cocoindex 0.1.58__cp312-cp312-manylinux_2_28_x86_64.whl → 0.1.60__cp312-cp312-manylinux_2_28_x86_64.whl

cocoindex/__init__.py

@@ -17,7 +17,16 @@ from .llm import LlmSpec, LlmApiType
 from .index import VectorSimilarityMetric, VectorIndexDef, IndexOptions
 from .setting import DatabaseConnectionSpec, Settings, ServerSettings
 from .setting import get_app_namespace
-from .typing import Float32, Float64, LocalDateTime, OffsetDateTime, Range, Vector, Json
+from .typing import (
+    Int64,
+    Float32,
+    Float64,
+    LocalDateTime,
+    OffsetDateTime,
+    Range,
+    Vector,
+    Json,
+)
 
 __all__ = [
     # Submodules
@@ -64,6 +73,7 @@ __all__ = [
     "ServerSettings",
     "get_app_namespace",
     # Typing
+    "Int64",
     "Float32",
     "Float64",
     "LocalDateTime",

cocoindex/convert.py

@@ -5,7 +5,6 @@ Utilities to convert between Python and engine values.
 import dataclasses
 import datetime
 import inspect
-import uuid
 from enum import Enum
 from typing import Any, Callable, Mapping, get_origin
 
@@ -14,7 +13,6 @@ import numpy as np
 from .typing import (
     KEY_FIELD_NAME,
     TABLE_TYPES,
-    AnalyzedTypeInfo,
     DtypeRegistry,
     analyze_type_info,
     encode_enriched_type,
@@ -74,23 +72,16 @@ def make_engine_value_decoder(
     Returns:
         A decoder from an engine value to a Python value.
     """
-
     src_type_kind = src_type["kind"]
 
-    dst_type_info: AnalyzedTypeInfo | None = None
-    if (
-        dst_annotation is not None
-        and dst_annotation is not inspect.Parameter.empty
-        and dst_annotation is not Any
-    ):
-        dst_type_info = analyze_type_info(dst_annotation)
-        if not _is_type_kind_convertible_to(src_type_kind, dst_type_info.kind):
-            raise ValueError(
-                f"Type mismatch for `{''.join(field_path)}`: "
-                f"passed in {src_type_kind}, declared {dst_annotation} ({dst_type_info.kind})"
-            )
-
-    if dst_type_info is None:
+    dst_is_any = (
+        dst_annotation is None
+        or dst_annotation is inspect.Parameter.empty
+        or dst_annotation is Any
+    )
+    if dst_is_any:
+        if src_type_kind == "Union":
+            return lambda value: value[1]
         if src_type_kind == "Struct" or src_type_kind in TABLE_TYPES:
             raise ValueError(
                 f"Missing type annotation for `{''.join(field_path)}`."
@@ -98,6 +89,41 @@ def make_engine_value_decoder(
             )
         return lambda value: value
 
+    dst_type_info = analyze_type_info(dst_annotation)
+
+    if src_type_kind == "Union":
+        dst_type_variants = (
+            dst_type_info.union_variant_types
+            if dst_type_info.union_variant_types is not None
+            else [dst_annotation]
+        )
+        src_type_variants = src_type["types"]
+        decoders = []
+        for i, src_type_variant in enumerate(src_type_variants):
+            src_field_path = field_path + [f"[{i}]"]
+            decoder = None
+            for dst_type_variant in dst_type_variants:
+                try:
+                    decoder = make_engine_value_decoder(
+                        src_field_path, src_type_variant, dst_type_variant
+                    )
+                    break
+                except ValueError:
+                    pass
+            if decoder is None:
+                raise ValueError(
+                    f"Type mismatch for `{''.join(field_path)}`: "
+                    f"cannot find matched target type for source type variant {src_type_variant}"
+                )
+            decoders.append(decoder)
+        return lambda value: decoders[value[0]](value[1])
+
+    if not _is_type_kind_convertible_to(src_type_kind, dst_type_info.kind):
+        raise ValueError(
+            f"Type mismatch for `{''.join(field_path)}`: "
+            f"passed in {src_type_kind}, declared {dst_annotation} ({dst_type_info.kind})"
+        )
+
     if dst_type_info.kind in ("Float32", "Float64", "Int64"):
         dst_core_type = dst_type_info.core_type
 
@@ -196,9 +222,6 @@ def make_engine_value_decoder(
         field_path.pop()
         return decode
 
-    if src_type_kind == "Union":
-        return lambda value: value[1]
-
     return lambda value: value
 
 

cocoindex/llm.py

@@ -12,6 +12,7 @@ class LlmApiType(Enum):
     LITE_LLM = "LiteLlm"
     OPEN_ROUTER = "OpenRouter"
     VOYAGE = "Voyage"
+    VLLM = "Vllm"
 
 
 @dataclass

cocoindex/tests/test_convert.py

@@ -78,16 +78,15 @@ def build_engine_value_decoder(
     return make_engine_value_decoder([], engine_type, python_type or engine_type_in_py)
 
 
-def validate_full_roundtrip(
+def validate_full_roundtrip_to(
     value: Any,
-    value_type: Any = None,
-    *other_decoded_values: tuple[Any, Any],
+    value_type: Any,
+    *decoded_values: tuple[Any, Any],
 ) -> None:
     """
-    Validate the given value doesn't change after encoding, sending to engine (using output_type), receiving back and decoding (using input_type).
+    Validate the given value becomes specific values after encoding, sending to engine (using output_type), receiving back and decoding (using input_type).
 
-    `other_decoded_values` is a tuple of (value, type) pairs.
-    If provided, also validate the value can be decoded to the other types.
+    `decoded_values` is a tuple of (value, type) pairs.
     """
     from cocoindex import _engine  # type: ignore
 
@@ -102,15 +101,27 @@ def validate_full_roundtrip(
     value_from_engine = _engine.testutil.seder_roundtrip(
         encoded_value, encoded_output_type
     )
-    decoder = make_engine_value_decoder([], encoded_output_type, value_type)
-    decoded_value = decoder(value_from_engine)
-    assert eq(decoded_value, value)
 
-    if other_decoded_values is not None:
-        for other_value, other_type in other_decoded_values:
-            decoder = make_engine_value_decoder([], encoded_output_type, other_type)
-            other_decoded_value = decoder(value_from_engine)
-            assert eq(other_decoded_value, other_value)
+    for other_value, other_type in decoded_values:
+        decoder = make_engine_value_decoder([], encoded_output_type, other_type)
+        other_decoded_value = decoder(value_from_engine)
+        assert eq(other_decoded_value, other_value)
+
+
+def validate_full_roundtrip(
+    value: Any,
+    value_type: Any,
+    *other_decoded_values: tuple[Any, Any],
+) -> None:
+    """
+    Validate the given value doesn't change after encoding, sending to engine (using output_type), receiving back and decoding (using input_type).
+
+    `other_decoded_values` is a tuple of (value, type) pairs.
+    If provided, also validate the value can be decoded to the other types.
+    """
+    validate_full_roundtrip_to(
+        value, value_type, (value, value_type), *other_decoded_values
+    )
 
 
 def test_encode_engine_value_basic_types() -> None:
@@ -218,17 +229,33 @@ def test_encode_engine_value_none() -> None:
 
 
 def test_roundtrip_basic_types() -> None:
-    validate_full_roundtrip(42, int, (42, None))
-    validate_full_roundtrip(3.25, float, (3.25, Float64))
     validate_full_roundtrip(
-        3.25, Float64, (3.25, float), (np.float64(3.25), np.float64)
+        42, cocoindex.Int64, (42, int), (np.int64(42), np.int64), (42, None)
+    )
+    validate_full_roundtrip(42, int, (42, cocoindex.Int64))
+    validate_full_roundtrip(np.int64(42), np.int64, (42, cocoindex.Int64))
+
+    validate_full_roundtrip(
+        3.25, Float64, (3.25, float), (np.float64(3.25), np.float64), (3.25, None)
     )
+    validate_full_roundtrip(3.25, float, (3.25, Float64))
+    validate_full_roundtrip(np.float64(3.25), np.float64, (3.25, Float64))
+
     validate_full_roundtrip(
-        3.25, Float32, (3.25, float), (np.float32(3.25), np.float32)
+        3.25,
+        Float32,
+        (3.25, float),
+        (np.float32(3.25), np.float32),
+        (np.float64(3.25), np.float64),
+        (3.25, Float64),
+        (3.25, None),
     )
+    validate_full_roundtrip(np.float32(3.25), np.float32, (3.25, Float32))
+
     validate_full_roundtrip("hello", str, ("hello", None))
     validate_full_roundtrip(True, bool, (True, None))
     validate_full_roundtrip(False, bool, (False, None))
+    validate_full_roundtrip((1, 2), cocoindex.Range, ((1, 2), None))
     validate_full_roundtrip(
         datetime.date(2025, 1, 1), datetime.date, (datetime.date(2025, 1, 1), None)
     )
@@ -238,14 +265,37 @@ def test_roundtrip_basic_types() -> None:
         cocoindex.LocalDateTime,
         (datetime.datetime(2025, 1, 2, 3, 4, 5, 123456), datetime.datetime),
     )
+
+    tz = datetime.timezone(datetime.timedelta(hours=5))
     validate_full_roundtrip(
-        datetime.datetime(2025, 1, 2, 3, 4, 5, 123456, datetime.UTC),
+        datetime.datetime(2025, 1, 2, 3, 4, 5, 123456, tz),
+        cocoindex.OffsetDateTime,
+        (
+            datetime.datetime(2025, 1, 2, 3, 4, 5, 123456, tz),
+            datetime.datetime,
+        ),
+    )
+    validate_full_roundtrip(
+        datetime.datetime(2025, 1, 2, 3, 4, 5, 123456, tz),
+        datetime.datetime,
+        (datetime.datetime(2025, 1, 2, 3, 4, 5, 123456, tz), cocoindex.OffsetDateTime),
+    )
+    validate_full_roundtrip_to(
+        datetime.datetime(2025, 1, 2, 3, 4, 5, 123456),
         cocoindex.OffsetDateTime,
         (
             datetime.datetime(2025, 1, 2, 3, 4, 5, 123456, datetime.UTC),
             datetime.datetime,
         ),
     )
+    validate_full_roundtrip_to(
+        datetime.datetime(2025, 1, 2, 3, 4, 5, 123456),
+        datetime.datetime,
+        (
+            datetime.datetime(2025, 1, 2, 3, 4, 5, 123456, datetime.UTC),
+            cocoindex.OffsetDateTime,
+        ),
+    )
 
     uuid_value = uuid.uuid4()
     validate_full_roundtrip(uuid_value, uuid.UUID, (uuid_value, None))
@@ -613,6 +663,18 @@ def test_roundtrip_union_timedelta() -> None:
     validate_full_roundtrip(value, t)
 
 
+def test_roundtrip_vector_of_union() -> None:
+    t = list[str | int]
+    value = ["a", 1]
+    validate_full_roundtrip(value, t)
+
+
+def test_roundtrip_union_with_vector() -> None:
+    t = NDArray[np.float32] | str
+    value = np.array([1.0, 2.0, 3.0], dtype=np.float32)
+    validate_full_roundtrip(value, t, ([1.0, 2.0, 3.0], list[float] | str))
+
+
 def test_roundtrip_ltable() -> None:
     t = list[Order]
     value = [Order("O1", "item1", 10.0), Order("O2", "item2", 20.0)]

cocoindex/typing.py

@@ -40,6 +40,7 @@ class TypeAttr:
 
 Annotation = TypeKind | TypeAttr | VectorInfo
 
+Int64 = Annotated[int, TypeKind("Int64")]
 Float32 = Annotated[float, TypeKind("Float32")]
 Float64 = Annotated[float, TypeKind("Float64")]
 Range = Annotated[tuple[int, int], TypeKind("Range")]

{cocoindex-0.1.58.dist-info → cocoindex-0.1.60.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: cocoindex
-Version: 0.1.58
+Version: 0.1.60
 Requires-Dist: click>=8.1.8
 Requires-Dist: rich>=14.0.0
 Requires-Dist: python-dotenv>=1.1.0

{cocoindex-0.1.58.dist-info → cocoindex-0.1.60.dist-info}/RECORD

@@ -1,17 +1,17 @@
-cocoindex-0.1.58.dist-info/METADATA,sha256=0JbA9gaANI6bomrbj-zoE5B4lW19k8wco3EC0mmdL5A,10020
-cocoindex-0.1.58.dist-info/WHEEL,sha256=_CzojsfcNnIHPHWM3rF2h1_Qa3VRJKIX8HSir-fDxGE,108
-cocoindex-0.1.58.dist-info/entry_points.txt,sha256=_NretjYVzBdNTn7dK-zgwr7YfG2afz1u1uSE-5bZXF8,46
-cocoindex-0.1.58.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-cocoindex/__init__.py,sha256=MFm-QJzrr0ODJCsAAsPUzJXh8KH1WdZod8F60B1iUYw,1877
-cocoindex/_engine.cpython-312-x86_64-linux-gnu.so,sha256=CHlzJ6rPQiHmqfnHrvY26KU8Dfq-PudqqIZjapinNaE,63200200
+cocoindex-0.1.60.dist-info/METADATA,sha256=TLP5AmkWol7oAVwBcHJfzpQNbilcmrczI6MJX_oVBD0,10020
+cocoindex-0.1.60.dist-info/WHEEL,sha256=_CzojsfcNnIHPHWM3rF2h1_Qa3VRJKIX8HSir-fDxGE,108
+cocoindex-0.1.60.dist-info/entry_points.txt,sha256=_NretjYVzBdNTn7dK-zgwr7YfG2afz1u1uSE-5bZXF8,46
+cocoindex-0.1.60.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+cocoindex/__init__.py,sha256=dSSqO5g2EPoPM3Yo7meIpIh8hWPDArsXFG8kXaQrCwk,1934
+cocoindex/_engine.cpython-312-x86_64-linux-gnu.so,sha256=WS_4ebHkiGkGpTajRgOP1SOq6B7c-MgPQpouNYYg0Hk,63204856
 cocoindex/auth_registry.py,sha256=1XqO7ibjmBBd8i11XSJTvTgdz8p1ptW-ZpuSgo_5zzk,716
 cocoindex/cli.py,sha256=8bDL-Qmd9NYtn1DsDfvUMk45xfAqNf9YTyM7H9KRuNU,21345
-cocoindex/convert.py,sha256=XeSr0ykudBrB-RWRmcbdbt3WCihlLDSBIYUcDtPbTdA,10228
+cocoindex/convert.py,sha256=FsKb2Pfbm7e1VQDOs_AsoiW9PbIUuyHQuqUlrENXmUY,11199
 cocoindex/flow.py,sha256=UQviW2O6nzCKt3jf7N1xAXULvK7YRCZDAj5hy5VLkjM,33041
 cocoindex/functions.py,sha256=IBwvdPpGR-S5mk53HvHpT2GVs15MI9wQznxgOdxA0ac,3202
 cocoindex/index.py,sha256=j93B9jEvvLXHtpzKWL88SY6wCGEoPgpsQhEGHlyYGFg,540
 cocoindex/lib.py,sha256=BeRUn3RqE_wSsVtsgCzbFFKe1LXgRyRmMOcmwWBuEXo,2940
-cocoindex/llm.py,sha256=hpwvHjWiLkoV018TkC191rztq62NQeewwF3kDsHUols,430
+cocoindex/llm.py,sha256=0ri8ZRg9_Zf2gyC5xuQ1Kq6kdZUO8r-A5WLnxit5S_4,448
 cocoindex/op.py,sha256=r_Usx7Jqh49Cck3tsYLx2vLRNUZArkQP_g7bIID6LPU,11809
 cocoindex/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 cocoindex/runtime.py,sha256=bAdHYaXFWiiUWyAgzmKTeaAaRR0D_AmaqVCIdPO-v00,1056
@@ -20,9 +20,9 @@ cocoindex/setup.py,sha256=7uIHKN4FOCuoidPXcKyGTrkqpkl9luL49-6UcnMxYzw,3068
 cocoindex/sources.py,sha256=JCnOhv1w4o28e03i7yvo4ESicWYAhckkBg5bQlxNH4U,1330
 cocoindex/targets.py,sha256=Nfh_tpFd1goTnS_cxBjIs4j9zl3Z4Z1JomAQ1dl3Sic,2796
 cocoindex/tests/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-cocoindex/tests/test_convert.py,sha256=1IoJzfPtM5zoy0YKhTsZTX-fF4xA3w_4usyS4UN44vE,36150
+cocoindex/tests/test_convert.py,sha256=WmgbaQTYPDnjBfMlIrZWBHipPW7SlrvHYsLLqb--KH0,38025
 cocoindex/tests/test_optional_database.py,sha256=snAmkNa6wtOSaxoZE1HgjvL5v_ylitt3Jt_9df4Cgdc,8506
 cocoindex/tests/test_typing.py,sha256=t6UCYShcfonTfjBlGRWPiFGMZ8DGFfABXo6idekPoJE,14757
-cocoindex/typing.py,sha256=kPMFVKs2i4SCLzW1Tn5NP_Ev9DAc-2qW6eJ68gpLexU,12580
+cocoindex/typing.py,sha256=s_Hk4Npi8pxJUM2h_7Km0VFrojA8wJU9VDtRqVSL6C0,12622
 cocoindex/utils.py,sha256=hUhX-XV6XGCtJSEIpBOuDv6VvqImwPlgBxztBTw7u0U,598
-cocoindex-0.1.58.dist-info/RECORD,,
+cocoindex-0.1.60.dist-info/RECORD,,