From 9954f7fa0c50409ae21107b783eacc938d773c32 Mon Sep 17 00:00:00 2001 From: Edwin Jose Date: Fri, 20 Jun 2025 14:53:40 -0500 Subject: [PATCH] feat: update input_value field to use MultilineInput (#8583) * Update structured_output.py * [autofix.ci] apply automated fixes * Update Image Sentiment Analysis.json * [autofix.ci] apply automated fixes * Update Image Sentiment Analysis.json --------- Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com> --- .../processing/structured_output.py | 2 +- .../Financial Report Parser.json | 2 +- .../starter_projects/Hybrid Search RAG.json | 2 +- .../Image Sentiment Analysis.json | 133 +++++++++--------- .../starter_projects/Market Research.json | 2 +- .../Portfolio Website Code Generator.json | 2 +- 6 files changed, 72 insertions(+), 71 deletions(-) diff --git a/src/backend/base/langflow/components/processing/structured_output.py b/src/backend/base/langflow/components/processing/structured_output.py index f898c6854..27112cc52 100644 --- a/src/backend/base/langflow/components/processing/structured_output.py +++ b/src/backend/base/langflow/components/processing/structured_output.py @@ -29,7 +29,7 @@ class StructuredOutputComponent(Component): input_types=["LanguageModel"], required=True, ), - MessageTextInput( + MultilineInput( name="input_value", display_name="Input Message", info="The input message to the language model.", diff --git a/src/backend/base/langflow/initial_setup/starter_projects/Financial Report Parser.json b/src/backend/base/langflow/initial_setup/starter_projects/Financial Report Parser.json index d6cef59f0..1db1b4d83 100644 --- a/src/backend/base/langflow/initial_setup/starter_projects/Financial Report Parser.json +++ b/src/backend/base/langflow/initial_setup/starter_projects/Financial Report Parser.json @@ -1254,7 +1254,7 @@ "show": true, "title_case": false, "type": "code", - "value": "from pydantic import BaseModel, Field, create_model\nfrom trustcall import create_extractor\n\nfrom langflow.base.models.chat_result import get_chat_result\nfrom langflow.custom.custom_component.component import Component\nfrom langflow.helpers.base_model import build_model_from_schema\nfrom langflow.io import (\n HandleInput,\n MessageTextInput,\n MultilineInput,\n Output,\n TableInput,\n)\nfrom langflow.schema.data import Data\nfrom langflow.schema.table import EditMode\n\n\nclass StructuredOutputComponent(Component):\n display_name = \"Structured Output\"\n description = \"Uses an LLM to generate structured data. Ideal for extraction and consistency.\"\n name = \"StructuredOutput\"\n icon = \"braces\"\n\n inputs = [\n HandleInput(\n name=\"llm\",\n display_name=\"Language Model\",\n info=\"The language model to use to generate the structured output.\",\n input_types=[\"LanguageModel\"],\n required=True,\n ),\n MessageTextInput(\n name=\"input_value\",\n display_name=\"Input Message\",\n info=\"The input message to the language model.\",\n tool_mode=True,\n required=True,\n ),\n MultilineInput(\n name=\"system_prompt\",\n display_name=\"Format Instructions\",\n info=\"The instructions to the language model for formatting the output.\",\n value=(\n \"You are an AI system designed to extract structured information from unstructured text.\"\n \"Given the input_text, return a JSON object with predefined keys based on the expected structure.\"\n \"Extract values accurately and format them according to the specified type \"\n \"(e.g., string, integer, float, date).\"\n \"If a value is missing or cannot be determined, return a default \"\n \"(e.g., null, 0, or 'N/A').\"\n \"If multiple instances of the expected structure exist within the input_text, \"\n \"stream each as a separate JSON object.\"\n ),\n required=True,\n advanced=True,\n ),\n MessageTextInput(\n name=\"schema_name\",\n display_name=\"Schema Name\",\n info=\"Provide a name for the output data schema.\",\n advanced=True,\n ),\n TableInput(\n name=\"output_schema\",\n display_name=\"Output Schema\",\n info=\"Define the structure and data types for the model's output.\",\n required=True,\n # TODO: remove deault value\n table_schema=[\n {\n \"name\": \"name\",\n \"display_name\": \"Name\",\n \"type\": \"str\",\n \"description\": \"Specify the name of the output field.\",\n \"default\": \"field\",\n \"edit_mode\": EditMode.INLINE,\n },\n {\n \"name\": \"description\",\n \"display_name\": \"Description\",\n \"type\": \"str\",\n \"description\": \"Describe the purpose of the output field.\",\n \"default\": \"description of field\",\n \"edit_mode\": EditMode.POPOVER,\n },\n {\n \"name\": \"type\",\n \"display_name\": \"Type\",\n \"type\": \"str\",\n \"edit_mode\": EditMode.INLINE,\n \"description\": (\"Indicate the data type of the output field (e.g., str, int, float, bool, dict).\"),\n \"options\": [\"str\", \"int\", \"float\", \"bool\", \"dict\"],\n \"default\": \"str\",\n },\n {\n \"name\": \"multiple\",\n \"display_name\": \"As List\",\n \"type\": \"boolean\",\n \"description\": \"Set to True if this output field should be a list of the specified type.\",\n \"default\": \"False\",\n \"edit_mode\": EditMode.INLINE,\n },\n ],\n value=[\n {\n \"name\": \"field\",\n \"description\": \"description of field\",\n \"type\": \"str\",\n \"multiple\": \"False\",\n }\n ],\n ),\n ]\n\n outputs = [\n Output(\n name=\"structured_output\",\n display_name=\"Structured Output\",\n method=\"build_structured_output\",\n ),\n ]\n\n def build_structured_output_base(self):\n schema_name = self.schema_name or \"OutputModel\"\n\n if not hasattr(self.llm, \"with_structured_output\"):\n msg = \"Language model does not support structured output.\"\n raise TypeError(msg)\n if not self.output_schema:\n msg = \"Output schema cannot be empty\"\n raise ValueError(msg)\n\n output_model_ = build_model_from_schema(self.output_schema)\n\n output_model = create_model(\n schema_name,\n __doc__=f\"A list of {schema_name}.\",\n objects=(list[output_model_], Field(description=f\"A list of {schema_name}.\")), # type: ignore[valid-type]\n )\n\n try:\n llm_with_structured_output = create_extractor(self.llm, tools=[output_model])\n except NotImplementedError as exc:\n msg = f\"{self.llm.__class__.__name__} does not support structured output.\"\n raise TypeError(msg) from exc\n\n config_dict = {\n \"run_name\": self.display_name,\n \"project_name\": self.get_project_name(),\n \"callbacks\": self.get_langchain_callbacks(),\n }\n result = get_chat_result(\n runnable=llm_with_structured_output,\n system_message=self.system_prompt,\n input_value=self.input_value,\n config=config_dict,\n )\n\n # OPTIMIZATION NOTE: Simplified processing based on trustcall response structure\n # Handle non-dict responses (shouldn't happen with trustcall, but defensive)\n if not isinstance(result, dict):\n return result\n\n # Extract first response and convert BaseModel to dict\n responses = result.get(\"responses\", [])\n if not responses:\n return result\n\n # Convert BaseModel to dict (creates the \"objects\" key)\n first_response = responses[0]\n structured_data = first_response.model_dump() if isinstance(first_response, BaseModel) else first_response\n\n # Extract the objects array (guaranteed to exist due to our Pydantic model structure)\n return structured_data.get(\"objects\", structured_data)\n\n def build_structured_output(self) -> Data:\n output = self.build_structured_output_base()\n if not isinstance(output, list) or not output:\n # handle empty or unexpected type case\n msg = \"No structured output returned\"\n raise ValueError(msg)\n if len(output) != 1:\n msg = \"Multiple structured outputs returned\"\n raise ValueError(msg)\n return Data(data=output[0])\n" + "value": "from pydantic import BaseModel, Field, create_model\nfrom trustcall import create_extractor\n\nfrom langflow.base.models.chat_result import get_chat_result\nfrom langflow.custom.custom_component.component import Component\nfrom langflow.helpers.base_model import build_model_from_schema\nfrom langflow.io import (\n HandleInput,\n MessageTextInput,\n MultilineInput,\n Output,\n TableInput,\n)\nfrom langflow.schema.data import Data\nfrom langflow.schema.table import EditMode\n\n\nclass StructuredOutputComponent(Component):\n display_name = \"Structured Output\"\n description = \"Uses an LLM to generate structured data. Ideal for extraction and consistency.\"\n name = \"StructuredOutput\"\n icon = \"braces\"\n\n inputs = [\n HandleInput(\n name=\"llm\",\n display_name=\"Language Model\",\n info=\"The language model to use to generate the structured output.\",\n input_types=[\"LanguageModel\"],\n required=True,\n ),\n MultilineInput(\n name=\"input_value\",\n display_name=\"Input Message\",\n info=\"The input message to the language model.\",\n tool_mode=True,\n required=True,\n ),\n MultilineInput(\n name=\"system_prompt\",\n display_name=\"Format Instructions\",\n info=\"The instructions to the language model for formatting the output.\",\n value=(\n \"You are an AI system designed to extract structured information from unstructured text.\"\n \"Given the input_text, return a JSON object with predefined keys based on the expected structure.\"\n \"Extract values accurately and format them according to the specified type \"\n \"(e.g., string, integer, float, date).\"\n \"If a value is missing or cannot be determined, return a default \"\n \"(e.g., null, 0, or 'N/A').\"\n \"If multiple instances of the expected structure exist within the input_text, \"\n \"stream each as a separate JSON object.\"\n ),\n required=True,\n advanced=True,\n ),\n MessageTextInput(\n name=\"schema_name\",\n display_name=\"Schema Name\",\n info=\"Provide a name for the output data schema.\",\n advanced=True,\n ),\n TableInput(\n name=\"output_schema\",\n display_name=\"Output Schema\",\n info=\"Define the structure and data types for the model's output.\",\n required=True,\n # TODO: remove deault value\n table_schema=[\n {\n \"name\": \"name\",\n \"display_name\": \"Name\",\n \"type\": \"str\",\n \"description\": \"Specify the name of the output field.\",\n \"default\": \"field\",\n \"edit_mode\": EditMode.INLINE,\n },\n {\n \"name\": \"description\",\n \"display_name\": \"Description\",\n \"type\": \"str\",\n \"description\": \"Describe the purpose of the output field.\",\n \"default\": \"description of field\",\n \"edit_mode\": EditMode.POPOVER,\n },\n {\n \"name\": \"type\",\n \"display_name\": \"Type\",\n \"type\": \"str\",\n \"edit_mode\": EditMode.INLINE,\n \"description\": (\"Indicate the data type of the output field (e.g., str, int, float, bool, dict).\"),\n \"options\": [\"str\", \"int\", \"float\", \"bool\", \"dict\"],\n \"default\": \"str\",\n },\n {\n \"name\": \"multiple\",\n \"display_name\": \"As List\",\n \"type\": \"boolean\",\n \"description\": \"Set to True if this output field should be a list of the specified type.\",\n \"default\": \"False\",\n \"edit_mode\": EditMode.INLINE,\n },\n ],\n value=[\n {\n \"name\": \"field\",\n \"description\": \"description of field\",\n \"type\": \"str\",\n \"multiple\": \"False\",\n }\n ],\n ),\n ]\n\n outputs = [\n Output(\n name=\"structured_output\",\n display_name=\"Structured Output\",\n method=\"build_structured_output\",\n ),\n ]\n\n def build_structured_output_base(self):\n schema_name = self.schema_name or \"OutputModel\"\n\n if not hasattr(self.llm, \"with_structured_output\"):\n msg = \"Language model does not support structured output.\"\n raise TypeError(msg)\n if not self.output_schema:\n msg = \"Output schema cannot be empty\"\n raise ValueError(msg)\n\n output_model_ = build_model_from_schema(self.output_schema)\n\n output_model = create_model(\n schema_name,\n __doc__=f\"A list of {schema_name}.\",\n objects=(list[output_model_], Field(description=f\"A list of {schema_name}.\")), # type: ignore[valid-type]\n )\n\n try:\n llm_with_structured_output = create_extractor(self.llm, tools=[output_model])\n except NotImplementedError as exc:\n msg = f\"{self.llm.__class__.__name__} does not support structured output.\"\n raise TypeError(msg) from exc\n\n config_dict = {\n \"run_name\": self.display_name,\n \"project_name\": self.get_project_name(),\n \"callbacks\": self.get_langchain_callbacks(),\n }\n result = get_chat_result(\n runnable=llm_with_structured_output,\n system_message=self.system_prompt,\n input_value=self.input_value,\n config=config_dict,\n )\n\n # OPTIMIZATION NOTE: Simplified processing based on trustcall response structure\n # Handle non-dict responses (shouldn't happen with trustcall, but defensive)\n if not isinstance(result, dict):\n return result\n\n # Extract first response and convert BaseModel to dict\n responses = result.get(\"responses\", [])\n if not responses:\n return result\n\n # Convert BaseModel to dict (creates the \"objects\" key)\n first_response = responses[0]\n structured_data = first_response.model_dump() if isinstance(first_response, BaseModel) else first_response\n\n # Extract the objects array (guaranteed to exist due to our Pydantic model structure)\n return structured_data.get(\"objects\", structured_data)\n\n def build_structured_output(self) -> Data:\n output = self.build_structured_output_base()\n if not isinstance(output, list) or not output:\n # handle empty or unexpected type case\n msg = \"No structured output returned\"\n raise ValueError(msg)\n if len(output) != 1:\n msg = \"Multiple structured outputs returned\"\n raise ValueError(msg)\n return Data(data=output[0])\n" }, "input_value": { "_input_type": "MessageTextInput", diff --git a/src/backend/base/langflow/initial_setup/starter_projects/Hybrid Search RAG.json b/src/backend/base/langflow/initial_setup/starter_projects/Hybrid Search RAG.json index 16404de3d..c53da20a9 100644 --- a/src/backend/base/langflow/initial_setup/starter_projects/Hybrid Search RAG.json +++ b/src/backend/base/langflow/initial_setup/starter_projects/Hybrid Search RAG.json @@ -609,7 +609,7 @@ "show": true, "title_case": false, "type": "code", - "value": "from pydantic import BaseModel, Field, create_model\nfrom trustcall import create_extractor\n\nfrom langflow.base.models.chat_result import get_chat_result\nfrom langflow.custom.custom_component.component import Component\nfrom langflow.helpers.base_model import build_model_from_schema\nfrom langflow.io import (\n HandleInput,\n MessageTextInput,\n MultilineInput,\n Output,\n TableInput,\n)\nfrom langflow.schema.data import Data\nfrom langflow.schema.table import EditMode\n\n\nclass StructuredOutputComponent(Component):\n display_name = \"Structured Output\"\n description = \"Uses an LLM to generate structured data. Ideal for extraction and consistency.\"\n name = \"StructuredOutput\"\n icon = \"braces\"\n\n inputs = [\n HandleInput(\n name=\"llm\",\n display_name=\"Language Model\",\n info=\"The language model to use to generate the structured output.\",\n input_types=[\"LanguageModel\"],\n required=True,\n ),\n MessageTextInput(\n name=\"input_value\",\n display_name=\"Input Message\",\n info=\"The input message to the language model.\",\n tool_mode=True,\n required=True,\n ),\n MultilineInput(\n name=\"system_prompt\",\n display_name=\"Format Instructions\",\n info=\"The instructions to the language model for formatting the output.\",\n value=(\n \"You are an AI system designed to extract structured information from unstructured text.\"\n \"Given the input_text, return a JSON object with predefined keys based on the expected structure.\"\n \"Extract values accurately and format them according to the specified type \"\n \"(e.g., string, integer, float, date).\"\n \"If a value is missing or cannot be determined, return a default \"\n \"(e.g., null, 0, or 'N/A').\"\n \"If multiple instances of the expected structure exist within the input_text, \"\n \"stream each as a separate JSON object.\"\n ),\n required=True,\n advanced=True,\n ),\n MessageTextInput(\n name=\"schema_name\",\n display_name=\"Schema Name\",\n info=\"Provide a name for the output data schema.\",\n advanced=True,\n ),\n TableInput(\n name=\"output_schema\",\n display_name=\"Output Schema\",\n info=\"Define the structure and data types for the model's output.\",\n required=True,\n # TODO: remove deault value\n table_schema=[\n {\n \"name\": \"name\",\n \"display_name\": \"Name\",\n \"type\": \"str\",\n \"description\": \"Specify the name of the output field.\",\n \"default\": \"field\",\n \"edit_mode\": EditMode.INLINE,\n },\n {\n \"name\": \"description\",\n \"display_name\": \"Description\",\n \"type\": \"str\",\n \"description\": \"Describe the purpose of the output field.\",\n \"default\": \"description of field\",\n \"edit_mode\": EditMode.POPOVER,\n },\n {\n \"name\": \"type\",\n \"display_name\": \"Type\",\n \"type\": \"str\",\n \"edit_mode\": EditMode.INLINE,\n \"description\": (\"Indicate the data type of the output field (e.g., str, int, float, bool, dict).\"),\n \"options\": [\"str\", \"int\", \"float\", \"bool\", \"dict\"],\n \"default\": \"str\",\n },\n {\n \"name\": \"multiple\",\n \"display_name\": \"As List\",\n \"type\": \"boolean\",\n \"description\": \"Set to True if this output field should be a list of the specified type.\",\n \"default\": \"False\",\n \"edit_mode\": EditMode.INLINE,\n },\n ],\n value=[\n {\n \"name\": \"field\",\n \"description\": \"description of field\",\n \"type\": \"str\",\n \"multiple\": \"False\",\n }\n ],\n ),\n ]\n\n outputs = [\n Output(\n name=\"structured_output\",\n display_name=\"Structured Output\",\n method=\"build_structured_output\",\n ),\n ]\n\n def build_structured_output_base(self):\n schema_name = self.schema_name or \"OutputModel\"\n\n if not hasattr(self.llm, \"with_structured_output\"):\n msg = \"Language model does not support structured output.\"\n raise TypeError(msg)\n if not self.output_schema:\n msg = \"Output schema cannot be empty\"\n raise ValueError(msg)\n\n output_model_ = build_model_from_schema(self.output_schema)\n\n output_model = create_model(\n schema_name,\n __doc__=f\"A list of {schema_name}.\",\n objects=(list[output_model_], Field(description=f\"A list of {schema_name}.\")), # type: ignore[valid-type]\n )\n\n try:\n llm_with_structured_output = create_extractor(self.llm, tools=[output_model])\n except NotImplementedError as exc:\n msg = f\"{self.llm.__class__.__name__} does not support structured output.\"\n raise TypeError(msg) from exc\n\n config_dict = {\n \"run_name\": self.display_name,\n \"project_name\": self.get_project_name(),\n \"callbacks\": self.get_langchain_callbacks(),\n }\n result = get_chat_result(\n runnable=llm_with_structured_output,\n system_message=self.system_prompt,\n input_value=self.input_value,\n config=config_dict,\n )\n\n # OPTIMIZATION NOTE: Simplified processing based on trustcall response structure\n # Handle non-dict responses (shouldn't happen with trustcall, but defensive)\n if not isinstance(result, dict):\n return result\n\n # Extract first response and convert BaseModel to dict\n responses = result.get(\"responses\", [])\n if not responses:\n return result\n\n # Convert BaseModel to dict (creates the \"objects\" key)\n first_response = responses[0]\n structured_data = first_response.model_dump() if isinstance(first_response, BaseModel) else first_response\n\n # Extract the objects array (guaranteed to exist due to our Pydantic model structure)\n return structured_data.get(\"objects\", structured_data)\n\n def build_structured_output(self) -> Data:\n output = self.build_structured_output_base()\n if not isinstance(output, list) or not output:\n # handle empty or unexpected type case\n msg = \"No structured output returned\"\n raise ValueError(msg)\n if len(output) != 1:\n msg = \"Multiple structured outputs returned\"\n raise ValueError(msg)\n return Data(data=output[0])\n" + "value": "from pydantic import BaseModel, Field, create_model\nfrom trustcall import create_extractor\n\nfrom langflow.base.models.chat_result import get_chat_result\nfrom langflow.custom.custom_component.component import Component\nfrom langflow.helpers.base_model import build_model_from_schema\nfrom langflow.io import (\n HandleInput,\n MessageTextInput,\n MultilineInput,\n Output,\n TableInput,\n)\nfrom langflow.schema.data import Data\nfrom langflow.schema.table import EditMode\n\n\nclass StructuredOutputComponent(Component):\n display_name = \"Structured Output\"\n description = \"Uses an LLM to generate structured data. Ideal for extraction and consistency.\"\n name = \"StructuredOutput\"\n icon = \"braces\"\n\n inputs = [\n HandleInput(\n name=\"llm\",\n display_name=\"Language Model\",\n info=\"The language model to use to generate the structured output.\",\n input_types=[\"LanguageModel\"],\n required=True,\n ),\n MultilineInput(\n name=\"input_value\",\n display_name=\"Input Message\",\n info=\"The input message to the language model.\",\n tool_mode=True,\n required=True,\n ),\n MultilineInput(\n name=\"system_prompt\",\n display_name=\"Format Instructions\",\n info=\"The instructions to the language model for formatting the output.\",\n value=(\n \"You are an AI system designed to extract structured information from unstructured text.\"\n \"Given the input_text, return a JSON object with predefined keys based on the expected structure.\"\n \"Extract values accurately and format them according to the specified type \"\n \"(e.g., string, integer, float, date).\"\n \"If a value is missing or cannot be determined, return a default \"\n \"(e.g., null, 0, or 'N/A').\"\n \"If multiple instances of the expected structure exist within the input_text, \"\n \"stream each as a separate JSON object.\"\n ),\n required=True,\n advanced=True,\n ),\n MessageTextInput(\n name=\"schema_name\",\n display_name=\"Schema Name\",\n info=\"Provide a name for the output data schema.\",\n advanced=True,\n ),\n TableInput(\n name=\"output_schema\",\n display_name=\"Output Schema\",\n info=\"Define the structure and data types for the model's output.\",\n required=True,\n # TODO: remove deault value\n table_schema=[\n {\n \"name\": \"name\",\n \"display_name\": \"Name\",\n \"type\": \"str\",\n \"description\": \"Specify the name of the output field.\",\n \"default\": \"field\",\n \"edit_mode\": EditMode.INLINE,\n },\n {\n \"name\": \"description\",\n \"display_name\": \"Description\",\n \"type\": \"str\",\n \"description\": \"Describe the purpose of the output field.\",\n \"default\": \"description of field\",\n \"edit_mode\": EditMode.POPOVER,\n },\n {\n \"name\": \"type\",\n \"display_name\": \"Type\",\n \"type\": \"str\",\n \"edit_mode\": EditMode.INLINE,\n \"description\": (\"Indicate the data type of the output field (e.g., str, int, float, bool, dict).\"),\n \"options\": [\"str\", \"int\", \"float\", \"bool\", \"dict\"],\n \"default\": \"str\",\n },\n {\n \"name\": \"multiple\",\n \"display_name\": \"As List\",\n \"type\": \"boolean\",\n \"description\": \"Set to True if this output field should be a list of the specified type.\",\n \"default\": \"False\",\n \"edit_mode\": EditMode.INLINE,\n },\n ],\n value=[\n {\n \"name\": \"field\",\n \"description\": \"description of field\",\n \"type\": \"str\",\n \"multiple\": \"False\",\n }\n ],\n ),\n ]\n\n outputs = [\n Output(\n name=\"structured_output\",\n display_name=\"Structured Output\",\n method=\"build_structured_output\",\n ),\n ]\n\n def build_structured_output_base(self):\n schema_name = self.schema_name or \"OutputModel\"\n\n if not hasattr(self.llm, \"with_structured_output\"):\n msg = \"Language model does not support structured output.\"\n raise TypeError(msg)\n if not self.output_schema:\n msg = \"Output schema cannot be empty\"\n raise ValueError(msg)\n\n output_model_ = build_model_from_schema(self.output_schema)\n\n output_model = create_model(\n schema_name,\n __doc__=f\"A list of {schema_name}.\",\n objects=(list[output_model_], Field(description=f\"A list of {schema_name}.\")), # type: ignore[valid-type]\n )\n\n try:\n llm_with_structured_output = create_extractor(self.llm, tools=[output_model])\n except NotImplementedError as exc:\n msg = f\"{self.llm.__class__.__name__} does not support structured output.\"\n raise TypeError(msg) from exc\n\n config_dict = {\n \"run_name\": self.display_name,\n \"project_name\": self.get_project_name(),\n \"callbacks\": self.get_langchain_callbacks(),\n }\n result = get_chat_result(\n runnable=llm_with_structured_output,\n system_message=self.system_prompt,\n input_value=self.input_value,\n config=config_dict,\n )\n\n # OPTIMIZATION NOTE: Simplified processing based on trustcall response structure\n # Handle non-dict responses (shouldn't happen with trustcall, but defensive)\n if not isinstance(result, dict):\n return result\n\n # Extract first response and convert BaseModel to dict\n responses = result.get(\"responses\", [])\n if not responses:\n return result\n\n # Convert BaseModel to dict (creates the \"objects\" key)\n first_response = responses[0]\n structured_data = first_response.model_dump() if isinstance(first_response, BaseModel) else first_response\n\n # Extract the objects array (guaranteed to exist due to our Pydantic model structure)\n return structured_data.get(\"objects\", structured_data)\n\n def build_structured_output(self) -> Data:\n output = self.build_structured_output_base()\n if not isinstance(output, list) or not output:\n # handle empty or unexpected type case\n msg = \"No structured output returned\"\n raise ValueError(msg)\n if len(output) != 1:\n msg = \"Multiple structured outputs returned\"\n raise ValueError(msg)\n return Data(data=output[0])\n" }, "input_value": { "_input_type": "MessageTextInput", diff --git a/src/backend/base/langflow/initial_setup/starter_projects/Image Sentiment Analysis.json b/src/backend/base/langflow/initial_setup/starter_projects/Image Sentiment Analysis.json index 30a76e6fb..97bd390a0 100644 --- a/src/backend/base/langflow/initial_setup/starter_projects/Image Sentiment Analysis.json +++ b/src/backend/base/langflow/initial_setup/starter_projects/Image Sentiment Analysis.json @@ -7,7 +7,7 @@ "data": { "sourceHandle": { "dataType": "parser", - "id": "parser-sIcOW", + "id": "parser-o6H8E", "name": "parsed_text", "output_types": [ "Message" @@ -15,7 +15,7 @@ }, "targetHandle": { "fieldName": "input_value", - "id": "ChatOutput-1VmAz", + "id": "ChatOutput-WVlr5", "inputTypes": [ "Data", "DataFrame", @@ -24,12 +24,12 @@ "type": "str" } }, - "id": "reactflow__edge-parser-sIcOW{œdataTypeœ:œparserœ,œidœ:œparser-sIcOWœ,œnameœ:œparsed_textœ,œoutput_typesœ:[œMessageœ]}-ChatOutput-1VmAz{œfieldNameœ:œinput_valueœ,œidœ:œChatOutput-1VmAzœ,œinputTypesœ:[œDataœ,œDataFrameœ,œMessageœ],œtypeœ:œstrœ}", + "id": "reactflow__edge-parser-o6H8E{œdataTypeœ:œparserœ,œidœ:œparser-o6H8Eœ,œnameœ:œparsed_textœ,œoutput_typesœ:[œMessageœ]}-ChatOutput-WVlr5{œfieldNameœ:œinput_valueœ,œidœ:œChatOutput-WVlr5œ,œinputTypesœ:[œDataœ,œDataFrameœ,œMessageœ],œtypeœ:œstrœ}", "selected": false, - "source": "parser-sIcOW", - "sourceHandle": "{œdataTypeœ: œparserœ, œidœ: œparser-sIcOWœ, œnameœ: œparsed_textœ, œoutput_typesœ: [œMessageœ]}", - "target": "ChatOutput-1VmAz", - "targetHandle": "{œfieldNameœ: œinput_valueœ, œidœ: œChatOutput-1VmAzœ, œinputTypesœ: [œDataœ, œDataFrameœ, œMessageœ], œtypeœ: œstrœ}" + "source": "parser-o6H8E", + "sourceHandle": "{œdataTypeœ: œparserœ, œidœ: œparser-o6H8Eœ, œnameœ: œparsed_textœ, œoutput_typesœ: [œMessageœ]}", + "target": "ChatOutput-WVlr5", + "targetHandle": "{œfieldNameœ: œinput_valueœ, œidœ: œChatOutput-WVlr5œ, œinputTypesœ: [œDataœ, œDataFrameœ, œMessageœ], œtypeœ: œstrœ}" }, { "animated": false, @@ -37,7 +37,7 @@ "data": { "sourceHandle": { "dataType": "StructuredOutput", - "id": "StructuredOutput-J6Rvk", + "id": "StructuredOutput-eoFyT", "name": "structured_output", "output_types": [ "Data" @@ -45,7 +45,7 @@ }, "targetHandle": { "fieldName": "input_data", - "id": "parser-sIcOW", + "id": "parser-o6H8E", "inputTypes": [ "DataFrame", "Data" @@ -53,12 +53,12 @@ "type": "other" } }, - "id": "reactflow__edge-StructuredOutput-J6Rvk{œdataTypeœ:œStructuredOutputœ,œidœ:œStructuredOutput-J6Rvkœ,œnameœ:œstructured_outputœ,œoutput_typesœ:[œDataœ]}-parser-sIcOW{œfieldNameœ:œinput_dataœ,œidœ:œparser-sIcOWœ,œinputTypesœ:[œDataFrameœ,œDataœ],œtypeœ:œotherœ}", + "id": "reactflow__edge-StructuredOutput-eoFyT{œdataTypeœ:œStructuredOutputœ,œidœ:œStructuredOutput-eoFyTœ,œnameœ:œstructured_outputœ,œoutput_typesœ:[œDataœ]}-parser-o6H8E{œfieldNameœ:œinput_dataœ,œidœ:œparser-o6H8Eœ,œinputTypesœ:[œDataFrameœ,œDataœ],œtypeœ:œotherœ}", "selected": false, - "source": "StructuredOutput-J6Rvk", - "sourceHandle": "{œdataTypeœ: œStructuredOutputœ, œidœ: œStructuredOutput-J6Rvkœ, œnameœ: œstructured_outputœ, œoutput_typesœ: [œDataœ]}", - "target": "parser-sIcOW", - "targetHandle": "{œfieldNameœ: œinput_dataœ, œidœ: œparser-sIcOWœ, œinputTypesœ: [œDataFrameœ, œDataœ], œtypeœ: œotherœ}" + "source": "StructuredOutput-eoFyT", + "sourceHandle": "{œdataTypeœ: œStructuredOutputœ, œidœ: œStructuredOutput-eoFyTœ, œnameœ: œstructured_outputœ, œoutput_typesœ: [œDataœ]}", + "target": "parser-o6H8E", + "targetHandle": "{œfieldNameœ: œinput_dataœ, œidœ: œparser-o6H8Eœ, œinputTypesœ: [œDataFrameœ, œDataœ], œtypeœ: œotherœ}" }, { "animated": false, @@ -66,7 +66,7 @@ "data": { "sourceHandle": { "dataType": "LanguageModelComponent", - "id": "LanguageModelComponent-x2hKm", + "id": "LanguageModelComponent-2h2qm", "name": "model_output", "output_types": [ "LanguageModel" @@ -74,19 +74,19 @@ }, "targetHandle": { "fieldName": "llm", - "id": "StructuredOutput-J6Rvk", + "id": "StructuredOutput-eoFyT", "inputTypes": [ "LanguageModel" ], "type": "other" } }, - "id": "reactflow__edge-LanguageModelComponent-x2hKm{œdataTypeœ:œLanguageModelComponentœ,œidœ:œLanguageModelComponent-x2hKmœ,œnameœ:œmodel_outputœ,œoutput_typesœ:[œLanguageModelœ]}-StructuredOutput-J6Rvk{œfieldNameœ:œllmœ,œidœ:œStructuredOutput-J6Rvkœ,œinputTypesœ:[œLanguageModelœ],œtypeœ:œotherœ}", + "id": "reactflow__edge-LanguageModelComponent-2h2qm{œdataTypeœ:œLanguageModelComponentœ,œidœ:œLanguageModelComponent-2h2qmœ,œnameœ:œmodel_outputœ,œoutput_typesœ:[œLanguageModelœ]}-StructuredOutput-eoFyT{œfieldNameœ:œllmœ,œidœ:œStructuredOutput-eoFyTœ,œinputTypesœ:[œLanguageModelœ],œtypeœ:œotherœ}", "selected": false, - "source": "LanguageModelComponent-x2hKm", - "sourceHandle": "{œdataTypeœ: œLanguageModelComponentœ, œidœ: œLanguageModelComponent-x2hKmœ, œnameœ: œmodel_outputœ, œoutput_typesœ: [œLanguageModelœ]}", - "target": "StructuredOutput-J6Rvk", - "targetHandle": "{œfieldNameœ: œllmœ, œidœ: œStructuredOutput-J6Rvkœ, œinputTypesœ: [œLanguageModelœ], œtypeœ: œotherœ}" + "source": "LanguageModelComponent-2h2qm", + "sourceHandle": "{œdataTypeœ: œLanguageModelComponentœ, œidœ: œLanguageModelComponent-2h2qmœ, œnameœ: œmodel_outputœ, œoutput_typesœ: [œLanguageModelœ]}", + "target": "StructuredOutput-eoFyT", + "targetHandle": "{œfieldNameœ: œllmœ, œidœ: œStructuredOutput-eoFyTœ, œinputTypesœ: [œLanguageModelœ], œtypeœ: œotherœ}" }, { "animated": false, @@ -94,7 +94,7 @@ "data": { "sourceHandle": { "dataType": "Prompt", - "id": "Prompt-AGdss", + "id": "Prompt-ElcZb", "name": "prompt", "output_types": [ "Message" @@ -102,19 +102,19 @@ }, "targetHandle": { "fieldName": "system_message", - "id": "LanguageModelComponent-tAUaX", + "id": "LanguageModelComponent-YHUqJ", "inputTypes": [ "Message" ], "type": "str" } }, - "id": "reactflow__edge-Prompt-AGdss{œdataTypeœ:œPromptœ,œidœ:œPrompt-AGdssœ,œnameœ:œpromptœ,œoutput_typesœ:[œMessageœ]}-LanguageModelComponent-tAUaX{œfieldNameœ:œsystem_messageœ,œidœ:œLanguageModelComponent-tAUaXœ,œinputTypesœ:[œMessageœ],œtypeœ:œstrœ}", + "id": "reactflow__edge-Prompt-ElcZb{œdataTypeœ:œPromptœ,œidœ:œPrompt-ElcZbœ,œnameœ:œpromptœ,œoutput_typesœ:[œMessageœ]}-LanguageModelComponent-YHUqJ{œfieldNameœ:œsystem_messageœ,œidœ:œLanguageModelComponent-YHUqJœ,œinputTypesœ:[œMessageœ],œtypeœ:œstrœ}", "selected": false, - "source": "Prompt-AGdss", - "sourceHandle": "{œdataTypeœ: œPromptœ, œidœ: œPrompt-AGdssœ, œnameœ: œpromptœ, œoutput_typesœ: [œMessageœ]}", - "target": "LanguageModelComponent-tAUaX", - "targetHandle": "{œfieldNameœ: œsystem_messageœ, œidœ: œLanguageModelComponent-tAUaXœ, œinputTypesœ: [œMessageœ], œtypeœ: œstrœ}" + "source": "Prompt-ElcZb", + "sourceHandle": "{œdataTypeœ: œPromptœ, œidœ: œPrompt-ElcZbœ, œnameœ: œpromptœ, œoutput_typesœ: [œMessageœ]}", + "target": "LanguageModelComponent-YHUqJ", + "targetHandle": "{œfieldNameœ: œsystem_messageœ, œidœ: œLanguageModelComponent-YHUqJœ, œinputTypesœ: [œMessageœ], œtypeœ: œstrœ}" }, { "animated": false, @@ -122,7 +122,7 @@ "data": { "sourceHandle": { "dataType": "ChatInput", - "id": "ChatInput-osYcH", + "id": "ChatInput-CbbPG", "name": "message", "output_types": [ "Message" @@ -130,19 +130,19 @@ }, "targetHandle": { "fieldName": "input_value", - "id": "LanguageModelComponent-tAUaX", + "id": "LanguageModelComponent-YHUqJ", "inputTypes": [ "Message" ], "type": "str" } }, - "id": "reactflow__edge-ChatInput-osYcH{œdataTypeœ:œChatInputœ,œidœ:œChatInput-osYcHœ,œnameœ:œmessageœ,œoutput_typesœ:[œMessageœ]}-LanguageModelComponent-tAUaX{œfieldNameœ:œinput_valueœ,œidœ:œLanguageModelComponent-tAUaXœ,œinputTypesœ:[œMessageœ],œtypeœ:œstrœ}", + "id": "reactflow__edge-ChatInput-CbbPG{œdataTypeœ:œChatInputœ,œidœ:œChatInput-CbbPGœ,œnameœ:œmessageœ,œoutput_typesœ:[œMessageœ]}-LanguageModelComponent-YHUqJ{œfieldNameœ:œinput_valueœ,œidœ:œLanguageModelComponent-YHUqJœ,œinputTypesœ:[œMessageœ],œtypeœ:œstrœ}", "selected": false, - "source": "ChatInput-osYcH", - "sourceHandle": "{œdataTypeœ: œChatInputœ, œidœ: œChatInput-osYcHœ, œnameœ: œmessageœ, œoutput_typesœ: [œMessageœ]}", - "target": "LanguageModelComponent-tAUaX", - "targetHandle": "{œfieldNameœ: œinput_valueœ, œidœ: œLanguageModelComponent-tAUaXœ, œinputTypesœ: [œMessageœ], œtypeœ: œstrœ}" + "source": "ChatInput-CbbPG", + "sourceHandle": "{œdataTypeœ: œChatInputœ, œidœ: œChatInput-CbbPGœ, œnameœ: œmessageœ, œoutput_typesœ: [œMessageœ]}", + "target": "LanguageModelComponent-YHUqJ", + "targetHandle": "{œfieldNameœ: œinput_valueœ, œidœ: œLanguageModelComponent-YHUqJœ, œinputTypesœ: [œMessageœ], œtypeœ: œstrœ}" }, { "animated": false, @@ -150,7 +150,7 @@ "data": { "sourceHandle": { "dataType": "LanguageModelComponent", - "id": "LanguageModelComponent-tAUaX", + "id": "LanguageModelComponent-YHUqJ", "name": "text_output", "output_types": [ "Message" @@ -158,19 +158,19 @@ }, "targetHandle": { "fieldName": "input_value", - "id": "StructuredOutput-J6Rvk", + "id": "StructuredOutput-eoFyT", "inputTypes": [ "Message" ], "type": "str" } }, - "id": "reactflow__edge-LanguageModelComponent-tAUaX{œdataTypeœ:œLanguageModelComponentœ,œidœ:œLanguageModelComponent-tAUaXœ,œnameœ:œtext_outputœ,œoutput_typesœ:[œMessageœ]}-StructuredOutput-J6Rvk{œfieldNameœ:œinput_valueœ,œidœ:œStructuredOutput-J6Rvkœ,œinputTypesœ:[œMessageœ],œtypeœ:œstrœ}", + "id": "reactflow__edge-LanguageModelComponent-YHUqJ{œdataTypeœ:œLanguageModelComponentœ,œidœ:œLanguageModelComponent-YHUqJœ,œnameœ:œtext_outputœ,œoutput_typesœ:[œMessageœ]}-StructuredOutput-eoFyT{œfieldNameœ:œinput_valueœ,œidœ:œStructuredOutput-eoFyTœ,œinputTypesœ:[œMessageœ],œtypeœ:œstrœ}", "selected": false, - "source": "LanguageModelComponent-tAUaX", - "sourceHandle": "{œdataTypeœ: œLanguageModelComponentœ, œidœ: œLanguageModelComponent-tAUaXœ, œnameœ: œtext_outputœ, œoutput_typesœ: [œMessageœ]}", - "target": "StructuredOutput-J6Rvk", - "targetHandle": "{œfieldNameœ: œinput_valueœ, œidœ: œStructuredOutput-J6Rvkœ, œinputTypesœ: [œMessageœ], œtypeœ: œstrœ}" + "source": "LanguageModelComponent-YHUqJ", + "sourceHandle": "{œdataTypeœ: œLanguageModelComponentœ, œidœ: œLanguageModelComponent-YHUqJœ, œnameœ: œtext_outputœ, œoutput_typesœ: [œMessageœ]}", + "target": "StructuredOutput-eoFyT", + "targetHandle": "{œfieldNameœ: œinput_valueœ, œidœ: œStructuredOutput-eoFyTœ, œinputTypesœ: [œMessageœ], œtypeœ: œstrœ}" } ], "nodes": [ @@ -178,7 +178,7 @@ "data": { "description": "Get chat inputs from the Playground.", "display_name": "Chat Input", - "id": "ChatInput-osYcH", + "id": "ChatInput-CbbPG", "node": { "base_classes": [ "Message" @@ -462,7 +462,7 @@ }, "dragging": false, "height": 234, - "id": "ChatInput-osYcH", + "id": "ChatInput-CbbPG", "measured": { "height": 234, "width": 320 @@ -483,7 +483,7 @@ "data": { "description": "Display a chat message in the Playground.", "display_name": "Chat Output", - "id": "ChatOutput-1VmAz", + "id": "ChatOutput-WVlr5", "node": { "base_classes": [ "Message" @@ -766,7 +766,7 @@ }, "dragging": false, "height": 234, - "id": "ChatOutput-1VmAz", + "id": "ChatOutput-WVlr5", "measured": { "height": 234, "width": 320 @@ -779,13 +779,13 @@ "x": 2742.72534045604, "y": 681.9098282545469 }, - "selected": true, + "selected": false, "type": "genericNode", "width": 320 }, { "data": { - "id": "note-uscMT", + "id": "note-k9mrI", "node": { "description": "# Image Sentiment Analysis\nWelcome to the Image Sentiment Classifier - an AI tool for quick image sentiment analysis!\n\n## Instructions\n\n1. **Prepare Your Image**\n - Image should be clear and visible\n\n2. **Upload Options**\n - Open the Playground\n - Click file attachment icon\n - Or drag and drop into playground\n\n3. **Wait for Analysis**\n - System will process the image\n - Uses zero-shot learning\n - Classification happens automatically\n\n4. **Review Results**\n - Get classification: Positive/Negative/Neutral\n - Review confidence level\n - Check reasoning if provided\n\n5. **Expected Classifications**\n - Positive: Happy scenes, smiles, celebrations\n - Negative: Sad scenes, problems, conflicts\n - Neutral: Objects, landscapes, neutral scenes\n\nRemember: The clearer the image, the more accurate the classification! 📸✨", "display_name": "", @@ -796,7 +796,7 @@ }, "dragging": false, "height": 583, - "id": "note-uscMT", + "id": "note-k9mrI", "measured": { "height": 583, "width": 325 @@ -822,7 +822,7 @@ "data": { "description": "Create a prompt template with dynamic variables.", "display_name": "Prompt", - "id": "Prompt-AGdss", + "id": "Prompt-ElcZb", "node": { "base_classes": [ "Message" @@ -929,7 +929,7 @@ }, "dragging": false, "height": 260, - "id": "Prompt-AGdss", + "id": "Prompt-ElcZb", "measured": { "height": 260, "width": 320 @@ -948,7 +948,7 @@ }, { "data": { - "id": "StructuredOutput-J6Rvk", + "id": "StructuredOutput-eoFyT", "node": { "base_classes": [ "Data" @@ -1009,7 +1009,7 @@ "show": true, "title_case": false, "type": "code", - "value": "from pydantic import BaseModel, Field, create_model\nfrom trustcall import create_extractor\n\nfrom langflow.base.models.chat_result import get_chat_result\nfrom langflow.custom.custom_component.component import Component\nfrom langflow.helpers.base_model import build_model_from_schema\nfrom langflow.io import (\n HandleInput,\n MessageTextInput,\n MultilineInput,\n Output,\n TableInput,\n)\nfrom langflow.schema.data import Data\nfrom langflow.schema.table import EditMode\n\n\nclass StructuredOutputComponent(Component):\n display_name = \"Structured Output\"\n description = \"Uses an LLM to generate structured data. Ideal for extraction and consistency.\"\n name = \"StructuredOutput\"\n icon = \"braces\"\n\n inputs = [\n HandleInput(\n name=\"llm\",\n display_name=\"Language Model\",\n info=\"The language model to use to generate the structured output.\",\n input_types=[\"LanguageModel\"],\n required=True,\n ),\n MessageTextInput(\n name=\"input_value\",\n display_name=\"Input Message\",\n info=\"The input message to the language model.\",\n tool_mode=True,\n required=True,\n ),\n MultilineInput(\n name=\"system_prompt\",\n display_name=\"Format Instructions\",\n info=\"The instructions to the language model for formatting the output.\",\n value=(\n \"You are an AI system designed to extract structured information from unstructured text.\"\n \"Given the input_text, return a JSON object with predefined keys based on the expected structure.\"\n \"Extract values accurately and format them according to the specified type \"\n \"(e.g., string, integer, float, date).\"\n \"If a value is missing or cannot be determined, return a default \"\n \"(e.g., null, 0, or 'N/A').\"\n \"If multiple instances of the expected structure exist within the input_text, \"\n \"stream each as a separate JSON object.\"\n ),\n required=True,\n advanced=True,\n ),\n MessageTextInput(\n name=\"schema_name\",\n display_name=\"Schema Name\",\n info=\"Provide a name for the output data schema.\",\n advanced=True,\n ),\n TableInput(\n name=\"output_schema\",\n display_name=\"Output Schema\",\n info=\"Define the structure and data types for the model's output.\",\n required=True,\n # TODO: remove deault value\n table_schema=[\n {\n \"name\": \"name\",\n \"display_name\": \"Name\",\n \"type\": \"str\",\n \"description\": \"Specify the name of the output field.\",\n \"default\": \"field\",\n \"edit_mode\": EditMode.INLINE,\n },\n {\n \"name\": \"description\",\n \"display_name\": \"Description\",\n \"type\": \"str\",\n \"description\": \"Describe the purpose of the output field.\",\n \"default\": \"description of field\",\n \"edit_mode\": EditMode.POPOVER,\n },\n {\n \"name\": \"type\",\n \"display_name\": \"Type\",\n \"type\": \"str\",\n \"edit_mode\": EditMode.INLINE,\n \"description\": (\"Indicate the data type of the output field (e.g., str, int, float, bool, dict).\"),\n \"options\": [\"str\", \"int\", \"float\", \"bool\", \"dict\"],\n \"default\": \"str\",\n },\n {\n \"name\": \"multiple\",\n \"display_name\": \"As List\",\n \"type\": \"boolean\",\n \"description\": \"Set to True if this output field should be a list of the specified type.\",\n \"default\": \"False\",\n \"edit_mode\": EditMode.INLINE,\n },\n ],\n value=[\n {\n \"name\": \"field\",\n \"description\": \"description of field\",\n \"type\": \"str\",\n \"multiple\": \"False\",\n }\n ],\n ),\n ]\n\n outputs = [\n Output(\n name=\"structured_output\",\n display_name=\"Structured Output\",\n method=\"build_structured_output\",\n ),\n ]\n\n def build_structured_output_base(self):\n schema_name = self.schema_name or \"OutputModel\"\n\n if not hasattr(self.llm, \"with_structured_output\"):\n msg = \"Language model does not support structured output.\"\n raise TypeError(msg)\n if not self.output_schema:\n msg = \"Output schema cannot be empty\"\n raise ValueError(msg)\n\n output_model_ = build_model_from_schema(self.output_schema)\n\n output_model = create_model(\n schema_name,\n __doc__=f\"A list of {schema_name}.\",\n objects=(list[output_model_], Field(description=f\"A list of {schema_name}.\")), # type: ignore[valid-type]\n )\n\n try:\n llm_with_structured_output = create_extractor(self.llm, tools=[output_model])\n except NotImplementedError as exc:\n msg = f\"{self.llm.__class__.__name__} does not support structured output.\"\n raise TypeError(msg) from exc\n\n config_dict = {\n \"run_name\": self.display_name,\n \"project_name\": self.get_project_name(),\n \"callbacks\": self.get_langchain_callbacks(),\n }\n result = get_chat_result(\n runnable=llm_with_structured_output,\n system_message=self.system_prompt,\n input_value=self.input_value,\n config=config_dict,\n )\n\n # OPTIMIZATION NOTE: Simplified processing based on trustcall response structure\n # Handle non-dict responses (shouldn't happen with trustcall, but defensive)\n if not isinstance(result, dict):\n return result\n\n # Extract first response and convert BaseModel to dict\n responses = result.get(\"responses\", [])\n if not responses:\n return result\n\n # Convert BaseModel to dict (creates the \"objects\" key)\n first_response = responses[0]\n structured_data = first_response.model_dump() if isinstance(first_response, BaseModel) else first_response\n\n # Extract the objects array (guaranteed to exist due to our Pydantic model structure)\n return structured_data.get(\"objects\", structured_data)\n\n def build_structured_output(self) -> Data:\n output = self.build_structured_output_base()\n if not isinstance(output, list) or not output:\n # handle empty or unexpected type case\n msg = \"No structured output returned\"\n raise ValueError(msg)\n if len(output) != 1:\n msg = \"Multiple structured outputs returned\"\n raise ValueError(msg)\n return Data(data=output[0])\n" + "value": "from pydantic import BaseModel, Field, create_model\nfrom trustcall import create_extractor\n\nfrom langflow.base.models.chat_result import get_chat_result\nfrom langflow.custom.custom_component.component import Component\nfrom langflow.helpers.base_model import build_model_from_schema\nfrom langflow.io import (\n HandleInput,\n MessageTextInput,\n MultilineInput,\n Output,\n TableInput,\n)\nfrom langflow.schema.data import Data\nfrom langflow.schema.table import EditMode\n\n\nclass StructuredOutputComponent(Component):\n display_name = \"Structured Output\"\n description = \"Uses an LLM to generate structured data. Ideal for extraction and consistency.\"\n name = \"StructuredOutput\"\n icon = \"braces\"\n\n inputs = [\n HandleInput(\n name=\"llm\",\n display_name=\"Language Model\",\n info=\"The language model to use to generate the structured output.\",\n input_types=[\"LanguageModel\"],\n required=True,\n ),\n MultilineInput(\n name=\"input_value\",\n display_name=\"Input Message\",\n info=\"The input message to the language model.\",\n tool_mode=True,\n required=True,\n ),\n MultilineInput(\n name=\"system_prompt\",\n display_name=\"Format Instructions\",\n info=\"The instructions to the language model for formatting the output.\",\n value=(\n \"You are an AI system designed to extract structured information from unstructured text.\"\n \"Given the input_text, return a JSON object with predefined keys based on the expected structure.\"\n \"Extract values accurately and format them according to the specified type \"\n \"(e.g., string, integer, float, date).\"\n \"If a value is missing or cannot be determined, return a default \"\n \"(e.g., null, 0, or 'N/A').\"\n \"If multiple instances of the expected structure exist within the input_text, \"\n \"stream each as a separate JSON object.\"\n ),\n required=True,\n advanced=True,\n ),\n MessageTextInput(\n name=\"schema_name\",\n display_name=\"Schema Name\",\n info=\"Provide a name for the output data schema.\",\n advanced=True,\n ),\n TableInput(\n name=\"output_schema\",\n display_name=\"Output Schema\",\n info=\"Define the structure and data types for the model's output.\",\n required=True,\n # TODO: remove deault value\n table_schema=[\n {\n \"name\": \"name\",\n \"display_name\": \"Name\",\n \"type\": \"str\",\n \"description\": \"Specify the name of the output field.\",\n \"default\": \"field\",\n \"edit_mode\": EditMode.INLINE,\n },\n {\n \"name\": \"description\",\n \"display_name\": \"Description\",\n \"type\": \"str\",\n \"description\": \"Describe the purpose of the output field.\",\n \"default\": \"description of field\",\n \"edit_mode\": EditMode.POPOVER,\n },\n {\n \"name\": \"type\",\n \"display_name\": \"Type\",\n \"type\": \"str\",\n \"edit_mode\": EditMode.INLINE,\n \"description\": (\"Indicate the data type of the output field (e.g., str, int, float, bool, dict).\"),\n \"options\": [\"str\", \"int\", \"float\", \"bool\", \"dict\"],\n \"default\": \"str\",\n },\n {\n \"name\": \"multiple\",\n \"display_name\": \"As List\",\n \"type\": \"boolean\",\n \"description\": \"Set to True if this output field should be a list of the specified type.\",\n \"default\": \"False\",\n \"edit_mode\": EditMode.INLINE,\n },\n ],\n value=[\n {\n \"name\": \"field\",\n \"description\": \"description of field\",\n \"type\": \"str\",\n \"multiple\": \"False\",\n }\n ],\n ),\n ]\n\n outputs = [\n Output(\n name=\"structured_output\",\n display_name=\"Structured Output\",\n method=\"build_structured_output\",\n ),\n ]\n\n def build_structured_output_base(self):\n schema_name = self.schema_name or \"OutputModel\"\n\n if not hasattr(self.llm, \"with_structured_output\"):\n msg = \"Language model does not support structured output.\"\n raise TypeError(msg)\n if not self.output_schema:\n msg = \"Output schema cannot be empty\"\n raise ValueError(msg)\n\n output_model_ = build_model_from_schema(self.output_schema)\n\n output_model = create_model(\n schema_name,\n __doc__=f\"A list of {schema_name}.\",\n objects=(list[output_model_], Field(description=f\"A list of {schema_name}.\")), # type: ignore[valid-type]\n )\n\n try:\n llm_with_structured_output = create_extractor(self.llm, tools=[output_model])\n except NotImplementedError as exc:\n msg = f\"{self.llm.__class__.__name__} does not support structured output.\"\n raise TypeError(msg) from exc\n\n config_dict = {\n \"run_name\": self.display_name,\n \"project_name\": self.get_project_name(),\n \"callbacks\": self.get_langchain_callbacks(),\n }\n result = get_chat_result(\n runnable=llm_with_structured_output,\n system_message=self.system_prompt,\n input_value=self.input_value,\n config=config_dict,\n )\n\n # OPTIMIZATION NOTE: Simplified processing based on trustcall response structure\n # Handle non-dict responses (shouldn't happen with trustcall, but defensive)\n if not isinstance(result, dict):\n return result\n\n # Extract first response and convert BaseModel to dict\n responses = result.get(\"responses\", [])\n if not responses:\n return result\n\n # Convert BaseModel to dict (creates the \"objects\" key)\n first_response = responses[0]\n structured_data = first_response.model_dump() if isinstance(first_response, BaseModel) else first_response\n\n # Extract the objects array (guaranteed to exist due to our Pydantic model structure)\n return structured_data.get(\"objects\", structured_data)\n\n def build_structured_output(self) -> Data:\n output = self.build_structured_output_base()\n if not isinstance(output, list) or not output:\n # handle empty or unexpected type case\n msg = \"No structured output returned\"\n raise ValueError(msg)\n if len(output) != 1:\n msg = \"Multiple structured outputs returned\"\n raise ValueError(msg)\n return Data(data=output[0])\n" }, "input_value": { "_input_type": "MessageTextInput", @@ -1207,7 +1207,7 @@ "type": "StructuredOutput" }, "dragging": false, - "id": "StructuredOutput-J6Rvk", + "id": "StructuredOutput-eoFyT", "measured": { "height": 349, "width": 320 @@ -1221,7 +1221,7 @@ }, { "data": { - "id": "parser-sIcOW", + "id": "parser-o6H8E", "node": { "base_classes": [ "Message" @@ -1382,7 +1382,7 @@ "type": "parser" }, "dragging": false, - "id": "parser-sIcOW", + "id": "parser-o6H8E", "measured": { "height": 361, "width": 320 @@ -1396,7 +1396,7 @@ }, { "data": { - "id": "LanguageModelComponent-x2hKm", + "id": "LanguageModelComponent-2h2qm", "node": { "base_classes": [ "LanguageModel", @@ -1473,7 +1473,7 @@ "dynamic": false, "info": "Model Provider API key", "input_types": [], - "load_from_db": true, + "load_from_db": false, "name": "api_key", "password": true, "placeholder": "", @@ -1667,11 +1667,12 @@ }, "tool_mode": false }, + "selected_output": "model_output", "showNode": true, "type": "LanguageModelComponent" }, "dragging": false, - "id": "LanguageModelComponent-x2hKm", + "id": "LanguageModelComponent-2h2qm", "measured": { "height": 451, "width": 320 @@ -1685,7 +1686,7 @@ }, { "data": { - "id": "LanguageModelComponent-tAUaX", + "id": "LanguageModelComponent-YHUqJ", "node": { "base_classes": [ "LanguageModel", @@ -1762,7 +1763,7 @@ "dynamic": false, "info": "Model Provider API key", "input_types": [], - "load_from_db": true, + "load_from_db": false, "name": "api_key", "password": true, "placeholder": "", @@ -1960,7 +1961,7 @@ "type": "LanguageModelComponent" }, "dragging": false, - "id": "LanguageModelComponent-tAUaX", + "id": "LanguageModelComponent-YHUqJ", "measured": { "height": 534, "width": 320 @@ -1974,14 +1975,14 @@ } ], "viewport": { - "x": -394.48978663654907, - "y": 83.15122658160072, - "zoom": 0.5803621249210289 + "x": -779.9735306873563, + "y": 114.7787669766526, + "zoom": 0.6873063891390285 } }, "description": "Analyzes images and categorizes them as positive, negative, or neutral using zero-shot learning.", "endpoint_name": null, - "id": "1a6e7961-5c68-450d-a5b3-db6d0bd90bfe", + "id": "0cccdf9b-c4ae-4e14-8d19-1b1ac4dd1edf", "is_component": false, "last_tested_version": "1.4.3", "name": "Image Sentiment Analysis", diff --git a/src/backend/base/langflow/initial_setup/starter_projects/Market Research.json b/src/backend/base/langflow/initial_setup/starter_projects/Market Research.json index 3a2cd2bf4..3626710f3 100644 --- a/src/backend/base/langflow/initial_setup/starter_projects/Market Research.json +++ b/src/backend/base/langflow/initial_setup/starter_projects/Market Research.json @@ -879,7 +879,7 @@ "show": true, "title_case": false, "type": "code", - "value": "from pydantic import BaseModel, Field, create_model\nfrom trustcall import create_extractor\n\nfrom langflow.base.models.chat_result import get_chat_result\nfrom langflow.custom.custom_component.component import Component\nfrom langflow.helpers.base_model import build_model_from_schema\nfrom langflow.io import (\n HandleInput,\n MessageTextInput,\n MultilineInput,\n Output,\n TableInput,\n)\nfrom langflow.schema.data import Data\nfrom langflow.schema.table import EditMode\n\n\nclass StructuredOutputComponent(Component):\n display_name = \"Structured Output\"\n description = \"Uses an LLM to generate structured data. Ideal for extraction and consistency.\"\n name = \"StructuredOutput\"\n icon = \"braces\"\n\n inputs = [\n HandleInput(\n name=\"llm\",\n display_name=\"Language Model\",\n info=\"The language model to use to generate the structured output.\",\n input_types=[\"LanguageModel\"],\n required=True,\n ),\n MessageTextInput(\n name=\"input_value\",\n display_name=\"Input Message\",\n info=\"The input message to the language model.\",\n tool_mode=True,\n required=True,\n ),\n MultilineInput(\n name=\"system_prompt\",\n display_name=\"Format Instructions\",\n info=\"The instructions to the language model for formatting the output.\",\n value=(\n \"You are an AI system designed to extract structured information from unstructured text.\"\n \"Given the input_text, return a JSON object with predefined keys based on the expected structure.\"\n \"Extract values accurately and format them according to the specified type \"\n \"(e.g., string, integer, float, date).\"\n \"If a value is missing or cannot be determined, return a default \"\n \"(e.g., null, 0, or 'N/A').\"\n \"If multiple instances of the expected structure exist within the input_text, \"\n \"stream each as a separate JSON object.\"\n ),\n required=True,\n advanced=True,\n ),\n MessageTextInput(\n name=\"schema_name\",\n display_name=\"Schema Name\",\n info=\"Provide a name for the output data schema.\",\n advanced=True,\n ),\n TableInput(\n name=\"output_schema\",\n display_name=\"Output Schema\",\n info=\"Define the structure and data types for the model's output.\",\n required=True,\n # TODO: remove deault value\n table_schema=[\n {\n \"name\": \"name\",\n \"display_name\": \"Name\",\n \"type\": \"str\",\n \"description\": \"Specify the name of the output field.\",\n \"default\": \"field\",\n \"edit_mode\": EditMode.INLINE,\n },\n {\n \"name\": \"description\",\n \"display_name\": \"Description\",\n \"type\": \"str\",\n \"description\": \"Describe the purpose of the output field.\",\n \"default\": \"description of field\",\n \"edit_mode\": EditMode.POPOVER,\n },\n {\n \"name\": \"type\",\n \"display_name\": \"Type\",\n \"type\": \"str\",\n \"edit_mode\": EditMode.INLINE,\n \"description\": (\"Indicate the data type of the output field (e.g., str, int, float, bool, dict).\"),\n \"options\": [\"str\", \"int\", \"float\", \"bool\", \"dict\"],\n \"default\": \"str\",\n },\n {\n \"name\": \"multiple\",\n \"display_name\": \"As List\",\n \"type\": \"boolean\",\n \"description\": \"Set to True if this output field should be a list of the specified type.\",\n \"default\": \"False\",\n \"edit_mode\": EditMode.INLINE,\n },\n ],\n value=[\n {\n \"name\": \"field\",\n \"description\": \"description of field\",\n \"type\": \"str\",\n \"multiple\": \"False\",\n }\n ],\n ),\n ]\n\n outputs = [\n Output(\n name=\"structured_output\",\n display_name=\"Structured Output\",\n method=\"build_structured_output\",\n ),\n ]\n\n def build_structured_output_base(self):\n schema_name = self.schema_name or \"OutputModel\"\n\n if not hasattr(self.llm, \"with_structured_output\"):\n msg = \"Language model does not support structured output.\"\n raise TypeError(msg)\n if not self.output_schema:\n msg = \"Output schema cannot be empty\"\n raise ValueError(msg)\n\n output_model_ = build_model_from_schema(self.output_schema)\n\n output_model = create_model(\n schema_name,\n __doc__=f\"A list of {schema_name}.\",\n objects=(list[output_model_], Field(description=f\"A list of {schema_name}.\")), # type: ignore[valid-type]\n )\n\n try:\n llm_with_structured_output = create_extractor(self.llm, tools=[output_model])\n except NotImplementedError as exc:\n msg = f\"{self.llm.__class__.__name__} does not support structured output.\"\n raise TypeError(msg) from exc\n\n config_dict = {\n \"run_name\": self.display_name,\n \"project_name\": self.get_project_name(),\n \"callbacks\": self.get_langchain_callbacks(),\n }\n result = get_chat_result(\n runnable=llm_with_structured_output,\n system_message=self.system_prompt,\n input_value=self.input_value,\n config=config_dict,\n )\n\n # OPTIMIZATION NOTE: Simplified processing based on trustcall response structure\n # Handle non-dict responses (shouldn't happen with trustcall, but defensive)\n if not isinstance(result, dict):\n return result\n\n # Extract first response and convert BaseModel to dict\n responses = result.get(\"responses\", [])\n if not responses:\n return result\n\n # Convert BaseModel to dict (creates the \"objects\" key)\n first_response = responses[0]\n structured_data = first_response.model_dump() if isinstance(first_response, BaseModel) else first_response\n\n # Extract the objects array (guaranteed to exist due to our Pydantic model structure)\n return structured_data.get(\"objects\", structured_data)\n\n def build_structured_output(self) -> Data:\n output = self.build_structured_output_base()\n if not isinstance(output, list) or not output:\n # handle empty or unexpected type case\n msg = \"No structured output returned\"\n raise ValueError(msg)\n if len(output) != 1:\n msg = \"Multiple structured outputs returned\"\n raise ValueError(msg)\n return Data(data=output[0])\n" + "value": "from pydantic import BaseModel, Field, create_model\nfrom trustcall import create_extractor\n\nfrom langflow.base.models.chat_result import get_chat_result\nfrom langflow.custom.custom_component.component import Component\nfrom langflow.helpers.base_model import build_model_from_schema\nfrom langflow.io import (\n HandleInput,\n MessageTextInput,\n MultilineInput,\n Output,\n TableInput,\n)\nfrom langflow.schema.data import Data\nfrom langflow.schema.table import EditMode\n\n\nclass StructuredOutputComponent(Component):\n display_name = \"Structured Output\"\n description = \"Uses an LLM to generate structured data. Ideal for extraction and consistency.\"\n name = \"StructuredOutput\"\n icon = \"braces\"\n\n inputs = [\n HandleInput(\n name=\"llm\",\n display_name=\"Language Model\",\n info=\"The language model to use to generate the structured output.\",\n input_types=[\"LanguageModel\"],\n required=True,\n ),\n MultilineInput(\n name=\"input_value\",\n display_name=\"Input Message\",\n info=\"The input message to the language model.\",\n tool_mode=True,\n required=True,\n ),\n MultilineInput(\n name=\"system_prompt\",\n display_name=\"Format Instructions\",\n info=\"The instructions to the language model for formatting the output.\",\n value=(\n \"You are an AI system designed to extract structured information from unstructured text.\"\n \"Given the input_text, return a JSON object with predefined keys based on the expected structure.\"\n \"Extract values accurately and format them according to the specified type \"\n \"(e.g., string, integer, float, date).\"\n \"If a value is missing or cannot be determined, return a default \"\n \"(e.g., null, 0, or 'N/A').\"\n \"If multiple instances of the expected structure exist within the input_text, \"\n \"stream each as a separate JSON object.\"\n ),\n required=True,\n advanced=True,\n ),\n MessageTextInput(\n name=\"schema_name\",\n display_name=\"Schema Name\",\n info=\"Provide a name for the output data schema.\",\n advanced=True,\n ),\n TableInput(\n name=\"output_schema\",\n display_name=\"Output Schema\",\n info=\"Define the structure and data types for the model's output.\",\n required=True,\n # TODO: remove deault value\n table_schema=[\n {\n \"name\": \"name\",\n \"display_name\": \"Name\",\n \"type\": \"str\",\n \"description\": \"Specify the name of the output field.\",\n \"default\": \"field\",\n \"edit_mode\": EditMode.INLINE,\n },\n {\n \"name\": \"description\",\n \"display_name\": \"Description\",\n \"type\": \"str\",\n \"description\": \"Describe the purpose of the output field.\",\n \"default\": \"description of field\",\n \"edit_mode\": EditMode.POPOVER,\n },\n {\n \"name\": \"type\",\n \"display_name\": \"Type\",\n \"type\": \"str\",\n \"edit_mode\": EditMode.INLINE,\n \"description\": (\"Indicate the data type of the output field (e.g., str, int, float, bool, dict).\"),\n \"options\": [\"str\", \"int\", \"float\", \"bool\", \"dict\"],\n \"default\": \"str\",\n },\n {\n \"name\": \"multiple\",\n \"display_name\": \"As List\",\n \"type\": \"boolean\",\n \"description\": \"Set to True if this output field should be a list of the specified type.\",\n \"default\": \"False\",\n \"edit_mode\": EditMode.INLINE,\n },\n ],\n value=[\n {\n \"name\": \"field\",\n \"description\": \"description of field\",\n \"type\": \"str\",\n \"multiple\": \"False\",\n }\n ],\n ),\n ]\n\n outputs = [\n Output(\n name=\"structured_output\",\n display_name=\"Structured Output\",\n method=\"build_structured_output\",\n ),\n ]\n\n def build_structured_output_base(self):\n schema_name = self.schema_name or \"OutputModel\"\n\n if not hasattr(self.llm, \"with_structured_output\"):\n msg = \"Language model does not support structured output.\"\n raise TypeError(msg)\n if not self.output_schema:\n msg = \"Output schema cannot be empty\"\n raise ValueError(msg)\n\n output_model_ = build_model_from_schema(self.output_schema)\n\n output_model = create_model(\n schema_name,\n __doc__=f\"A list of {schema_name}.\",\n objects=(list[output_model_], Field(description=f\"A list of {schema_name}.\")), # type: ignore[valid-type]\n )\n\n try:\n llm_with_structured_output = create_extractor(self.llm, tools=[output_model])\n except NotImplementedError as exc:\n msg = f\"{self.llm.__class__.__name__} does not support structured output.\"\n raise TypeError(msg) from exc\n\n config_dict = {\n \"run_name\": self.display_name,\n \"project_name\": self.get_project_name(),\n \"callbacks\": self.get_langchain_callbacks(),\n }\n result = get_chat_result(\n runnable=llm_with_structured_output,\n system_message=self.system_prompt,\n input_value=self.input_value,\n config=config_dict,\n )\n\n # OPTIMIZATION NOTE: Simplified processing based on trustcall response structure\n # Handle non-dict responses (shouldn't happen with trustcall, but defensive)\n if not isinstance(result, dict):\n return result\n\n # Extract first response and convert BaseModel to dict\n responses = result.get(\"responses\", [])\n if not responses:\n return result\n\n # Convert BaseModel to dict (creates the \"objects\" key)\n first_response = responses[0]\n structured_data = first_response.model_dump() if isinstance(first_response, BaseModel) else first_response\n\n # Extract the objects array (guaranteed to exist due to our Pydantic model structure)\n return structured_data.get(\"objects\", structured_data)\n\n def build_structured_output(self) -> Data:\n output = self.build_structured_output_base()\n if not isinstance(output, list) or not output:\n # handle empty or unexpected type case\n msg = \"No structured output returned\"\n raise ValueError(msg)\n if len(output) != 1:\n msg = \"Multiple structured outputs returned\"\n raise ValueError(msg)\n return Data(data=output[0])\n" }, "input_value": { "_input_type": "MessageTextInput", diff --git a/src/backend/base/langflow/initial_setup/starter_projects/Portfolio Website Code Generator.json b/src/backend/base/langflow/initial_setup/starter_projects/Portfolio Website Code Generator.json index 4895e94e4..f0bfc7e45 100644 --- a/src/backend/base/langflow/initial_setup/starter_projects/Portfolio Website Code Generator.json +++ b/src/backend/base/langflow/initial_setup/starter_projects/Portfolio Website Code Generator.json @@ -1509,7 +1509,7 @@ "show": true, "title_case": false, "type": "code", - "value": "from pydantic import BaseModel, Field, create_model\nfrom trustcall import create_extractor\n\nfrom langflow.base.models.chat_result import get_chat_result\nfrom langflow.custom.custom_component.component import Component\nfrom langflow.helpers.base_model import build_model_from_schema\nfrom langflow.io import (\n HandleInput,\n MessageTextInput,\n MultilineInput,\n Output,\n TableInput,\n)\nfrom langflow.schema.data import Data\nfrom langflow.schema.table import EditMode\n\n\nclass StructuredOutputComponent(Component):\n display_name = \"Structured Output\"\n description = \"Uses an LLM to generate structured data. Ideal for extraction and consistency.\"\n name = \"StructuredOutput\"\n icon = \"braces\"\n\n inputs = [\n HandleInput(\n name=\"llm\",\n display_name=\"Language Model\",\n info=\"The language model to use to generate the structured output.\",\n input_types=[\"LanguageModel\"],\n required=True,\n ),\n MessageTextInput(\n name=\"input_value\",\n display_name=\"Input Message\",\n info=\"The input message to the language model.\",\n tool_mode=True,\n required=True,\n ),\n MultilineInput(\n name=\"system_prompt\",\n display_name=\"Format Instructions\",\n info=\"The instructions to the language model for formatting the output.\",\n value=(\n \"You are an AI system designed to extract structured information from unstructured text.\"\n \"Given the input_text, return a JSON object with predefined keys based on the expected structure.\"\n \"Extract values accurately and format them according to the specified type \"\n \"(e.g., string, integer, float, date).\"\n \"If a value is missing or cannot be determined, return a default \"\n \"(e.g., null, 0, or 'N/A').\"\n \"If multiple instances of the expected structure exist within the input_text, \"\n \"stream each as a separate JSON object.\"\n ),\n required=True,\n advanced=True,\n ),\n MessageTextInput(\n name=\"schema_name\",\n display_name=\"Schema Name\",\n info=\"Provide a name for the output data schema.\",\n advanced=True,\n ),\n TableInput(\n name=\"output_schema\",\n display_name=\"Output Schema\",\n info=\"Define the structure and data types for the model's output.\",\n required=True,\n # TODO: remove deault value\n table_schema=[\n {\n \"name\": \"name\",\n \"display_name\": \"Name\",\n \"type\": \"str\",\n \"description\": \"Specify the name of the output field.\",\n \"default\": \"field\",\n \"edit_mode\": EditMode.INLINE,\n },\n {\n \"name\": \"description\",\n \"display_name\": \"Description\",\n \"type\": \"str\",\n \"description\": \"Describe the purpose of the output field.\",\n \"default\": \"description of field\",\n \"edit_mode\": EditMode.POPOVER,\n },\n {\n \"name\": \"type\",\n \"display_name\": \"Type\",\n \"type\": \"str\",\n \"edit_mode\": EditMode.INLINE,\n \"description\": (\"Indicate the data type of the output field (e.g., str, int, float, bool, dict).\"),\n \"options\": [\"str\", \"int\", \"float\", \"bool\", \"dict\"],\n \"default\": \"str\",\n },\n {\n \"name\": \"multiple\",\n \"display_name\": \"As List\",\n \"type\": \"boolean\",\n \"description\": \"Set to True if this output field should be a list of the specified type.\",\n \"default\": \"False\",\n \"edit_mode\": EditMode.INLINE,\n },\n ],\n value=[\n {\n \"name\": \"field\",\n \"description\": \"description of field\",\n \"type\": \"str\",\n \"multiple\": \"False\",\n }\n ],\n ),\n ]\n\n outputs = [\n Output(\n name=\"structured_output\",\n display_name=\"Structured Output\",\n method=\"build_structured_output\",\n ),\n ]\n\n def build_structured_output_base(self):\n schema_name = self.schema_name or \"OutputModel\"\n\n if not hasattr(self.llm, \"with_structured_output\"):\n msg = \"Language model does not support structured output.\"\n raise TypeError(msg)\n if not self.output_schema:\n msg = \"Output schema cannot be empty\"\n raise ValueError(msg)\n\n output_model_ = build_model_from_schema(self.output_schema)\n\n output_model = create_model(\n schema_name,\n __doc__=f\"A list of {schema_name}.\",\n objects=(list[output_model_], Field(description=f\"A list of {schema_name}.\")), # type: ignore[valid-type]\n )\n\n try:\n llm_with_structured_output = create_extractor(self.llm, tools=[output_model])\n except NotImplementedError as exc:\n msg = f\"{self.llm.__class__.__name__} does not support structured output.\"\n raise TypeError(msg) from exc\n\n config_dict = {\n \"run_name\": self.display_name,\n \"project_name\": self.get_project_name(),\n \"callbacks\": self.get_langchain_callbacks(),\n }\n result = get_chat_result(\n runnable=llm_with_structured_output,\n system_message=self.system_prompt,\n input_value=self.input_value,\n config=config_dict,\n )\n\n # OPTIMIZATION NOTE: Simplified processing based on trustcall response structure\n # Handle non-dict responses (shouldn't happen with trustcall, but defensive)\n if not isinstance(result, dict):\n return result\n\n # Extract first response and convert BaseModel to dict\n responses = result.get(\"responses\", [])\n if not responses:\n return result\n\n # Convert BaseModel to dict (creates the \"objects\" key)\n first_response = responses[0]\n structured_data = first_response.model_dump() if isinstance(first_response, BaseModel) else first_response\n\n # Extract the objects array (guaranteed to exist due to our Pydantic model structure)\n return structured_data.get(\"objects\", structured_data)\n\n def build_structured_output(self) -> Data:\n output = self.build_structured_output_base()\n if not isinstance(output, list) or not output:\n # handle empty or unexpected type case\n msg = \"No structured output returned\"\n raise ValueError(msg)\n if len(output) != 1:\n msg = \"Multiple structured outputs returned\"\n raise ValueError(msg)\n return Data(data=output[0])\n" + "value": "from pydantic import BaseModel, Field, create_model\nfrom trustcall import create_extractor\n\nfrom langflow.base.models.chat_result import get_chat_result\nfrom langflow.custom.custom_component.component import Component\nfrom langflow.helpers.base_model import build_model_from_schema\nfrom langflow.io import (\n HandleInput,\n MessageTextInput,\n MultilineInput,\n Output,\n TableInput,\n)\nfrom langflow.schema.data import Data\nfrom langflow.schema.table import EditMode\n\n\nclass StructuredOutputComponent(Component):\n display_name = \"Structured Output\"\n description = \"Uses an LLM to generate structured data. Ideal for extraction and consistency.\"\n name = \"StructuredOutput\"\n icon = \"braces\"\n\n inputs = [\n HandleInput(\n name=\"llm\",\n display_name=\"Language Model\",\n info=\"The language model to use to generate the structured output.\",\n input_types=[\"LanguageModel\"],\n required=True,\n ),\n MultilineInput(\n name=\"input_value\",\n display_name=\"Input Message\",\n info=\"The input message to the language model.\",\n tool_mode=True,\n required=True,\n ),\n MultilineInput(\n name=\"system_prompt\",\n display_name=\"Format Instructions\",\n info=\"The instructions to the language model for formatting the output.\",\n value=(\n \"You are an AI system designed to extract structured information from unstructured text.\"\n \"Given the input_text, return a JSON object with predefined keys based on the expected structure.\"\n \"Extract values accurately and format them according to the specified type \"\n \"(e.g., string, integer, float, date).\"\n \"If a value is missing or cannot be determined, return a default \"\n \"(e.g., null, 0, or 'N/A').\"\n \"If multiple instances of the expected structure exist within the input_text, \"\n \"stream each as a separate JSON object.\"\n ),\n required=True,\n advanced=True,\n ),\n MessageTextInput(\n name=\"schema_name\",\n display_name=\"Schema Name\",\n info=\"Provide a name for the output data schema.\",\n advanced=True,\n ),\n TableInput(\n name=\"output_schema\",\n display_name=\"Output Schema\",\n info=\"Define the structure and data types for the model's output.\",\n required=True,\n # TODO: remove deault value\n table_schema=[\n {\n \"name\": \"name\",\n \"display_name\": \"Name\",\n \"type\": \"str\",\n \"description\": \"Specify the name of the output field.\",\n \"default\": \"field\",\n \"edit_mode\": EditMode.INLINE,\n },\n {\n \"name\": \"description\",\n \"display_name\": \"Description\",\n \"type\": \"str\",\n \"description\": \"Describe the purpose of the output field.\",\n \"default\": \"description of field\",\n \"edit_mode\": EditMode.POPOVER,\n },\n {\n \"name\": \"type\",\n \"display_name\": \"Type\",\n \"type\": \"str\",\n \"edit_mode\": EditMode.INLINE,\n \"description\": (\"Indicate the data type of the output field (e.g., str, int, float, bool, dict).\"),\n \"options\": [\"str\", \"int\", \"float\", \"bool\", \"dict\"],\n \"default\": \"str\",\n },\n {\n \"name\": \"multiple\",\n \"display_name\": \"As List\",\n \"type\": \"boolean\",\n \"description\": \"Set to True if this output field should be a list of the specified type.\",\n \"default\": \"False\",\n \"edit_mode\": EditMode.INLINE,\n },\n ],\n value=[\n {\n \"name\": \"field\",\n \"description\": \"description of field\",\n \"type\": \"str\",\n \"multiple\": \"False\",\n }\n ],\n ),\n ]\n\n outputs = [\n Output(\n name=\"structured_output\",\n display_name=\"Structured Output\",\n method=\"build_structured_output\",\n ),\n ]\n\n def build_structured_output_base(self):\n schema_name = self.schema_name or \"OutputModel\"\n\n if not hasattr(self.llm, \"with_structured_output\"):\n msg = \"Language model does not support structured output.\"\n raise TypeError(msg)\n if not self.output_schema:\n msg = \"Output schema cannot be empty\"\n raise ValueError(msg)\n\n output_model_ = build_model_from_schema(self.output_schema)\n\n output_model = create_model(\n schema_name,\n __doc__=f\"A list of {schema_name}.\",\n objects=(list[output_model_], Field(description=f\"A list of {schema_name}.\")), # type: ignore[valid-type]\n )\n\n try:\n llm_with_structured_output = create_extractor(self.llm, tools=[output_model])\n except NotImplementedError as exc:\n msg = f\"{self.llm.__class__.__name__} does not support structured output.\"\n raise TypeError(msg) from exc\n\n config_dict = {\n \"run_name\": self.display_name,\n \"project_name\": self.get_project_name(),\n \"callbacks\": self.get_langchain_callbacks(),\n }\n result = get_chat_result(\n runnable=llm_with_structured_output,\n system_message=self.system_prompt,\n input_value=self.input_value,\n config=config_dict,\n )\n\n # OPTIMIZATION NOTE: Simplified processing based on trustcall response structure\n # Handle non-dict responses (shouldn't happen with trustcall, but defensive)\n if not isinstance(result, dict):\n return result\n\n # Extract first response and convert BaseModel to dict\n responses = result.get(\"responses\", [])\n if not responses:\n return result\n\n # Convert BaseModel to dict (creates the \"objects\" key)\n first_response = responses[0]\n structured_data = first_response.model_dump() if isinstance(first_response, BaseModel) else first_response\n\n # Extract the objects array (guaranteed to exist due to our Pydantic model structure)\n return structured_data.get(\"objects\", structured_data)\n\n def build_structured_output(self) -> Data:\n output = self.build_structured_output_base()\n if not isinstance(output, list) or not output:\n # handle empty or unexpected type case\n msg = \"No structured output returned\"\n raise ValueError(msg)\n if len(output) != 1:\n msg = \"Multiple structured outputs returned\"\n raise ValueError(msg)\n return Data(data=output[0])\n" }, "input_value": { "_input_type": "MessageTextInput",