cocoindex 0.1.50__cp312-cp312-macosx_11_0_arm64.whl → 0.1.52__cp312-cp312-macosx_11_0_arm64.whl

cocoindex/__init__.py

@@ -2,7 +2,9 @@
 Cocoindex is a framework for building and running indexing pipelines.
 """
 
-from . import functions, sources, storages, cli, utils
+from . import functions, sources, targets, cli, utils
+
+from . import targets as storages  # Deprecated: Use targets instead
 
 from .auth_registry import AuthEntryReference, add_auth_entry, ref_auth_entry
 from .flow import FlowBuilder, DataScope, DataSlice, Flow, transform_flow
@@ -15,3 +17,55 @@ 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
+
+__all__ = [
+    # Submodules
+    "_engine",
+    "functions",
+    "sources",
+    "targets",
+    "storages",
+    "cli",
+    "utils",
+    # Auth registry
+    "AuthEntryReference",
+    "add_auth_entry",
+    "ref_auth_entry",
+    # Flow
+    "FlowBuilder",
+    "DataScope",
+    "DataSlice",
+    "Flow",
+    "transform_flow",
+    "flow_def",
+    "EvaluateAndDumpOptions",
+    "GeneratedField",
+    "update_all_flows_async",
+    "FlowLiveUpdater",
+    "FlowLiveUpdaterOptions",
+    # Lib
+    "init",
+    "start_server",
+    "stop",
+    "main_fn",
+    # LLM
+    "LlmSpec",
+    "LlmApiType",
+    # Index
+    "VectorSimilarityMetric",
+    "VectorIndexDef",
+    "IndexOptions",
+    # Settings
+    "DatabaseConnectionSpec",
+    "Settings",
+    "ServerSettings",
+    "get_app_namespace",
+    # Typing
+    "Float32",
+    "Float64",
+    "LocalDateTime",
+    "OffsetDateTime",
+    "Range",
+    "Vector",
+    "Json",
+]

cocoindex/cli.py

@@ -237,10 +237,17 @@ def show(app_flow_specifier: str, color: bool, verbose: bool) -> None:
 
 @cli.command()
 @click.argument("app_target", type=str)
-def setup(app_target: str) -> None:
+@click.option(
+    "-f",
+    "--force",
+    is_flag=True,
+    show_default=True,
+    default=False,
+    help="Force setup without confirmation prompts.",
+)
+def setup(app_target: str, force: bool) -> None:
     """
-    Check and apply backend setup changes for flows, including the internal and target storage
-    (to export).
+    Check and apply backend setup changes for flows, including the internal storage and target (to export to).
 
     APP_TARGET: path/to/app.py or installed_module.
     """
@@ -252,7 +259,7 @@ def setup(app_target: str) -> None:
     if setup_status.is_up_to_date():
         click.echo("No changes need to be pushed.")
         return
-    if not click.confirm(
+    if not force and not click.confirm(
         "Changes need to be pushed. Continue? [yes/N]",
         default=False,
         show_default=False,
@@ -275,7 +282,17 @@ def setup(app_target: str) -> None:
     "even if not defined in the current process."
     "If used, APP_TARGET and any listed flow names are ignored.",
 )
-def drop(app_target: str | None, flow_name: tuple[str, ...], drop_all: bool) -> None:
+@click.option(
+    "-f",
+    "--force",
+    is_flag=True,
+    show_default=True,
+    default=False,
+    help="Force drop without confirmation prompts.",
+)
+def drop(
+    app_target: str | None, flow_name: tuple[str, ...], drop_all: bool, force: bool
+) -> None:
     """
     Drop the backend setup for flows.
 
@@ -328,7 +345,7 @@ def drop(app_target: str | None, flow_name: tuple[str, ...], drop_all: bool) ->
     if setup_status.is_up_to_date():
         click.echo("No flows need to be dropped.")
         return
-    if not click.confirm(
+    if not force and not click.confirm(
         f"\nThis will apply changes to drop setup for: {', '.join(flow_names)}. Continue? [yes/N]",
         default=False,
         show_default=False,

cocoindex/convert.py

@@ -127,8 +127,7 @@ def make_engine_value_decoder(
         return lambda value: uuid.UUID(bytes=value)
 
     if src_type_kind == "Vector":
-        elem_coco_type_info = analyze_type_info(dst_type_info.elem_type)
-        dtype_info = DtypeRegistry.get_by_kind(elem_coco_type_info.kind)
+        dtype_info = DtypeRegistry.get_by_dtype(dst_type_info.np_number_type)
 
         def decode_vector(value: Any) -> Any | None:
             if value is None:
@@ -237,7 +236,12 @@ def dump_engine_object(v: Any) -> Any:
         nanos = int((total_secs - secs) * 1e9)
         return {"secs": secs, "nanos": nanos}
     elif hasattr(v, "__dict__"):
-        s = {k: dump_engine_object(v) for k, v in v.__dict__.items()}
+        s = {}
+        for k, val in v.__dict__.items():
+            if val is None:
+                # Skip None values
+                continue
+            s[k] = dump_engine_object(val)
         if hasattr(v, "kind") and "kind" not in s:
             s["kind"] = v.kind
         return s

cocoindex/flow.py

@@ -327,7 +327,7 @@ class DataCollector:
     def export(
         self,
         name: str,
-        target_spec: op.StorageSpec,
+        target_spec: op.TargetSpec,
         /,
         *,
         primary_key_fields: Sequence[str],
@@ -340,7 +340,7 @@ class DataCollector:
 
         `vector_index` is for backward compatibility only. Please use `vector_indexes` instead.
         """
-        if not isinstance(target_spec, op.StorageSpec):
+        if not isinstance(target_spec, op.TargetSpec):
             raise ValueError(
                 "export() can only be called on a CocoIndex target storage"
             )

cocoindex/op.py

@@ -19,7 +19,7 @@ class OpCategory(Enum):
 
     FUNCTION = "function"
     SOURCE = "source"
-    STORAGE = "storage"
+    TARGET = "target"
     DECLARATION = "declaration"
 
 
@@ -52,8 +52,8 @@ class FunctionSpec(metaclass=SpecMeta, category=OpCategory.FUNCTION):  # pylint:
     """A function spec. All its subclass can be instantiated similar to a dataclass, i.e. ClassName(field1=value1, field2=value2, ...)"""
 
 
-class StorageSpec(metaclass=SpecMeta, category=OpCategory.STORAGE):  # pylint: disable=too-few-public-methods
-    """A storage spec. All its subclass can be instantiated similar to a dataclass, i.e. ClassName(field1=value1, field2=value2, ...)"""
+class TargetSpec(metaclass=SpecMeta, category=OpCategory.TARGET):  # pylint: disable=too-few-public-methods
+    """A target spec. All its subclass can be instantiated similar to a dataclass, i.e. ClassName(field1=value1, field2=value2, ...)"""
 
 
 class DeclarationSpec(metaclass=SpecMeta, category=OpCategory.DECLARATION):  # pylint: disable=too-few-public-methods

cocoindex/setting.py

@@ -62,18 +62,22 @@ def _load_field(
 class Settings:
     """Settings for the cocoindex library."""
 
-    database: DatabaseConnectionSpec
+    database: DatabaseConnectionSpec | None = None
     app_namespace: str = ""
 
     @classmethod
     def from_env(cls) -> Self:
         """Load settings from environment variables."""
 
-        db_kwargs: dict[str, str] = dict()
-        _load_field(db_kwargs, "url", "COCOINDEX_DATABASE_URL", required=True)
-        _load_field(db_kwargs, "user", "COCOINDEX_DATABASE_USER")
-        _load_field(db_kwargs, "password", "COCOINDEX_DATABASE_PASSWORD")
-        database = DatabaseConnectionSpec(**db_kwargs)
+        database_url = os.getenv("COCOINDEX_DATABASE_URL")
+        if database_url is not None:
+            db_kwargs: dict[str, str] = dict()
+            _load_field(db_kwargs, "url", "COCOINDEX_DATABASE_URL", required=True)
+            _load_field(db_kwargs, "user", "COCOINDEX_DATABASE_USER")
+            _load_field(db_kwargs, "password", "COCOINDEX_DATABASE_PASSWORD")
+            database = DatabaseConnectionSpec(**db_kwargs)
+        else:
+            database = None
 
         app_namespace = os.getenv("COCOINDEX_APP_NAMESPACE", "")
 

cocoindex/{storages.py → targets.py}

@@ -1,4 +1,4 @@
-"""All builtin storages."""
+"""All builtin targets."""
 
 from dataclasses import dataclass
 from typing import Sequence
@@ -9,8 +9,8 @@ from .auth_registry import AuthEntryReference
 from .setting import DatabaseConnectionSpec
 
 
-class Postgres(op.StorageSpec):
-    """Storage powered by Postgres and pgvector."""
+class Postgres(op.TargetSpec):
+    """Target powered by Postgres and pgvector."""
 
     database: AuthEntryReference[DatabaseConnectionSpec] | None = None
     table_name: str | None = None
@@ -25,8 +25,8 @@ class QdrantConnection:
 
 
 @dataclass
-class Qdrant(op.StorageSpec):
-    """Storage powered by Qdrant - https://qdrant.tech/."""
+class Qdrant(op.TargetSpec):
+    """Target powered by Qdrant - https://qdrant.tech/."""
 
     collection_name: str
     connection: AuthEntryReference[QdrantConnection] | None = None
@@ -52,7 +52,7 @@ class NodeFromFields:
 
 @dataclass
 class ReferencedNode:
-    """Storage spec for a graph node."""
+    """Target spec for a graph node."""
 
     label: str
     primary_key_fields: Sequence[str]
@@ -95,7 +95,7 @@ class Neo4jConnection:
     db: str | None = None
 
 
-class Neo4j(op.StorageSpec):
+class Neo4j(op.TargetSpec):
     """Graph storage powered by Neo4j."""
 
     connection: AuthEntryReference[Neo4jConnection]
@@ -119,7 +119,7 @@ class KuzuConnection:
     api_server_url: str
 
 
-class Kuzu(op.StorageSpec):
+class Kuzu(op.TargetSpec):
     """Graph storage powered by Kuzu."""
 
     connection: AuthEntryReference[KuzuConnection]