331 lines
11 KiB
Python
331 lines
11 KiB
Python
"""Tests du runner G2 : process_fn injectée, vrais fichiers tmp, aucun moteur réel."""
|
|
|
|
from __future__ import annotations
|
|
|
|
import threading
|
|
from pathlib import Path
|
|
|
|
import pytest
|
|
|
|
from gui_v6.processing_runner import (
|
|
ProcessingRunner,
|
|
RunSummary,
|
|
default_output_dir,
|
|
discover_documents,
|
|
)
|
|
|
|
_EXTS = (".pdf", ".txt")
|
|
|
|
|
|
def _touch(path: Path) -> Path:
|
|
path.parent.mkdir(parents=True, exist_ok=True)
|
|
path.write_text("x", encoding="utf-8")
|
|
return path
|
|
|
|
|
|
# -- découverte & chemins --------------------------------------------------
|
|
|
|
def test_discover_single_file(tmp_path):
|
|
f = _touch(tmp_path / "doc.pdf")
|
|
assert discover_documents(f, _EXTS) == [f]
|
|
|
|
|
|
def test_discover_single_file_unsupported(tmp_path):
|
|
f = _touch(tmp_path / "doc.xyz")
|
|
assert discover_documents(f, _EXTS) == []
|
|
|
|
|
|
def test_discover_folder_sorted_and_skips_output(tmp_path):
|
|
_touch(tmp_path / "b.pdf")
|
|
_touch(tmp_path / "a.pdf")
|
|
_touch(tmp_path / "note.txt")
|
|
_touch(tmp_path / "anonymise" / "already.pdf") # sous-arbre de sortie ignoré
|
|
found = discover_documents(tmp_path, _EXTS)
|
|
names = [p.name for p in found]
|
|
assert names == ["a.pdf", "b.pdf", "note.txt"]
|
|
|
|
|
|
def test_default_output_dir_file_and_dir(tmp_path):
|
|
f = _touch(tmp_path / "doc.pdf")
|
|
assert default_output_dir(f) == tmp_path / "anonymise"
|
|
assert default_output_dir(tmp_path) == tmp_path / "anonymise"
|
|
|
|
|
|
# -- exécution -------------------------------------------------------------
|
|
|
|
def test_run_processes_all_docs(tmp_path):
|
|
_touch(tmp_path / "a.pdf")
|
|
_touch(tmp_path / "b.pdf")
|
|
calls = []
|
|
runner = ProcessingRunner(process_fn=lambda d, o: calls.append((d, o)) or {}, extensions=_EXTS)
|
|
|
|
summary = runner.run(tmp_path)
|
|
|
|
assert isinstance(summary, RunSummary)
|
|
assert summary.total == 2
|
|
assert summary.succeeded == 2
|
|
assert summary.failed == 0
|
|
assert summary.ok is True
|
|
assert len(calls) == 2
|
|
# Le dossier de sortie par défaut a été créé.
|
|
assert (tmp_path / "anonymise").is_dir()
|
|
|
|
|
|
def test_run_single_file_uses_output_dir(tmp_path):
|
|
f = _touch(tmp_path / "doc.pdf")
|
|
out = tmp_path / "sortie"
|
|
seen = {}
|
|
runner = ProcessingRunner(process_fn=lambda d, o: seen.update(doc=d, out=o) or {}, extensions=_EXTS)
|
|
|
|
summary = runner.run(f, output_dir=out)
|
|
|
|
assert summary.total == 1 and summary.succeeded == 1
|
|
assert seen["doc"] == f
|
|
assert seen["out"] == out
|
|
assert out.is_dir()
|
|
|
|
|
|
def test_run_continues_after_failure(tmp_path):
|
|
_touch(tmp_path / "a.pdf")
|
|
_touch(tmp_path / "boom.pdf")
|
|
_touch(tmp_path / "c.pdf")
|
|
|
|
def proc(doc, out):
|
|
if doc.name == "boom.pdf":
|
|
raise RuntimeError("explosion")
|
|
return {}
|
|
|
|
runner = ProcessingRunner(process_fn=proc, extensions=_EXTS)
|
|
summary = runner.run(tmp_path)
|
|
|
|
assert summary.total == 3
|
|
assert summary.succeeded == 2
|
|
assert summary.failed == 1
|
|
assert summary.ok is False
|
|
assert summary.errors[0][0] == "boom.pdf"
|
|
assert "explosion" in summary.errors[0][1]
|
|
|
|
|
|
def test_run_marks_quarantined_engine_result_as_failure(tmp_path):
|
|
f = _touch(tmp_path / "scan.pdf")
|
|
logs = []
|
|
|
|
def proc(doc, out):
|
|
return {"status": "quarantined", "reason": "preflight_text_too_short"}
|
|
|
|
runner = ProcessingRunner(process_fn=proc, extensions=_EXTS)
|
|
summary = runner.run(f, on_log=logs.append)
|
|
|
|
assert summary.succeeded == 0
|
|
assert summary.failed == 1
|
|
assert summary.ok is False
|
|
assert summary.documents[0].status == "failed"
|
|
assert "preflight_text_too_short" in summary.errors[0][1]
|
|
assert any("ÉCHEC : scan.pdf" in item for item in logs)
|
|
|
|
|
|
def test_run_marks_missing_pdf_output_as_failure(tmp_path):
|
|
f = _touch(tmp_path / "doc.pdf")
|
|
out = tmp_path / "sortie"
|
|
|
|
def proc(doc, out_dir):
|
|
txt = out_dir / "doc.pseudonymise.txt"
|
|
audit = out_dir / "doc.audit.jsonl"
|
|
txt.write_text("ok", encoding="utf-8")
|
|
audit.write_text("{}", encoding="utf-8")
|
|
return {"text": str(txt), "audit": str(audit)}
|
|
|
|
runner = ProcessingRunner(process_fn=proc, extensions=_EXTS)
|
|
summary = runner.run(f, output_dir=out)
|
|
|
|
assert summary.succeeded == 0
|
|
assert summary.failed == 1
|
|
assert summary.documents[0].status == "failed"
|
|
assert "Aucune sortie PDF" in summary.errors[0][1]
|
|
|
|
|
|
def test_run_accepts_existing_pdf_output(tmp_path):
|
|
f = _touch(tmp_path / "doc.pdf")
|
|
out = tmp_path / "sortie"
|
|
|
|
def proc(doc, out_dir):
|
|
pdf = out_dir / "doc.redacted_raster.pdf"
|
|
pdf.write_bytes(b"%PDF-1.4\n")
|
|
return {"pdf_raster": str(pdf)}
|
|
|
|
runner = ProcessingRunner(process_fn=proc, extensions=_EXTS)
|
|
summary = runner.run(f, output_dir=out)
|
|
|
|
assert summary.succeeded == 1
|
|
assert summary.failed == 0
|
|
assert summary.documents[0].status == "success"
|
|
|
|
|
|
def test_run_empty_folder(tmp_path):
|
|
logs = []
|
|
runner = ProcessingRunner(process_fn=lambda d, o: {}, extensions=_EXTS)
|
|
summary = runner.run(tmp_path, on_log=logs.append)
|
|
assert summary.total == 0
|
|
assert any("Aucun document" in m for m in logs)
|
|
|
|
|
|
def test_stop_event_interrupts_between_docs(tmp_path):
|
|
for name in ("a.pdf", "b.pdf", "c.pdf"):
|
|
_touch(tmp_path / name)
|
|
stop = threading.Event()
|
|
processed = []
|
|
|
|
def proc(doc, out):
|
|
processed.append(doc.name)
|
|
stop.set() # demande l'arrêt après le 1er document
|
|
return {}
|
|
|
|
runner = ProcessingRunner(process_fn=proc, extensions=_EXTS)
|
|
summary = runner.run(tmp_path, stop_event=stop)
|
|
|
|
assert summary.stopped is True
|
|
assert summary.succeeded == 1
|
|
assert len(processed) == 1 # arrêt effectif entre deux documents
|
|
|
|
|
|
def test_progress_callbacks(tmp_path):
|
|
_touch(tmp_path / "a.pdf")
|
|
_touch(tmp_path / "b.pdf")
|
|
events = []
|
|
runner = ProcessingRunner(process_fn=lambda d, o: {}, extensions=_EXTS)
|
|
runner.run(tmp_path, on_progress=lambda done, total, name: events.append((done, total)))
|
|
assert (2, 2) in events # progression finale atteinte
|
|
|
|
|
|
def test_run_fails_fast_when_output_not_writable(tmp_path, monkeypatch):
|
|
from gui_v6.processing_runner import ProcessingRunner, OutputNotWritableError
|
|
src = tmp_path / "in"
|
|
src.mkdir()
|
|
(src / "a.txt").write_text("x", encoding="utf-8")
|
|
out = tmp_path / "ro"
|
|
out.mkdir()
|
|
|
|
def boom(*a, **k):
|
|
raise PermissionError("read-only")
|
|
|
|
monkeypatch.setattr("gui_v6.processing_runner.Path.mkdir", boom)
|
|
runner = ProcessingRunner(process_fn=lambda d, o: {})
|
|
with pytest.raises(OutputNotWritableError):
|
|
runner.run(src, out)
|
|
|
|
|
|
def test_no_double_run(tmp_path):
|
|
_touch(tmp_path / "a.pdf")
|
|
started = threading.Event()
|
|
release = threading.Event()
|
|
result = {}
|
|
|
|
def proc(doc, out):
|
|
started.set()
|
|
release.wait(timeout=2)
|
|
return {}
|
|
|
|
runner = ProcessingRunner(process_fn=proc, extensions=_EXTS)
|
|
worker = threading.Thread(target=lambda: runner.run(tmp_path))
|
|
worker.start()
|
|
assert started.wait(timeout=2)
|
|
# Pendant le run, un second lancement est refusé.
|
|
with pytest.raises(RuntimeError):
|
|
runner.run(tmp_path)
|
|
release.set()
|
|
worker.join(timeout=2)
|
|
assert runner.is_running is False
|
|
|
|
|
|
# -- détails par document (télémétrie) -------------------------------------
|
|
|
|
def test_run_records_per_document_details(tmp_path):
|
|
_touch(tmp_path / "a.pdf")
|
|
_touch(tmp_path / "b.pdf")
|
|
|
|
def fake(doc, out):
|
|
if doc.name == "b.pdf":
|
|
raise RuntimeError("boom")
|
|
return {}
|
|
|
|
runner = ProcessingRunner(process_fn=fake, extensions=_EXTS)
|
|
summary = runner.run(tmp_path)
|
|
|
|
assert len(summary.documents) == 2
|
|
statuses = {doc.ordinal: doc.status for doc in summary.documents}
|
|
assert statuses == {0: "success", 1: "failed"}
|
|
for doc in summary.documents:
|
|
assert doc.extension == "pdf"
|
|
assert isinstance(doc.duration_ms, int)
|
|
# RGPD : aucun nom/chemin de fichier dans les détails
|
|
assert not hasattr(doc, "path")
|
|
assert not hasattr(doc, "filename")
|
|
assert not hasattr(doc, "name")
|
|
|
|
|
|
# -- diagnostics d'erreur RGPD-safe (E2) -----------------------------------
|
|
|
|
def test_failed_doc_carries_rgpd_safe_error_fields(tmp_path):
|
|
from gui_v6.processing_runner import ProcessingRunner
|
|
|
|
secret = "Dupont Jean 1980" # simulacre de PII dans un message d'exception
|
|
|
|
def boom(_inp, _out):
|
|
raise ValueError(f"échec sur patient {secret}")
|
|
|
|
inp = tmp_path / "in"; inp.mkdir()
|
|
(inp / "a.pdf").write_bytes(b"%PDF-1.4\n")
|
|
out = tmp_path / "out"; out.mkdir()
|
|
runner = ProcessingRunner(process_fn=boom)
|
|
summary = runner.run(inp, out)
|
|
|
|
assert summary.failed == 1
|
|
doc = summary.documents[0]
|
|
assert doc.error_type == "ValueError"
|
|
assert doc.error_code in {"ner_unavailable", "quarantined", "no_output", "processing_error"}
|
|
blob = repr(vars(doc)).lower()
|
|
assert "dupont" not in blob and "patient" not in blob and secret.lower() not in blob
|
|
|
|
|
|
def test_success_doc_has_no_error_fields(tmp_path):
|
|
from gui_v6.processing_runner import ProcessingRunner
|
|
|
|
def ok(_inp, out_dir):
|
|
# process_fn reçoit le DOSSIER de sortie : on y écrit un PDF livrable.
|
|
pdf = out_dir / "a.redacted_raster.pdf"
|
|
pdf.write_bytes(b"%PDF-1.4\n")
|
|
return {"status": "ok", "pdf_raster": str(pdf)}
|
|
|
|
inp = tmp_path / "in"; inp.mkdir()
|
|
(inp / "a.pdf").write_bytes(b"%PDF-1.4\n")
|
|
out = tmp_path / "out"; out.mkdir()
|
|
summary = ProcessingRunner(process_fn=ok).run(inp, out)
|
|
doc = summary.documents[0]
|
|
assert doc.status == "success"
|
|
assert doc.error_type is None and doc.error_code is None
|
|
|
|
|
|
# -- classification d'erreur : une assertion par branche (mapping vérifié) -
|
|
|
|
def test_classify_error_code_ner_unavailable():
|
|
from gui_v6.processing_runner import classify_error_code
|
|
from gui_v6.engine_bridge import EngineUnavailableError # import the REAL class
|
|
# importing the real class means a future rename breaks this test (intended guard)
|
|
assert classify_error_code(EngineUnavailableError("modèle indispo")) == "ner_unavailable"
|
|
|
|
|
|
def test_classify_error_code_quarantined():
|
|
from gui_v6.processing_runner import classify_error_code
|
|
assert classify_error_code(RuntimeError("Document mis en quarantaine : texte trop court")) == "quarantined"
|
|
|
|
|
|
def test_classify_error_code_no_output():
|
|
from gui_v6.processing_runner import classify_error_code
|
|
assert classify_error_code(RuntimeError("Aucune sortie PDF anonymisée produite")) == "no_output"
|
|
|
|
|
|
def test_classify_error_code_processing_error_default():
|
|
from gui_v6.processing_runner import classify_error_code, _ERROR_CODES
|
|
assert classify_error_code(ValueError("patient Dupont")) == "processing_error"
|
|
assert classify_error_code(ValueError("x")) in _ERROR_CODES
|