phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-10-12 05:54:38 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

chore: refactor (chat)completions endpoints to use shared params struct (#3761)

Browse source 
# What does this PR do?

Converts openai(_chat)_completions params to pydantic BaseModel to
reduce code duplication across all providers.

## Test Plan
CI









---
[//]: # (BEGIN SAPLING FOOTER)
Stack created with [Sapling](https://sapling-scm.com). Best reviewed
with
[ReviewStack](https://reviewstack.dev/llamastack/llama-stack/pull/3761).
* #3777
* __->__ #3761
This commit is contained in:
ehhuang 2025-10-10 15:46:34 -07:00 • committed by GitHub
parent 6954fe2274
commit 80d58ab519
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
33 changed files with 599 additions and 890 deletions
Download patch file Download diff file Expand all files Collapse all files

41
docs/openapi_generator/pyopenapi/generator.py
View file

@ -23,6 +23,7 @@ from llama_stack.strong_typing.inspection import (
is_generic_list,
is_type_optional,
is_type_union,
is_unwrapped_body_param,
unwrap_generic_list,
unwrap_optional_type,
unwrap_union_types,
@ -769,24 +770,30 @@ class Generator:
first = next(iter(op.request_params))
request_name, request_type = first
op_name = "".join(word.capitalize() for word in op.name.split("_"))
request_name = f"{op_name}Request"
fields = [
(
name,
type_,
)
for name, type_ in op.request_params
]
request_type = make_dataclass(
request_name,
fields,
namespace={
"__doc__": create_docstring_for_request(
request_name, fields, doc_params
# Special case: if there's a single parameter with Body(embed=False) that's a BaseModel,
# unwrap it to show the flat structure in the OpenAPI spec
# Example: openai_chat_completion()
if (len(op.request_params) == 1 and is_unwrapped_body_param(request_type)):
pass
else:
op_name = "".join(word.capitalize() for word in op.name.split("_"))
request_name = f"{op_name}Request"
fields = [
(
name,
type_,
)
},
)
for name, type_ in op.request_params
]
request_type = make_dataclass(
request_name,
fields,
namespace={
"__doc__": create_docstring_for_request(
request_name, fields, doc_params
)
},
)
requestBody = RequestBody(
content={

13
docs/openapi_generator/pyopenapi/utility.py
View file

@ -8,10 +8,11 @@ import json
import typing
import inspect
from pathlib import Path
from typing import TextIO
from typing import Any, List, Optional, Union, get_type_hints, get_origin, get_args
from typing import Any, List, Optional, TextIO, Union, get_type_hints, get_origin, get_args
from pydantic import BaseModel
from llama_stack.strong_typing.schema import object_to_json, StrictJsonType
from llama_stack.strong_typing.inspection import is_unwrapped_body_param
from llama_stack.core.resolver import api_protocol_map
from .generator import Generator
@ -205,6 +206,14 @@ def _validate_has_return_in_docstring(method) -> str | None:
def _validate_has_params_in_docstring(method) -> str | None:
source = inspect.getsource(method)
sig = inspect.signature(method)
params_list = [p for p in sig.parameters.values() if p.name != "self"]
if len(params_list) == 1:
param = params_list[0]
param_type = param.annotation
if is_unwrapped_body_param(param_type):
return
# Only check if the method has more than one parameter
if len(sig.parameters) > 1 and ":param" not in source:
return "does not have a ':param' in its docstring"

20
docs/static/deprecated-llama-stack-spec.html vendored
View file

@ -1527,7 +1527,7 @@
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/OpenaiChatCompletionRequest"
"$ref": "#/components/schemas/OpenAIChatCompletionRequest"
}
}
},
@ -1617,7 +1617,7 @@
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/OpenaiCompletionRequest"
"$ref": "#/components/schemas/OpenAICompletionRequest"
}
}
},
@ -7522,7 +7522,7 @@
"title": "OpenAIResponseFormatText",
"description": "Text response format for OpenAI-compatible chat completion requests."
},
"OpenaiChatCompletionRequest": {
"OpenAIChatCompletionRequest": {
"type": "object",
"properties": {
"model": {
@ -7769,7 +7769,8 @@
"model",
"messages"
],
"title": "OpenaiChatCompletionRequest"
"title": "OpenAIChatCompletionRequest",
"description": "Request parameters for OpenAI-compatible chat completion endpoint."
},
"OpenAIChatCompletion": {
"type": "object",
@ -7965,7 +7966,7 @@
],
"title": "OpenAICompletionWithInputMessages"
},
"OpenaiCompletionRequest": {
"OpenAICompletionRequest": {
"type": "object",
"properties": {
"model": {
@ -8100,10 +8101,12 @@
"type": "array",
"items": {
"type": "string"
}
},
"description": "(Optional) vLLM-specific parameter for guided generation with a list of choices."
},
"prompt_logprobs": {
"type": "integer"
"type": "integer",
"description": "(Optional) vLLM-specific parameter for number of log probabilities to return for prompt tokens."
},
"suffix": {
"type": "string",
@ -8115,7 +8118,8 @@
"model",
"prompt"
],
"title": "OpenaiCompletionRequest"
"title": "OpenAICompletionRequest",
"description": "Request parameters for OpenAI-compatible completion endpoint."
},
"OpenAICompletion": {
"type": "object",

22
docs/static/deprecated-llama-stack-spec.yaml vendored
View file

@ -1098,7 +1098,7 @@ paths:
content:
application/json:
schema:
$ref: '#/components/schemas/OpenaiChatCompletionRequest'
$ref: '#/components/schemas/OpenAIChatCompletionRequest'
required: true
deprecated: true
/v1/openai/v1/chat/completions/{completion_id}:
@ -1167,7 +1167,7 @@ paths:
content:
application/json:
schema:
$ref: '#/components/schemas/OpenaiCompletionRequest'
$ref: '#/components/schemas/OpenAICompletionRequest'
required: true
deprecated: true
/v1/openai/v1/embeddings:
@ -5575,7 +5575,7 @@ components:
title: OpenAIResponseFormatText
description: >-
Text response format for OpenAI-compatible chat completion requests.
OpenaiChatCompletionRequest:
OpenAIChatCompletionRequest:
type: object
properties:
model:
@ -5717,7 +5717,9 @@ components:
required:
- model
- messages
title: OpenaiChatCompletionRequest
title: OpenAIChatCompletionRequest
description: >-
Request parameters for OpenAI-compatible chat completion endpoint.
OpenAIChatCompletion:
type: object
properties:
@ -5883,7 +5885,7 @@ components:
- model
- input_messages
title: OpenAICompletionWithInputMessages
OpenaiCompletionRequest:
OpenAICompletionRequest:
type: object
properties:
model:
@ -5975,8 +5977,14 @@ components:
type: array
items:
type: string
description: >-
(Optional) vLLM-specific parameter for guided generation with a list of
choices.
prompt_logprobs:
type: integer
description: >-
(Optional) vLLM-specific parameter for number of log probabilities to
return for prompt tokens.
suffix:
type: string
description: >-
@ -5985,7 +5993,9 @@ components:
required:
- model
- prompt
title: OpenaiCompletionRequest
title: OpenAICompletionRequest
description: >-
Request parameters for OpenAI-compatible completion endpoint.
OpenAICompletion:
type: object
properties:

20
docs/static/llama-stack-spec.html vendored
View file

@ -153,7 +153,7 @@
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/OpenaiChatCompletionRequest"
"$ref": "#/components/schemas/OpenAIChatCompletionRequest"
}
}
},
@ -243,7 +243,7 @@
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/OpenaiCompletionRequest"
"$ref": "#/components/schemas/OpenAICompletionRequest"
}
}
},
@ -5018,7 +5018,7 @@
"title": "OpenAIResponseFormatText",
"description": "Text response format for OpenAI-compatible chat completion requests."
},
"OpenaiChatCompletionRequest": {
"OpenAIChatCompletionRequest": {
"type": "object",
"properties": {
"model": {
@ -5265,7 +5265,8 @@
"model",
"messages"
],
"title": "OpenaiChatCompletionRequest"
"title": "OpenAIChatCompletionRequest",
"description": "Request parameters for OpenAI-compatible chat completion endpoint."
},
"OpenAIChatCompletion": {
"type": "object",
@ -5461,7 +5462,7 @@
],
"title": "OpenAICompletionWithInputMessages"
},
"OpenaiCompletionRequest": {
"OpenAICompletionRequest": {
"type": "object",
"properties": {
"model": {
@ -5596,10 +5597,12 @@
"type": "array",
"items": {
"type": "string"
}
},
"description": "(Optional) vLLM-specific parameter for guided generation with a list of choices."
},
"prompt_logprobs": {
"type": "integer"
"type": "integer",
"description": "(Optional) vLLM-specific parameter for number of log probabilities to return for prompt tokens."
},
"suffix": {
"type": "string",
@ -5611,7 +5614,8 @@
"model",
"prompt"
],
"title": "OpenaiCompletionRequest"
"title": "OpenAICompletionRequest",
"description": "Request parameters for OpenAI-compatible completion endpoint."
},
"OpenAICompletion": {
"type": "object",

22
docs/static/llama-stack-spec.yaml vendored
View file

@ -98,7 +98,7 @@ paths:
content:
application/json:
schema:
$ref: '#/components/schemas/OpenaiChatCompletionRequest'
$ref: '#/components/schemas/OpenAIChatCompletionRequest'
required: true
deprecated: false
/v1/chat/completions/{completion_id}:
@ -167,7 +167,7 @@ paths:
content:
application/json:
schema:
$ref: '#/components/schemas/OpenaiCompletionRequest'
$ref: '#/components/schemas/OpenAICompletionRequest'
required: true
deprecated: false
/v1/conversations:
@ -3824,7 +3824,7 @@ components:
title: OpenAIResponseFormatText
description: >-
Text response format for OpenAI-compatible chat completion requests.
OpenaiChatCompletionRequest:
OpenAIChatCompletionRequest:
type: object
properties:
model:
@ -3966,7 +3966,9 @@ components:
required:
- model
- messages
title: OpenaiChatCompletionRequest
title: OpenAIChatCompletionRequest
description: >-
Request parameters for OpenAI-compatible chat completion endpoint.
OpenAIChatCompletion:
type: object
properties:
@ -4132,7 +4134,7 @@ components:
- model
- input_messages
title: OpenAICompletionWithInputMessages
OpenaiCompletionRequest:
OpenAICompletionRequest:
type: object
properties:
model:
@ -4224,8 +4226,14 @@ components:
type: array
items:
type: string
description: >-
(Optional) vLLM-specific parameter for guided generation with a list of
choices.
prompt_logprobs:
type: integer
description: >-
(Optional) vLLM-specific parameter for number of log probabilities to
return for prompt tokens.
suffix:
type: string
description: >-
@ -4234,7 +4242,9 @@ components:
required:
- model
- prompt
title: OpenaiCompletionRequest
title: OpenAICompletionRequest
description: >-
Request parameters for OpenAI-compatible completion endpoint.
OpenAICompletion:
type: object
properties:

20
docs/static/stainless-llama-stack-spec.html vendored
View file

@ -153,7 +153,7 @@
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/OpenaiChatCompletionRequest"
"$ref": "#/components/schemas/OpenAIChatCompletionRequest"
}
}
},
@ -243,7 +243,7 @@
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/OpenaiCompletionRequest"
"$ref": "#/components/schemas/OpenAICompletionRequest"
}
}
},
@ -7027,7 +7027,7 @@
"title": "OpenAIResponseFormatText",
"description": "Text response format for OpenAI-compatible chat completion requests."
},
"OpenaiChatCompletionRequest": {
"OpenAIChatCompletionRequest": {
"type": "object",
"properties": {
"model": {
@ -7274,7 +7274,8 @@
"model",
"messages"
],
"title": "OpenaiChatCompletionRequest"
"title": "OpenAIChatCompletionRequest",
"description": "Request parameters for OpenAI-compatible chat completion endpoint."
},
"OpenAIChatCompletion": {
"type": "object",
@ -7470,7 +7471,7 @@
],
"title": "OpenAICompletionWithInputMessages"
},
"OpenaiCompletionRequest": {
"OpenAICompletionRequest": {
"type": "object",
"properties": {
"model": {
@ -7605,10 +7606,12 @@
"type": "array",
"items": {
"type": "string"
}
},
"description": "(Optional) vLLM-specific parameter for guided generation with a list of choices."
},
"prompt_logprobs": {
"type": "integer"
"type": "integer",
"description": "(Optional) vLLM-specific parameter for number of log probabilities to return for prompt tokens."
},
"suffix": {
"type": "string",
@ -7620,7 +7623,8 @@
"model",
"prompt"
],
"title": "OpenaiCompletionRequest"
"title": "OpenAICompletionRequest",
"description": "Request parameters for OpenAI-compatible completion endpoint."
},
"OpenAICompletion": {
"type": "object",

22
docs/static/stainless-llama-stack-spec.yaml vendored
View file

@ -101,7 +101,7 @@ paths:
content:
application/json:
schema:
$ref: '#/components/schemas/OpenaiChatCompletionRequest'
$ref: '#/components/schemas/OpenAIChatCompletionRequest'
required: true
deprecated: false
/v1/chat/completions/{completion_id}:
@ -170,7 +170,7 @@ paths:
content:
application/json:
schema:
$ref: '#/components/schemas/OpenaiCompletionRequest'
$ref: '#/components/schemas/OpenAICompletionRequest'
required: true
deprecated: false
/v1/conversations:
@ -5269,7 +5269,7 @@ components:
title: OpenAIResponseFormatText
description: >-
Text response format for OpenAI-compatible chat completion requests.
OpenaiChatCompletionRequest:
OpenAIChatCompletionRequest:
type: object
properties:
model:
@ -5411,7 +5411,9 @@ components:
required:
- model
- messages
title: OpenaiChatCompletionRequest
title: OpenAIChatCompletionRequest
description: >-
Request parameters for OpenAI-compatible chat completion endpoint.
OpenAIChatCompletion:
type: object
properties:
@ -5577,7 +5579,7 @@ components:
- model
- input_messages
title: OpenAICompletionWithInputMessages
OpenaiCompletionRequest:
OpenAICompletionRequest:
type: object
properties:
model:
@ -5669,8 +5671,14 @@ components:
type: array
items:
type: string
description: >-
(Optional) vLLM-specific parameter for guided generation with a list of
choices.
prompt_logprobs:
type: integer
description: >-
(Optional) vLLM-specific parameter for number of log probabilities to
return for prompt tokens.
suffix:
type: string
description: >-
@ -5679,7 +5687,9 @@ components:
required:
- model
- prompt
title: OpenaiCompletionRequest
title: OpenAICompletionRequest
description: >-
Request parameters for OpenAI-compatible completion endpoint.
OpenAICompletion:
type: object
properties: