.39.0-beta.0"
version = "0.39.0-beta.1"
dependencies = [
"arrow",
"async-trait",
+1 -1
View File
@@ -14,7 +14,7 @@ Add the following dependency to your `pom.xml`:
<dependency>
<groupId>com.lancedb</groupId>
<artifactId>lancedb-core</artifactId>
<version>0.39.0-beta.0</version>
<version>0.39.0-beta.1</version>
</dependency>
```
+1 -1
View File
@@ -8,7 +8,7 @@
<parent>
<groupId>com.lancedb</groupId>
<artifactId>lancedb-parent</artifactId>
<version>0.39.0-beta.0</version>
<version>0.39.0-beta.1</version>
<relativePath>../pom.xml</relativePath>
</parent>
+1 -1
View File
@@ -6,7 +6,7 @@
<groupId>com.lancedb</groupId>
<artifactId>lancedb-parent</artifactId>
<version>0.39.0-beta.0</version>
<version>0.39.0-beta.1</version>
<packaging>pom</packaging>
<name>${project.artifactId}</name>
<description>LanceDB Java SDK Parent POM</description>
+1 -1
View File
@@ -1,7 +1,7 @@
[package]
name = "lancedb-nodejs"
edition.workspace = true
version = "0.39.0-beta.0"
version = "0.39.0-beta.1"
publish = false
license.workspace = true
description.workspace = true
+1 -1
View File
@@ -1,6 +1,6 @@
{
"name": "@lancedb/lancedb-darwin-arm64",
"version": "0.39.0-beta.0",
"version": "0.39.0-beta.1",
"os": ["darwin"],
"cpu": ["arm64"],
"main": "lancedb.darwin-arm64.node",
+1 -1
View File
@@ -1,6 +1,6 @@
{
"name": "@lancedb/lancedb-linux-arm64-gnu",
"version": "0.39.0-beta.0",
"version": "0.39.0-beta.1",
"os": ["linux"],
"cpu": ["arm64"],
"main": "lancedb.linux-arm64-gnu.node",
+1 -1
View File
@@ -1,6 +1,6 @@
{
"name": "@lancedb/lancedb-linux-arm64-musl",
"version": "0.39.0-beta.0",
"version": "0.39.0-beta.1",
"os": ["linux"],
"cpu": ["arm64"],
"main": "lancedb.linux-arm64-musl.node",
+1 -1
View File
@@ -1,6 +1,6 @@
{
"name": "@lancedb/lancedb-linux-x64-gnu",
"version": "0.39.0-beta.0",
"version": "0.39.0-beta.1",
"os": ["linux"],
"cpu": ["x64"],
"main": "lancedb.linux-x64-gnu.node",
+1 -1
View File
@@ -1,6 +1,6 @@
{
"name": "@lancedb/lancedb-linux-x64-musl",
"version": "0.39.0-beta.0",
"version": "0.39.0-beta.1",
"os": ["linux"],
"cpu": ["x64"],
"main": "lancedb.linux-x64-musl.node",
+1 -1
View File
@@ -1,6 +1,6 @@
{
"name": "@lancedb/lancedb-win32-arm64-msvc",
"version": "0.39.0-beta.0",
"version": "0.39.0-beta.1",
"os": [
"win32"
],
+1 -1
View File
@@ -1,6 +1,6 @@
{
"name": "@lancedb/lancedb-win32-x64-msvc",
"version": "0.39.0-beta.0",
"version": "0.39.0-beta.1",
"os": ["win32"],
"cpu": ["x64"],
"main": "lancedb.win32-x64-msvc.node",
+1 -1
View File
@@ -11,7 +11,7 @@
"ann"
],
"private": false,
"version": "0.39.0-beta.0",
"version": "0.39.0-beta.1",
"main": "dist/index.js",
"exports": {
".": "./dist/index.js",
+1 -1
View File
@@ -1,6 +1,6 @@
[package]
name = "lancedb-python"
version = "0.39.0-beta.0"
version = "0.39.0-beta.1"
publish = false
edition.workspace = true
description = "Python bindings for LanceDB"
+1 -1
View File
@@ -1,6 +1,6 @@
[package]
name = "lancedb"
version = "0.39.0-beta.0"
version = "0.39.0-beta.1"
edition.workspace = true
description = "LanceDB: A serverless, low-latency vector database for AI applications"
license.workspace = true
+78 -2
View File
@@ -133,7 +133,7 @@ impl NativeTable {
),
});
}
(resolved.canonical_path, resolved.field)
(resolved.canonical_path, resolved.terminal_field)
} else {
Self::resolve_index_field(dataset.schema(), &opts.columns[0])?
};
@@ -439,7 +439,8 @@ mod tests {
use arrow_array::record_batch;
use arrow_array::{
Array, ArrayRef, BinaryArray, BooleanArray, FixedSizeListArray, Float32Array, Int32Array,
LargeBinaryArray, LargeStringArray, RecordBatch, StringArray, StructArray,
LargeBinaryArray, LargeStringArray, ListArray, RecordBatch, StringArray, StructArray,
UInt32Array,
};
use arrow_data::ArrayDataBuilder;
use arrow_schema::{DataType, Field, Schema};
@@ -458,6 +459,7 @@ mod tests {
use crate::query::{ExecutableQuery, QueryBase};
use crate::table::optimize::{CompactionOptions, OptimizeAction};
use lance_index::scalar::FullTextSearchQuery;
use lance_index::scalar::inverted::query::{FtsQuery, MatchQuery};
fn create_fixed_size_list<T: Array>(
values: T,
@@ -599,6 +601,80 @@ mod tests {
assert!(invalid_granularity.is_err());
}
#[tokio::test]
async fn test_nested_list_fts_uses_deepest_document_coordinates() {
let conn = connect("memory://").execute().await.unwrap();
let mut docs = ListBuilder::new(ListBuilder::new(StringBuilder::new()));
docs.values().values().append_value("alpha");
docs.values().values().append_value("beta");
docs.values().append(true);
docs.values().values().append_value("gamma");
docs.values().values().append_value("alpha delta");
docs.values().append(true);
docs.append(true);
docs.values().append(true);
docs.values().values().append_value("alpha");
docs.values().append(true);
docs.append(true);
let batch = RecordBatch::try_from_iter(vec![
("id", Arc::new(Int32Array::from(vec![0, 1])) as ArrayRef),
("docs", Arc::new(docs.finish()) as ArrayRef),
])
.unwrap();
let table = conn.create_table("nested", batch).execute().await.unwrap();
let job = table
.create_index(
&["docs"],
Index::FTS(
FtsIndexBuilder::default()
.document_granularity(DocumentGranularity::ListElement),
),
)
.execute_async()
.await
.unwrap();
job.wait().await.unwrap();
let query = FullTextSearchQuery::new_query(FtsQuery::Match(
MatchQuery::new("alpha".to_string())
.with_column(Some("docs".to_string()))
.with_document_granularity(DocumentGranularity::ListElement),
));
let batches = table
.query()
.full_text_search(query)
.limit(10)
.execute()
.await
.unwrap()
.try_collect::<Vec<_>>()
.await
.unwrap();
let mut hits = Vec::new();
for batch in batches {
let ids = batch["id"].as_any().downcast_ref::<Int32Array>().unwrap();
let coordinates = batch["_doc_index"]
.as_any()
.downcast_ref::<ListArray>()
.unwrap();
for row in 0..batch.num_rows() {
let coordinate = coordinates.value(row);
let coordinate = coordinate.as_any().downcast_ref::<UInt32Array>().unwrap();
hits.push((ids.value(row), coordinate.values().to_vec()));
}
}
hits.sort_unstable();
assert_eq!(
hits,
vec![(0, vec![0, 0]), (0, vec![1, 1]), (1, vec![1, 0])]
);
}
/// Concurrent waiters, and a wait issued after the job settled, all
/// succeed once the build does.
#[tokio::test]
+5 -4
View File
@@ -227,7 +227,7 @@ pub(crate) fn resolve_arrow_field_path(schema: &Schema, column: &str) -> Result<
pub(crate) struct ResolvedFtsField {
pub canonical_path: String,
pub field: Field,
pub terminal_field: Field,
pub list_depth: usize,
}
@@ -309,7 +309,7 @@ pub(crate) fn resolve_lance_fts_field_path(
);
Ok(ResolvedFtsField {
canonical_path,
field: Field::from(field),
terminal_field: Field::from(terminal),
list_depth,
})
}
@@ -375,7 +375,7 @@ pub(crate) fn resolve_arrow_fts_field_path(
message: format!("Invalid schema: {}", e),
})?;
let resolved = resolve_lance_fts_field_path(&lance_schema, column)?;
Ok((resolved.canonical_path, resolved.field))
Ok((resolved.canonical_path, resolved.terminal_field))
}
pub fn supported_btree_data_type(dtype: &DataType) -> bool {
@@ -647,8 +647,9 @@ mod tests {
Field::new("docs", text_list(), true),
]);
let (path, _) = resolve_arrow_fts_field_path(&schema, "docs.content").unwrap();
let (path, field) = resolve_arrow_fts_field_path(&schema, "docs.content").unwrap();
assert_eq!(path, "docs.content");
assert_eq!(field.data_type(), &DataType::Utf8);
let lance_schema = lance_core::datatypes::Schema::try_from(&schema).unwrap();
let field_id = lance_schema