fix(preprocess): publish DWH artifacts atomically

This commit is contained in:
2026-07-12 05:43:57 +02:00
parent 4028ef7821
commit 24f6a5daa6
6 changed files with 438 additions and 46 deletions
+19 -3
View File
@@ -37,6 +37,11 @@ def _load_config_or_exit(config: Path):
def physical_path(cfg) -> Path:
from tht.jobs.dwh_pipeline import active_generation_dir
active = active_generation_dir(cfg.paths.artifacts.parent)
if active is not None:
return active / "physical.yaml"
return cfg.paths.artifacts / "mschema" / "physical.yaml"
@@ -44,13 +49,13 @@ def annotations_path(cfg) -> Path:
return cfg.paths.artifacts / "mschema" / "annotations.yaml"
def refresh_catalog(cfg, *, dwh=None):
def refresh_catalog(cfg, *, dwh=None, output_path: Path | None = None):
"""Run the existing catalog algorithm and persist its canonical output."""
target = dwh if dwh is not None else build_dwh(cfg)
physical = target.introspect()
_add_examples(target, physical, cfg.examples)
classify_all(physical, cfg.eligibility)
physical.to_yaml(physical_path(cfg))
physical.to_yaml(output_path or (cfg.paths.artifacts / "mschema" / "physical.yaml"))
return physical
@@ -90,7 +95,18 @@ def introspect_cmd(
)
return
try:
phys = refresh_catalog(cfg)
from tht.jobs.dwh_pipeline import active_generation_dir
if active_generation_dir(cfg.paths.artifacts.parent) is None:
phys = refresh_catalog(cfg)
else:
from tht.cli.preprocess_cmd import run_dwh_from_config
from tht.mschema.models import PhysicalSchema
report = run_dwh_from_config(config, steps=("introspect",))
if report.status != "succeeded":
raise RuntimeError("DWH preprocessing failed")
phys = PhysicalSchema.from_yaml(physical_path(cfg))
except Exception as e:
typer.secho(f"ERRORE: {e}", fg=typer.colors.RED, err=True)
raise typer.Exit(code=1)