mptyl/ThothII
1
0
Fork 0
Code Issues 25 Pull Requests Actions Packages Projects Releases 1 Wiki Activity

Chiusura fase di ristrutturazione e modularizzazione del workflow per favorire sviluppo modulare

Browse Source
This commit is contained in:
marcopan
2026-08-24 13:32:20 +02:00
parent fa2298653b
commit 6062cb010e
96 changed files with 304 additions and 259 deletions
Download Patch File Download Diff File Expand all files Collapse all files
harness/tests/test_corpus_pipeline.py
+20 -17
View File
@@ -2,11 +2,11 @@ from datetime import UTC, datetime, timedelta
import pytest
from tht.evidence.contracts import AcquiredDocument, SourceObject
from tht.evidence.corpus.chunk import ChunkPolicy
from tht.evidence.corpus.models import CanonicalChunk, CanonicalDocument, CorpusManifest
from tht.evidence.corpus.pipeline import CorpusPipeline, PipelineError, PipelineResult
from tht.evidence.corpus.store import CorpusStore
from tht.evidence.corpus.models import CanonicalChunk, CanonicalDocument, CorpusManifest
from tht.evidence.contracts import AcquiredDocument, SourceObject
from tht.ports.vector import VectorCapabilities, VectorHealth
@@ -273,6 +273,7 @@ def test_gc_preserves_vector_dependencies_of_retained_manifests(tmp_path):
def test_active_searcher_without_active_fails_closed_for_evidence(tmp_path):
from types import SimpleNamespace
from tht.evidence.search import active_searcher
class Delegate:
@@ -287,6 +288,7 @@ def test_active_searcher_without_active_fails_closed_for_evidence(tmp_path):
def test_active_searcher_splits_default_and_mixed_kinds_before_global_limit(tmp_path):
from types import SimpleNamespace
from tht.evidence.search import ActiveEvidenceSearcher
store = CorpusStore(tmp_path / "corpus")
@@ -319,6 +321,7 @@ def test_active_searcher_splits_default_and_mixed_kinds_before_global_limit(tmp_
def test_active_evidence_query_holds_lock_against_publish(tmp_path):
import threading
from types import SimpleNamespace
from tht.evidence.search import ActiveEvidenceSearcher
first_pipeline = pipeline(tmp_path, Source([(item("one", "a"), "old")]), vectors=Vectors())
@@ -519,9 +522,9 @@ def test_unchanged_job_reuses_active_generation_without_new_directory(tmp_path):
"hello",
)])
candidate = pipeline(tmp_path, source)
args = dict(workspace_id="demo", workspace_root=tmp_path,
config_fingerprint="sha256:" + "1" * 64,
input_fingerprint="sha256:" + "2" * 64)
args = {"workspace_id": "demo", "workspace_root": tmp_path,
"config_fingerprint": "sha256:" + "1" * 64,
"input_fingerprint": "sha256:" + "2" * 64}
first = candidate.run_as_job(**args)
snapshot = first.manifest.metadata["source_snapshot"]["fs:one"]
assert snapshot == {
@@ -546,9 +549,9 @@ def test_job_source_snapshot_change_forces_publish_with_same_fingerprint(tmp_pat
metadata={"media_type": "text/markdown", "size": 5, "label": "original"},
)
vectors = Vectors()
args = dict(workspace_id="demo", workspace_root=tmp_path,
config_fingerprint="sha256:" + "1" * 64,
input_fingerprint="sha256:" + "2" * 64)
args = {"workspace_id": "demo", "workspace_root": tmp_path,
"config_fingerprint": "sha256:" + "1" * 64,
"input_fingerprint": "sha256:" + "2" * 64}
first = pipeline(tmp_path, Source([(original, "hello")]), vectors=vectors).run_as_job(**args)
updates = {
"uri": "file:///safe/renamed.md",
@@ -567,9 +570,9 @@ def test_job_source_snapshot_change_forces_publish_with_same_fingerprint(tmp_pat
def test_job_binding_change_forces_publish(tmp_path, fingerprint_name):
vectors = Vectors()
source_object = item("one", "a")
args = dict(workspace_id="demo", workspace_root=tmp_path,
config_fingerprint="sha256:" + "1" * 64,
input_fingerprint="sha256:" + "2" * 64)
args = {"workspace_id": "demo", "workspace_root": tmp_path,
"config_fingerprint": "sha256:" + "1" * 64,
"input_fingerprint": "sha256:" + "2" * 64}
first = pipeline(tmp_path, Source([(source_object, "hello")]), vectors=vectors).run_as_job(**args)
args[fingerprint_name] = "sha256:" + "3" * 64
source = Source([(source_object, "hello")])
@@ -587,9 +590,9 @@ def test_job_incomplete_active_contract_never_noops(tmp_path, damage):
vectors = Vectors()
source_object = item("one", "a")
args = dict(workspace_id="demo", workspace_root=tmp_path,
config_fingerprint="sha256:" + "1" * 64,
input_fingerprint="sha256:" + "2" * 64)
args = {"workspace_id": "demo", "workspace_root": tmp_path,
"config_fingerprint": "sha256:" + "1" * 64,
"input_fingerprint": "sha256:" + "2" * 64}
candidate = pipeline(tmp_path, Source([(source_object, "hello")]), vectors=vectors)
first = candidate.run_as_job(**args)
if damage == "legacy_metadata":
@@ -628,9 +631,9 @@ def test_job_corrupt_canonical_document_or_chunk_never_noops(tmp_path, damage):
modified_at=datetime(2026, 1, 1, tzinfo=UTC),
metadata={"media_type": "text/markdown", "size": 11, "owner": "docs"},
)
args = dict(workspace_id="demo", workspace_root=tmp_path,
config_fingerprint="sha256:" + "1" * 64,
input_fingerprint="sha256:" + "2" * 64)
args = {"workspace_id": "demo", "workspace_root": tmp_path,
"config_fingerprint": "sha256:" + "1" * 64,
"input_fingerprint": "sha256:" + "2" * 64}
candidate = pipeline(
tmp_path, Source([(source_object, "hello world")]), vectors=vectors,
policy=ChunkPolicy(version="chunk-v1", max_chars=6),