V4.4 Complete

2026-08-15 14:30:33 -05:00
parent 63373bf24d
commit 7db4df1729
32 changed files with 1529 additions and 716 deletions
@@ -0,0 +1,103 @@
+from pathlib import Path
+
+import pytest
+
+from transcription.db import session_scope
+from transcription.db.models import Document
+from transcription.db.models import Source
+from transcription.ui.pages.print_preview_page import reflow_transcription
+
+
+def test_reflow_transcription_preserves_paragraph_boundaries():
+    assert reflow_transcription("first line\nsecond line\n\nnext paragraph") == [
+        "first line second line",
+        "next paragraph",
+    ]
+
+
+@pytest.mark.integration
+@pytest.mark.asyncio
+async def test_document_print_preview_and_safe_media_route(app_client):
+    app, client = app_client
+    media_path = app.state.settings.upload_dir / "documents" / "print-page.png"
+    pdf_path = app.state.settings.upload_dir / "documents" / "print-page.pdf"
+    media_path.parent.mkdir(parents=True, exist_ok=True)
+    media_path.write_bytes(b"\x89PNG\r\n\x1a\n")
+    pdf_path.write_bytes(b"%PDF-1.4\n%%EOF")
+
+    async with session_scope() as session:
+        document = Document(name="<Print & Preserve>", notes="<script>unsafe()</script>")
+        session.add(document)
+        await session.flush()
+        source = Source(
+            document_id=document.id,
+            page_number=1,
+            upload_name="print-page.png",
+            filename="print-page.png",
+            file_path=str(media_path),
+            file_hash="a" * 64,
+            file_size_bytes=media_path.stat().st_size,
+            raw_transcription="line one\nline two",
+        )
+        session.add(source)
+        session.add(
+            Source(
+                document_id=document.id,
+                page_number=2,
+                upload_name="print-page.pdf",
+                filename="print-page.pdf",
+                file_path=str(pdf_path),
+                file_hash="c" * 64,
+                file_size_bytes=pdf_path.stat().st_size,
+                raw_transcription="PDF source",
+            )
+        )
+        await session.commit()
+        document_id = document.id
+        source_id = source.id
+
+    response = client.get(f"/ui/documents/{document_id}/print")
+    assert response.status_code == 200
+    assert "Print &amp; Preserve" in response.text
+    assert "&lt;script&gt;unsafe()&lt;/script&gt;" in response.text
+    assert "Facsimile" in response.text
+    assert "Text only" in response.text
+    assert "print-page-break" in response.text
+    assert "print-source-pdf" in response.text
+    assert str(media_path) not in response.text
+    assert str(pdf_path) not in response.text
+
+    media_response = client.get(f"/api/v4/documents/{document_id}/sources/{source_id}/media")
+    assert media_response.status_code == 200
+    assert media_response.headers["content-type"] == "image/png"
+
+
+@pytest.mark.integration
+@pytest.mark.asyncio
+async def test_document_source_media_rejects_cross_document_access(app_client):
+    app, client = app_client
+    media_path = Path(app.state.settings.upload_dir) / "documents" / "other.png"
+    media_path.parent.mkdir(parents=True, exist_ok=True)
+    media_path.write_bytes(b"\x89PNG\r\n\x1a\n")
+
+    async with session_scope() as session:
+        owner = Document(name="Owner")
+        other = Document(name="Other")
+        session.add_all([owner, other])
+        await session.flush()
+        source = Source(
+            document_id=owner.id,
+            page_number=1,
+            upload_name="other.png",
+            filename="other.png",
+            file_path=str(media_path),
+            file_hash="b" * 64,
+            file_size_bytes=media_path.stat().st_size,
+        )
+        session.add(source)
+        await session.commit()
+        other_id = other.id
+        source_id = source.id
+
+    response = client.get(f"/api/v4/documents/{other_id}/sources/{source_id}/media")
+    assert response.status_code == 404