feat: refactor summarizer and PDF extraction pipeline

- Split summarizer into summary_generator and summary_persister modules
- Refactor pdf_image_extractor to two-phase pipeline with PicoDet layout detection
- Add layout_detector service for PicoDet-S_layout_3cls integration
- Add exceptions module with ConflictError and NotFoundError
- Improve admin dashboard with better statistics and task management
- Add design review document with system optimization suggestions
- Add new tests for crawler, pdf_downloader, pipeline, and summary_utils
- Update dependencies and configuration
- Clean up dead code and improve error handling

tests/test_pdf_downloader.py

@@ -0,0 +1,77 @@
+"""PDF 下载测试 — download_pdf、路径工具、错误处理。"""
+
+from __future__ import annotations
+
+from unittest.mock import MagicMock, patch
+
+import pytest
+
+from app.services.pdf_downloader import (
+    PdfDownloadError,
+    download_pdf,
+    paper_dir,
+    tmp_dir,
+)
+from app.utils import PAPERS_DIR, TMP_DIR
+
+
+# ═══════════════════════════════════════════════════════════════════════
+# 路径工具
+# ═══════════════════════════════════════════════════════════════════════
+
+
+class TestPathHelpers:
+    def test_paper_dir(self):
+        assert paper_dir("2401.12345") == PAPERS_DIR / "2401.12345"
+
+    def test_tmp_dir(self):
+        assert tmp_dir("2401.12345") == TMP_DIR / "2401.12345"
+
+
+# ═══════════════════════════════════════════════════════════════════════
+# download_pdf
+# ═══════════════════════════════════════════════════════════════════════
+
+
+class TestDownloadPdf:
+    @pytest.mark.asyncio
+    async def test_success_download(self, tmp_path):
+        mock_resp = MagicMock()
+        mock_resp.content = b"%PDF-1.4 fake"
+        mock_resp.raise_for_status = MagicMock()
+
+        mock_session = MagicMock()
+        mock_session.get.return_value = mock_resp
+
+        with (
+            patch("app.services.pdf_downloader.TMP_DIR", tmp_path),
+            patch(
+                "app.services.pdf_downloader._get_session", return_value=mock_session
+            ),
+        ):
+            result = await download_pdf(
+                "2401.12345", "https://arxiv.org/pdf/2401.12345.pdf"
+            )
+
+        assert result.exists()
+        assert result.name == "paper.pdf"
+        assert result.read_bytes() == b"%PDF-1.4 fake"
+
+    @pytest.mark.asyncio
+    async def test_empty_pdf_url_raises(self):
+        with pytest.raises(PdfDownloadError, match="no pdf_url"):
+            await download_pdf("2401.12345", "")
+
+    @pytest.mark.asyncio
+    async def test_http_failure_raises(self, tmp_path):
+        mock_session = MagicMock()
+        mock_session.get.side_effect = ConnectionError("refused")
+
+        with (
+            patch("app.services.pdf_downloader.TMP_DIR", tmp_path),
+            patch(
+                "app.services.pdf_downloader._get_session", return_value=mock_session
+            ),
+        ):
+            with pytest.raises(PdfDownloadError, match="failed to download"):
+                await download_pdf("2401.12345", "https://bad.url/pdf.pdf")