chore: drop the dead sunlei/denylist community text list source

The repo answers 404 (gone, probed 2026-10-10), so every import printed a
skip and the source contributed nothing while the manifest still listed it.
The fetch, its manifest entry and its source name are removed. The block
also wrote its counts under community_reports, which the merge owns, so the
importer no longer claims that source fresh. A test pins the manifest's
source list and keeps the retired URL out of the importer.

Author: SysAdminDoc

CHANGELOG.md

@@ -160,6 +160,9 @@ All notable changes to CallShield will be documented in this file.
   block those 10,000 numbers, with Undo. It never offers your own exchange
   or one where a contact's number is, so it needs access to your contacts
   to check. North American numbers only.
+- The data import no longer asks for the sunlei/denylist text list. That
+  GitHub repository is gone, so every import got a 404 and skipped it, and
+  it never added a number. Its source entry is gone too.
 
 ## v1.11.0 (2026-09-30)
 

data/source-manifest.json

@@ -115,20 +115,6 @@
       "stale_after_days": 30,
       "evidence_ttl_days": 3650
     },
-    {
-      "id": "community_text_lists",
-      "access_mode": "public_feed",
-      "geography": "global",
-      "license": "Upstream list terms; verify before redistribution",
-      "attribution": "Community-maintained text blocklists",
-      "cadence": "on demand",
-      "parser_version": "community-text-v1",
-      "evidence_type": "community_curated",
-      "confidence_tier": "unverified",
-      "redistributable": false,
-      "stale_after_days": 30,
-      "evidence_ttl_days": 365
-    },
     {
       "id": "maintainer_review",
       "access_mode": "maintainer_review",

scripts/import_all_sources.py

@@ -873,47 +873,6 @@ def fetch_toastedspam(allow_insecure: bool = False) -> list[dict]:
     return list(numbers.values())
 
 
-# ── Source 5: ScamCallers community text lists (GitHub mirror) ─────────
-def fetch_community_text_lists() -> list[dict]:
-    print("\n[Community Text Blocklists]")
-    sources = [
-        # Repo-hosted plain-text spam number lists (one number per line, +1XXXXXXXXXX)
-        ("https://raw.githubusercontent.com/sunlei/denylist/master/denylist.txt", "sunlei/denylist"),
-    ]
-    numbers = {}
-    for url, name in sources:
-        try:
-            resp = requests.get(url, timeout=20)
-            if not resp.ok:
-                print(f"  {name}: HTTP {resp.status_code}, skipping")
-                continue
-            count = 0
-            for line in resp.text.splitlines():
-                line = line.strip()
-                if not line or line.startswith('#'):
-                    continue
-                normalized = normalize_phone(line)
-                if normalized:
-                    if normalized in numbers:
-                        numbers[normalized]["reports"] += 1
-                    else:
-                        numbers[normalized] = {
-                            "number": normalized,
-                            "type": "robocall",
-                            "reports": 1,
-                            "first_seen": datetime.now().strftime("%Y-%m-%d"),
-                            "last_seen": datetime.now().strftime("%Y-%m-%d"),
-                            "description": f"Community list: {name}",
-                        }
-                        count += 1
-            print(f"  {name}: {count:,} numbers")
-        except Exception as e:
-            print(f"  {name}: Error — {e}")
-
-    print(f"  Total from community lists: {len(numbers):,} unique numbers")
-    return list(numbers.values())
-
-
 COMPLAINT_SOURCE_IDS = {"ftc_complaints", "fcc_complaints"}
 
 
@@ -1289,17 +1248,6 @@ def main():
         "last_success_at": retrieved_at if ts else None,
     }
 
-    # Source 5: Community text lists
-    cl = fetch_community_text_lists()
-    attach_source_evidence(cl, manifest, "community_text_lists", retrieved_at=retrieved_at)
-    all_numbers.extend(cl)
-    source_stats["community_reports"] = {
-        "status": "ok",
-        "accepted": len(cl),
-        "checksum": payload_checksum(cl),
-        "last_success_at": retrieved_at,
-    }
-
     # Deduplicate across all sources (accumulate reports)
     deduped = {}
     for n in all_numbers:
@@ -1325,7 +1273,6 @@ def main():
         "fcc_complaints",
         "toastedspam",
         "community_reports",
-        "community_text_lists",
     }
     if pb:
         source_names.add("phoneblock_bulk")

scripts/test_release_drift.py

@@ -37,7 +37,7 @@ def test_checkout_is_synchronized(self) -> None:
         self.assertEqual([], report["issues"], report)
         self.assertEqual("1.11.0", report["version_name"])
         self.assertEqual(71, report["version_code"])
-        self.assertEqual(10, report["sources"]["source_count"])
+        self.assertEqual(9, report["sources"]["source_count"])
         self.assertEqual(3, len(report["advisories"]))
 
     def test_readme_resource_counts_must_match_the_resource_file(self) -> None:

scripts/test_source_registry.py

@@ -226,6 +226,30 @@ def test_real_manifest_keeps_number_evidence_for_at_least_a_year(self):
             self.assertGreaterEqual(ttl[source_id], 365, source_id)
         self.assertGreaterEqual(ttl["saracroche_prefixes"], 90)
 
+    def test_real_manifest_lists_exactly_the_sources_the_pipeline_uses(self):
+        # sunlei/denylist answered 404 for every import and stayed listed until
+        # 2026-10-10. Adding or dropping a source changes this list on purpose.
+        manifest = source_registry.load_source_manifest(Path(__file__).parent.parent / "data" / "source-manifest.json")
+        self.assertEqual(
+            [source["id"] for source in manifest["sources"]],
+            [
+                "ftc_complaints",
+                "fcc_complaints",
+                "phoneblock_bulk",
+                "saracroche_prefixes",
+                "nomorobo_irs",
+                "toastedspam",
+                "community_reports",
+                "github_database",
+                "maintainer_review",
+            ],
+        )
+
+    def test_importer_requests_no_retired_source(self):
+        importer = (Path(__file__).parent / "import_all_sources.py").read_text(encoding="utf-8")
+        for retired in ("sunlei/denylist", "community_text_lists"):
+            self.assertNotIn(retired, importer)
+
     def test_refresh_recomputes_expiry_from_retrieval_time_once(self):
         manifest = {"version": 1, "sources": [self._source("fcc", 14, ttl=365)]}
         rows = [