101 lines
3.9 KiB
Python
101 lines
3.9 KiB
Python
from __future__ import annotations
|
|
|
|
import json
|
|
import sys
|
|
from pathlib import Path
|
|
|
|
import scripts.runtime.materials as materials
|
|
from reportlab.pdfgen import canvas
|
|
|
|
REPO_ROOT = Path(__file__).resolve().parents[1]
|
|
if str(REPO_ROOT) not in sys.path:
|
|
sys.path.insert(0, str(REPO_ROOT))
|
|
|
|
from scripts.runtime.phase1 import create_project, render_framework
|
|
|
|
|
|
def make_text_pdf(path: Path, text: str) -> None:
|
|
c = canvas.Canvas(str(path))
|
|
c.drawString(72, 720, text)
|
|
c.save()
|
|
|
|
|
|
def test_init_ingests_pdf_material_into_phase0(tmp_path: Path) -> None:
|
|
pdf = tmp_path / "audit.pdf"
|
|
make_text_pdf(pdf, "GMP audit finding: deviation management is incomplete.")
|
|
|
|
project = create_project(
|
|
topic="白帆生物 GMP 与运营诊断",
|
|
slug="baifan-test",
|
|
projects_dir=tmp_path / "projects",
|
|
method_key="gmp_quality_operations_diagnosis",
|
|
input_materials=[str(pdf), "补充说明：运营团队需要同步诊断"],
|
|
)
|
|
|
|
manifest = json.loads((project / "manifest.json").read_text(encoding="utf-8"))
|
|
inventory = manifest["material_inventory"]
|
|
|
|
assert inventory[0]["kind"] == "pdf"
|
|
assert inventory[0]["copied_to"] == "phase0/inputs/audit.pdf"
|
|
assert inventory[0]["extracted_to"] == "phase0/extracted/audit.md"
|
|
assert inventory[0]["ocr_required"] is False
|
|
assert "deviation management" in (project / "phase0/extracted/audit.md").read_text(encoding="utf-8")
|
|
assert inventory[1]["kind"] == "note"
|
|
material_brief = project / "phase1" / "material_brief.md"
|
|
assert material_brief.exists()
|
|
assert "Phase 0 材料简报" in material_brief.read_text(encoding="utf-8")
|
|
assert "待用户确认" in material_brief.read_text(encoding="utf-8")
|
|
assert manifest["phase1"]["requires_user_interview"] is True
|
|
|
|
|
|
def test_framework_mentions_ingested_materials(tmp_path: Path) -> None:
|
|
pdf = tmp_path / "audit.pdf"
|
|
make_text_pdf(pdf, "Quality system audit.")
|
|
project = create_project(
|
|
topic="白帆生物 GMP 与运营诊断",
|
|
slug="baifan-test",
|
|
projects_dir=tmp_path / "projects",
|
|
method_key="gmp_quality_operations_diagnosis",
|
|
input_materials=[str(pdf)],
|
|
)
|
|
|
|
render_framework(project, method_key="gmp_quality_operations_diagnosis")
|
|
|
|
framework = (project / "phase1/framework.md").read_text(encoding="utf-8")
|
|
assert "phase0/extracted/audit.md" in framework
|
|
assert "NMPA、FDA、EMA、ICH、WHO" in framework
|
|
assert "请先确认 `phase1/material_brief.md`" in framework
|
|
|
|
|
|
def test_pdf_requiring_ocr_uses_firered_and_records_result(tmp_path: Path, monkeypatch) -> None:
|
|
pdf = tmp_path / "scan.pdf"
|
|
c = canvas.Canvas(str(pdf))
|
|
c.showPage()
|
|
c.save()
|
|
|
|
def fake_ocr_pdf(*, pdf_path: Path, output_dir: Path, endpoint: str, max_pages: int) -> materials.OcrResult:
|
|
assert pdf_path == pdf
|
|
assert endpoint == materials.DEFAULT_FIRERED_OCR_ENDPOINT
|
|
out = output_dir / "scan.ocr.md"
|
|
out.write_text("# OCR\n\n扫描审计发现：偏差管理未闭环。\n", encoding="utf-8")
|
|
return materials.OcrResult(text="扫描审计发现：偏差管理未闭环。", pages_processed=1, output_path=out)
|
|
|
|
monkeypatch.setattr(materials, "ocr_pdf_with_firered", fake_ocr_pdf)
|
|
|
|
project = create_project(
|
|
topic="白帆生物 GMP 与运营诊断",
|
|
slug="baifan-test",
|
|
projects_dir=tmp_path / "projects",
|
|
method_key="gmp_quality_operations_diagnosis",
|
|
input_materials=[str(pdf)],
|
|
)
|
|
|
|
manifest = json.loads((project / "manifest.json").read_text(encoding="utf-8"))
|
|
item = manifest["material_inventory"][0]
|
|
|
|
assert item["ocr_required"] is True
|
|
assert item["ocr_status"] == "completed"
|
|
assert item["ocr_text_chars"] > 0
|
|
assert item["ocr_extracted_to"] == "phase0/extracted/scan.ocr.md"
|
|
assert "扫描审计发现" in (project / "phase0/extracted/scan.md").read_text(encoding="utf-8")
|