∞ Flat¶
Documentation¶
- Class name:
LLMFlatReader
- Category:
SALT/Language Toolkit/Readers
- Output node:
False
The LLMFlatReader node is designed to read and process 'flat' files, converting them into a format suitable for indexing and further analysis by the llama_index Document system. It leverages the foundational capabilities of the FlatReader class to facilitate the ingestion of simple text files into more structured document representations.
Input types¶
Required¶
path
- Specifies the filesystem path to the 'flat' file that needs to be read. This path is essential for locating and accessing the file for processing.
- Comfy dtype:
STRING
- Python dtype:
str
Optional¶
extra_info
- Provides additional, optional information in a string format that can be used to influence the reading process or to pass extra parameters that might be needed for specific use cases.
- Comfy dtype:
STRING
- Python dtype:
str
Output types¶
documents
- Comfy dtype:
DOCUMENT
- The processed data from the 'flat' file, structured as a document suitable for indexing and further analysis.
- Python dtype:
tuple
- Comfy dtype:
Usage tips¶
- Infra type:
CPU
- Common nodes: unknown
Source code¶
class LLMFlatReader(FlatReader):
"""
@NOTE: Reads 'flat' files into a llama_index Document
@Source: https://github.com/run-llama/llama_index/blob/main/llama-index-integrations/readers/llama-index-readers-file/llama_index/readers/file/flat/base.py
@Documentation: https://docs.llamaindex.ai/en/latest/api_reference/readers/file/#llama_index.readers.file.FlatReader
"""
def __init__(self):
super().__init__()
@classmethod
def INPUT_TYPES(cls):
return {
"required": {
"path": ("STRING", {"default": ""}),
},
"optional": {
"extra_info": ("STRING", {"multiline": True, "dynamicPrompts": False, "default": "{}"}),
}
}
RETURN_TYPES = ("DOCUMENT", )
RETURN_NAMES = ("documents",)
FUNCTION = "execute"
CATEGORY = f"{MENU_NAME}/{SUB_MENU_NAME}/Readers"
def execute(self, path:str, extra_info:str, fs = None):
get_full_path(1, path)
if not os.path.exists(path):
raise FileNotFoundError(f"No file available at: {path}")
path = Path(path)
extra_info = read_extra_info(extra_info)
data = self.load_data(path, extra_info)
return (data, )