SQL Text Search

sql_text_search retrieves chunk text written by PersistChunks into SQLStore.

It is SQL text retrieval, not a dedicated full-text search index. For BM25, phrase query, analyzers, or Elasticsearch/OpenSearch-style indexing, use IndexFullText and full_text_search.

Required Asset

PersistChunks declares:

SearchAsset(
    kind="chunk_text_index",
    name="chunks",
    store="stores.sql",
    metadata={
        "table": "chunks",
        "dialect": "generic",
        "text_field": "content_text",
    },
)

When this asset exists in the latest run record, the default query registry enables:

sql_text_search

Usage

response = await kb.query(
    "flight control fault",
    mode="sql_text_search",
    top_k=5,
)

for result in response.results:
    print(result.score, result.text)

Each QueryResult represents one chunk and includes:

id
text
score
source
metadata

source includes:

document_id
source_key
source_chunk
page_index
chunk_index
token_start
token_end

SQL Strategy

generic uses:

LOWER(content_text) LIKE LOWER(:pattern)

It works across SQLite, MySQL, and minimal SQL environments. Recall is conservative but portable.

postgresql uses:

content_tsv @@ plainto_tsquery('simple', :query)
ts_rank(content_tsv, plainto_tsquery('simple', :query))

PersistChunks(dialect="postgresql") creates content_tsv and a GIN index.

Scope

sql_text_search only performs SQL text recall. It does not embed, rerank, rewrite queries, fuse hybrid results, or generate answers.

sql_text_search and full_text_search are parallel capabilities. They do not depend on each other and do not automatically fall back to each other.