polyllm-anthropic 0.1.0__tar.gz

polyllm_anthropic-0.1.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 PolyLLM contributors
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

polyllm_anthropic-0.1.0/PKG-INFO

@@ -0,0 +1,35 @@
+Metadata-Version: 2.4
+Name: polyllm-anthropic
+Version: 0.1.0
+Summary: Anthropic plugin for PolyLLM.
+Author: PolyLLM contributors
+License-Expression: MIT
+Project-URL: Homepage, https://polyllm.raingpt.top
+Project-URL: Repository, https://github.com/you-want/PolyLLM
+Project-URL: Issues, https://github.com/you-want/PolyLLM/issues
+Keywords: llm,polyllm,anthropic,claude
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
+Classifier: Topic :: Software Development :: Libraries :: Python Modules
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: polyllm-core>=0.1.0
+Dynamic: license-file
+
+# PolyLLM Anthropic Plugin for Python
+
+```python
+from polyllm_core import create_llm
+from polyllm_anthropic import AnthropicPlugin
+
+llm = create_llm(plugins=[AnthropicPlugin()], providers={"anthropic": {"api_key": "..."}})
+response = await llm.chat("anthropic:claude-haiku-4-5", messages=[{"role": "user", "content": "你好"}])
+```

polyllm_anthropic-0.1.0/README.md

@@ -0,0 +1,9 @@
+# PolyLLM Anthropic Plugin for Python
+
+```python
+from polyllm_core import create_llm
+from polyllm_anthropic import AnthropicPlugin
+
+llm = create_llm(plugins=[AnthropicPlugin()], providers={"anthropic": {"api_key": "..."}})
+response = await llm.chat("anthropic:claude-haiku-4-5", messages=[{"role": "user", "content": "你好"}])
+```

polyllm_anthropic-0.1.0/pyproject.toml

@@ -0,0 +1,38 @@
+[build-system]
+requires = ["setuptools>=68"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "polyllm-anthropic"
+version = "0.1.0"
+description = "Anthropic plugin for PolyLLM."
+readme = "README.md"
+requires-python = ">=3.10"
+dependencies = ["polyllm-core>=0.1.0"]
+license = "MIT"
+license-files = ["LICENSE"]
+authors = [{ name = "PolyLLM contributors" }]
+keywords = ["llm", "polyllm", "anthropic", "claude"]
+classifiers = [
+  "Development Status :: 3 - Alpha",
+  "Intended Audience :: Developers",
+  "Operating System :: OS Independent",
+  "Programming Language :: Python :: 3",
+  "Programming Language :: Python :: 3.10",
+  "Programming Language :: Python :: 3.11",
+  "Programming Language :: Python :: 3.12",
+  "Programming Language :: Python :: 3.13",
+  "Programming Language :: Python :: 3.14",
+  "Topic :: Software Development :: Libraries :: Python Modules",
+]
+
+[project.urls]
+Homepage = "https://polyllm.raingpt.top"
+Repository = "https://github.com/you-want/PolyLLM"
+Issues = "https://github.com/you-want/PolyLLM/issues"
+
+[tool.setuptools.packages.find]
+where = ["src"]
+
+[tool.setuptools.package-data]
+polyllm_anthropic = ["py.typed"]

polyllm_anthropic-0.1.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

polyllm_anthropic-0.1.0/src/polyllm_anthropic/__init__.py

@@ -0,0 +1,4 @@
+from .plugin import AnthropicPlugin, parse_sse
+from .specs import models
+
+__all__ = ["AnthropicPlugin", "models", "parse_sse"]

polyllm_anthropic-0.1.0/src/polyllm_anthropic/plugin.py

@@ -0,0 +1,235 @@
+from __future__ import annotations
+
+import asyncio
+import json
+from typing import Any, AsyncIterator, Callable, cast
+from urllib.error import HTTPError
+from urllib.request import Request, urlopen
+
+from polyllm_core.config import resolve_provider_runtime
+from polyllm_core.errors import PolyLLMError
+from polyllm_core.types import (
+    ChatContext,
+    FinishReason,
+    LLMPlugin,
+    ModelSpec,
+    ProviderConfig,
+    ProviderModelInfo,
+    UnifiedChoice,
+    UnifiedMessage,
+    UnifiedRequest,
+    UnifiedResponse,
+    UnifiedStreamChunk,
+    UnifiedToolCall,
+    UnifiedUsage,
+)
+
+from .specs import capabilities, models
+
+Transport = Callable[[str, str, dict[str, str], dict[str, Any] | None, int], tuple[int, str]]
+
+
+class AnthropicPlugin(LLMPlugin):
+    provider = "anthropic"
+    models = models
+
+    def __init__(self, transport: Transport | None = None):
+        self._transport = transport or urllib_transport
+
+    async def chat(
+        self,
+        request: UnifiedRequest,
+        config: ProviderConfig,
+        context: ChatContext,
+    ) -> UnifiedResponse:
+        status, body = await self._request("POST", "/v1/messages", self._build_body(request), config, context)
+        data = self._parse_json(status, body)
+        content = data.get("content", [])
+        text = "".join(part.get("text", "") for part in content if part.get("type") == "text")
+        tool_calls = tuple(
+            UnifiedToolCall(
+                id=part["id"],
+                type="function",
+                function={"name": part["name"], "arguments": json.dumps(part.get("input", {}), ensure_ascii=False)},
+            )
+            for part in content
+            if part.get("type") == "tool_use"
+        )
+        finish_reason: FinishReason = (
+            "tool_calls" if tool_calls else cast(FinishReason, data.get("stop_reason", "other"))
+        )
+        usage = data.get("usage", {})
+        return UnifiedResponse(
+            id=data["id"],
+            model=data["model"],
+            choices=[
+                UnifiedChoice(
+                    index=0,
+                    message=UnifiedMessage(role="assistant", content=text, tool_calls=tool_calls),
+                    finish_reason=finish_reason,
+                )
+            ],
+            usage=UnifiedUsage(
+                prompt_tokens=usage.get("input_tokens", 0),
+                completion_tokens=usage.get("output_tokens", 0),
+                total_tokens=usage.get("input_tokens", 0) + usage.get("output_tokens", 0),
+            ),
+            provider_metadata=data.get("metadata"),
+        )
+
+    async def chat_stream(
+        self,
+        request: UnifiedRequest,
+        config: ProviderConfig,
+        context: ChatContext,
+    ) -> AsyncIterator[UnifiedStreamChunk]:
+        body = self._build_body(request)
+        body["stream"] = True
+        status, text = await self._request("POST", "/v1/messages", body, config, context)
+        response_id = "anthropic-stream"
+        model = request.model
+        if status < 200 or status >= 300:
+            raise PolyLLMError("PROVIDER_ERROR", f"Anthropic API 错误: {status} {text}", {"status": status})
+        for event in parse_sse(text):
+            event_type = event.get("type")
+            if event_type == "content_block_delta":
+                delta = event.get("delta", {})
+                yield UnifiedStreamChunk(id=response_id, model=model, delta={"content": delta.get("text", "")})
+            elif event_type == "message_delta":
+                yield UnifiedStreamChunk(
+                    id=response_id,
+                    model=model,
+                    delta={},
+                    finish_reason=cast(
+                        FinishReason,
+                        event.get("delta", {}).get("stop_reason", "other"),
+                    ),
+                )
+
+    async def list_models(self, config: ProviderConfig) -> list[ProviderModelInfo]:
+        status, body = await self._request("GET", "/v1/models", None, config, None)
+        data = self._parse_json(status, body)
+        return [
+            ProviderModelInfo(
+                id=item["id"],
+                display_name=item.get("display_name"),
+                metadata=item.get("metadata"),
+            )
+            for item in data.get("data", [])
+        ]
+
+    def create_model_spec(self, model_id: str) -> ModelSpec:
+        return ModelSpec(
+            id=model_id,
+            provider=self.provider,
+            capabilities=capabilities,
+            context_window=200000,
+            max_output=8192,
+        )
+
+    def _build_body(self, request: UnifiedRequest) -> dict[str, Any]:
+        system_parts = [message.content for message in request.messages if message.role == "system"]
+        messages = [
+            {
+                "role": message.role,
+                "content": message.content,
+                **({"tool_call_id": message.tool_call_id} if message.tool_call_id else {}),
+                **(
+                    {
+                        "tool_calls": [
+                            {
+                                "id": call.id,
+                                "type": "function",
+                                "function": dict(call.function),
+                            }
+                            for call in message.tool_calls
+                        ]
+                    }
+                    if message.tool_calls
+                    else {}
+                ),
+            }
+            for message in request.messages
+            if message.role in ("user", "assistant", "tool")
+        ]
+        body: dict[str, Any] = {
+            "model": request.model,
+            "messages": messages,
+            "max_tokens": request.max_tokens or 4096,
+        }
+        if system_parts:
+            body["system"] = "\n".join(part for part in system_parts if isinstance(part, str))
+        if request.temperature is not None:
+            body["temperature"] = request.temperature
+        if request.top_p is not None:
+            body["top_p"] = request.top_p
+        if request.tools:
+            body["tools"] = [
+                {
+                    "name": tool.function["name"],
+                    "description": tool.function.get("description", ""),
+                    "input_schema": tool.function.get("parameters", {"type": "object"}),
+                }
+                for tool in request.tools
+            ]
+        return body
+
+    async def _request(
+        self,
+        method: str,
+        path: str,
+        payload: dict[str, Any] | None,
+        config: ProviderConfig,
+        context: ChatContext | None,
+    ) -> tuple[int, str]:
+        runtime = resolve_provider_runtime(self.provider, config)
+        base_url = (runtime.base_url or "https://api.anthropic.com").rstrip("/")
+        headers = {
+            "x-api-key": runtime.api_key,
+            "anthropic-version": "2023-06-01",
+            **runtime.headers,
+        }
+        if payload is not None:
+            headers["Content-Type"] = "application/json"
+        timeout_ms = context.timeout_ms if context is not None and context.timeout_ms is not None else 120000
+        return await asyncio.to_thread(
+            self._transport, method, f"{base_url}{path}", headers, payload, timeout_ms
+        )
+
+    @staticmethod
+    def _parse_json(status: int, body: str) -> dict[str, Any]:
+        try:
+            data = cast(dict[str, Any], json.loads(body))
+        except json.JSONDecodeError as error:
+            raise PolyLLMError("PROVIDER_ERROR", f"Anthropic 返回无效 JSON: {body[:200]}", {"status": status}) from error
+        if status < 200 or status >= 300:
+            raise PolyLLMError("PROVIDER_ERROR", f"Anthropic API 错误: {status} {body}", {"status": status, "body": data})
+        return data
+
+
+def urllib_transport(
+    method: str,
+    url: str,
+    headers: dict[str, str],
+    payload: dict[str, Any] | None,
+    timeout_ms: int,
+) -> tuple[int, str]:
+    data = json.dumps(payload).encode("utf-8") if payload is not None else None
+    request = Request(url, data=data, headers=headers, method=method)
+    try:
+        with urlopen(request, timeout=timeout_ms / 1000) as response:
+            return response.status, response.read().decode("utf-8")
+    except HTTPError as error:
+        return error.code, error.read().decode("utf-8")
+
+
+def parse_sse(text: str) -> list[dict[str, Any]]:
+    events: list[dict[str, Any]] = []
+    for line in text.splitlines():
+        if not line.startswith("data:"):
+            continue
+        value = line.removeprefix("data:").strip()
+        if not value:
+            continue
+        events.append(cast(dict[str, Any], json.loads(value)))
+    return events

polyllm_anthropic-0.1.0/src/polyllm_anthropic/specs.py

@@ -0,0 +1,43 @@
+from polyllm_core.types import ModelCapabilities, ModelSpec, NumericParamRange
+
+capabilities = ModelCapabilities(
+    streaming=True,
+    vision=True,
+    function_calling=True,
+    json_mode=False,
+    system_prompt=True,
+)
+
+models = (
+    ModelSpec(
+        id="claude-sonnet-4-5",
+        provider="anthropic",
+        aliases=("claude-sonnet", "sonnet"),
+        version="2025-09-29",
+        snapshot="claude-sonnet-4-5-20250929",
+        is_latest=True,
+        capabilities=capabilities,
+        params={
+            "temperature": NumericParamRange(0, 1, 1),
+            "max_tokens": NumericParamRange(1, 8192, 4096, field="max_tokens"),
+            "top_p": NumericParamRange(0, 1, 1),
+        },
+        context_window=200000,
+        max_output=8192,
+    ),
+    ModelSpec(
+        id="claude-haiku-4-5",
+        provider="anthropic",
+        aliases=("claude-haiku", "haiku"),
+        version="2025-10-01",
+        snapshot="claude-haiku-4-5-20251001",
+        capabilities=capabilities,
+        params={
+            "temperature": NumericParamRange(0, 1, 1),
+            "max_tokens": NumericParamRange(1, 8192, 4096, field="max_tokens"),
+            "top_p": NumericParamRange(0, 1, 1),
+        },
+        context_window=200000,
+        max_output=8192,
+    ),
+)

polyllm_anthropic-0.1.0/src/polyllm_anthropic.egg-info/PKG-INFO

@@ -0,0 +1,35 @@
+Metadata-Version: 2.4
+Name: polyllm-anthropic
+Version: 0.1.0
+Summary: Anthropic plugin for PolyLLM.
+Author: PolyLLM contributors
+License-Expression: MIT
+Project-URL: Homepage, https://polyllm.raingpt.top
+Project-URL: Repository, https://github.com/you-want/PolyLLM
+Project-URL: Issues, https://github.com/you-want/PolyLLM/issues
+Keywords: llm,polyllm,anthropic,claude
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
+Classifier: Topic :: Software Development :: Libraries :: Python Modules
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: polyllm-core>=0.1.0
+Dynamic: license-file
+
+# PolyLLM Anthropic Plugin for Python
+
+```python
+from polyllm_core import create_llm
+from polyllm_anthropic import AnthropicPlugin
+
+llm = create_llm(plugins=[AnthropicPlugin()], providers={"anthropic": {"api_key": "..."}})
+response = await llm.chat("anthropic:claude-haiku-4-5", messages=[{"role": "user", "content": "你好"}])
+```

polyllm_anthropic-0.1.0/src/polyllm_anthropic.egg-info/SOURCES.txt

@@ -0,0 +1,14 @@
+LICENSE
+README.md
+pyproject.toml
+src/polyllm_anthropic/__init__.py
+src/polyllm_anthropic/plugin.py
+src/polyllm_anthropic/py.typed
+src/polyllm_anthropic/specs.py
+src/polyllm_anthropic.egg-info/PKG-INFO
+src/polyllm_anthropic.egg-info/SOURCES.txt
+src/polyllm_anthropic.egg-info/dependency_links.txt
+src/polyllm_anthropic.egg-info/requires.txt
+src/polyllm_anthropic.egg-info/top_level.txt
+tests/test_anthropic.py
+tests/test_stream.py

polyllm_anthropic-0.1.0/src/polyllm_anthropic.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

polyllm_anthropic-0.1.0/src/polyllm_anthropic.egg-info/requires.txt

@@ -0,0 +1 @@
+polyllm-core>=0.1.0

polyllm_anthropic-0.1.0/src/polyllm_anthropic.egg-info/top_level.txt

@@ -0,0 +1 @@
+polyllm_anthropic

polyllm_anthropic-0.1.0/tests/test_anthropic.py

@@ -0,0 +1,58 @@
+import asyncio
+import json
+import unittest
+
+from polyllm_core.types import ProviderConfig, UnifiedMessage, UnifiedRequest
+from polyllm_anthropic import AnthropicPlugin
+from polyllm_core import ModelRegistry
+
+
+class AnthropicTests(unittest.TestCase):
+    def test_registry_aliases_and_snapshots(self):
+        registry = ModelRegistry()
+        for spec in AnthropicPlugin.models:
+            registry.register(spec)
+        self.assertEqual(registry.resolve("anthropic:sonnet").id, "claude-sonnet-4-5")
+        self.assertEqual(registry.resolve("anthropic:haiku").id, "claude-haiku-4-5")
+        self.assertEqual(
+            registry.resolve("anthropic:claude-sonnet-4-5-20250929").id,
+            "claude-sonnet-4-5",
+        )
+
+    def test_chat_wire_and_response(self):
+        calls = []
+
+        def transport(method, url, headers, payload, timeout):
+            calls.append((method, url, headers, payload))
+            return 200, json.dumps(
+                {
+                    "id": "msg",
+                    "model": "claude-haiku-4-5",
+                    "content": [{"type": "text", "text": "ok"}],
+                    "stop_reason": "end_turn",
+                    "usage": {"input_tokens": 2, "output_tokens": 3},
+                }
+            )
+
+        response = asyncio.run(
+            AnthropicPlugin(transport).chat(
+                UnifiedRequest(
+                    model="claude-haiku-4-5",
+                    messages=[
+                        UnifiedMessage(role="system", content="be brief"),
+                        UnifiedMessage(role="user", content="hi"),
+                    ],
+                    max_tokens=128,
+                ),
+                ProviderConfig(api_key="k"),
+                None,
+            )
+        )
+        self.assertEqual(response.choices[0].message.content, "ok")
+        self.assertEqual(response.usage.total_tokens, 5)
+        self.assertEqual(calls[0][3]["system"], "be brief")
+        self.assertEqual(calls[0][2]["anthropic-version"], "2023-06-01")
+
+
+if __name__ == "__main__":
+    unittest.main()

polyllm_anthropic-0.1.0/tests/test_stream.py

@@ -0,0 +1,33 @@
+import asyncio
+import unittest
+
+from polyllm_anthropic import AnthropicPlugin
+from polyllm_core.types import ProviderConfig, UnifiedMessage, UnifiedRequest
+
+
+class AnthropicStreamTests(unittest.TestCase):
+    def test_stream_parses_sse(self):
+        sse = (
+            'data: {"type":"content_block_delta","delta":{"type":"text_delta","text":"he"}}\n\n'
+            'data: {"type":"content_block_delta","delta":{"type":"text_delta","text":"llo"}}\n\n'
+            'data: {"type":"message_delta","delta":{"stop_reason":"end_turn"}}\n\n'
+        )
+
+        async def run():
+            chunks = []
+            plugin = AnthropicPlugin(lambda *args: (200, sse))
+            async for chunk in plugin.chat_stream(
+                UnifiedRequest(model="claude-haiku-4-5", messages=[UnifiedMessage(role="user", content="hi")]),
+                ProviderConfig(api_key="k"),
+                None,
+            ):
+                chunks.append(chunk)
+            return chunks
+
+        chunks = asyncio.run(run())
+        self.assertEqual([chunk.delta.get("content") for chunk in chunks], ["he", "llo", None])
+        self.assertEqual(chunks[-1].finish_reason, "end_turn")
+
+
+if __name__ == "__main__":
+    unittest.main()