📘 Overview

CodeAutoGeneratedSampleEvaluator is an operator that evaluates code samples based on auto-generation markers. It scans for common phrases indicating automatically generated code within file headers and assigns scores, which can be used to filter out such files from a dataset.

__init__ function

def __init__(self, is_generated_func: Optional[Callable[[], bool]] = None):

Parameter Name	Type	Default	Description
is_generated_func	Optional[Callable[[], bool]]	None	An optional external function that can be provided to perform a custom check for auto-generated code.

Prompt Template Descriptions

Prompt Template Name	Primary Purpose	Applicable Scenarios	Feature Description

run function

def run(self, storage: DataFlowStorage, input_key: str):

Name	Type	Default	Description
storage	DataFlowStorage	Required	The data flow storage instance, responsible for reading and writing data.
input_key	str	Required	The name of the input column containing the code to be evaluated.

🧠 Example Usage

🧾 Default Output Format (Output Format)

Field	Type	Description
input_key	any	The original input data.
CodeAutoGeneratedMarkerCount	int	The number of auto-generation markers detected in the code sample.
CodeAutoGeneratedScore	float	A score from 0.0 to 1.0, where 1.0 indicates the code is likely not auto-generated, and 0.0 indicates it is.

Example Input:

{
    "code_lines": [
        "# This file is automatically generated by the build system.",
        "def some_function():",
        "    return True"
    ]
}

Example Output:

{
    "code_lines": [
        "# This file is automatically generated by the build system.",
        "def some_function():",
        "    return True"
    ],
    "CodeAutoGeneratedMarkerCount": 1,
    "CodeAutoGeneratedScore": 0.0
}