athena-intelligence 0.1.122__tar.gz → 0.1.123__tar.gz

{athena_intelligence-0.1.122 → athena_intelligence-0.1.123}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: athena-intelligence
-Version: 0.1.122
+Version: 0.1.123
 Summary: Athena Intelligence Python Library
 Requires-Python: >=3.8,<4.0
 Classifier: Intended Audience :: Developers

{athena_intelligence-0.1.122 → athena_intelligence-0.1.123}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "athena-intelligence"
-version = "0.1.122"
+version = "0.1.123"
 description = "Athena Intelligence Python Library"
 readme = "README.md"
 authors = []

{athena_intelligence-0.1.122 → athena_intelligence-0.1.123}/src/athena/__init__.py

@@ -2,6 +2,12 @@
 
 from .types import (
     AssetNotFoundError,
+    Chunk,
+    ChunkContentItem,
+    ChunkContentItem_ImageUrl,
+    ChunkContentItem_Text,
+    ChunkResult,
+    ChunkResultChunkId,
     CustomAgentResponse,
     DataFrameRequestOut,
     DataFrameRequestOutColumnsItem,
@@ -16,12 +22,16 @@ from .types import (
     GeneralAgentConfigEnabledToolsItem,
     GeneralAgentRequest,
     GeneralAgentResponse,
+    ImageUrlContent,
     ParentFolderError,
+    PromptMessage,
     ResearchAgentResponse,
     SaveAssetRequestOut,
     SqlAgentResponse,
-    StructuredDataExtractorReponse,
+    StructuredDataExtractorResponse,
+    TextContent,
     Tool,
+    Type,
 )
 from .errors import (
     BadRequestError,
@@ -42,6 +52,12 @@ __all__ = [
     "AssetNotFoundError",
     "AthenaEnvironment",
     "BadRequestError",
+    "Chunk",
+    "ChunkContentItem",
+    "ChunkContentItem_ImageUrl",
+    "ChunkContentItem_Text",
+    "ChunkResult",
+    "ChunkResultChunkId",
     "ContentTooLargeError",
     "CustomAgentResponse",
     "DataFrameRequestOut",
@@ -57,16 +73,20 @@ __all__ = [
     "GeneralAgentConfigEnabledToolsItem",
     "GeneralAgentRequest",
     "GeneralAgentResponse",
+    "ImageUrlContent",
     "InternalServerError",
     "NotFoundError",
     "ParentFolderError",
+    "PromptMessage",
     "QueryExecuteRequestDatabaseAssetIds",
     "ResearchAgentResponse",
     "SaveAssetRequestOut",
     "SqlAgentResponse",
-    "StructuredDataExtractorReponse",
+    "StructuredDataExtractorResponse",
+    "TextContent",
     "Tool",
     "ToolsDataFrameRequestColumnsItem",
+    "Type",
     "UnauthorizedError",
     "UnprocessableEntityError",
     "UnsupportedMediaTypeError",

{athena_intelligence-0.1.122 → athena_intelligence-0.1.123}/src/athena/core/client_wrapper.py

@@ -17,7 +17,7 @@ class BaseClientWrapper:
         headers: typing.Dict[str, str] = {
             "X-Fern-Language": "Python",
             "X-Fern-SDK-Name": "athena-intelligence",
-            "X-Fern-SDK-Version": "0.1.122",
+            "X-Fern-SDK-Version": "0.1.123",
         }
         headers["X-API-KEY"] = self.api_key
         return headers

athena_intelligence-0.1.123/src/athena/tools/structured_data_extractor/client.py

@@ -0,0 +1,264 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+from json.decoder import JSONDecodeError
+
+from ...core.api_error import ApiError
+from ...core.client_wrapper import AsyncClientWrapper, SyncClientWrapper
+from ...core.pydantic_utilities import pydantic_v1
+from ...core.request_options import RequestOptions
+from ...errors.unprocessable_entity_error import UnprocessableEntityError
+from ...types.chunk import Chunk
+from ...types.prompt_message import PromptMessage
+from ...types.structured_data_extractor_response import StructuredDataExtractorResponse
+
+# this is used as the default value for optional parameters
+OMIT = typing.cast(typing.Any, ...)
+
+
+class StructuredDataExtractorClient:
+    def __init__(self, *, client_wrapper: SyncClientWrapper):
+        self._client_wrapper = client_wrapper
+
+    def invoke(
+        self,
+        *,
+        chunks: typing.Sequence[Chunk],
+        json_schema: typing.Dict[str, typing.Any],
+        chunk_messages: typing.Optional[typing.Sequence[PromptMessage]] = OMIT,
+        reduce: typing.Optional[bool] = OMIT,
+        reduce_messages: typing.Optional[typing.Sequence[PromptMessage]] = OMIT,
+        request_options: typing.Optional[RequestOptions] = None
+    ) -> StructuredDataExtractorResponse:
+        """
+        Extract structured data.
+
+        tl;dr:
+
+        - pass a valid JSON schema in `json_schema`
+        - pass the page chunks as a list of `Chunk` objects, by default: {"type": "text", "content": "..."}
+        - leave all other fields as default
+
+        Detailed configuration (only relevant for complex use cases):
+
+        The structured data extractor's architecture follows the map-reduce pattern,
+        where the asset is divided into chunks, the schema is extracted from each chunk,
+        and the chunks are then reduced to a single structured data object.
+
+        In some applications, you may not want to: - map (if your input asset is small enough) - reduce (if your output object is large enough that it will overflow the output length;
+        if you're extracting a long list of entities; if youre )
+        to extract all instances of the schema).
+
+        You can configure these behaviors with the `map` and `reduce` fields.
+
+        Parameters
+        ----------
+        chunks : typing.Sequence[Chunk]
+            The chunks from which to extract structured data.
+
+        json_schema : typing.Dict[str, typing.Any]
+            The JSON schema to use for validation (version draft 2020-12). See the docs [here](https://json-schema.org/learn/getting-started-step-by-step).
+
+        chunk_messages : typing.Optional[typing.Sequence[PromptMessage]]
+            The prompt to use for the data extraction over *each individual chunk*. It must be a list of messages.  The chunk content will be appended as a list of human messages.
+
+        reduce : typing.Optional[bool]
+            If `map`, whether to reduce the chunks to a single structured object (true) or return the full list (false).  Use True unless you want to preserve duplicates from each page or expect the object to overflow the output context.
+
+        reduce_messages : typing.Optional[typing.Sequence[PromptMessage]]
+            The prompt to use for the reduce steps. It must be a list of messages. The two extraction attempts will be appended as a list of human messages.
+
+        request_options : typing.Optional[RequestOptions]
+            Request-specific configuration.
+
+        Returns
+        -------
+        StructuredDataExtractorResponse
+            Successful Response
+
+        Examples
+        --------
+        from athena import Chunk, ChunkContentItem_Text
+        from athena.client import Athena
+
+        client = Athena(
+            api_key="YOUR_API_KEY",
+        )
+        client.tools.structured_data_extractor.invoke(
+            chunks=[
+                Chunk(
+                    chunk_id="1",
+                    content=[
+                        ChunkContentItem_Text(
+                            text="John Smith is a 35 year old developer. You can reach him at john.smith@example.com",
+                        )
+                    ],
+                ),
+                Chunk(
+                    chunk_id="2",
+                    content=[
+                        ChunkContentItem_Text(
+                            text="Jane Doe is a 25 year old developer. You can reach her at jane@example.com",
+                        )
+                    ],
+                ),
+            ],
+            json_schema={
+                "description": "A person",
+                "properties": {
+                    "age": {"type": "integer"},
+                    "email": {"type": "string"},
+                    "name": {"type": "string"},
+                },
+                "required": ["name"],
+                "title": "Person",
+                "type": "object",
+            },
+        )
+        """
+        _response = self._client_wrapper.httpx_client.request(
+            "api/v0/tools/structured-data-extractor/invoke",
+            method="POST",
+            json={
+                "chunk_messages": chunk_messages,
+                "chunks": chunks,
+                "json_schema": json_schema,
+                "reduce": reduce,
+                "reduce_messages": reduce_messages,
+            },
+            request_options=request_options,
+            omit=OMIT,
+        )
+        if 200 <= _response.status_code < 300:
+            return pydantic_v1.parse_obj_as(StructuredDataExtractorResponse, _response.json())  # type: ignore
+        if _response.status_code == 422:
+            raise UnprocessableEntityError(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+        try:
+            _response_json = _response.json()
+        except JSONDecodeError:
+            raise ApiError(status_code=_response.status_code, body=_response.text)
+        raise ApiError(status_code=_response.status_code, body=_response_json)
+
+
+class AsyncStructuredDataExtractorClient:
+    def __init__(self, *, client_wrapper: AsyncClientWrapper):
+        self._client_wrapper = client_wrapper
+
+    async def invoke(
+        self,
+        *,
+        chunks: typing.Sequence[Chunk],
+        json_schema: typing.Dict[str, typing.Any],
+        chunk_messages: typing.Optional[typing.Sequence[PromptMessage]] = OMIT,
+        reduce: typing.Optional[bool] = OMIT,
+        reduce_messages: typing.Optional[typing.Sequence[PromptMessage]] = OMIT,
+        request_options: typing.Optional[RequestOptions] = None
+    ) -> StructuredDataExtractorResponse:
+        """
+        Extract structured data.
+
+        tl;dr:
+
+        - pass a valid JSON schema in `json_schema`
+        - pass the page chunks as a list of `Chunk` objects, by default: {"type": "text", "content": "..."}
+        - leave all other fields as default
+
+        Detailed configuration (only relevant for complex use cases):
+
+        The structured data extractor's architecture follows the map-reduce pattern,
+        where the asset is divided into chunks, the schema is extracted from each chunk,
+        and the chunks are then reduced to a single structured data object.
+
+        In some applications, you may not want to: - map (if your input asset is small enough) - reduce (if your output object is large enough that it will overflow the output length;
+        if you're extracting a long list of entities; if youre )
+        to extract all instances of the schema).
+
+        You can configure these behaviors with the `map` and `reduce` fields.
+
+        Parameters
+        ----------
+        chunks : typing.Sequence[Chunk]
+            The chunks from which to extract structured data.
+
+        json_schema : typing.Dict[str, typing.Any]
+            The JSON schema to use for validation (version draft 2020-12). See the docs [here](https://json-schema.org/learn/getting-started-step-by-step).
+
+        chunk_messages : typing.Optional[typing.Sequence[PromptMessage]]
+            The prompt to use for the data extraction over *each individual chunk*. It must be a list of messages.  The chunk content will be appended as a list of human messages.
+
+        reduce : typing.Optional[bool]
+            If `map`, whether to reduce the chunks to a single structured object (true) or return the full list (false).  Use True unless you want to preserve duplicates from each page or expect the object to overflow the output context.
+
+        reduce_messages : typing.Optional[typing.Sequence[PromptMessage]]
+            The prompt to use for the reduce steps. It must be a list of messages. The two extraction attempts will be appended as a list of human messages.
+
+        request_options : typing.Optional[RequestOptions]
+            Request-specific configuration.
+
+        Returns
+        -------
+        StructuredDataExtractorResponse
+            Successful Response
+
+        Examples
+        --------
+        from athena import Chunk, ChunkContentItem_Text
+        from athena.client import AsyncAthena
+
+        client = AsyncAthena(
+            api_key="YOUR_API_KEY",
+        )
+        await client.tools.structured_data_extractor.invoke(
+            chunks=[
+                Chunk(
+                    chunk_id="1",
+                    content=[
+                        ChunkContentItem_Text(
+                            text="John Smith is a 35 year old developer. You can reach him at john.smith@example.com",
+                        )
+                    ],
+                ),
+                Chunk(
+                    chunk_id="2",
+                    content=[
+                        ChunkContentItem_Text(
+                            text="Jane Doe is a 25 year old developer. You can reach her at jane@example.com",
+                        )
+                    ],
+                ),
+            ],
+            json_schema={
+                "description": "A person",
+                "properties": {
+                    "age": {"type": "integer"},
+                    "email": {"type": "string"},
+                    "name": {"type": "string"},
+                },
+                "required": ["name"],
+                "title": "Person",
+                "type": "object",
+            },
+        )
+        """
+        _response = await self._client_wrapper.httpx_client.request(
+            "api/v0/tools/structured-data-extractor/invoke",
+            method="POST",
+            json={
+                "chunk_messages": chunk_messages,
+                "chunks": chunks,
+                "json_schema": json_schema,
+                "reduce": reduce,
+                "reduce_messages": reduce_messages,
+            },
+            request_options=request_options,
+            omit=OMIT,
+        )
+        if 200 <= _response.status_code < 300:
+            return pydantic_v1.parse_obj_as(StructuredDataExtractorResponse, _response.json())  # type: ignore
+        if _response.status_code == 422:
+            raise UnprocessableEntityError(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+        try:
+            _response_json = _response.json()
+        except JSONDecodeError:
+            raise ApiError(status_code=_response.status_code, body=_response.text)
+        raise ApiError(status_code=_response.status_code, body=_response_json)

{athena_intelligence-0.1.122 → athena_intelligence-0.1.123}/src/athena/types/__init__.py

@@ -1,6 +1,10 @@
 # This file was auto-generated by Fern from our API Definition.
 
 from .asset_not_found_error import AssetNotFoundError
+from .chunk import Chunk
+from .chunk_content_item import ChunkContentItem, ChunkContentItem_ImageUrl, ChunkContentItem_Text
+from .chunk_result import ChunkResult
+from .chunk_result_chunk_id import ChunkResultChunkId
 from .custom_agent_response import CustomAgentResponse
 from .data_frame_request_out import DataFrameRequestOut
 from .data_frame_request_out_columns_item import DataFrameRequestOutColumnsItem
@@ -15,15 +19,25 @@ from .general_agent_config import GeneralAgentConfig
 from .general_agent_config_enabled_tools_item import GeneralAgentConfigEnabledToolsItem
 from .general_agent_request import GeneralAgentRequest
 from .general_agent_response import GeneralAgentResponse
+from .image_url_content import ImageUrlContent
 from .parent_folder_error import ParentFolderError
+from .prompt_message import PromptMessage
 from .research_agent_response import ResearchAgentResponse
 from .save_asset_request_out import SaveAssetRequestOut
 from .sql_agent_response import SqlAgentResponse
-from .structured_data_extractor_reponse import StructuredDataExtractorReponse
+from .structured_data_extractor_response import StructuredDataExtractorResponse
+from .text_content import TextContent
 from .tool import Tool
+from .type import Type
 
 __all__ = [
     "AssetNotFoundError",
+    "Chunk",
+    "ChunkContentItem",
+    "ChunkContentItem_ImageUrl",
+    "ChunkContentItem_Text",
+    "ChunkResult",
+    "ChunkResultChunkId",
     "CustomAgentResponse",
     "DataFrameRequestOut",
     "DataFrameRequestOutColumnsItem",
@@ -38,10 +52,14 @@ __all__ = [
     "GeneralAgentConfigEnabledToolsItem",
     "GeneralAgentRequest",
     "GeneralAgentResponse",
+    "ImageUrlContent",
     "ParentFolderError",
+    "PromptMessage",
     "ResearchAgentResponse",
     "SaveAssetRequestOut",
     "SqlAgentResponse",
-    "StructuredDataExtractorReponse",
+    "StructuredDataExtractorResponse",
+    "TextContent",
     "Tool",
+    "Type",
 ]

athena_intelligence-0.1.123/src/athena/types/chunk.py

@@ -0,0 +1,36 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import datetime as dt
+import typing
+
+from ..core.datetime_utils import serialize_datetime
+from ..core.pydantic_utilities import deep_union_pydantic_dicts, pydantic_v1
+from .chunk_content_item import ChunkContentItem
+
+
+class Chunk(pydantic_v1.BaseModel):
+    """
+    A chunk of content to extract data from.
+    """
+
+    chunk_id: str
+    content: typing.List[ChunkContentItem]
+    metadata: typing.Optional[typing.Dict[str, typing.Optional[str]]] = None
+
+    def json(self, **kwargs: typing.Any) -> str:
+        kwargs_with_defaults: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
+        return super().json(**kwargs_with_defaults)
+
+    def dict(self, **kwargs: typing.Any) -> typing.Dict[str, typing.Any]:
+        kwargs_with_defaults_exclude_unset: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
+        kwargs_with_defaults_exclude_none: typing.Any = {"by_alias": True, "exclude_none": True, **kwargs}
+
+        return deep_union_pydantic_dicts(
+            super().dict(**kwargs_with_defaults_exclude_unset), super().dict(**kwargs_with_defaults_exclude_none)
+        )
+
+    class Config:
+        frozen = True
+        smart_union = True
+        extra = pydantic_v1.Extra.allow
+        json_encoders = {dt.datetime: serialize_datetime}

athena_intelligence-0.1.123/src/athena/types/chunk_content_item.py

@@ -0,0 +1,58 @@
+# This file was auto-generated by Fern from our API Definition.
+
+from __future__ import annotations
+
+import datetime as dt
+import typing
+
+from ..core.datetime_utils import serialize_datetime
+from ..core.pydantic_utilities import deep_union_pydantic_dicts, pydantic_v1
+
+
+class ChunkContentItem_Text(pydantic_v1.BaseModel):
+    text: str
+    type: typing.Literal["text"] = "text"
+
+    def json(self, **kwargs: typing.Any) -> str:
+        kwargs_with_defaults: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
+        return super().json(**kwargs_with_defaults)
+
+    def dict(self, **kwargs: typing.Any) -> typing.Dict[str, typing.Any]:
+        kwargs_with_defaults_exclude_unset: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
+        kwargs_with_defaults_exclude_none: typing.Any = {"by_alias": True, "exclude_none": True, **kwargs}
+
+        return deep_union_pydantic_dicts(
+            super().dict(**kwargs_with_defaults_exclude_unset), super().dict(**kwargs_with_defaults_exclude_none)
+        )
+
+    class Config:
+        frozen = True
+        smart_union = True
+        extra = pydantic_v1.Extra.allow
+        json_encoders = {dt.datetime: serialize_datetime}
+
+
+class ChunkContentItem_ImageUrl(pydantic_v1.BaseModel):
+    image_url: typing.Dict[str, str]
+    type: typing.Literal["image_url"] = "image_url"
+
+    def json(self, **kwargs: typing.Any) -> str:
+        kwargs_with_defaults: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
+        return super().json(**kwargs_with_defaults)
+
+    def dict(self, **kwargs: typing.Any) -> typing.Dict[str, typing.Any]:
+        kwargs_with_defaults_exclude_unset: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
+        kwargs_with_defaults_exclude_none: typing.Any = {"by_alias": True, "exclude_none": True, **kwargs}
+
+        return deep_union_pydantic_dicts(
+            super().dict(**kwargs_with_defaults_exclude_unset), super().dict(**kwargs_with_defaults_exclude_none)
+        )
+
+    class Config:
+        frozen = True
+        smart_union = True
+        extra = pydantic_v1.Extra.allow
+        json_encoders = {dt.datetime: serialize_datetime}
+
+
+ChunkContentItem = typing.Union[ChunkContentItem_Text, ChunkContentItem_ImageUrl]

athena_intelligence-0.1.123/src/athena/types/chunk_result.py

@@ -0,0 +1,35 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import datetime as dt
+import typing
+
+from ..core.datetime_utils import serialize_datetime
+from ..core.pydantic_utilities import deep_union_pydantic_dicts, pydantic_v1
+from .chunk_result_chunk_id import ChunkResultChunkId
+
+
+class ChunkResult(pydantic_v1.BaseModel):
+    """
+    The result of a chunk extraction.
+    """
+
+    chunk_id: ChunkResultChunkId
+    chunk_result: typing.Optional[typing.Dict[str, typing.Any]] = None
+
+    def json(self, **kwargs: typing.Any) -> str:
+        kwargs_with_defaults: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
+        return super().json(**kwargs_with_defaults)
+
+    def dict(self, **kwargs: typing.Any) -> typing.Dict[str, typing.Any]:
+        kwargs_with_defaults_exclude_unset: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
+        kwargs_with_defaults_exclude_none: typing.Any = {"by_alias": True, "exclude_none": True, **kwargs}
+
+        return deep_union_pydantic_dicts(
+            super().dict(**kwargs_with_defaults_exclude_unset), super().dict(**kwargs_with_defaults_exclude_none)
+        )
+
+    class Config:
+        frozen = True
+        smart_union = True
+        extra = pydantic_v1.Extra.allow
+        json_encoders = {dt.datetime: serialize_datetime}

athena_intelligence-0.1.123/src/athena/types/chunk_result_chunk_id.py

@@ -0,0 +1,5 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+ChunkResultChunkId = typing.Union[str, int]

athena_intelligence-0.1.123/src/athena/types/image_url_content.py

@@ -0,0 +1,33 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import datetime as dt
+import typing
+
+from ..core.datetime_utils import serialize_datetime
+from ..core.pydantic_utilities import deep_union_pydantic_dicts, pydantic_v1
+
+
+class ImageUrlContent(pydantic_v1.BaseModel):
+    """
+    An image content item.
+    """
+
+    image_url: typing.Dict[str, str]
+
+    def json(self, **kwargs: typing.Any) -> str:
+        kwargs_with_defaults: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
+        return super().json(**kwargs_with_defaults)
+
+    def dict(self, **kwargs: typing.Any) -> typing.Dict[str, typing.Any]:
+        kwargs_with_defaults_exclude_unset: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
+        kwargs_with_defaults_exclude_none: typing.Any = {"by_alias": True, "exclude_none": True, **kwargs}
+
+        return deep_union_pydantic_dicts(
+            super().dict(**kwargs_with_defaults_exclude_unset), super().dict(**kwargs_with_defaults_exclude_none)
+        )
+
+    class Config:
+        frozen = True
+        smart_union = True
+        extra = pydantic_v1.Extra.allow
+        json_encoders = {dt.datetime: serialize_datetime}

athena_intelligence-0.1.123/src/athena/types/prompt_message.py

@@ -0,0 +1,35 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import datetime as dt
+import typing
+
+from ..core.datetime_utils import serialize_datetime
+from ..core.pydantic_utilities import deep_union_pydantic_dicts, pydantic_v1
+from .type import Type
+
+
+class PromptMessage(pydantic_v1.BaseModel):
+    """
+    A message to use for the structured data extractor.
+    """
+
+    content: str
+    type: Type
+
+    def json(self, **kwargs: typing.Any) -> str:
+        kwargs_with_defaults: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
+        return super().json(**kwargs_with_defaults)
+
+    def dict(self, **kwargs: typing.Any) -> typing.Dict[str, typing.Any]:
+        kwargs_with_defaults_exclude_unset: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
+        kwargs_with_defaults_exclude_none: typing.Any = {"by_alias": True, "exclude_none": True, **kwargs}
+
+        return deep_union_pydantic_dicts(
+            super().dict(**kwargs_with_defaults_exclude_unset), super().dict(**kwargs_with_defaults_exclude_none)
+        )
+
+    class Config:
+        frozen = True
+        smart_union = True
+        extra = pydantic_v1.Extra.allow
+        json_encoders = {dt.datetime: serialize_datetime}

athena_intelligence-0.1.122/src/athena/types/structured_data_extractor_reponse.py → athena_intelligence-0.1.123/src/athena/types/structured_data_extractor_response.py

@@ -5,16 +5,22 @@ import typing
 
 from ..core.datetime_utils import serialize_datetime
 from ..core.pydantic_utilities import deep_union_pydantic_dicts, pydantic_v1
+from .chunk_result import ChunkResult
 
 
-class StructuredDataExtractorReponse(pydantic_v1.BaseModel):
+class StructuredDataExtractorResponse(pydantic_v1.BaseModel):
     """
     The agent's response.
     """
 
-    data: typing.Dict[str, typing.Any] = pydantic_v1.Field()
+    chunk_by_chunk_data: typing.Optional[typing.List[ChunkResult]] = pydantic_v1.Field(default=None)
     """
-    The extracted structured data. Guaranteed to match `json_schema`.
+    The extracted structured data for each chunk. A list where each element is guaranteed to match `json_schema`.
+    """
+
+    reduced_data: typing.Optional[typing.Dict[str, typing.Any]] = pydantic_v1.Field(default=None)
+    """
+    If reduce is True, the reduced structured data, otherwise null. Guaranteed to match `json_schema`.
     """
 
     def json(self, **kwargs: typing.Any) -> str:

athena_intelligence-0.1.123/src/athena/types/text_content.py

@@ -0,0 +1,33 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import datetime as dt
+import typing
+
+from ..core.datetime_utils import serialize_datetime
+from ..core.pydantic_utilities import deep_union_pydantic_dicts, pydantic_v1
+
+
+class TextContent(pydantic_v1.BaseModel):
+    """
+    A text content item in a multimodal message content.
+    """
+
+    text: str
+
+    def json(self, **kwargs: typing.Any) -> str:
+        kwargs_with_defaults: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
+        return super().json(**kwargs_with_defaults)
+
+    def dict(self, **kwargs: typing.Any) -> typing.Dict[str, typing.Any]:
+        kwargs_with_defaults_exclude_unset: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
+        kwargs_with_defaults_exclude_none: typing.Any = {"by_alias": True, "exclude_none": True, **kwargs}
+
+        return deep_union_pydantic_dicts(
+            super().dict(**kwargs_with_defaults_exclude_unset), super().dict(**kwargs_with_defaults_exclude_none)
+        )
+
+    class Config:
+        frozen = True
+        smart_union = True
+        extra = pydantic_v1.Extra.allow
+        json_encoders = {dt.datetime: serialize_datetime}

athena_intelligence-0.1.123/src/athena/types/type.py

@@ -0,0 +1,25 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import enum
+import typing
+
+T_Result = typing.TypeVar("T_Result")
+
+
+class Type(str, enum.Enum):
+    SYSTEM = "system"
+    HUMAN = "human"
+    USER = "user"
+
+    def visit(
+        self,
+        system: typing.Callable[[], T_Result],
+        human: typing.Callable[[], T_Result],
+        user: typing.Callable[[], T_Result],
+    ) -> T_Result:
+        if self is Type.SYSTEM:
+            return system()
+        if self is Type.HUMAN:
+            return human()
+        if self is Type.USER:
+            return user()

athena_intelligence-0.1.122/src/athena/tools/structured_data_extractor/client.py

@@ -1,150 +0,0 @@
-# This file was auto-generated by Fern from our API Definition.
-
-import typing
-from json.decoder import JSONDecodeError
-
-from ...core.api_error import ApiError
-from ...core.client_wrapper import AsyncClientWrapper, SyncClientWrapper
-from ...core.pydantic_utilities import pydantic_v1
-from ...core.request_options import RequestOptions
-from ...errors.unprocessable_entity_error import UnprocessableEntityError
-from ...types.structured_data_extractor_reponse import StructuredDataExtractorReponse
-
-# this is used as the default value for optional parameters
-OMIT = typing.cast(typing.Any, ...)
-
-
-class StructuredDataExtractorClient:
-    def __init__(self, *, client_wrapper: SyncClientWrapper):
-        self._client_wrapper = client_wrapper
-
-    def invoke(
-        self,
-        *,
-        asset_ids: typing.Sequence[str],
-        json_schema: typing.Dict[str, typing.Any],
-        map_: typing.Optional[bool] = OMIT,
-        reduce: typing.Optional[bool] = OMIT,
-        request_options: typing.Optional[RequestOptions] = None
-    ) -> StructuredDataExtractorReponse:
-        """
-        Coming soon! Extract structured data from assets of arbitrary length.
-
-        Parameters
-        ----------
-        asset_ids : typing.Sequence[str]
-            The IDs of the assets from which to extract structured data matching `json_schema`.
-
-        json_schema : typing.Dict[str, typing.Any]
-            The JSON schema to use for validation (version draft 2020-12).  See the docs [here](https://json-schema.org/learn/getting-started-step-by-step).
-
-        map_ : typing.Optional[bool]
-            Whether to split the asset into chunks and attempt to extract the schema from each chunk.  Set to false if you know the asset is small.
-
-        reduce : typing.Optional[bool]
-            If `map`, whether to reduce the chunks to a single structured object (true) or return the full list (false).  Use True unless you want to preserve duplicates from each page or expect the object to overflow the output context.
-
-        request_options : typing.Optional[RequestOptions]
-            Request-specific configuration.
-
-        Returns
-        -------
-        StructuredDataExtractorReponse
-            Successful Response
-
-        Examples
-        --------
-        from athena.client import Athena
-
-        client = Athena(
-            api_key="YOUR_API_KEY",
-        )
-        client.tools.structured_data_extractor.invoke(
-            asset_ids=["asset_ids"],
-            json_schema={"key": "value"},
-        )
-        """
-        _response = self._client_wrapper.httpx_client.request(
-            "api/v0/tools/structured-data-extractor/invoke",
-            method="POST",
-            json={"asset_ids": asset_ids, "json_schema": json_schema, "map": map_, "reduce": reduce},
-            request_options=request_options,
-            omit=OMIT,
-        )
-        if 200 <= _response.status_code < 300:
-            return pydantic_v1.parse_obj_as(StructuredDataExtractorReponse, _response.json())  # type: ignore
-        if _response.status_code == 422:
-            raise UnprocessableEntityError(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
-        try:
-            _response_json = _response.json()
-        except JSONDecodeError:
-            raise ApiError(status_code=_response.status_code, body=_response.text)
-        raise ApiError(status_code=_response.status_code, body=_response_json)
-
-
-class AsyncStructuredDataExtractorClient:
-    def __init__(self, *, client_wrapper: AsyncClientWrapper):
-        self._client_wrapper = client_wrapper
-
-    async def invoke(
-        self,
-        *,
-        asset_ids: typing.Sequence[str],
-        json_schema: typing.Dict[str, typing.Any],
-        map_: typing.Optional[bool] = OMIT,
-        reduce: typing.Optional[bool] = OMIT,
-        request_options: typing.Optional[RequestOptions] = None
-    ) -> StructuredDataExtractorReponse:
-        """
-        Coming soon! Extract structured data from assets of arbitrary length.
-
-        Parameters
-        ----------
-        asset_ids : typing.Sequence[str]
-            The IDs of the assets from which to extract structured data matching `json_schema`.
-
-        json_schema : typing.Dict[str, typing.Any]
-            The JSON schema to use for validation (version draft 2020-12).  See the docs [here](https://json-schema.org/learn/getting-started-step-by-step).
-
-        map_ : typing.Optional[bool]
-            Whether to split the asset into chunks and attempt to extract the schema from each chunk.  Set to false if you know the asset is small.
-
-        reduce : typing.Optional[bool]
-            If `map`, whether to reduce the chunks to a single structured object (true) or return the full list (false).  Use True unless you want to preserve duplicates from each page or expect the object to overflow the output context.
-
-        request_options : typing.Optional[RequestOptions]
-            Request-specific configuration.
-
-        Returns
-        -------
-        StructuredDataExtractorReponse
-            Successful Response
-
-        Examples
-        --------
-        from athena.client import AsyncAthena
-
-        client = AsyncAthena(
-            api_key="YOUR_API_KEY",
-        )
-        await client.tools.structured_data_extractor.invoke(
-            asset_ids=["asset_ids"],
-            json_schema={"key": "value"},
-        )
-        """
-        _response = await self._client_wrapper.httpx_client.request(
-            "api/v0/tools/structured-data-extractor/invoke",
-            method="POST",
-            json={"asset_ids": asset_ids, "json_schema": json_schema, "map": map_, "reduce": reduce},
-            request_options=request_options,
-            omit=OMIT,
-        )
-        if 200 <= _response.status_code < 300:
-            return pydantic_v1.parse_obj_as(StructuredDataExtractorReponse, _response.json())  # type: ignore
-        if _response.status_code == 422:
-            raise UnprocessableEntityError(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
-        try:
-            _response_json = _response.json()
-        except JSONDecodeError:
-            raise ApiError(status_code=_response.status_code, body=_response.text)
-        raise ApiError(status_code=_response.status_code, body=_response_json)