5 cmd foglia portati con rename + drift fix: - memory_cmd: portato col modello registry INTATTO (TODO marker per il drop registry decisione spec 5 — task separato, richiede L2 per validare il rewrite su vectordb) - search_cmd: creata search_app sub-app (era funzione standalone in ChironeWp3), registrata come 'tht search find' - evidence_cmd, db_cmd, decision_cmd: port verbatim Drift fix decision_cmd: DECISION_MIN_PHASE.get(type,1) -> load_workflow().decision_min_phase(type). Check grep-per-file: ~15 residui nsp/PSD_SSL_CA nei messaggi utente fixati (nsp <cmd> -> tht <cmd>, nsp.yaml -> workspace yaml, PSD_SSL_CA -> THT_SSL_CA). Suite: 165 passed. tht --help ora mostra 10 sottocomandi.
77 lines
2.5 KiB
Python
77 lines
2.5 KiB
Python
from pathlib import Path
|
|
|
|
import typer
|
|
|
|
from tht.cli.config_cmd import CONFIG_OPT
|
|
from tht.cli.schema_cmd import _load_config_or_exit
|
|
from tht.cli._guards import require_vector_write_allowed
|
|
|
|
evidence_app = typer.Typer(help="Generazione e gestione delle evidence")
|
|
|
|
|
|
def evidence_root(cfg) -> Path:
|
|
return cfg.paths.artifacts / "evidence"
|
|
|
|
|
|
def _require_evidence_cfg(cfg):
|
|
if cfg.evidence is None:
|
|
typer.secho(
|
|
"ERRORE: sezione `evidence` mancante nel workspace yaml (serve `source_root`).",
|
|
fg=typer.colors.RED, err=True,
|
|
)
|
|
raise typer.Exit(code=1)
|
|
return cfg.evidence
|
|
|
|
|
|
@evidence_app.command("extract")
|
|
def extract_cmd(config: Path = CONFIG_OPT) -> None:
|
|
"""Rispecchia la cartella curata dell'ETL in artifacts/evidence/.
|
|
|
|
artifacts/evidence/ e' un artefatto derivato e rigenerabile: viene riallineato a
|
|
ogni estrazione (le evidence rimosse a monte spariscono anche qui). La gerarchia
|
|
per dominio della cartella sorgente viene preservata.
|
|
"""
|
|
import shutil
|
|
|
|
from tht.evidence.extract import load_curated
|
|
|
|
cfg = _load_config_or_exit(config)
|
|
sources = _require_evidence_cfg(cfg)
|
|
out_dir = evidence_root(cfg)
|
|
if out_dir.exists():
|
|
shutil.rmtree(out_dir)
|
|
written = 0
|
|
for rel, doc in load_curated(sources):
|
|
doc.save(out_dir / rel)
|
|
written += 1
|
|
if written == 0:
|
|
typer.secho(
|
|
f"ATTENZIONE: nessuna evidence trovata in "
|
|
f"{sources.source_root / sources.evidence_dir}",
|
|
fg=typer.colors.YELLOW, err=True,
|
|
)
|
|
raise typer.Exit(code=1)
|
|
typer.secho(f"OK: {written} evidence rispecchiate in {out_dir}", fg=typer.colors.GREEN)
|
|
|
|
|
|
@evidence_app.command("index")
|
|
def index_cmd(config: Path = CONFIG_OPT) -> None:
|
|
"""Embedda e sincronizza su pgvector tutte le evidence presenti in artifacts/."""
|
|
from tht.cli.vector_cmd import (
|
|
_print_stats,
|
|
make_embedder,
|
|
open_store,
|
|
require_vector_cfg,
|
|
)
|
|
from tht.evidence.model import load_evidence_dir
|
|
from tht.vectorstore.records import evidence_records
|
|
|
|
cfg = _load_config_or_exit(config)
|
|
require_vector_write_allowed(cfg, "evidence index")
|
|
require_vector_cfg(cfg)
|
|
docs = load_evidence_dir(evidence_root(cfg))
|
|
records = evidence_records(docs, cfg.vector.max_chunk_chars)
|
|
store = open_store(cfg, "evidence")
|
|
stats = store.sync(records, make_embedder(cfg.embeddings), kinds={"evidence"})
|
|
_print_stats(stats)
|