identifiers: Properly define the behavior of raw_extrinsic_metadata on negative timestamps.

The rounding algorithm wasn't specified

swh/model/identifiers.py

@@ -759,9 +759,9 @@ def raw_extrinsic_metadata_identifier(metadata: Dict[str, Any]) -> str:
     $ExtendedSwhid is a core SWHID, with extra types allowed ('ori' for
     origins and 'emd' for raw extrinsic metadata)
 
-    $Timestamp is a decimal representation of the integer number of seconds since
-    the UNIX epoch (1970-01-01 00:00:00 UTC), with no leading '0'
-    (unless the timestamp value is zero) and no timezone.
+    $Timestamp is a decimal representation of the rounded-down integer number of
+    seconds since the UNIX epoch (1970-01-01 00:00:00 UTC),
+    with no leading '0' (unless the timestamp value is zero) and no timezone.
     It may be negative by prefixing it with a '-', which must not be followed
     by a '0'.
 
@@ -772,7 +772,19 @@ def raw_extrinsic_metadata_identifier(metadata: Dict[str, Any]) -> str:
       str: the intrinsic identifier for `metadata`
 
     """
-    timestamp = metadata["discovery_date"].timestamp()
+    # equivalent to using math.floor(dt.timestamp()) to round down,
+    # as int(dt.timestamp()) rounds toward zero,
+    # which would map two seconds on the 0 timestamp.
+    #
+    # This should never be an issue in practice as Software Heritage didn't
+    # start collecting metadata before 2015.
+    timestamp = (
+        metadata["discovery_date"]
+        .astimezone(datetime.timezone.utc)
+        .replace(microsecond=0)
+        .timestamp()
+    )
+    assert timestamp.is_integer()
 
     headers = [
         (b"target", str(metadata["target"]).encode()),

swh/model/tests/test_identifiers.py

@@ -920,6 +920,26 @@ class RawExtrinsicMetadataIdentifier(unittest.TestCase):
             "5c13f20ba336e44549baf3d7b9305b027ec9f43d",
         )
 
+    def test_noninteger_timezone(self):
+        """Checks the discovery_date is translated to UTC before truncating
+        microseconds"""
+        tz = datetime.timezone(datetime.timedelta(microseconds=-42))
+        metadata = {
+            **self.minimal,
+            "discovery_date": datetime.datetime(
+                2021, 1, 25, 11, 27, 50, 1_000_000 - 42, tzinfo=tz,
+            ),
+        }
+
+        self.assertEqual(
+            identifiers.raw_extrinsic_metadata_identifier(self.minimal),
+            identifiers.raw_extrinsic_metadata_identifier(metadata),
+        )
+        self.assertEqual(
+            identifiers.raw_extrinsic_metadata_identifier(metadata),
+            "5c13f20ba336e44549baf3d7b9305b027ec9f43d",
+        )
+
     def test_negative_timestamp(self):
         metadata = {
             **self.minimal,
@@ -948,6 +968,62 @@ class RawExtrinsicMetadataIdentifier(unittest.TestCase):
             "895d0821a2991dd376ddc303424aceb7c68280f9",
         )
 
+    def test_epoch(self):
+        metadata = {
+            **self.minimal,
+            "discovery_date": datetime.datetime(
+                1970, 1, 1, 0, 0, 0, tzinfo=datetime.timezone.utc,
+            ),
+        }
+
+        manifest = (
+            b"raw_extrinsic_metadata 201\0"
+            b"target swh:1:cnt:568aaf43d83b2c3df8067f3bedbb97d83260be6d\n"
+            b"discovery_date 0\n"
+            b"authority forge https://forge.softwareheritage.org/\n"
+            b"fetcher swh-phabricator-metadata-fetcher 0.0.1\n"
+            b"format json\n"
+            b"\n"
+            b'{"foo": "bar"}'
+        )
+
+        self.assertEqual(
+            identifiers.raw_extrinsic_metadata_identifier(metadata),
+            hashlib.sha1(manifest).hexdigest(),
+        )
+        self.assertEqual(
+            identifiers.raw_extrinsic_metadata_identifier(metadata),
+            "27a53df54ace35ebd910493cdc70b334d6b7cb88",
+        )
+
+    def test_negative_epoch(self):
+        metadata = {
+            **self.minimal,
+            "discovery_date": datetime.datetime(
+                1969, 12, 31, 23, 59, 59, 1, tzinfo=datetime.timezone.utc,
+            ),
+        }
+
+        manifest = (
+            b"raw_extrinsic_metadata 202\0"
+            b"target swh:1:cnt:568aaf43d83b2c3df8067f3bedbb97d83260be6d\n"
+            b"discovery_date -1\n"
+            b"authority forge https://forge.softwareheritage.org/\n"
+            b"fetcher swh-phabricator-metadata-fetcher 0.0.1\n"
+            b"format json\n"
+            b"\n"
+            b'{"foo": "bar"}'
+        )
+
+        self.assertEqual(
+            identifiers.raw_extrinsic_metadata_identifier(metadata),
+            hashlib.sha1(manifest).hexdigest(),
+        )
+        self.assertEqual(
+            identifiers.raw_extrinsic_metadata_identifier(metadata),
+            "be7154a8fd49d87f81547ea634d1e2152907d089",
+        )
+
 
 origin_example = {
     "url": "https://github.com/torvalds/linux",