Skip to content

saving_utils.md

Latest commit

 

History

History
56 lines (38 loc) · 1.48 KB

saving_utils.md

File metadata and controls

56 lines (38 loc) · 1.48 KB

loaders.saving_utils

Save DataFrame to file (CSV, TXT, XLSX).

Overview

The saving_utils module provides functions to save pandas DataFrames to disk. It complements loading_utils (which loads data) and is used by the View Data save dialog and Streamlit download functionality.

Key Functions

save_dataframe(data, file_path, file_type=None) -> str

Save a DataFrame to file.

Parameters:

  • data: DataFrame to save
  • file_path: Full path to the output file
  • file_type: Optional. 'csv', 'txt', or 'xlsx'. If None, inferred from path extension.

Returns:

  • The path where the file was saved

Raises:

  • ValueError: If file type is not supported

Example:

from loaders.saving_utils import save_dataframe
import pandas as pd

df = pd.DataFrame({'x': [1, 2, 3], 'y': [2, 4, 6]})
save_dataframe(df, 'output/processed.csv')
# Or with explicit type:
save_dataframe(df, 'output/processed.xlsx', file_type='xlsx')

get_default_save_directory() -> str

Return the default directory for saving (typically the input folder from config).

Returns:

  • Absolute path to the default save directory

Example:

from loaders.saving_utils import get_default_save_directory

default_dir = get_default_save_directory()
# Use in file picker: initialdir=default_dir

Supported Formats

  • CSV: Comma-separated; NaN written as 'no' (consistent with loading).
  • TXT: Tab-separated; NaN written as 'no'.
  • XLSX: Excel format; requires openpyxl.