zero-agent-core-embeddings 0.1.0__tar.gz

zero_agent_core_embeddings-0.1.0/.gitignore

@@ -0,0 +1,16 @@
+__pycache__/
+*.py[cod]
+*.egg-info/
+.pytest_cache/
+.ruff_cache/
+.mypy_cache/
+.coverage
+htmlcov/
+dist/
+build/
+
+site/
+*.db
+*.sqlite
+*.sqlite3
+.venv/*

zero_agent_core_embeddings-0.1.0/LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2026 Agent Core contributors
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

zero_agent_core_embeddings-0.1.0/PKG-INFO

@@ -0,0 +1,56 @@
+Metadata-Version: 2.5
+Name: zero-agent-core-embeddings
+Version: 0.1.0
+Summary: Handwritten Agent Core 的 Embedding Provider 扩展
+Project-URL: Homepage, https://github.com/qiuzixu/agent-core
+Project-URL: Repository, https://github.com/qiuzixu/agent-core
+License-Expression: Apache-2.0
+License-File: LICENSE
+Classifier: Development Status :: 3 - Alpha
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Typing :: Typed
+Requires-Python: <3.14,>=3.13
+Requires-Dist: zero-agent-core<0.3,>=0.2.0
+Provides-Extra: all
+Requires-Dist: google-genai<2,>=1.0; extra == 'all'
+Requires-Dist: httpx<1,>=0.28; extra == 'all'
+Requires-Dist: openai<2,>=1.0; extra == 'all'
+Provides-Extra: dev
+Requires-Dist: mypy<2,>=1.18; extra == 'dev'
+Requires-Dist: pytest-asyncio<2,>=1.3; extra == 'dev'
+Requires-Dist: pytest<10,>=9; extra == 'dev'
+Requires-Dist: ruff<1,>=0.14; extra == 'dev'
+Provides-Extra: gemini
+Requires-Dist: google-genai<2,>=1.0; extra == 'gemini'
+Provides-Extra: ollama
+Requires-Dist: httpx<1,>=0.28; extra == 'ollama'
+Provides-Extra: openai
+Requires-Dist: openai<2,>=1.0; extra == 'openai'
+Description-Content-Type: text/markdown
+
+# Handwritten Agent Core Embeddings
+
+`zero-agent-core-embeddings` 为 Agent Core 的 `Embeddings` 协议提供 OpenAI、Gemini 和
+Ollama 实现。Core 的 Retriever 和 VectorStore 不依赖任何厂商 SDK。
+
+```bash
+pip install "zero-agent-core-embeddings[openai]"
+pip install "zero-agent-core-embeddings[gemini]"
+pip install "zero-agent-core-embeddings[ollama]"
+```
+
+```python
+from agent_core_embeddings import OpenAIEmbeddings
+
+embeddings = OpenAIEmbeddings(
+    api_key="从安全配置读取",
+    model="text-embedding-3-small",
+)
+vectors = await embeddings.embed_documents(["第一段", "第二段"])
+query_vector = await embeddings.embed_query("查询内容")
+```
+
+三个适配器共享批处理、并发、有限重试、超时、有限值和维度校验。API Key 和模型选择由应用配置，
+扩展包不读取 `.env`。

zero_agent_core_embeddings-0.1.0/README.md

@@ -0,0 +1,24 @@
+# Handwritten Agent Core Embeddings
+
+`zero-agent-core-embeddings` 为 Agent Core 的 `Embeddings` 协议提供 OpenAI、Gemini 和
+Ollama 实现。Core 的 Retriever 和 VectorStore 不依赖任何厂商 SDK。
+
+```bash
+pip install "zero-agent-core-embeddings[openai]"
+pip install "zero-agent-core-embeddings[gemini]"
+pip install "zero-agent-core-embeddings[ollama]"
+```
+
+```python
+from agent_core_embeddings import OpenAIEmbeddings
+
+embeddings = OpenAIEmbeddings(
+    api_key="从安全配置读取",
+    model="text-embedding-3-small",
+)
+vectors = await embeddings.embed_documents(["第一段", "第二段"])
+query_vector = await embeddings.embed_query("查询内容")
+```
+
+三个适配器共享批处理、并发、有限重试、超时、有限值和维度校验。API Key 和模型选择由应用配置，
+扩展包不读取 `.env`。

zero_agent_core_embeddings-0.1.0/pyproject.toml

@@ -0,0 +1,67 @@
+[build-system]
+requires = ["hatchling>=1.27"]
+build-backend = "hatchling.build"
+
+[project]
+name = "zero-agent-core-embeddings"
+version = "0.1.0"
+description = "Handwritten Agent Core 的 Embedding Provider 扩展"
+readme = "README.md"
+requires-python = ">=3.13,<3.14"
+license = "Apache-2.0"
+dependencies = ["zero-agent-core>=0.2.0,<0.3"]
+classifiers = [
+  "Development Status :: 3 - Alpha",
+  "Operating System :: OS Independent",
+  "Programming Language :: Python :: 3",
+  "Programming Language :: Python :: 3.13",
+  "Typing :: Typed",
+]
+
+[project.urls]
+Homepage = "https://github.com/qiuzixu/agent-core"
+Repository = "https://github.com/qiuzixu/agent-core"
+
+[project.optional-dependencies]
+openai = ["openai>=1.0,<2"]
+gemini = ["google-genai>=1.0,<2"]
+ollama = ["httpx>=0.28,<1"]
+all = [
+  "openai>=1.0,<2",
+  "google-genai>=1.0,<2",
+  "httpx>=0.28,<1",
+]
+dev = [
+  "mypy>=1.18,<2",
+  "pytest>=9,<10",
+  "pytest-asyncio>=1.3,<2",
+  "ruff>=0.14,<1",
+]
+
+[tool.hatch.build.targets.wheel]
+packages = ["src/agent_core_embeddings"]
+
+[tool.uv.sources]
+zero-agent-core = { path = "../..", editable = true }
+
+[tool.pytest.ini_options]
+addopts = "-ra"
+asyncio_mode = "auto"
+testpaths = ["tests"]
+
+[tool.ruff]
+line-length = 110
+target-version = "py313"
+
+[tool.ruff.lint]
+select = ["E", "F", "I", "UP", "B", "ASYNC", "RUF"]
+ignore = ["RUF001", "RUF002", "RUF003"]
+
+[tool.mypy]
+python_version = "3.13"
+strict = true
+files = ["src"]
+
+[[tool.mypy.overrides]]
+module = ["google.*", "httpx", "openai"]
+ignore_missing_imports = true

zero_agent_core_embeddings-0.1.0/src/agent_core_embeddings/__init__.py

@@ -0,0 +1,36 @@
+"""Agent Core 的 Embedding Provider 扩展。"""
+
+from agent_core_embeddings.base import BaseProviderEmbeddings, EmbeddingExecutionPolicy
+from agent_core_embeddings.errors import (
+    EmbeddingAdapterError,
+    EmbeddingConfigurationError,
+    EmbeddingInvocationError,
+    EmbeddingResponseError,
+    UnknownEmbeddingProviderError,
+)
+from agent_core_embeddings.factory import (
+    EmbeddingProviderFactory,
+    EmbeddingProviderRegistry,
+    create_embeddings,
+    default_embedding_provider_registry,
+)
+from agent_core_embeddings.gemini import GeminiEmbeddings
+from agent_core_embeddings.ollama import OllamaEmbeddings
+from agent_core_embeddings.openai import OpenAIEmbeddings
+
+__all__ = [
+    "BaseProviderEmbeddings",
+    "EmbeddingAdapterError",
+    "EmbeddingConfigurationError",
+    "EmbeddingExecutionPolicy",
+    "EmbeddingInvocationError",
+    "EmbeddingProviderFactory",
+    "EmbeddingProviderRegistry",
+    "EmbeddingResponseError",
+    "GeminiEmbeddings",
+    "OllamaEmbeddings",
+    "OpenAIEmbeddings",
+    "UnknownEmbeddingProviderError",
+    "create_embeddings",
+    "default_embedding_provider_registry",
+]

zero_agent_core_embeddings-0.1.0/src/agent_core_embeddings/base.py

@@ -0,0 +1,159 @@
+"""Embedding Provider 共用执行策略和校验。"""
+
+from __future__ import annotations
+
+import asyncio
+import math
+from abc import ABC, abstractmethod
+from collections.abc import Sequence
+from dataclasses import dataclass, replace
+from typing import Literal
+
+from agent_core_embeddings.errors import (
+    EmbeddingConfigurationError,
+    EmbeddingInvocationError,
+    EmbeddingResponseError,
+)
+
+EmbeddingTask = Literal["document", "query"]
+
+
+@dataclass(frozen=True)
+class EmbeddingExecutionPolicy:
+    """跨 Provider 共用的批处理和调用策略。"""
+
+    batch_size: int = 64
+    max_concurrency: int = 2
+    timeout_seconds: float = 60.0
+    max_retries: int = 2
+    retry_base_seconds: float = 0.5
+    expected_dimensions: int | None = None
+
+    def __post_init__(self) -> None:
+        if self.batch_size <= 0:
+            raise EmbeddingConfigurationError("batch_size 必须大于 0")
+        if self.max_concurrency <= 0:
+            raise EmbeddingConfigurationError("max_concurrency 必须大于 0")
+        if self.timeout_seconds <= 0:
+            raise EmbeddingConfigurationError("timeout_seconds 必须大于 0")
+        if self.max_retries < 0:
+            raise EmbeddingConfigurationError("max_retries 不能小于 0")
+        if self.retry_base_seconds < 0:
+            raise EmbeddingConfigurationError("retry_base_seconds 不能小于 0")
+        if self.expected_dimensions is not None and self.expected_dimensions <= 0:
+            raise EmbeddingConfigurationError("expected_dimensions 必须大于 0")
+
+
+class BaseProviderEmbeddings(ABC):
+    """为具体 Provider 提供批处理、重试、超时和向量校验。"""
+
+    provider_name = "base"
+
+    def __init__(self, model: str, *, policy: EmbeddingExecutionPolicy | None = None) -> None:
+        if not model.strip():
+            raise EmbeddingConfigurationError("Embedding model 不能为空")
+        self.model = model
+        self.policy = policy or EmbeddingExecutionPolicy()
+        self._observed_dimensions: int | None = None
+        self._dimension_lock = asyncio.Lock()
+
+    @property
+    def dimensions(self) -> int | None:
+        """返回配置或最近一次成功响应的向量维度。"""
+        return self.policy.expected_dimensions or self._observed_dimensions
+
+    async def embed_documents(self, texts: Sequence[str]) -> list[list[float]]:
+        values = [str(text) for text in texts]
+        if not values:
+            return []
+        return await self._embed_many(values, task="document")
+
+    async def embed_query(self, text: str) -> list[float]:
+        if not text.strip():
+            raise EmbeddingConfigurationError("查询文本不能为空")
+        vectors = await self._embed_many([text], task="query")
+        return vectors[0]
+
+    async def _embed_many(self, texts: list[str], *, task: EmbeddingTask) -> list[list[float]]:
+        batches = [
+            texts[offset : offset + self.policy.batch_size]
+            for offset in range(0, len(texts), self.policy.batch_size)
+        ]
+        semaphore = asyncio.Semaphore(self.policy.max_concurrency)
+
+        async def execute(batch: list[str]) -> list[list[float]]:
+            async with semaphore:
+                return await self._execute_batch(batch, task=task)
+
+        results = await asyncio.gather(*(execute(batch) for batch in batches))
+        vectors = [vector for batch in results for vector in batch]
+        if len(vectors) != len(texts):
+            raise EmbeddingResponseError(
+                f"{self.provider_name} 返回 {len(vectors)} 个向量，期望 {len(texts)} 个"
+            )
+        await self._validate_vectors(vectors)
+        return vectors
+
+    async def _execute_batch(self, texts: list[str], *, task: EmbeddingTask) -> list[list[float]]:
+        last_error: Exception | None = None
+        attempts = self.policy.max_retries + 1
+        for attempt in range(attempts):
+            try:
+                vectors = await asyncio.wait_for(
+                    self._embed_batch(texts, task=task),
+                    timeout=self.policy.timeout_seconds,
+                )
+                if len(vectors) != len(texts):
+                    raise EmbeddingResponseError(
+                        f"{self.provider_name} 当前批次返回 {len(vectors)} 个向量，"
+                        f"期望 {len(texts)} 个"
+                    )
+                return vectors
+            except (EmbeddingConfigurationError, EmbeddingResponseError):
+                raise
+            except Exception as exc:
+                last_error = exc
+                if attempt + 1 >= attempts:
+                    break
+                delay = self.policy.retry_base_seconds * (2**attempt)
+                if delay > 0:
+                    await asyncio.sleep(delay)
+        raise EmbeddingInvocationError(
+            f"{self.provider_name} Embedding 调用失败，已尝试 {attempts} 次"
+        ) from last_error
+
+    async def _validate_vectors(self, vectors: list[list[float]]) -> None:
+        if not vectors:
+            raise EmbeddingResponseError(f"{self.provider_name} 没有返回向量")
+        async with self._dimension_lock:
+            expected = self.policy.expected_dimensions or self._observed_dimensions or len(vectors[0])
+            if expected <= 0:
+                raise EmbeddingResponseError("Embedding 向量不能为空")
+            for vector in vectors:
+                if len(vector) != expected:
+                    raise EmbeddingResponseError(
+                        f"Embedding 向量维度不一致：期望 {expected}，实际 {len(vector)}"
+                    )
+                if not all(math.isfinite(value) for value in vector):
+                    raise EmbeddingResponseError("Embedding 向量包含 NaN 或 Infinity")
+            self._observed_dimensions = expected
+
+    @abstractmethod
+    async def _embed_batch(self, texts: list[str], *, task: EmbeddingTask) -> list[list[float]]:
+        """调用具体 Provider 生成一批向量。"""
+
+
+def policy_with_expected_dimensions(
+    policy: EmbeddingExecutionPolicy | None,
+    dimensions: int | None,
+) -> EmbeddingExecutionPolicy:
+    """把 Provider 的维度参数合并进通用校验策略。"""
+    if dimensions is None:
+        return policy or EmbeddingExecutionPolicy()
+    if dimensions <= 0:
+        raise EmbeddingConfigurationError("dimensions 必须大于 0")
+    if policy is None:
+        return EmbeddingExecutionPolicy(expected_dimensions=dimensions)
+    if policy.expected_dimensions not in {None, dimensions}:
+        raise EmbeddingConfigurationError("dimensions 与 policy.expected_dimensions 不一致")
+    return replace(policy, expected_dimensions=dimensions)

zero_agent_core_embeddings-0.1.0/src/agent_core_embeddings/errors.py

@@ -0,0 +1,23 @@
+"""Embedding 扩展错误。"""
+
+from agent_core import AgentError
+
+
+class EmbeddingAdapterError(AgentError):
+    """Embedding 适配器基础错误。"""
+
+
+class EmbeddingConfigurationError(EmbeddingAdapterError):
+    """Embedding Provider 配置无效。"""
+
+
+class EmbeddingInvocationError(EmbeddingAdapterError):
+    """Embedding Provider 调用失败。"""
+
+
+class EmbeddingResponseError(EmbeddingAdapterError):
+    """Provider 返回的向量数量、维度或数值无效。"""
+
+
+class UnknownEmbeddingProviderError(EmbeddingConfigurationError):
+    """请求了尚未注册的 Embedding Provider。"""

zero_agent_core_embeddings-0.1.0/src/agent_core_embeddings/factory.py

@@ -0,0 +1,72 @@
+"""Embedding Provider 注册表和工厂。"""
+
+from __future__ import annotations
+
+from collections.abc import Callable
+from typing import Any
+
+from agent_core import Embeddings
+
+from agent_core_embeddings.errors import (
+    EmbeddingConfigurationError,
+    UnknownEmbeddingProviderError,
+)
+from agent_core_embeddings.gemini import GeminiEmbeddings
+from agent_core_embeddings.ollama import OllamaEmbeddings
+from agent_core_embeddings.openai import OpenAIEmbeddings
+
+type EmbeddingProviderFactory = Callable[..., Embeddings]
+
+
+def _normalize_provider(name: str) -> str:
+    normalized = name.strip().lower().replace("_", "-")
+    if not normalized:
+        raise EmbeddingConfigurationError("Embedding Provider 名称不能为空")
+    return normalized
+
+
+class EmbeddingProviderRegistry:
+    """保存 Provider 名称与构造函数，不动态导入配置中的类路径。"""
+
+    def __init__(self) -> None:
+        self._factories: dict[str, EmbeddingProviderFactory] = {}
+
+    def register(
+        self,
+        name: str,
+        factory: EmbeddingProviderFactory,
+        *,
+        replace: bool = False,
+    ) -> None:
+        normalized = _normalize_provider(name)
+        if normalized in self._factories and not replace:
+            raise EmbeddingConfigurationError(f"Embedding Provider 已注册：{normalized}")
+        self._factories[normalized] = factory
+
+    def create(self, name: str, **kwargs: Any) -> Embeddings:
+        normalized = _normalize_provider(name)
+        factory = self._factories.get(normalized)
+        if factory is None:
+            raise UnknownEmbeddingProviderError(
+                f"未知 Embedding Provider：{name}；可用值：{', '.join(self.names())}"
+            )
+        return factory(**kwargs)
+
+    def names(self) -> tuple[str, ...]:
+        return tuple(sorted(self._factories))
+
+
+default_embedding_provider_registry = EmbeddingProviderRegistry()
+default_embedding_provider_registry.register("openai", OpenAIEmbeddings)
+default_embedding_provider_registry.register("gemini", GeminiEmbeddings)
+default_embedding_provider_registry.register("ollama", OllamaEmbeddings)
+
+
+def create_embeddings(
+    provider: str,
+    *,
+    registry: EmbeddingProviderRegistry | None = None,
+    **kwargs: Any,
+) -> Embeddings:
+    """按注册名创建符合 Core ``Embeddings`` 协议的适配器。"""
+    return (registry or default_embedding_provider_registry).create(provider, **kwargs)

zero_agent_core_embeddings-0.1.0/src/agent_core_embeddings/gemini.py

@@ -0,0 +1,72 @@
+"""Gemini Embedding 适配器。"""
+
+from __future__ import annotations
+
+from typing import Any
+
+from agent_core_embeddings.base import (
+    BaseProviderEmbeddings,
+    EmbeddingExecutionPolicy,
+    EmbeddingTask,
+    policy_with_expected_dimensions,
+)
+from agent_core_embeddings.errors import EmbeddingConfigurationError, EmbeddingResponseError
+
+
+class GeminiEmbeddings(BaseProviderEmbeddings):
+    """使用 Google Gen AI SDK 生成文档和查询向量。"""
+
+    provider_name = "gemini"
+
+    def __init__(
+        self,
+        *,
+        api_key: str | None = None,
+        model: str = "gemini-embedding-001",
+        dimensions: int | None = None,
+        policy: EmbeddingExecutionPolicy | None = None,
+        client: Any | None = None,
+    ) -> None:
+        effective_policy = policy_with_expected_dimensions(policy, dimensions)
+        super().__init__(model, policy=effective_policy)
+        self._requested_dimensions = dimensions
+        if client is None:
+            try:
+                from google import genai
+            except ImportError as exc:
+                raise ImportError(
+                    "GeminiEmbeddings 需要 google-genai，请安装 "
+                    "zero-agent-core-embeddings[gemini]"
+                ) from exc
+            client = genai.Client(api_key=api_key)
+        aio = getattr(client, "aio", None)
+        models: Any = getattr(aio, "models", None)
+        if models is None:
+            raise EmbeddingConfigurationError("Gemini client 必须提供 aio.models")
+        self._models: Any = models
+
+    async def _embed_batch(self, texts: list[str], *, task: EmbeddingTask) -> list[list[float]]:
+        config: dict[str, Any] = {
+            "task_type": "RETRIEVAL_DOCUMENT" if task == "document" else "RETRIEVAL_QUERY"
+        }
+        if self._requested_dimensions is not None:
+            config["output_dimensionality"] = self._requested_dimensions
+        response = await self._models.embed_content(
+            model=self.model,
+            contents=texts,
+            config=config,
+        )
+        embeddings = _read_value(response, "embeddings")
+        if not isinstance(embeddings, (list, tuple)):
+            raise EmbeddingResponseError("Gemini 响应缺少 embeddings 数组")
+        vectors: list[list[float]] = []
+        for item in embeddings:
+            values = _read_value(item, "values")
+            if not isinstance(values, (list, tuple)):
+                raise EmbeddingResponseError("Gemini 响应缺少 values 数组")
+            vectors.append([float(value) for value in values])
+        return vectors
+def _read_value(value: Any, key: str, default: Any = None) -> Any:
+    if isinstance(value, dict):
+        return value.get(key, default)
+    return getattr(value, key, default)

zero_agent_core_embeddings-0.1.0/src/agent_core_embeddings/ollama.py

@@ -0,0 +1,92 @@
+"""Ollama Embedding 适配器。"""
+
+from __future__ import annotations
+
+from collections.abc import Mapping
+from typing import Any
+
+from agent_core_embeddings.base import BaseProviderEmbeddings, EmbeddingExecutionPolicy, EmbeddingTask
+from agent_core_embeddings.errors import EmbeddingInvocationError, EmbeddingResponseError
+
+
+class OllamaEmbeddings(BaseProviderEmbeddings):
+    """调用 Ollama 原生 ``/api/embed`` 接口。"""
+
+    provider_name = "ollama"
+
+    def __init__(
+        self,
+        *,
+        model: str = "nomic-embed-text",
+        base_url: str = "http://127.0.0.1:11434",
+        api_key: str | None = None,
+        truncate: bool = True,
+        keep_alive: str | int | None = None,
+        policy: EmbeddingExecutionPolicy | None = None,
+        client: Any | None = None,
+    ) -> None:
+        super().__init__(model, policy=policy)
+        self._endpoint = f"{base_url.rstrip('/')}/api/embed"
+        self._api_key = api_key
+        self._truncate = truncate
+        self._keep_alive = keep_alive
+        self._client = client
+
+    async def _embed_batch(self, texts: list[str], *, task: EmbeddingTask) -> list[list[float]]:
+        del task
+        payload: dict[str, Any] = {
+            "model": self.model,
+            "input": texts,
+            "truncate": self._truncate,
+        }
+        if self._keep_alive is not None:
+            payload["keep_alive"] = self._keep_alive
+        headers = {"Accept": "application/json"}
+        if self._api_key:
+            headers["Authorization"] = f"Bearer {self._api_key}"
+
+        if self._client is not None:
+            return await self._post(self._client, payload=payload, headers=headers)
+
+        try:
+            import httpx
+        except ImportError as exc:
+            raise ImportError(
+                "OllamaEmbeddings 需要 httpx，请安装 "
+                "zero-agent-core-embeddings[ollama]"
+            ) from exc
+        async with httpx.AsyncClient() as client:
+            return await self._post(client, payload=payload, headers=headers)
+
+    async def _post(
+        self,
+        client: Any,
+        *,
+        payload: Mapping[str, Any],
+        headers: Mapping[str, str],
+    ) -> list[list[float]]:
+        response = await client.post(
+            self._endpoint,
+            json=dict(payload),
+            headers=dict(headers),
+            timeout=self.policy.timeout_seconds,
+        )
+        status_code = int(getattr(response, "status_code", 0))
+        if status_code < 200 or status_code >= 300:
+            body = str(getattr(response, "text", ""))[:500]
+            raise EmbeddingInvocationError(f"Ollama 返回 HTTP {status_code}：{body}")
+        try:
+            value = response.json()
+        except Exception as exc:
+            raise EmbeddingResponseError("Ollama 返回了无效 JSON") from exc
+        if not isinstance(value, Mapping):
+            raise EmbeddingResponseError("Ollama 响应根节点必须是对象")
+        embeddings = value.get("embeddings")
+        if not isinstance(embeddings, list):
+            raise EmbeddingResponseError("Ollama 响应缺少 embeddings 数组")
+        vectors: list[list[float]] = []
+        for item in embeddings:
+            if not isinstance(item, (list, tuple)):
+                raise EmbeddingResponseError("Ollama embedding 必须是数组")
+            vectors.append([float(number) for number in item])
+        return vectors

zero_agent_core_embeddings-0.1.0/src/agent_core_embeddings/openai.py

@@ -0,0 +1,77 @@
+"""OpenAI Embedding 适配器。"""
+
+from __future__ import annotations
+
+from typing import Any
+
+from agent_core_embeddings.base import (
+    BaseProviderEmbeddings,
+    EmbeddingExecutionPolicy,
+    EmbeddingTask,
+    policy_with_expected_dimensions,
+)
+from agent_core_embeddings.errors import EmbeddingResponseError
+
+
+class OpenAIEmbeddings(BaseProviderEmbeddings):
+    """使用 OpenAI Embeddings API 生成文档和查询向量。"""
+
+    provider_name = "openai"
+
+    def __init__(
+        self,
+        *,
+        api_key: str | None = None,
+        model: str = "text-embedding-3-small",
+        base_url: str | None = None,
+        dimensions: int | None = None,
+        policy: EmbeddingExecutionPolicy | None = None,
+        client: Any | None = None,
+    ) -> None:
+        effective_policy = policy_with_expected_dimensions(policy, dimensions)
+        super().__init__(model, policy=effective_policy)
+        self._requested_dimensions = dimensions
+        if client is None:
+            try:
+                from openai import AsyncOpenAI
+            except ImportError as exc:
+                raise ImportError(
+                    "OpenAIEmbeddings 需要 openai，请安装 "
+                    "zero-agent-core-embeddings[openai]"
+                ) from exc
+            kwargs: dict[str, Any] = {}
+            if api_key is not None:
+                kwargs["api_key"] = api_key
+            if base_url is not None:
+                kwargs["base_url"] = base_url
+            client = AsyncOpenAI(**kwargs)
+        self._client = client
+
+    async def _embed_batch(self, texts: list[str], *, task: EmbeddingTask) -> list[list[float]]:
+        del task
+        kwargs: dict[str, Any] = {
+            "model": self.model,
+            "input": texts,
+            "encoding_format": "float",
+        }
+        if self._requested_dimensions is not None:
+            kwargs["dimensions"] = self._requested_dimensions
+        response = await self._client.embeddings.create(**kwargs)
+        data = _read_value(response, "data")
+        if not isinstance(data, list):
+            try:
+                data = list(data)
+            except TypeError as exc:
+                raise EmbeddingResponseError("OpenAI 响应缺少 data 数组") from exc
+        ordered = sorted(data, key=lambda item: int(_read_value(item, "index", 0)))
+        vectors: list[list[float]] = []
+        for item in ordered:
+            embedding = _read_value(item, "embedding")
+            if not isinstance(embedding, (list, tuple)):
+                raise EmbeddingResponseError("OpenAI 响应缺少 embedding 数组")
+            vectors.append([float(value) for value in embedding])
+        return vectors
+def _read_value(value: Any, key: str, default: Any = None) -> Any:
+    if isinstance(value, dict):
+        return value.get(key, default)
+    return getattr(value, key, default)

zero_agent_core_embeddings-0.1.0/src/agent_core_embeddings/py.typed

@@ -0,0 +1 @@
+

zero_agent_core_embeddings-0.1.0/tests/test_adapters.py

@@ -0,0 +1,234 @@
+"""Embedding Provider 扩展契约测试。"""
+
+from __future__ import annotations
+
+import asyncio
+import math
+import unittest
+from types import SimpleNamespace
+from typing import Any
+
+from agent_core import Embeddings
+
+from agent_core_embeddings import (
+    BaseProviderEmbeddings,
+    EmbeddingConfigurationError,
+    EmbeddingExecutionPolicy,
+    EmbeddingProviderRegistry,
+    EmbeddingResponseError,
+    GeminiEmbeddings,
+    OllamaEmbeddings,
+    OpenAIEmbeddings,
+    UnknownEmbeddingProviderError,
+    create_embeddings,
+)
+from agent_core_embeddings.base import EmbeddingTask
+
+
+class _OpenAIEndpoint:
+    def __init__(self) -> None:
+        self.calls: list[dict[str, Any]] = []
+
+    async def create(self, **kwargs: Any) -> Any:
+        self.calls.append(kwargs)
+        data = [
+            SimpleNamespace(index=index, embedding=[float(len(text)), float(index), 1.0])
+            for index, text in enumerate(kwargs["input"])
+        ]
+        return SimpleNamespace(data=list(reversed(data)))
+
+
+class _OpenAIClient:
+    def __init__(self) -> None:
+        self.embeddings = _OpenAIEndpoint()
+
+
+class _GeminiModels:
+    def __init__(self) -> None:
+        self.calls: list[dict[str, Any]] = []
+
+    async def embed_content(self, **kwargs: Any) -> Any:
+        self.calls.append(kwargs)
+        return SimpleNamespace(
+            embeddings=[
+                SimpleNamespace(values=[float(len(text)), 1.0, 0.0])
+                for text in kwargs["contents"]
+            ]
+        )
+
+
+class _GeminiClient:
+    def __init__(self) -> None:
+        self.models = _GeminiModels()
+        self.aio = SimpleNamespace(models=self.models)
+
+
+class _HttpResponse:
+    def __init__(self, payload: object, *, status_code: int = 200, text: str = "") -> None:
+        self.payload = payload
+        self.status_code = status_code
+        self.text = text
+
+    def json(self) -> object:
+        return self.payload
+
+
+class _HttpClient:
+    def __init__(self, response: _HttpResponse) -> None:
+        self.response = response
+        self.calls: list[dict[str, Any]] = []
+
+    async def post(self, url: str, **kwargs: Any) -> _HttpResponse:
+        self.calls.append({"url": url, **kwargs})
+        return self.response
+
+
+class _FlakyEmbeddings(BaseProviderEmbeddings):
+    provider_name = "flaky"
+
+    def __init__(self, failures: int, vectors: list[list[float]], **kwargs: Any) -> None:
+        super().__init__("fake", **kwargs)
+        self.failures = failures
+        self.vectors = vectors
+        self.calls = 0
+
+    async def _embed_batch(self, texts: list[str], *, task: EmbeddingTask) -> list[list[float]]:
+        del texts, task
+        self.calls += 1
+        if self.calls <= self.failures:
+            raise RuntimeError("temporary")
+        return self.vectors
+
+
+class _SlowEmbeddings(BaseProviderEmbeddings):
+    provider_name = "slow"
+
+    async def _embed_batch(self, texts: list[str], *, task: EmbeddingTask) -> list[list[float]]:
+        del texts, task
+        await asyncio.sleep(0.1)
+        return [[1.0]]
+
+
+class ProviderAdapterTests(unittest.IsolatedAsyncioTestCase):
+    async def test_openai_batches_and_restores_response_order(self) -> None:
+        client = _OpenAIClient()
+        embeddings = OpenAIEmbeddings(
+            client=client,
+            dimensions=3,
+            policy=EmbeddingExecutionPolicy(
+                batch_size=2,
+                max_concurrency=1,
+                expected_dimensions=3,
+            ),
+        )
+
+        vectors = await embeddings.embed_documents(["a", "bb", "ccc"])
+
+        self.assertEqual(vectors, [[1.0, 0.0, 1.0], [2.0, 1.0, 1.0], [3.0, 0.0, 1.0]])
+        self.assertEqual(len(client.embeddings.calls), 2)
+        self.assertTrue(all(call["dimensions"] == 3 for call in client.embeddings.calls))
+        self.assertIsInstance(embeddings, Embeddings)
+
+    async def test_gemini_distinguishes_document_and_query_tasks(self) -> None:
+        client = _GeminiClient()
+        embeddings = GeminiEmbeddings(client=client, dimensions=3)
+
+        documents = await embeddings.embed_documents(["文档"])
+        query = await embeddings.embed_query("查询")
+
+        self.assertEqual(len(documents[0]), 3)
+        self.assertEqual(len(query), 3)
+        self.assertEqual(
+            client.models.calls[0]["config"]["task_type"],
+            "RETRIEVAL_DOCUMENT",
+        )
+        self.assertEqual(
+            client.models.calls[1]["config"]["task_type"],
+            "RETRIEVAL_QUERY",
+        )
+
+    async def test_ollama_uses_native_embed_endpoint(self) -> None:
+        client = _HttpClient(_HttpResponse({"embeddings": [[1, 0], [0, 1]]}))
+        embeddings = OllamaEmbeddings(
+            client=client,
+            model="nomic-embed-text",
+            base_url="http://ollama.local/",
+            api_key="secret",
+            policy=EmbeddingExecutionPolicy(expected_dimensions=2),
+        )
+
+        vectors = await embeddings.embed_documents(["one", "two"])
+
+        self.assertEqual(vectors, [[1.0, 0.0], [0.0, 1.0]])
+        call = client.calls[0]
+        self.assertEqual(call["url"], "http://ollama.local/api/embed")
+        self.assertEqual(call["json"]["input"], ["one", "two"])
+        self.assertEqual(call["headers"]["Authorization"], "Bearer secret")
+
+    async def test_retry_and_observed_dimensions(self) -> None:
+        embeddings = _FlakyEmbeddings(
+            1,
+            [[1.0, 2.0]],
+            policy=EmbeddingExecutionPolicy(max_retries=1, retry_base_seconds=0),
+        )
+
+        vector = await embeddings.embed_query("query")
+
+        self.assertEqual(vector, [1.0, 2.0])
+        self.assertEqual(embeddings.calls, 2)
+        self.assertEqual(embeddings.dimensions, 2)
+
+    async def test_invalid_vector_and_timeout_are_rejected(self) -> None:
+        invalid = _FlakyEmbeddings(0, [[math.nan]])
+        slow = _SlowEmbeddings(
+            "slow",
+            policy=EmbeddingExecutionPolicy(timeout_seconds=0.01, max_retries=0),
+        )
+
+        with self.assertRaises(EmbeddingResponseError):
+            await invalid.embed_query("query")
+        with self.assertRaises(Exception) as caught:
+            await slow.embed_query("query")
+        self.assertEqual(caught.exception.__class__.__name__, "EmbeddingInvocationError")
+
+    async def test_dimension_mismatch_is_rejected(self) -> None:
+        embeddings = _FlakyEmbeddings(
+            0,
+            [[1.0, 2.0]],
+            policy=EmbeddingExecutionPolicy(expected_dimensions=3),
+        )
+
+        with self.assertRaises(EmbeddingResponseError):
+            await embeddings.embed_query("query")
+
+        with self.assertRaises(EmbeddingConfigurationError):
+            OpenAIEmbeddings(
+                client=_OpenAIClient(),
+                dimensions=2,
+                policy=EmbeddingExecutionPolicy(expected_dimensions=3),
+            )
+
+
+class FactoryTests(unittest.TestCase):
+    def test_default_factory_and_custom_registry(self) -> None:
+        adapter = create_embeddings("OPENAI", client=_OpenAIClient())
+        self.assertIsInstance(adapter, OpenAIEmbeddings)
+
+        registry = EmbeddingProviderRegistry()
+        registry.register("custom", lambda **_: _FlakyEmbeddings(0, [[1.0]]))
+        custom = registry.create("custom")
+        self.assertIsInstance(custom, Embeddings)
+
+    def test_duplicate_unknown_and_invalid_policy_are_explicit(self) -> None:
+        registry = EmbeddingProviderRegistry()
+        registry.register("custom", lambda **_: _FlakyEmbeddings(0, [[1.0]]))
+        with self.assertRaises(EmbeddingConfigurationError):
+            registry.register("custom", lambda **_: _FlakyEmbeddings(0, [[1.0]]))
+        with self.assertRaises(UnknownEmbeddingProviderError):
+            registry.create("missing")
+        with self.assertRaises(EmbeddingConfigurationError):
+            EmbeddingExecutionPolicy(batch_size=0)
+
+
+if __name__ == "__main__":
+    unittest.main()