feat: mark files a core finds by content

DuckStation identifies its BIOS by MD5 in a folder it scans
(bios.cpp:208), so its entries' names are labels. The export wrote them
into Batocera's and ROCKNIX's lists as files to hold: a user with only
scph5501.bin read 104 missing BIOS where DuckStation boots. Labels also
lent their md5 to a name another core opens, and paired by name they
replaced System.dat's scph5000.bin hashes.

found_by: name | content, on a profile or a file and keyed by build like
load_from. The truth records which cores open each name (_named_by). A
label pairs with a platform entry by hash only, is never added, merges
with an opened name only on a shared hash, and is counted in the export
report rather than listed as missing.

tests/test_exporters.py

@@ -330,6 +330,49 @@ class Reconciliation(unittest.TestCase):
         scraped_sys = {"files": [{"name": "scph1001.bin", "md5": "b" * 32}]}
         self.assertNotIn("hash_mismatch", _diff_system(truth_sys, scraped_sys))
 
+    def test_a_label_is_never_added(self):
+        """A name only a content-scanning core gives is no file to hold."""
+        truth = {"systems": {"sony-playstation": {"files": [
+            {"name": "scph1001_v20.bin", "md5": "a" * 32,
+             "_cores": ["duckstation"], "_named_by": []},
+            {"name": "scph5903.bin", "md5": "c" * 32,
+             "_cores": ["beetle_psx"], "_named_by": ["beetle_psx"]},
+        ]}}}
+        scraped = {"systems": {"sony-playstation": {"native_id": "psx", "files": [
+            {"name": "scph5501.bin", "destination": "scph5501.bin",
+             "native_system": "psx", "md5": "d" * 32}]}}}
+        systems, report = build_native_model(truth, scraped)
+        self.assertEqual([fe.name for fe in systems["psx"].files],
+                         ["scph5501.bin", "scph5903.bin"])
+        self.assertEqual(report.files_added, 1)
+
+    def test_a_label_pairs_by_content_only(self):
+        """DuckStation's and rustation's scph5000.bin labels carried another
+        dump's md5 and, paired by name, replaced System.dat's scph5000.bin."""
+        truth = {"systems": {"sony-playstation": {"files": [
+            {"name": "scph5000.bin", "md5": "a" * 32,
+             "_cores": ["duckstation"], "_named_by": []},
+            {"name": "other_label.bin", "md5": "e" * 32, "sha256": "f" * 64,
+             "_cores": ["duckstation"], "_named_by": []},
+        ]}}}
+        scraped = {"systems": {"sony-playstation": {"native_id": "psx", "files": [
+            {"name": "scph5000.bin", "destination": "scph5000.bin",
+             "native_system": "psx", "md5": "b" * 32},
+            {"name": "scph7001.bin", "destination": "scph7001.bin",
+             "native_system": "psx", "md5": "e" * 32},
+        ]}}}
+        systems, report = build_native_model(truth, scraped)
+        scph5000, scph7001 = systems["psx"].files
+        self.assertIsNone(scph5000.truth)
+        self.assertEqual(scph7001.truth["name"], "other_label.bin")
+        self.assertEqual(report.hashes_corrected, [])
+
+    def test_the_diff_names_no_missing_label(self):
+        from truth import _diff_system
+        truth_sys = {"files": [{"name": "scph1001_v20.bin", "md5": "a" * 32,
+                                "_cores": ["duckstation"], "_named_by": []}]}
+        self.assertNotIn("missing", _diff_system(truth_sys, {"files": []}))
+
     def test_the_diff_pairs_by_the_same_rule(self):
         """diff_truth reads the same pairing, or it names a mismatch the
         export does not correct."""

tests/test_truth_layout.py

@@ -294,5 +294,50 @@ class ASizeOutsideARangeIsAnotherFile(unittest.TestCase):
         self.assertEqual(len(system["files"]), 2)
 
 
+class TruthRecordsWhoOpensByName(unittest.TestCase):
+    """DuckStation finds its BIOS by MD5 in a folder it scans (bios.cpp:208,
+    397-427): its names are labels. The export wrote them into Batocera's
+    list as files to hold, and a user with only scph5501.bin read 104
+    missing BIOS where DuckStation boots."""
+
+    DUCK = {"type": "libretro", "systems": ["sony-playstation"], "found_by": "content",
+            "files": [{"name": "scph1001_v20.bin", "md5": "a" * 32},
+                      {"name": "scph1001.bin", "md5": "b" * 32}]}
+    PCSX = {"type": "libretro", "systems": ["sony-playstation"],
+            "files": [{"name": "scph1001.bin", "size": 524288}]}
+
+    def truth(self, profiles: dict) -> dict:
+        config = {"cores": list(profiles), "systems": {"sony-playstation": {"files": []}}}
+        return generate_platform_truth("p", config, {}, profiles)
+
+    def entries(self, truth: dict) -> list[dict]:
+        return truth["systems"]["sony-playstation"]["files"]
+
+    def test_a_content_scanning_core_names_nothing(self):
+        truth = self.truth({"duckstation": self.DUCK})
+        self.assertEqual({f["name"]: f["_named_by"] for f in self.entries(truth)},
+                         {"scph1001_v20.bin": [], "scph1001.bin": []})
+
+    def test_a_label_lends_no_hash_to_a_name_another_core_opens(self):
+        """pcsx_rearmed opens scph1001.bin by name, with no hash: it took the
+        md5 of DuckStation's label of that name."""
+        truth = self.truth({"duckstation": self.DUCK, "pcsx": self.PCSX})
+        named = [f for f in self.entries(truth) if f["_named_by"] == ["pcsx"]]
+        self.assertEqual(len(named), 1)
+        self.assertNotIn("md5", named[0])
+
+    def test_the_build_decides(self):
+        ymir = {"type": "standalone + libretro", "systems": ["sega-saturn"],
+                "found_by": {"standalone": "content"},
+                "files": [{"name": "sega_101.bin", "md5": "c" * 32}]}
+        config = {"cores": ["ymir"], "standalone_cores": [],
+                  "systems": {"sega-saturn": {"files": []}}}
+        libretro = generate_platform_truth("p", config, {}, {"ymir": ymir})
+        self.assertEqual(libretro["systems"]["sega-saturn"]["files"][0]["_named_by"], ["ymir"])
+        config["standalone_cores"] = ["ymir"]
+        standalone = generate_platform_truth("p", config, {}, {"ymir": ymir})
+        self.assertEqual(standalone["systems"]["sega-saturn"]["files"][0]["_named_by"], [])
+
+
 if __name__ == "__main__":
     unittest.main()