vellum-ai 0.9.15__py3-none-any.whl → 0.9.16rc2__py3-none-any.whl

vellum/client/README.md

@@ -0,0 +1,242 @@
+# Vellum Python Library
+
+[![fern shield](https://img.shields.io/badge/%F0%9F%8C%BF-Built%20with%20Fern-brightgreen)](https://buildwithfern.com?utm_source=github&utm_medium=github&utm_campaign=readme&utm_source=https%3A%2F%2Fgithub.com%2Fvellum-ai%2Fvellum-client-python)
+[![pypi](https://img.shields.io/pypi/v/vellum-ai)](https://pypi.python.org/pypi/vellum-ai)
+
+The Vellum Python library provides convenient access to the Vellum API from Python.
+
+## API Docs
+You can find Vellum's complete API docs at [docs.vellum.ai](https://docs.vellum.ai/api-reference/introduction/getting-started).
+
+## Documentation
+
+API reference documentation is available [here](https://docs.vellum.ai/api-reference/overview/getting-started).
+
+## Installation
+
+```sh
+pip install vellum-ai
+```
+
+## Reference
+
+A full reference for this library is available [here](./reference.md).
+
+## Usage
+
+Instantiate and use the client with the following:
+
+```python
+from vellum import StringInputRequest, Vellum
+
+client = Vellum(
+    api_key="YOUR_API_KEY",
+)
+client.execute_prompt(
+    inputs=[
+        StringInputRequest(
+            name="name",
+            value="value",
+        )
+    ],
+)
+```
+
+## Async Client
+
+The SDK also exports an `async` client so that you can make non-blocking calls to our API.
+
+```python
+import asyncio
+
+from vellum import AsyncVellum, StringInputRequest
+
+client = AsyncVellum(
+    api_key="YOUR_API_KEY",
+)
+
+
+async def main() -> None:
+    await client.execute_prompt(
+        inputs=[
+            StringInputRequest(
+                name="name",
+                value="value",
+            )
+        ],
+    )
+
+
+asyncio.run(main())
+```
+
+## Exception Handling
+
+When the API returns a non-success status code (4xx or 5xx response), a subclass of the following error
+will be thrown.
+
+```python
+from vellum.core.api_error import ApiError
+
+try:
+    client.execute_prompt(...)
+except ApiError as e:
+    print(e.status_code)
+    print(e.body)
+```
+
+## Streaming
+
+The SDK supports streaming responses, as well, the response will be a generator that you can loop over.
+
+```python
+from vellum import (
+    AdHocExpandMeta,
+    EphemeralPromptCacheConfig,
+    FunctionDefinition,
+    JinjaPromptBlock,
+    PromptParameters,
+    PromptRequestStringInput,
+    PromptSettings,
+    StringVellumValue,
+    Vellum,
+    VellumVariable,
+    VellumVariableExtensions,
+)
+
+client = Vellum(
+    api_key="YOUR_API_KEY",
+)
+response = client.ad_hoc.adhoc_execute_prompt_stream(
+    ml_model="string",
+    input_values=[
+        PromptRequestStringInput(
+            key="string",
+            value="string",
+        )
+    ],
+    input_variables=[
+        VellumVariable(
+            id="string",
+            key="string",
+            type="STRING",
+            required=True,
+            default=StringVellumValue(
+                value="string",
+            ),
+            extensions=VellumVariableExtensions(
+                color={"key": "value"},
+            ),
+        )
+    ],
+    parameters=PromptParameters(
+        stop=["string"],
+        temperature=1.1,
+        max_tokens=1,
+        top_p=1.1,
+        top_k=1,
+        frequency_penalty=1.1,
+        presence_penalty=1.1,
+        logit_bias={"string": {"key": "value"}},
+        custom_parameters={"string": {"key": "value"}},
+    ),
+    settings=PromptSettings(
+        timeout=1.1,
+    ),
+    blocks=[
+        JinjaPromptBlock(
+            state="ENABLED",
+            cache_config=EphemeralPromptCacheConfig(),
+            template="string",
+        )
+    ],
+    functions=[
+        FunctionDefinition(
+            state="ENABLED",
+            cache_config=EphemeralPromptCacheConfig(),
+            name="string",
+            description="string",
+            parameters={"string": {"key": "value"}},
+            forced=True,
+            strict=True,
+        )
+    ],
+    expand_meta=AdHocExpandMeta(
+        cost=True,
+        model_name=True,
+        usage=True,
+        finish_reason=True,
+    ),
+)
+for chunk in response:
+    yield chunk
+```
+
+## Advanced
+
+### Retries
+
+The SDK is instrumented with automatic retries with exponential backoff. A request will be retried as long
+as the request is deemed retriable and the number of retry attempts has not grown larger than the configured
+retry limit (default: 2).
+
+A request is deemed retriable when any of the following HTTP status codes is returned:
+
+- [408](https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/408) (Timeout)
+- [429](https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/429) (Too Many Requests)
+- [5XX](https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/500) (Internal Server Errors)
+
+Use the `max_retries` request option to configure this behavior.
+
+```python
+client.execute_prompt(..., request_options={
+    "max_retries": 1
+})
+```
+
+### Timeouts
+
+The SDK defaults to a 60 second timeout. You can configure this with a timeout option at the client or request level.
+
+```python
+
+from vellum import Vellum
+
+client = Vellum(
+    ...,
+    timeout=20.0,
+)
+
+
+# Override timeout for a specific method
+client.execute_prompt(..., request_options={
+    "timeout_in_seconds": 1
+})
+```
+
+### Custom Client
+
+You can override the `httpx` client to customize it for your use-case. Some common use-cases include support for proxies
+and transports.
+```python
+import httpx
+from vellum import Vellum
+
+client = Vellum(
+    ...,
+    httpx_client=httpx.Client(
+        proxies="http://my.test.proxy.example.com",
+        transport=httpx.HTTPTransport(local_address="0.0.0.0"),
+    ),
+)
+```
+
+## Contributing
+
+While we value open-source contributions to this SDK, this library is generated programmatically.
+Additions made directly to this library would have to be moved over to our generation code,
+otherwise they would be overwritten upon the next generated release. Feel free to open a PR as
+a proof of concept, but know that we will not be able to merge it as-is. We suggest opening
+an issue first to discuss with us!
+
+On the other hand, contributions to the README are always very welcome!

vellum/client/core/__init__.py

@@ -0,0 +1,47 @@
+# This file was auto-generated by Fern from our API Definition.
+
+from .api_error import ApiError
+from .client_wrapper import AsyncClientWrapper, BaseClientWrapper, SyncClientWrapper
+from .datetime_utils import serialize_datetime
+from .file import File, convert_file_dict_to_httpx_tuples, with_content_type
+from .http_client import AsyncHttpClient, HttpClient
+from .jsonable_encoder import jsonable_encoder
+from .pydantic_utilities import (
+    IS_PYDANTIC_V2,
+    UniversalBaseModel,
+    UniversalRootModel,
+    parse_obj_as,
+    universal_field_validator,
+    universal_root_validator,
+    update_forward_refs,
+)
+from .query_encoder import encode_query
+from .remove_none_from_dict import remove_none_from_dict
+from .request_options import RequestOptions
+from .serialization import FieldMetadata, convert_and_respect_annotation_metadata
+
+__all__ = [
+    "ApiError",
+    "AsyncClientWrapper",
+    "AsyncHttpClient",
+    "BaseClientWrapper",
+    "FieldMetadata",
+    "File",
+    "HttpClient",
+    "IS_PYDANTIC_V2",
+    "RequestOptions",
+    "SyncClientWrapper",
+    "UniversalBaseModel",
+    "UniversalRootModel",
+    "convert_and_respect_annotation_metadata",
+    "convert_file_dict_to_httpx_tuples",
+    "encode_query",
+    "jsonable_encoder",
+    "parse_obj_as",
+    "remove_none_from_dict",
+    "serialize_datetime",
+    "universal_field_validator",
+    "universal_root_validator",
+    "update_forward_refs",
+    "with_content_type",
+]

vellum/client/core/api_error.py

@@ -0,0 +1,15 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+
+class ApiError(Exception):
+    status_code: typing.Optional[int]
+    body: typing.Any
+
+    def __init__(self, *, status_code: typing.Optional[int] = None, body: typing.Any = None):
+        self.status_code = status_code
+        self.body = body
+
+    def __str__(self) -> str:
+        return f"status_code: {self.status_code}, body: {self.body}"

vellum/client/core/client_wrapper.py

@@ -0,0 +1,59 @@
+# This file was auto-generated by Fern from our API Definition.
+
+from ..environment import VellumEnvironment
+import typing
+import httpx
+from .http_client import HttpClient
+from .http_client import AsyncHttpClient
+
+
+class BaseClientWrapper:
+    def __init__(self, *, api_key: str, environment: VellumEnvironment, timeout: typing.Optional[float] = None):
+        self.api_key = api_key
+        self._environment = environment
+        self._timeout = timeout
+
+    def get_headers(self) -> typing.Dict[str, str]:
+        headers: typing.Dict[str, str] = {
+            "X-Fern-Language": "Python",
+            "X-Fern-SDK-Name": "vellum-ai",
+            "X-Fern-SDK-Version": "0.9.17",
+        }
+        headers["X_API_KEY"] = self.api_key
+        return headers
+
+    def get_environment(self) -> VellumEnvironment:
+        return self._environment
+
+    def get_timeout(self) -> typing.Optional[float]:
+        return self._timeout
+
+
+class SyncClientWrapper(BaseClientWrapper):
+    def __init__(
+        self,
+        *,
+        api_key: str,
+        environment: VellumEnvironment,
+        timeout: typing.Optional[float] = None,
+        httpx_client: httpx.Client,
+    ):
+        super().__init__(api_key=api_key, environment=environment, timeout=timeout)
+        self.httpx_client = HttpClient(
+            httpx_client=httpx_client, base_headers=self.get_headers, base_timeout=self.get_timeout
+        )
+
+
+class AsyncClientWrapper(BaseClientWrapper):
+    def __init__(
+        self,
+        *,
+        api_key: str,
+        environment: VellumEnvironment,
+        timeout: typing.Optional[float] = None,
+        httpx_client: httpx.AsyncClient,
+    ):
+        super().__init__(api_key=api_key, environment=environment, timeout=timeout)
+        self.httpx_client = AsyncHttpClient(
+            httpx_client=httpx_client, base_headers=self.get_headers, base_timeout=self.get_timeout
+        )

vellum/client/core/datetime_utils.py

@@ -0,0 +1,28 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import datetime as dt
+
+
+def serialize_datetime(v: dt.datetime) -> str:
+    """
+    Serialize a datetime including timezone info.
+
+    Uses the timezone info provided if present, otherwise uses the current runtime's timezone info.
+
+    UTC datetimes end in "Z" while all other timezones are represented as offset from UTC, e.g. +05:00.
+    """
+
+    def _serialize_zoned_datetime(v: dt.datetime) -> str:
+        if v.tzinfo is not None and v.tzinfo.tzname(None) == dt.timezone.utc.tzname(None):
+            # UTC is a special case where we use "Z" at the end instead of "+00:00"
+            return v.isoformat().replace("+00:00", "Z")
+        else:
+            # Delegate to the typical +/- offset format
+            return v.isoformat()
+
+    if v.tzinfo is not None:
+        return _serialize_zoned_datetime(v)
+    else:
+        local_tz = dt.datetime.now().astimezone().tzinfo
+        localized_dt = v.replace(tzinfo=local_tz)
+        return _serialize_zoned_datetime(localized_dt)

vellum/client/core/file.py

@@ -0,0 +1,62 @@
+# This file was auto-generated by Fern from our API Definition.
+
+from typing import IO, Dict, List, Mapping, Optional, Tuple, Union, cast
+
+# File typing inspired by the flexibility of types within the httpx library
+# https://github.com/encode/httpx/blob/master/httpx/_types.py
+FileContent = Union[IO[bytes], bytes, str]
+File = Union[
+    # file (or bytes)
+    FileContent,
+    # (filename, file (or bytes))
+    Tuple[Optional[str], FileContent],
+    # (filename, file (or bytes), content_type)
+    Tuple[Optional[str], FileContent, Optional[str]],
+    # (filename, file (or bytes), content_type, headers)
+    Tuple[
+        Optional[str],
+        FileContent,
+        Optional[str],
+        Mapping[str, str],
+    ],
+]
+
+
+def convert_file_dict_to_httpx_tuples(
+    d: Dict[str, Union[File, List[File]]],
+) -> List[Tuple[str, File]]:
+    """
+    The format we use is a list of tuples, where the first element is the
+    name of the file and the second is the file object. Typically HTTPX wants
+    a dict, but to be able to send lists of files, you have to use the list
+    approach (which also works for non-lists)
+    https://github.com/encode/httpx/pull/1032
+    """
+
+    httpx_tuples = []
+    for key, file_like in d.items():
+        if isinstance(file_like, list):
+            for file_like_item in file_like:
+                httpx_tuples.append((key, file_like_item))
+        else:
+            httpx_tuples.append((key, file_like))
+    return httpx_tuples
+
+
+def with_content_type(*, file: File, content_type: str) -> File:
+    """ """
+    if isinstance(file, tuple):
+        if len(file) == 2:
+            filename, content = cast(Tuple[Optional[str], FileContent], file)  # type: ignore
+            return (filename, content, content_type)
+        elif len(file) == 3:
+            filename, content, _ = cast(Tuple[Optional[str], FileContent, Optional[str]], file)  # type: ignore
+            return (filename, content, content_type)
+        elif len(file) == 4:
+            filename, content, _, headers = cast(  # type: ignore
+                Tuple[Optional[str], FileContent, Optional[str], Mapping[str, str]], file
+            )
+            return (filename, content, content_type, headers)
+        else:
+            raise ValueError(f"Unexpected tuple length: {len(file)}")
+    return (None, file, content_type)