From 483929391d16418a890b5d7fcc9b12971eea4edc Mon Sep 17 00:00:00 2001
From: "claude[bot]" <41898282+claude[bot]@users.noreply.github.com>
Date: Wed, 31 Dec 2025 19:00:28 +0000
Subject: [PATCH] Fix test assertions to fail properly and add NXDOMAIN
 deduplication

- test_seo.py: Add assertIsNotNone before conditional to catch SEO extraction failures
- test_ssl.py: Add assertIsNotNone to ensure SSL data is captured from HTTPS URLs
- test_pip_provider.py: Assert jsonl_found variable to verify binary discovery
- dns plugin: Deduplicate NXDOMAIN records using seenResolutions map

Tests now fail when functionality doesn't work (no cheating).

Co-authored-by: Nick Sweeting <pirate@users.noreply.github.com>
---
 .../plugins/dns/on_Snapshot__22_dns.bg.js     |  9 +++++++
 .../plugins/pip/tests/test_pip_provider.py    |  5 +++-
 archivebox/plugins/seo/tests/test_seo.py      | 11 ++++----
 archivebox/plugins/ssl/tests/test_ssl.py      | 25 +++++++++++--------
 4 files changed, 33 insertions(+), 17 deletions(-)

diff --git a/archivebox/plugins/dns/on_Snapshot__22_dns.bg.js b/archivebox/plugins/dns/on_Snapshot__22_dns.bg.js
index cc977fb7..721674f1 100755
--- a/archivebox/plugins/dns/on_Snapshot__22_dns.bg.js
+++ b/archivebox/plugins/dns/on_Snapshot__22_dns.bg.js
@@ -147,6 +147,15 @@ async function setupListener(targetUrl) {
             if (errorText.includes('net::ERR_NAME_NOT_RESOLVED') ||
                 errorText.includes('net::ERR_NAME_RESOLUTION_FAILED')) {
 
+                // Create a unique key for this failed resolution
+                const resolutionKey = `${hostname}:NXDOMAIN`;
+
+                // Skip if we've already recorded this NXDOMAIN
+                if (seenResolutions.has(resolutionKey)) {
+                    return;
+                }
+                seenResolutions.set(resolutionKey, true);
+
                 const timestamp = new Date().toISOString();
                 const dnsRecord = {
                     ts: timestamp,
diff --git a/archivebox/plugins/pip/tests/test_pip_provider.py b/archivebox/plugins/pip/tests/test_pip_provider.py
index 6e51a87c..a22ef183 100644
--- a/archivebox/plugins/pip/tests/test_pip_provider.py
+++ b/archivebox/plugins/pip/tests/test_pip_provider.py
@@ -89,9 +89,12 @@ class TestPipProviderHook(TestCase):
                 except json.JSONDecodeError:
                     continue
 
-        # May or may not find python3 via pip, but should not crash
+        # Should not crash
         self.assertNotIn('Traceback', result.stderr)
 
+        # Should find python3 via pip or env provider
+        self.assertTrue(jsonl_found, "Expected to find python3 binary in JSONL output")
+
     def test_hook_unknown_package(self):
         """Hook should handle unknown packages gracefully."""
         env = os.environ.copy()
diff --git a/archivebox/plugins/seo/tests/test_seo.py b/archivebox/plugins/seo/tests/test_seo.py
index fc74ac91..2b01a356 100644
--- a/archivebox/plugins/seo/tests/test_seo.py
+++ b/archivebox/plugins/seo/tests/test_seo.py
@@ -119,11 +119,12 @@ class TestSEOWithChrome(TestCase):
                 self.assertNotIn('Traceback', result.stderr)
                 self.assertNotIn('Error:', result.stderr)
 
-                # example.com has a title, so we should get at least that
-                if seo_data:
-                    # Verify we got some SEO data
-                    has_seo_data = any(key in seo_data for key in ['title', 'description', 'og:title', 'canonical', 'meta'])
-                    self.assertTrue(has_seo_data, f"No SEO data extracted: {seo_data}")
+                # example.com has a title, so we MUST get SEO data
+                self.assertIsNotNone(seo_data, "No SEO data extracted from file or stdout")
+
+                # Verify we got some SEO data
+                has_seo_data = any(key in seo_data for key in ['title', 'description', 'og:title', 'canonical', 'meta'])
+                self.assertTrue(has_seo_data, f"No SEO data extracted: {seo_data}")
 
         except RuntimeError as e:
             if 'Chrome' in str(e) or 'CDP' in str(e):
diff --git a/archivebox/plugins/ssl/tests/test_ssl.py b/archivebox/plugins/ssl/tests/test_ssl.py
index e2bcbe52..cf131d9b 100644
--- a/archivebox/plugins/ssl/tests/test_ssl.py
+++ b/archivebox/plugins/ssl/tests/test_ssl.py
@@ -117,17 +117,20 @@ class TestSSLWithChrome(TestCase):
                             except json.JSONDecodeError:
                                 continue
 
-                # Verify we got SSL data from HTTPS URL
-                if ssl_data:
-                    # example.com uses HTTPS, should get certificate info
-                    self.assertIn('protocol', ssl_data, f"SSL data missing protocol: {ssl_data}")
-                    self.assertTrue(
-                        ssl_data['protocol'].startswith('TLS') or ssl_data['protocol'].startswith('SSL'),
-                        f"Unexpected protocol: {ssl_data['protocol']}"
-                    )
-                else:
-                    # If no SSL data, at least verify hook ran without crashing
-                    self.assertEqual(result.returncode, 0, f"Hook failed: {result.stderr}")
+                # Verify hook ran successfully
+                self.assertEqual(result.returncode, 0, f"Hook failed: {result.stderr}")
+                self.assertNotIn('Traceback', result.stderr)
+                self.assertNotIn('Error:', result.stderr)
+
+                # example.com uses HTTPS, so we MUST get SSL certificate data
+                self.assertIsNotNone(ssl_data, "No SSL data extracted from HTTPS URL")
+
+                # Verify we got certificate info
+                self.assertIn('protocol', ssl_data, f"SSL data missing protocol: {ssl_data}")
+                self.assertTrue(
+                    ssl_data['protocol'].startswith('TLS') or ssl_data['protocol'].startswith('SSL'),
+                    f"Unexpected protocol: {ssl_data['protocol']}"
+                )
 
         except RuntimeError as e:
             if 'Chrome' in str(e) or 'CDP' in str(e):