"""Local fixture tests. All browser HTTP requests are intercepted; no site is read. Run separately: py test_browser_pdf.py --output Requires installed Chromium and the normal requirements. """ import argparse import copy import json from pathlib import Path from playwright.sync_api import sync_playwright from pypdf import PdfReader import heilind_sku_pdf as tool from test_heilind_sku_pdf import candidate, fake_manifest def fixture(sku): rows = "".join(f"TEST ATTRIBUTE {i:03d}Fixture value {i} - test data only" for i in range(1, 91)) return f'''LOCAL TEST FIXTURE {sku}

LOCAL TEST FIXTURE - NOT A LIVE HEILIND PAGE

{sku}

Test fixture manufacturer. This text ensures the product page has enough rendered content for the capture validation checks.

{rows}

Attribute	Value

Additional test content

NATIVE DETAILS CONTENT INCLUDED

''' def main(): parser = argparse.ArgumentParser() parser.add_argument("--output", type=Path, required=True) args = parser.parse_args() out = args.output.resolve() out.mkdir(parents=True, exist_ok=False) (out / "product_pdfs").mkdir() settings = copy.deepcopy(tool.DEFAULTS["capture"]) settings.update(settle_seconds=0, between_pages_seconds=0) manifest = fake_manifest() rows = [] for i in range(25): row = candidate(f"TESTSKU{i:02d}", f"TEST MFG {i % 12:02d}", 4, 12 + i % 7) row.update(percent_increase=(row["conversion_count"] - 4) / 4 * 100, url=tool.product_url(row["sku"], settings), capture_status="pending") rows.append(row) manifest["selected"] = rows manifest["warnings"].append("LOCAL TEST FIXTURE ONLY - not live product or database evidence.") with sync_playwright() as p: browser = p.chromium.launch() context = browser.new_context(viewport={"width": 1440, "height": 1000}) def route(request): if "blocked" in request.request.url: request.fulfill(status=403, content_type="text/html", body="

Access denied

") elif "missing" in request.request.url: request.fulfill(status=200, content_type="text/html", body="

404 Not Found

") else: request.fulfill(status=200, content_type="text/html", body=fixture("TESTSKU00")) context.route("**/*", route) tool.capture_one(context, rows[0], 1, out, settings) assert rows[0]["capture_status"] == "saved", rows[0] pdf = PdfReader(str(out / rows[0]["pdf_file"])) assert len(pdf.pages) == 1, "Continuous output should be a single tall page" assert float(pdf.pages[0].mediabox.height) > 3000, "PDF is only viewport-sized" contents = "\n".join(page.extract_text() for page in pdf.pages) for marker in ("TEST ATTRIBUTE 090", "NATIVE DETAILS CONTENT INCLUDED", "END OF PRODUCT TEST FOOTER"): assert marker in contents, f"Missing full-page content: {marker}" # The single-page size guard must fall back to complete paginated output. page = context.new_page() page.set_content(fixture("TESTSKU00")) tool.prepare_page(page, settings) paged_settings = {**settings, "max_single_page_height_px": 1000} layout = tool.print_full_page(page, out / "fixture_paged.pdf", paged_settings) assert layout == "paged" paged = PdfReader(str(out / "fixture_paged.pdf")) assert len(paged.pages) >= 3 assert "END OF PRODUCT TEST FOOTER" in "\n".join(p.extract_text() for p in paged.pages) page.close() for kind, expected in (("blocked", tool.CaptureBlocked), ("missing", RuntimeError)): bad = candidate(kind, "TEST") bad["url"] = f"https://fixture.invalid/{kind}.html" try: tool.capture_one(context, bad, 99, out, settings) raise AssertionError(f"{kind} page incorrectly saved") except expected: pass assert "pdf_file" not in bad tool.checkpoint(out, manifest) # Only a local file navigation may bypass the route's synthetic HTTP response. context.unroute("**/*", route) context.route("http://**/*", lambda route: route.abort()) context.route("https://**/*", lambda route: route.abort()) tool.export_summary(context, out) context.close() browser.close() summary = PdfReader(str(out / "selection_summary.pdf")) summary_text = "\n".join(p.extract_text() for p in summary.pages) assert all(row["sku"] in summary_text for row in rows) tool.combine_pdfs(out, manifest) combined = PdfReader(str(out / "all_selected_product_pages.pdf")) assert len(combined.pages) == len(summary.pages) + 1 print(json.dumps({"result": "PASS", "fixture": "synthetic, all HTTP requests intercepted", "continuous_pdf_pages": len(pdf.pages), "continuous_height_points": float(pdf.pages[0].mediabox.height), "paged_pdf_pages": len(paged.pages), "summary_pages": len(summary.pages), "all_25_summary_skus_present": True, "end_of_page_present": True, "blocked_and_soft_404_rejected": True, "output": str(out)}, indent=2)) if __name__ == "__main__": main()