marcopanandClaude Fable 5
e24b41b156
Lever 1: Join-graph via FK logics in annotations + suggest-fks command
- TableAnnotation.foreign_keys field stores curated logical FKs (DWH has no FK constraints)
- tht schema suggest-fks: mine from approved SQL, heuristics (time_key → dim_time),
same-name discovery + explicit --assume flag for multi-owner PKs
- mschema renders 【Foreign keys】 section populated; validation in merge.py
- SKILL.md F4 now reads FKs from mschema-text, no custom data_time_key logic
Lever 2: Context-pack consolidation at kickoff (tht search pack)
- Single embedding of question, reused for schema + evidence + solved searches
- One command: tht search pack <question> --session <id> → retrieval_pack.md
- Graceful degradation when Ollama/vector store unreachable (exit 0, empty sections)
- SKILL.md F1 prescribes as first call; reduces model thinking turns via pre-retrieval
Lever 3: Phase-summary recap v2 auto-construction from session ledger
- tht session show --json includes full decisions ledger
- tht phase meta --json exports 'emits' (substantive decision types per phase)
- Gate appends deterministic 【Decisioni registrate in questa fase】 section (appendLedgerSection)
- Model authors only summary + checks; recap table comes from persisted state (exact by construction)
- SKILL.md Disciplina 6: brief model output, gate fills the rest
Tests: 358 Python (including 10 FK + 3 pack + 1 session-ledger tests) + 111 JS gate tests, all pass.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
95 lines
2.4 KiB
Python
95 lines
2.4 KiB
Python
from datetime import datetime
|
|
from pathlib import Path
|
|
from typing import Self
|
|
|
|
import yaml
|
|
from pydantic import BaseModel, Field
|
|
|
|
|
|
class _YamlModel(BaseModel):
|
|
def to_yaml(self, path: Path) -> None:
|
|
path.parent.mkdir(parents=True, exist_ok=True)
|
|
data = self.model_dump(by_alias=True, mode="json", exclude_defaults=False)
|
|
path.write_text(
|
|
yaml.safe_dump(data, sort_keys=False, allow_unicode=True, width=120)
|
|
)
|
|
|
|
@classmethod
|
|
def from_yaml(cls, path: Path) -> Self:
|
|
raw = yaml.safe_load(path.read_text())
|
|
return cls.model_validate(raw)
|
|
|
|
|
|
class ColumnPhysical(BaseModel):
|
|
type: str
|
|
nullable: bool = True
|
|
pk: bool = False
|
|
default: str | None = None
|
|
comment: str = ""
|
|
examples: list[str] = []
|
|
is_enum: bool = False
|
|
eligible: bool = True
|
|
eligibility_reason: str = ""
|
|
|
|
|
|
class ForeignKey(BaseModel):
|
|
columns: list[str]
|
|
ref_table: str
|
|
ref_columns: list[str]
|
|
name: str = ""
|
|
|
|
|
|
class Index(BaseModel):
|
|
name: str
|
|
columns: list[str]
|
|
unique: bool = False
|
|
primary: bool = False
|
|
type: str = "btree"
|
|
|
|
|
|
class TablePhysical(BaseModel):
|
|
comment: str = ""
|
|
row_count: int = 0 # stima da pg_class.reltuples
|
|
columns: dict[str, ColumnPhysical] = {}
|
|
foreign_keys: list[ForeignKey] = []
|
|
indexes: list[Index] = []
|
|
|
|
|
|
class PhysicalSchema(_YamlModel):
|
|
database: str
|
|
db_schema: str = Field(alias="schema")
|
|
introspected_at: datetime
|
|
tables: dict[str, TablePhysical] = {}
|
|
|
|
model_config = {"populate_by_name": True}
|
|
|
|
|
|
class ColumnAnnotation(BaseModel):
|
|
description: str = ""
|
|
synonyms: list[str] = []
|
|
concepts: list[str] = []
|
|
evidence: list[str] = []
|
|
notes: str = ""
|
|
eligible: bool | None = None
|
|
|
|
|
|
class TableAnnotation(BaseModel):
|
|
description: str = ""
|
|
concepts: list[str] = []
|
|
notes: str = ""
|
|
columns: dict[str, ColumnAnnotation] = {}
|
|
# FK "logiche" curate a mano: il DWH non dichiara vincoli, quindi i join
|
|
# noti (es. data_time_key -> dim_time.day_key) vivono qui e vengono fusi
|
|
# con le FK fisiche in tutte le viste renderizzate.
|
|
foreign_keys: list[ForeignKey] = []
|
|
|
|
|
|
class Annotations(_YamlModel):
|
|
tables: dict[str, TableAnnotation] = {}
|
|
|
|
@classmethod
|
|
def from_yaml(cls, path: Path) -> "Annotations":
|
|
if not path.exists():
|
|
return cls()
|
|
return super().from_yaml(path)
|