fix: drop bundled files from provenance needs

profile_needs read every profile entry, so --missing named 43 files the emulator installs itself among 50 needs and buried the six xm7 ROMs. A bundled entry is skipped unless a platform package leaves it out (not_bundled_on). The report now lists 7 needs.

scripts/provenance_report.py

@@ -66,7 +66,8 @@ def profile_needs(profiles: dict, db: dict) -> dict:
 
     Keyed by every hash a profile accepts, and, for an entry that declares
     a size and no hash, by its names casefolded. An entry with neither
-    cannot be told apart from any same-named dump and is left out.
+    cannot be told apart from any same-named dump and is left out, as is a
+    file the emulator installs itself (`bundled:`) on every platform.
     """
     files = db.get("files", {})
     indexes = db.get("indexes", {})
@@ -92,6 +93,10 @@ def profile_needs(profiles: dict, db: dict) -> dict:
             name = str(entry.get("name", ""))
             if not name or name.endswith("/"):
                 continue
+            # Installed with the emulator, so not an acquisition target,
+            # unless a platform package leaves it out.
+            if entry.get("bundled") and not entry.get("not_bundled_on"):
+                continue
             need = {"emulator": emulator, "name": name,
                     "required": bool(entry.get("required"))}
             hashes = {(field, v) for field in HASH_FIELDS for v in _values(entry, field)}

tests/test_provenance.py

@@ -623,6 +623,21 @@ class AMissingDumpNamesTheProfileItFills(unittest.TestCase):
         needs = provenance_report.profile_needs(self.PROFILES, db)
         self.assertNotIn("dicrom.rom", needs["by_name"])
 
+    def test_a_file_installed_with_the_emulator_is_not_a_need(self):
+        """43 of 50 needs were bundled entries (samcoupe.rom, zesarux's
+        tk90x.rom), burying the six xm7 ROMs a maintainer has to find."""
+        from scripts import provenance_report  # noqa: PLC0415
+
+        profiles = {
+            "samcoupe": {"files": [{"name": "samcoupe.rom", "required": True,
+                                    "bundled": True, "sha1": "f" * 40}]},
+            "hcl": {"files": [{"name": "kanji.rom", "required": True, "bundled": True,
+                               "not_bundled_on": ["batocera"], "sha1": "9" * 40}]},
+        }
+        needs = provenance_report.profile_needs(profiles, {"files": {}})
+        self.assertNotIn(("sha1", "f" * 40), needs["by_hash"])
+        self.assertIn(("sha1", "9" * 40), needs["by_hash"])
+
 
 class TheWantedPageAsksOnlyForWhatCannotBeMade(unittest.TestCase):
     """The site's Wanted page asked the public for 79 dumps the collection