feat(memory/vector): 向量检索引擎(SQLite BLOB 内积默认 + PG pgvector 真语义增强,配置后真生效)
This commit is contained in:
1 parent
ef79edd96a
commit
fb880572a4
6 files changed
+1009
No files matched your search
@@ -0,0 +1,164 @@
|
||||
package vector
|
||||
|
||||
import (
|
||||
"context"
|
||||
"database/sql"
|
||||
"errors"
|
||||
"io"
|
||||
"log/slog"
|
||||
"testing"
|
||||
|
||||
"git.yeij.top/AskaEth/Cyrene_v2/internal/config"
|
||||
"git.yeij.top/AskaEth/Cyrene_v2/internal/db"
|
||||
"git.yeij.top/AskaEth/Cyrene_v2/internal/runtime/embedding"
|
||||
)
|
||||
|
||||
// testLogger returns a logger that writes nowhere, keeping test output clean.
|
||||
func testLogger() *slog.Logger {
|
||||
return slog.New(slog.NewTextHandler(io.Discard, nil))
|
||||
}
|
||||
|
||||
// openTestDB opens an isolated in-memory SQLite DB. Each call owns a private DB,
|
||||
// so parallel tests are isolated by construction. The returned DB is closed by
|
||||
// t.Cleanup.
|
||||
func openTestDB(t *testing.T) *sql.DB {
|
||||
t.Helper()
|
||||
ctx := context.Background()
|
||||
d, err := db.Open(ctx, ":memory:", testLogger())
|
||||
if err != nil {
|
||||
t.Fatalf("open in-memory db: %v", err)
|
||||
}
|
||||
t.Cleanup(func() { _ = d.Close() })
|
||||
return d
|
||||
}
|
||||
|
||||
// newInMemorySQLite builds a SQLiteIndex on a fresh in-memory DB with dim
|
||||
// dimensions. The embedder is left unwired so tests control it.
|
||||
func newInMemorySQLite(t *testing.T, dim int) *SQLiteIndex {
|
||||
t.Helper()
|
||||
ctx := context.Background()
|
||||
idx, err := NewSQLite(ctx, openTestDB(t), dim, testLogger())
|
||||
if err != nil {
|
||||
t.Fatalf("NewSQLite: %v", err)
|
||||
}
|
||||
return idx
|
||||
}
|
||||
|
||||
// mockEmbedder is a scriptable embedding.Embedder for tests. With no fn it
|
||||
// returns one-hot vectors, one per input text, so tests can predict results.
|
||||
type mockEmbedder struct {
|
||||
dim int
|
||||
fn func(ctx context.Context, texts []string) ([]embedding.Vector, error)
|
||||
}
|
||||
|
||||
func (m *mockEmbedder) Dim() int { return m.dim }
|
||||
|
||||
func (m *mockEmbedder) Embed(ctx context.Context, texts []string) ([]embedding.Vector, error) {
|
||||
if err := ctx.Err(); err != nil {
|
||||
return nil, err
|
||||
}
|
||||
if m.fn != nil {
|
||||
return m.fn(ctx, texts)
|
||||
}
|
||||
out := make([]embedding.Vector, len(texts))
|
||||
for i := range texts {
|
||||
v := make(embedding.Vector, m.dim)
|
||||
v[i%m.dim] = 1
|
||||
out[i] = v
|
||||
}
|
||||
return out, nil
|
||||
}
|
||||
|
||||
func TestNew_NilEmbedder_异常(t *testing.T) {
|
||||
t.Parallel()
|
||||
if _, err := New(config.DBConfig{}, nil, testLogger()); !errors.Is(err, ErrNoEmbedder) {
|
||||
t.Fatalf("New(nil embedder) err = %v, want ErrNoEmbedder", err)
|
||||
}
|
||||
}
|
||||
|
||||
func TestNew_InvalidDim_异常(t *testing.T) {
|
||||
t.Parallel()
|
||||
emb := &mockEmbedder{dim: 0}
|
||||
if _, err := New(config.DBConfig{SQLitePath: ":memory:"}, emb, testLogger()); !errors.Is(err, ErrInvalidDim) {
|
||||
t.Fatalf("New(dim 0) err = %v, want ErrInvalidDim", err)
|
||||
}
|
||||
}
|
||||
|
||||
func TestNew_SelectsSQLite_WhenPGURLBlank(t *testing.T) {
|
||||
t.Parallel()
|
||||
ctx := context.Background()
|
||||
idx, err := New(config.DBConfig{SQLitePath: ":memory:"}, &mockEmbedder{dim: 4}, testLogger())
|
||||
if err != nil {
|
||||
t.Fatalf("New: %v", err)
|
||||
}
|
||||
if _, ok := idx.(*SQLiteIndex); !ok {
|
||||
t.Fatalf("New type = %T, want *SQLiteIndex", idx)
|
||||
}
|
||||
// New wires the embedder, so Embed works on the returned index.
|
||||
vs, err := idx.Embed(ctx, []string{"hello"})
|
||||
if err != nil {
|
||||
t.Fatalf("Embed: %v", err)
|
||||
}
|
||||
if len(vs) != 1 || len(vs[0]) != 4 {
|
||||
t.Fatalf("Embed => %v, want one 4-dim vector", vs)
|
||||
}
|
||||
if err := idx.Close(); err != nil {
|
||||
t.Fatalf("Close: %v", err)
|
||||
}
|
||||
}
|
||||
|
||||
func TestNew_PGURLWhitespace_SelectsSQLite(t *testing.T) {
|
||||
t.Parallel()
|
||||
idx, err := New(config.DBConfig{SQLitePath: ":memory:", PGURL: " \t "}, &mockEmbedder{dim: 4}, testLogger())
|
||||
if err != nil {
|
||||
t.Fatalf("New: %v", err)
|
||||
}
|
||||
if _, ok := idx.(*SQLiteIndex); !ok {
|
||||
t.Fatalf("New type = %T, want *SQLiteIndex for whitespace PGURL", idx)
|
||||
}
|
||||
if err := idx.Close(); err != nil {
|
||||
t.Fatalf("Close: %v", err)
|
||||
}
|
||||
}
|
||||
|
||||
func TestNew_PGSelection_AttemptsPGWhenURLSet(t *testing.T) {
|
||||
t.Parallel()
|
||||
// A loopback URL on a closed port fails fast, proving New chose the PG path
|
||||
// rather than silently falling back to SQLite. Port 1 (tcpmux) does not
|
||||
// listen on localhost, so the connection is refused immediately.
|
||||
idx, err := New(config.DBConfig{
|
||||
SQLitePath: ":memory:",
|
||||
PGURL: "postgres://127.0.0.1:1/mem?sslmode=disable&connect_timeout=1",
|
||||
}, &mockEmbedder{dim: 4}, testLogger())
|
||||
if err == nil {
|
||||
if idx != nil {
|
||||
_ = idx.Close()
|
||||
}
|
||||
t.Fatal("New with unreachable PGURL: expected error, got nil")
|
||||
}
|
||||
if idx != nil {
|
||||
t.Fatalf("New returned non-nil index %T on error", idx)
|
||||
}
|
||||
}
|
||||
|
||||
func TestUsePG(t *testing.T) {
|
||||
t.Parallel()
|
||||
tests := []struct {
|
||||
name string
|
||||
url string
|
||||
want bool
|
||||
}{
|
||||
{name: "empty", url: "", want: false},
|
||||
{name: "whitespace", url: " \t ", want: false},
|
||||
{name: "set", url: "postgres://user:pass@host:5432/db", want: true},
|
||||
}
|
||||
for _, tt := range tests {
|
||||
tt := tt
|
||||
t.Run(tt.name, func(t *testing.T) {
|
||||
t.Parallel()
|
||||
if got := usePG(tt.url); got != tt.want {
|
||||
t.Fatalf("usePG(%q) = %v, want %v", tt.url, got, tt.want)
|
||||
}
|
||||
})
|
||||
}
|
||||
}
|
||||
Reference in new issue
Block a user