chore: merge origin/main

mypy-baseline.txt

@@ -899,6 +899,8 @@ scripts/scraper/targets/rocknix_targets_scraper.py:0: error: Name "github_header
 scripts/scraper/targets/rocknix_targets_scraper.py:0: error: Name "standalone_emulators" already defined (possibly by an import)  [no-redef]
 scripts/sect233r1.py:0: error: Incompatible types in assignment (expression has type "Optional[tuple[int, int]]", variable has type "tuple[int, int]")  [assignment]
 scripts/site_insights.py:0: error: Item "list[Any]" of "Union[dict[Any, Any], list[Any]]" has no attribute "get"  [union-attr]
+scripts/site_insights.py:0: error: Item "list[Any]" of "Union[dict[Any, Any], list[Any]]" has no attribute "get"  [union-attr]
+scripts/site_insights.py:0: error: Item "list[Any]" of "Union[dict[Any, Any], list[Any]]" has no attribute "get"  [union-attr]
 scripts/sitediagrams.py:0: error: "AST" has no attribute "value"  [attr-defined]
 scripts/sitediagrams.py:0: error: Argument 1 to "lstrip" of "bytes" has incompatible type "str"; expected "Optional[Buffer]"  [arg-type]
 scripts/sitediagrams.py:0: error: If x = b'abc' then f"{x}" or "{}".format(x) produces "b'abc'", not "abc". If this is desired behavior, use f"{x!r}" or "{!r}".format(x). Otherwise, decode the bytes  [str-bytes-safe]

scripts/profile_sync.py

@@ -5018,7 +5018,7 @@ def _print_triage(
 ) -> bool:
     """Profiles ranked by drift, worst first. False when Ctrl-C cut the
     scoring short: the ranking then covers the profiles scored."""
-    ranked: list[tuple] = []
+    ranked: list[tuple[int, ProfileReport, str]] = []
     sampled = 0
     progress = _Progress(len(reports), "scored", args.offline or len(reports) < 2)
     complete = True

scripts/site_insights.py

@@ -123,15 +123,7 @@ def fetch_queries(start: dt.date, end: dt.date) -> list[Query]:
             return rows
 
 
-def _get_object(url: str, headers: dict[str, str] | None = None) -> dict:
-    """A JSON answer that must be an object: GoatCounter's stats endpoints."""
-    answer = _get_json(url, headers)
-    if not isinstance(answer, dict):
-        raise ValueError(f"{url} answered {type(answer).__name__}, not an object")
-    return answer
-
-
-def fetch_goatcounter(start: dt.date, end: dt.date) -> dict[str, list[tuple[str, int]] | int]:
+def fetch_goatcounter(start: dt.date, end: dt.date) -> dict:
     """Paths and events with their visit counts, and the referring domains."""
     token = (CONFIG / "goatcounter-api-token.txt").read_text(encoding="utf-8").strip()
     auth = {"Authorization": f"Bearer {token}"}
@@ -140,14 +132,16 @@ def fetch_goatcounter(start: dt.date, end: dt.date) -> dict[str, list[tuple[str,
     exclude: list[str] = []
     while True:
         params = urllib.parse.urlencode({**span, "limit": 100, "exclude_paths": ",".join(exclude)})
-        answer = _get_object(f"{GOATCOUNTER}/stats/hits?{params}", auth)
+        answer = _get_json(f"{GOATCOUNTER}/stats/hits?{params}", auth)
         for hit in answer.get("hits", []):
             hits.append((hit["path"] + (" [404]" if hit.get("title") == "404" else ""), int(hit["count"])))
             exclude.append(str(hit["path_id"]))
         if not answer.get("more"):
             break
-    refs = _get_object(f"{GOATCOUNTER}/stats/toprefs?{urllib.parse.urlencode({**span, 'limit': 50})}", auth)
-    total = _get_object(f"{GOATCOUNTER}/stats/total?{urllib.parse.urlencode(span)}", auth)
+    refs = _get_json(f"{GOATCOUNTER}/stats/toprefs?{urllib.parse.urlencode({**span, 'limit': 50})}", auth)
+    total = _get_json(f"{GOATCOUNTER}/stats/total?{urllib.parse.urlencode(span)}", auth)
+    if not isinstance(refs, dict) or not isinstance(total, dict):
+        raise SystemExit("GoatCounter answered with something other than an object")
     # An action carries no referrer, and GoatCounter files it as a direct
     # visit: 592 of the first day's 1,145 "direct" were clicks.
     events = int(total.get("total_events", 0))

tests/test_refresh_data_dirs.py

@@ -323,6 +323,7 @@ class TheManifestCarriesTheArchiveSize(unittest.TestCase):
                       _manifest_data_directories(members, refs, registry, "")}
         self.assertEqual(listed["pak"]["archive_size"], 21_360_579)
         self.assertNotIn("archive_size", listed["old"])
+        self.assertNotIn("optional", listed["pak"])
 
 
 if __name__ == "__main__":