ZacharyZhang-NY
a02b555e66
Sweep every first-party crate source (1956 .rs files) to the project comment guidelines: delete redundant restatements, decorative banners, change narration, and end-of-line comments; keep and tighten the crucial ones (invariants, bug rationale, SAFETY blocks, ported-source attribution). No functional code changed. Every edit is proven comment-only against the prior tree by a comment-stripping lexer (string/char/raw-string aware) plus a separate doctest-fence check. Where removing a comment made rustfmt or clippy want to re-lay-out adjacent code, the minimal triggering comment is restored so code tokens stay byte-identical. Gates green: cargo fmt --all --check (0 diffs), cargo check and cargo clippy --workspace --all-targets (0 warnings). Adds scripts/check_codegen_comment_guidelines.py — the enforcement gate for these guidelines (flags banners, end-of-line comments, change narration, and commented-out code).
88 lines
2.8 KiB
Rust
88 lines
2.8 KiB
Rust
//! SQL schema for the memory index: `meta` key-value settings, `chunks` text
|
|
//! with blake3 content hashes, a contentless FTS5 `chunks_fts` for BM25
|
|
//! keyword search, and — only when sqlite-vec loaded — a vec0 `chunks_vec`
|
|
//! for KNN vector search.
|
|
|
|
/// Bump on a breaking change that requires dropping and recreating tables.
|
|
pub const SCHEMA_VERSION: u32 = 1;
|
|
|
|
/// `dimensions` sizes the `chunks_vec` embedding column; when `vec_available`
|
|
/// is false that table is omitted entirely.
|
|
///
|
|
/// Connection pragmas (busy_timeout, journal_mode) deliberately live on the
|
|
/// open path (`kigi_sqlite_journal::JournalMode::open`) instead of here,
|
|
/// because the journal mode depends on the database's filesystem.
|
|
pub fn schema_sql(dimensions: usize, vec_available: bool) -> String {
|
|
let mut sql = r#"
|
|
CREATE TABLE IF NOT EXISTS meta (
|
|
key TEXT PRIMARY KEY,
|
|
value TEXT NOT NULL
|
|
);
|
|
|
|
CREATE TABLE IF NOT EXISTS chunks (
|
|
rowid INTEGER PRIMARY KEY AUTOINCREMENT,
|
|
id TEXT UNIQUE NOT NULL,
|
|
path TEXT NOT NULL,
|
|
start_line INTEGER NOT NULL,
|
|
end_line INTEGER NOT NULL,
|
|
text TEXT NOT NULL,
|
|
hash TEXT NOT NULL,
|
|
source TEXT NOT NULL,
|
|
created_at INTEGER NOT NULL,
|
|
updated_at INTEGER NOT NULL,
|
|
access_count INTEGER DEFAULT 0,
|
|
last_accessed INTEGER
|
|
);
|
|
|
|
CREATE INDEX IF NOT EXISTS idx_chunks_path ON chunks(path);
|
|
CREATE INDEX IF NOT EXISTS idx_chunks_hash ON chunks(hash);
|
|
|
|
CREATE VIRTUAL TABLE IF NOT EXISTS chunks_fts USING fts5(text, content='');
|
|
|
|
INSERT OR IGNORE INTO meta(key, value) VALUES ('reindex_claim', '');
|
|
"#
|
|
.to_string();
|
|
|
|
if vec_available {
|
|
sql.push_str(&format!(
|
|
"\nCREATE VIRTUAL TABLE IF NOT EXISTS chunks_vec USING vec0(\n \
|
|
chunk_id TEXT PRIMARY KEY,\n \
|
|
embedding FLOAT[{dimensions}]\n);\n"
|
|
));
|
|
}
|
|
|
|
sql
|
|
}
|
|
|
|
pub const UPSERT_META_SQL: &str = "INSERT OR REPLACE INTO meta(key, value) VALUES (?1, ?2)";
|
|
|
|
pub const GET_META_SQL: &str = "SELECT value FROM meta WHERE key = ?1";
|
|
|
|
#[cfg(test)]
|
|
mod tests {
|
|
use super::*;
|
|
|
|
#[test]
|
|
fn test_schema_sql_without_vec() {
|
|
let sql = schema_sql(1536, false);
|
|
assert!(sql.contains("CREATE TABLE IF NOT EXISTS chunks"));
|
|
assert!(sql.contains("CREATE VIRTUAL TABLE IF NOT EXISTS chunks_fts"));
|
|
assert!(!sql.contains("chunks_vec"));
|
|
// Connection pragmas live on the open path, not in the schema batch.
|
|
assert!(!sql.contains("PRAGMA"));
|
|
}
|
|
|
|
#[test]
|
|
fn test_schema_sql_with_vec() {
|
|
let sql = schema_sql(384, true);
|
|
assert!(sql.contains("chunks_vec"));
|
|
assert!(sql.contains("FLOAT[384]"));
|
|
}
|
|
|
|
#[test]
|
|
fn test_schema_sql_different_dimensions() {
|
|
let sql = schema_sql(768, true);
|
|
assert!(sql.contains("FLOAT[768]"));
|
|
}
|
|
}
|