[OPIK-568] Updated uuidv4_to_uuidv7 function to handle timestamps (#1204

)

* Updated uuidv4_to_uuidv7 function to handle timestamps

* Updated uuidv4_to_uuidv7 function to handle timestamps

* Fix issue following review

sdks/python/src/opik/id_helpers.py

@@ -2,42 +2,31 @@
 import uuid
 
 
-# STATIC NS
-UUIDV7_NS = 1733244176020523256
-
-
 def uuid4_to_uuid7(user_datetime: datetime, user_uuid: str) -> uuid.UUID:
-    """Convert an UUID4 in a UUID7 in a naive way"""
-    last = [0, 0, 0, 0]
-
-    # Use a hardcoded timestamp in ns for now
-    ns = UUIDV7_NS
-
-    # Simple uuid7 implementation
-    sixteen_secs = 16_000_000_000
-    t1, rest1 = divmod(ns, sixteen_secs)
-    t2, rest2 = divmod(rest1 << 16, sixteen_secs)
-    t3, _ = divmod(rest2 << 12, sixteen_secs)
-    t3 |= 7 << 12  # Put uuid version in top 4 bits, which are 0 in t3
-
-    # The next two bytes are an int (t4) with two bits for
-    # the variant 2 and a 14 bit sequence counter which increments
-    # if the time is unchanged.
-    if t1 == last[0] and t2 == last[1] and t3 == last[2]:
-        # Stop the seq counter wrapping past 0x3FFF.
-        # This won't happen in practice, but if it does,
-        # uuids after the 16383rd with that same timestamp
-        # will not longer be correctly ordered but
-        # are still unique due to the 6 random bytes.
-        if last[3] < 0x3FFF:
-            last[3] += 1
-    else:
-        last[:] = (t1, t2, t3, 0)
-    t4 = (2 << 14) | last[3]  # Put variant 0b10 in top two bits
-
-    # Six random bytes from the provided UUIDv4
+    """Convert a UUID v4 into a UUID v7 following RFC draft specification."""
+    # Get Unix timestamp in milliseconds
+    unix_ts_ms = int(user_datetime.timestamp() * 1000)
+
     uuidv4 = uuid.UUID(user_uuid)
-    assert uuidv4.version == 4
-    rand = uuidv4.bytes[-6:]
+    if uuidv4.version != 4:
+        raise ValueError("Input UUID must be version 4")
+
+    # Create the 16-byte array
+    uuid_bytes = bytearray(16)
+
+    # First 48 bits (6 bytes): Unix timestamp in milliseconds
+    uuid_bytes[0:6] = unix_ts_ms.to_bytes(6, byteorder="big")
+
+    # Next byte: Version 7 in top 4 bits
+    uuid_bytes[6] = 0x70 | (uuidv4.bytes[6] & 0x0F)
+
+    # Next byte: random from UUID v4
+    uuid_bytes[7] = uuidv4.bytes[7]
+
+    # Next byte: Variant bits (0b10) in top 2 bits
+    uuid_bytes[8] = 0x80 | (uuidv4.bytes[8] & 0x3F)
+
+    # Remaining bytes: random from UUID v4
+    uuid_bytes[9:16] = uuidv4.bytes[9:16]
 
-    return uuid.UUID(f"{t1:>08x}-{t2:>04x}-{t3:>04x}-{t4:>04x}-{rand.hex()}")
+    return uuid.UUID(bytes=bytes(uuid_bytes))

sdks/python/tests/unit/test_id_helpers.py

@@ -0,0 +1,72 @@
+from datetime import datetime
+import uuid
+from opik.id_helpers import uuid4_to_uuid7
+import time
+
+
+def test_uuid4_to_uuid7__generates_valid_uuidv7():
+    """
+    Test that uuid4_to_uuid7 generates valid UUIDv7.
+    """
+    uuid_v7 = uuid4_to_uuid7(datetime.now(), str(uuid.uuid4()))
+    assert uuid_v7.version == 7, f"Generated UUID {uuid_v7} is not a version 7 UUID"
+
+
+def test_uuid4_to_uuid7__generates_consistent_uuids():
+    """
+    Test that uuid4_to_uuid7 generates consistent UUIDs.
+    """
+    # Create test data with known timestamps and UUIDs
+    NB_ID = 5
+    test_uuids = []
+
+    timestamp = datetime.now()
+    uuid4 = str(uuid.uuid4())
+
+    for i in range(NB_ID):
+        test_uuids.append((timestamp, uuid4))
+
+    # Convert UUIDs
+    uuids_v7 = [str(uuid4_to_uuid7(ts, uuid4)) for ts, uuid4 in test_uuids]
+
+    # Check ids are distinct
+    assert len(set(uuids_v7)) == 1, "UUIDs are not distinct"
+
+
+def test_uuid4_to_uuid7__sequential_timestamps__maintains_temporal_ordering():
+    """
+    Test that uuid4_to_uuid7 maintains temporal ordering when given sequential timestamps.
+    """
+    # Create test data with known timestamps and UUIDs
+    NB_ID = 5
+    test_uuids = []
+
+    for i in range(NB_ID):
+        test_uuids.append((datetime.now(), str(uuid.uuid4())))
+        time.sleep(0.5)
+
+    # Convert UUIDs
+    uuids_v7 = [str(uuid4_to_uuid7(ts, uuid4)) for ts, uuid4 in test_uuids]
+
+    # Assert temporal ordering
+    assert uuids_v7 == sorted(uuids_v7), "UUIDs are not sorted"
+
+
+def test_uuid4_to_uuid7__different_uuid4_same_timestamp():
+    """
+    Test that uuid4_to_uuid7 creates different UUIDv7 when given the same timestamp and different UUIDv4.
+    """
+    # Create test data with known timestamps and UUIDs
+    NB_ID = 5
+    test_uuids = []
+    timestamp = datetime.now()
+
+    for i in range(NB_ID):
+        test_uuids.append((timestamp, str(uuid.uuid4())))
+        time.sleep(0.5)
+
+    # Convert UUIDs
+    uuids_v7 = [str(uuid4_to_uuid7(ts, uuid4)) for ts, uuid4 in test_uuids]
+
+    # Check ids are different
+    assert len(uuids_v7) == len(set(uuids_v7)), "UUIDs are not distinct"