feat(harness): port sql/cte/datamart/lsh cmd (Onda 4) — CLI completa F1→F8

Ultima onda CLI. 4 cmd portati con rename + grep-per-file (3 residui nsp nei messaggi fixati). Nessun drift costanti phase in questi cmd. La CLI tht e' ora COMPLETA: 14 gruppi di comandi (phase config schema session vector memory search evidence db decision sql cte datamart lsh). tht --help li list tutti. Suite: 165 passed. Il loop skill->LLM->gate ora ha tutti i comandi che la skill chiamera'. Resta: skill riscritta (S), setup pre-sessione (0b), sessione L2 manuale.
2026-06-27 14:16:35 +02:00
parent 99b01b0407
commit 91a374492c
5 changed files with 588 additions and 0 deletions
@@ -0,0 +1,97 @@
+from pathlib import Path
+
+import typer
+
+from tht.cli.config_cmd import CONFIG_OPT
+from tht.cli.schema_cmd import _load_config_or_exit, physical_path
+
+lsh_app = typer.Typer(help="Indice LSH su valori dei campi (derivato, rigenerabile)")
+
+
+def _lsh_dir(cfg) -> Path:
+    return cfg.paths.indexes / "lsh"
+
+
+@lsh_app.command("build")
+def build_cmd(config: Path = CONFIG_OPT) -> None:
+    """Costruisce l'indice LSH dai valori del database e lo salva su pickle."""
+    from tht.lshindex import build_index, save_index
+    from tht.mschema.models import Annotations, PhysicalSchema
+
+    cfg = _load_config_or_exit(config)
+    phys_file = physical_path(cfg)
+    if not phys_file.exists():
+        typer.secho(
+            f"ERRORE: {phys_file} non trovato. Esegui prima `tht schema introspect`.",
+            fg=typer.colors.RED, err=True,
+        )
+        raise typer.Exit(code=1)
+    physical = PhysicalSchema.from_yaml(phys_file)
+    from tht.cli.schema_cmd import annotations_path
+
+    annotations = Annotations.from_yaml(annotations_path(cfg))
+
+    typer.echo("Estrazione valori (i più frequenti) dalle colonne testuali eligible...")
+    if cfg.database.transport == "rest":
+        from tht.db.sampling import unique_values_for_lsh_rest
+        from tht.rest.client import RestClient
+
+        values, skipped, truncated = unique_values_for_lsh_rest(
+            RestClient(cfg.rest), physical, cfg.lsh, annotations
+        )
+    else:
+        from tht.db.connection import make_engine
+        from tht.db.sampling import unique_values_for_lsh
+
+        values, skipped, truncated = unique_values_for_lsh(
+            make_engine(cfg.database), physical, cfg.lsh, annotations
+        )
+    n_values = sum(len(v) for t in values.values() for v in t.values())
+    typer.echo(f"  {n_values} valori da {sum(len(t) for t in values.values())} colonne")
+    for s in skipped:
+        typer.secho(f"  saltata {s.table}.{s.column}: {s.reason}", fg=typer.colors.YELLOW)
+    for t in truncated:
+        typer.secho(
+            f"  troncata {t.table}.{t.column}: indicizzati i {t.indexed} valori più frequenti "
+            f"(limite max_values_per_column raggiunto; altri valori distinti NON indicizzati)",
+            fg=typer.colors.YELLOW,
+        )
+
+    lsh, minhashes = build_index(values, cfg.lsh, verbose=True)
+    save_index(lsh, minhashes, cfg.lsh, _lsh_dir(cfg), name=cfg.database.db_schema)
+    typer.secho(
+        f"OK: indice LSH ({len(minhashes)} entry) -> {_lsh_dir(cfg)}", fg=typer.colors.GREEN
+    )
+
+
+@lsh_app.command("query")
+def query_cmd(
+    keyword: str = typer.Argument(..., help="Termine da cercare, es. 'ablazione'."),
+    config: Path = CONFIG_OPT,
+    top: int = typer.Option(10, "--top", help="Numero massimo di risultati."),
+) -> None:
+    """Probe visuale: mostra i candidati LSH per un termine, con score Jaccard."""
+    from rich.console import Console
+    from rich.table import Table
+
+    from tht.lshindex import LshIndexError, load_index, query_index
+
+    cfg = _load_config_or_exit(config)
+    try:
+        lsh, minhashes, meta = load_index(_lsh_dir(cfg), name=cfg.database.db_schema)
+    except LshIndexError as e:
+        typer.secho(f"ERRORE: {e}", fg=typer.colors.RED)
+        raise typer.Exit(code=1)
+
+    hits = query_index(lsh, minhashes, keyword, meta, top_n=top)
+    if not hits:
+        typer.secho(f"Nessun candidato LSH per '{keyword}'.", fg=typer.colors.YELLOW)
+        return
+    table = Table(title=f"Candidati LSH per '{keyword}' ({len(hits)})")
+    table.add_column("Tabella")
+    table.add_column("Colonna")
+    table.add_column("Valore")
+    table.add_column("Score", justify="right")
+    for h in hits:
+        table.add_row(h.table, h.column, h.value, f"{h.score:.3f}")
+    Console().print(table)