simple-data-loader 1.0.0

A simple Python package for loading data from CSV and XLSX files

DataLoader

A Python class for loading data from CSV and XLSX files, with support for single files or concatenating multiple files from folders.

Features

• Single File Loading: Read CSV or XLSX files individually
• Folder Loading: Automatically concatenate all CSV/XLSX files in a folder
• Subfolder Support: Option to include files from subfolders
• Verbose Output: Control the level of detail in console output
• Error Handling: Graceful handling of file errors and unsupported formats
• Flexible Usage: Both class-based and function-based interfaces

Installation

1. Install the required dependencies:

pip install -r requirements.txt

2. Or install dependencies manually:

pip install pandas openpyxl xlrd

Dependencies

• pandas (>=1.3.0): For data manipulation and DataFrame operations
• openpyxl (>=3.0.0): For reading Excel (.xlsx) files
• xlrd (>=2.0.0): For reading legacy Excel (.xls) files

Quick Start

Basic Usage

from data_loader import DataLoader

# Load a single file
loader = DataLoader("data.csv")
df = loader.load()

# Load all files from a folder
loader = DataLoader("data_folder")
df = loader.load()

Using the Convenience Function

from data_loader import load_data

# Direct loading
df = load_data("data.csv")
df = load_data("data_folder")

Detailed Usage

Class Initialization

DataLoader(file_path, include_subfolders=False, verbose=True, column_consistency='error')

Parameters:

• file_path (str): Path to a file or folder
• include_subfolders (bool): Whether to include files from subfolders (default: False)
• verbose (bool): Whether to print detailed information (default: True)
• column_consistency (str): How to handle column consistency ('error', 'warning', 'ignore') (default: 'error')

Examples

1. Single File Loading

from data_loader import DataLoader

# Load a CSV file
loader = DataLoader("sales_data.csv")
df = loader.load()

# Load an Excel file
loader = DataLoader("financial_report.xlsx")
df = loader.load()

Output:

sales_data.csv is imported with 1000 rows and 5 columns

2. Folder Loading (No Subfolders)

loader = DataLoader("data_folder", include_subfolders=False)
df = loader.load()

Output:

Found 3 files to process
data_1.csv is imported with 500 rows and 4 columns
data_2.csv is imported with 300 rows and 4 columns
data_3.xlsx is imported with 200 rows and 4 columns

Summary:
Successfully loaded 3 files
Combined dataset has 1000 rows and 4 columns

3. Folder Loading (With Subfolders)

loader = DataLoader("data_folder", include_subfolders=True)
df = loader.load()

This will recursively search through all subfolders and load all CSV/XLSX files.

4. Quiet Mode

loader = DataLoader("data.csv", verbose=False)
df = loader.load()

No console output will be displayed.

5. Column Consistency Control

# Error mode (default) - stops if columns don't match
loader = DataLoader("data_folder", column_consistency='error')
df = loader.load()

# Warning mode - shows warning but continues
loader = DataLoader("data_folder", column_consistency='warning')
df = loader.load()

# Ignore mode - skips consistency check entirely
loader = DataLoader("data_folder", column_consistency='ignore')
df = loader.load()

Column Consistency Modes:

• 'error' (default): Raises an error if files have different column counts or names
• 'warning': Shows a warning but continues processing
• 'ignore': Skips consistency check entirely

6. Convenience Function

from data_loader import load_data

# All parameters are optional
df = load_data("data.csv")  # Uses defaults
df = load_data("data_folder", include_subfolders=True, verbose=False, column_consistency='warning')

Supported File Formats

• CSV files: .csv
• Excel files: .xlsx, .xls

Error Handling

The DataLoader handles various error scenarios:

• File not found: Raises FileNotFoundError
• Unsupported format: Raises ValueError with format information
• Invalid path: Raises ValueError if path is neither file nor directory
• Column consistency errors: Raises ValueError when column_consistency='error' and files have mismatched columns
• Individual file errors: Continues processing other files and reports errors in verbose mode

Example Project Structure

project/
├── data_loader.py
├── requirements.txt
├── example_usage.py
├── README.md
├── data/
│   ├── sales_2023.csv
│   ├── sales_2024.csv
│   └── reports/
│       ├── monthly_report.xlsx
│       └── quarterly_summary.csv
└── single_file.csv

Running Examples

To see the DataLoader in action, run the example script:

python example_usage.py

This will create sample data files and demonstrate various usage patterns.

API Reference

DataLoader Class

Methods

• load(): Load data from the specified path
  • Returns: pandas.DataFrame

Internal Methods

• _load_single_file(): Load data from a single file
• _load_folder(): Load and concatenate data from folder
• _load_single_file_from_path(): Internal method for loading individual files

Convenience Function

• load_data(file_path, include_subfolders=False, verbose=True, column_consistency='error'): Direct data loading function

Performance Notes

• Large files are loaded into memory entirely
• For very large datasets, consider processing files individually
• Concatenation happens in memory, so ensure sufficient RAM for large folder operations

License

This project is open source and available under the MIT License.