Files
kebab/crates/kebab-store-sqlite/tests/incremental_ingest.rs
altair823 58ac62d53a feat(search): provenance 출처 필터 — [[workspace.sources]] 멀티소스 + --source/--source-type
혼합 출처 KB(위키+jira 등)에서 색인은 전부 하되 질의 시 출처로 좁히는 provenance
레버. 전역 trust 곱셈가중(weighted-RRF)은 A/B 에서 반증(θ=0.85 만으로 incident MRR
0.918→0.340 절벽, 점수 압축) — 필터가 see-saw 없는 올바른 레버.

- config [[workspace.sources]] (각 id/root/exclude/trust_level/source_type);
  단일 root 는 implicit `default` source 로 정규화. validate: id 유일·비어있지 않음.
- config schema v3→v4 (step_3_to_4, root→[[workspace.sources]] id=default 미러, 멱등)
- V014 documents.source_id 컬럼+인덱스 (additive, DEFAULT 'default', 재색인 0)
- Metadata.source_id + BodyHints trust precedence(frontmatter > source 기본값 > Primary)
- ingest: --root 미지정 시 resolved_sources() 순회 + doc 마다 source_id/trust stamp
- 검색 SearchFilters.source_type/source_id → lexical + vector 두 site (IN, OR)
- CLI kebab search --source <id> / --source-type <type> (repeatable/comma-sep)

도그푸딩(620 doc, jira400+wiki220): --source wiki 로 개념 질의 MRR 0.780→0.810,
--source jira 로 incident 0.918→0.975. trust precedence 실측(jira=secondary 기본값).

version bump 0.28.0 → 0.29.0 (신규 CLI flag + config 키 + V014 migration → minor).
follow-up: MCP search 필터 미노출 · kebab list source_id 미표시 · RAG provenance 라벨.

자세한 내용: tasks/HOTFIXES.md (2026-06-21), docs/release-notes/v0.29.0-draft.md.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_012Mc6W1fgsrbFKTsqA6P8La
2026-06-21 08:35:19 +00:00

154 lines
5.0 KiB
Rust

//! Round-trip tests for `last_chunker_version` / `last_embedding_version`
//! columns added by the V006 migration (p9-fb-23 task 3).
use std::path::PathBuf;
use kebab_core::{
AssetId, AssetStorage, Block, CanonicalDocument, Checksum, ChunkerVersion, CommonBlock,
DocumentId, DocumentStore, EmbeddingVersion, HeadingBlock, Lang, MediaType, Metadata,
ParserVersion, Provenance, RawAsset, SourceSpan, SourceType, SourceUri, TrustLevel,
WorkspacePath,
};
use kebab_store_sqlite::SqliteStore;
use time::OffsetDateTime;
mod common;
fn make_asset() -> RawAsset {
let bytes = b"incremental-ingest-test";
RawAsset {
asset_id: AssetId("f".repeat(32)),
source_uri: SourceUri::File(PathBuf::from("/tmp/inc.md")),
workspace_path: WorkspacePath::new("notes/inc.md".into()).unwrap(),
media_type: MediaType::Markdown,
byte_len: bytes.len() as u64,
checksum: Checksum(blake3::hash(bytes).to_hex().to_string()),
discovered_at: OffsetDateTime::from_unix_timestamp(1_700_000_000).unwrap(),
stored: AssetStorage::Reference {
path: PathBuf::from("/tmp/inc.md"),
sha: Checksum(blake3::hash(bytes).to_hex().to_string()),
},
}
}
fn make_doc() -> CanonicalDocument {
let doc_id = DocumentId("d".repeat(32));
let block = Block::Heading(HeadingBlock {
common: CommonBlock {
block_id: kebab_core::BlockId("b".repeat(32)),
heading_path: vec![],
source_span: SourceSpan::Line { start: 1, end: 1 },
},
level: 1,
text: "Incremental Title".into(),
});
let metadata = Metadata {
aliases: vec![],
tags: vec![],
created_at: OffsetDateTime::from_unix_timestamp(1_700_000_000).unwrap(),
updated_at: OffsetDateTime::from_unix_timestamp(1_700_000_000).unwrap(),
source_type: SourceType::Markdown,
trust_level: TrustLevel::Primary,
user_id_alias: None,
user: Default::default(),
repo: None,
git_branch: None,
git_commit: None,
code_lang: None,
source_id: None,
};
CanonicalDocument {
doc_id,
source_asset_id: AssetId("f".repeat(32)),
workspace_path: WorkspacePath::new("notes/inc.md".into()).unwrap(),
title: "Incremental Title".into(),
lang: Lang("en".into()),
blocks: vec![block],
metadata,
provenance: Provenance { events: vec![] },
parser_version: ParserVersion("test-parser".into()),
schema_version: 1,
doc_version: 1,
last_chunker_version: None,
last_embedding_version: None,
}
}
#[test]
fn put_then_get_document_roundtrips_version_stamps() {
let env = common::TestEnv::new();
let store = SqliteStore::open(&env.config()).unwrap();
store.run_migrations().unwrap();
let asset = make_asset();
store.put_asset(&asset).unwrap();
let mut doc = make_doc();
doc.last_chunker_version = Some(ChunkerVersion("md-heading-v1".into()));
doc.last_embedding_version = Some(EmbeddingVersion("multilingual-e5-small@v1".into()));
store.put_document(&doc).unwrap();
let loaded = store
.get_document(&doc.doc_id)
.unwrap()
.expect("doc round-trips");
assert_eq!(loaded.last_chunker_version, doc.last_chunker_version);
assert_eq!(loaded.last_embedding_version, doc.last_embedding_version);
}
#[test]
fn put_then_get_document_roundtrips_none_stamps() {
let env = common::TestEnv::new();
let store = SqliteStore::open(&env.config()).unwrap();
store.run_migrations().unwrap();
let asset = make_asset();
store.put_asset(&asset).unwrap();
let doc = make_doc(); // both version stamps are None by default
store.put_document(&doc).unwrap();
let loaded = store
.get_document(&doc.doc_id)
.unwrap()
.expect("doc round-trips");
assert!(
loaded.last_chunker_version.is_none(),
"last_chunker_version must be None when not set"
);
assert!(
loaded.last_embedding_version.is_none(),
"last_embedding_version must be None when not set"
);
}
#[test]
fn get_asset_by_workspace_path_roundtrips() {
let env = common::TestEnv::new();
let store = SqliteStore::open(&env.config()).unwrap();
store.run_migrations().unwrap();
let asset = make_asset();
store.put_asset(&asset).unwrap();
let loaded = store
.get_asset_by_workspace_path(&asset.workspace_path)
.unwrap()
.expect("asset must round-trip");
assert_eq!(loaded.asset_id, asset.asset_id);
assert_eq!(loaded.checksum, asset.checksum);
assert_eq!(loaded.byte_len, asset.byte_len);
}
#[test]
fn get_asset_by_workspace_path_returns_none_for_unknown() {
let env = common::TestEnv::new();
let store = SqliteStore::open(&env.config()).unwrap();
store.run_migrations().unwrap();
let path = WorkspacePath::new("notes/missing.md".into()).unwrap();
assert!(store.get_asset_by_workspace_path(&path).unwrap().is_none());
}