How to disable SimpleDirectoryReader from parsing markdown files

[lifu963]

I hope it can split the markdown file like a txt file, instead of spliting it into nodes based on markdown headers.

[dosubot[bot]]

To disable SimpleDirectoryReader from parsing markdown files and instead split the markdown file like a text file, you can modify the file_extractor argument to exclude the MarkdownReader for .md files. Here is how you can do it:

from llama_index.legacy.readers.file.base import SimpleDirectoryReader

# Create a SimpleDirectoryReader instance without MarkdownReader
reader = SimpleDirectoryReader(
    input_dir="path/to/your/directory",
    file_extractor={'.md': None}  # This disables the MarkdownReader for .md files
)

# Load data
documents = reader.load_data()

By setting the file_extractor for .md files to None, the SimpleDirectoryReader will treat markdown files as plain text files ^[1][2][3][4].

To continue talking to Dosu, mention @dosu.