Convert CSV To Parquet (CC2P)

CC2P (Convert CSV To Parquet) is a high-performance command-line tool written in Rust that efficiently converts CSV files to the Apache Parquet format. Parquet is a columnar storage file format that offers efficient data compression and encoding schemes, making it ideal for big data processing.

Why Use CC2P?

• Performance: Leverages Rust's speed and multi-threading for fast conversions
• Memory Efficiency: Processes files with minimal memory footprint
• Flexibility: Supports various CSV formats with different delimiters and header options
• Schema Inference: Automatically detects column types from your data
• Batch Processing: Convert multiple CSV files in a single command

Installation

From Cargo (Recommended)

If you have Rust installed, you can install CC2P directly from crates.io:

cargo install cc2p

From GitHub Releases

You can download pre-built binaries from the GitHub Releases page.

From Source

To build from source:

# Clone the repository
git clone https://github.com/rayyildiz/cc2p.git
cd cc2p

# Build in release mode
cargo build --release

# The binary will be in target/release/cc2p

Usage

Basic usage:

cc2p [OPTIONS] [PATH]

Where PATH is the path to a CSV file or a glob pattern (default: *.csv).

Examples

Convert a single CSV file:

cc2p data.csv

Convert all CSV files in the current directory:

cc2p

Convert CSV files with semicolon delimiter:

cc2p --delimiter ";" *.csv

Convert CSV files without headers:

cc2p --no-header data_files/*.csv

Use 4 worker threads for faster processing:

cc2p --worker 4 large_data.csv

Options

• -d, --delimiter : Delimiter character used in CSV files (default: ,)
• -n, --no-header: Whether to include the header in the CSV search column (default: false)
• -w, --worker: Number of worker threads to use for performing the task (default: 1)
• -s, --sampling: Number of rows to sample for inferring the schema (default: 2048)

$ cc2p --help

Convert a CSV to parquet file format

Usage: cc2p [OPTIONS] [PATH]

Arguments:
  [PATH]  Represents the folder path for CSV search [default: *.csv]

Options:
  -d, --delimiter <DELIMITER>  Represents the delimiter used in CSV files [default: ,]
  -n, --no-header              Represents whether to include the header in the CSV search column
  -w, --worker <WORKER>        Number of worker threads to use for performing the task [default: 1]
  -s, --sampling <SAMPLING>    Number of rows to sample for inferring the schema. [default: 100]
  -h, --help                   Print help
  -V, --version                Print version

Features

Technical Features

• Columnar Storage: Parquet's columnar format provides better compression and faster query performance compared to row-based formats like CSV
• Efficient Compression: Uses Snappy compression for a good balance between compression ratio and speed
• Schema Handling: Automatically infers data types and handles duplicate column names
• Parallel Processing: Multi-threaded conversion using Tokio runtime
• Progress Tracking: Real-time progress indication with indicatif progress bars
• Error Handling: Robust error handling with detailed error messages

Performance Benefits

• Reduced Storage: Parquet files are typically much smaller than equivalent CSV files
• Faster Analytics: A columnar format allows for more efficient querying in data analysis tools
• Schema Enforcement: Parquet maintains schema information, unlike CSV which is schema-less
• Selective Column Reading: Analytics tools can read only the columns they need, improving performance

Platform-Specific Notes

macOS Users

NOTE for macOS Users: Our Apple signing/notarization is not entirely done yet, thus you have to run the following command once to run the application. Download the app and run this command:

xattr -c cc2p

Linux Users

On Linux, you can also install CC2P via Snap:

sudo snap install cc2p

Technical Requirements

• Rust Version: 1.85.0 or later
• Rust Edition: 2024
• Minimum Memory: Depends on the size of CSV files being processed

Contributing

If you wish to contribute, please feel free to fork the repository, make your changes, and submit a pull request. All contributions are welcome!

Development Setup

1. Clone the repository
2. Install Rust (1.85.0 or later)
3. Run cargo build to build the project
4. Run cargo test to run the tests

License

This project is licensed under MIT, see the LICENSE file for details.

Contact

• Project Link: https://github.com/rayyildiz/cc2p
• Report Issues: https://github.com/rayyildiz/cc2p/issues