10 releases

Uses old Rust 2015

0.3.2 Apr 10, 2019
0.3.1 Apr 3, 2019
0.3.0 Feb 21, 2019
0.2.0 Oct 14, 2018
0.1.1 Mar 31, 2018

#854 in Encoding

Custom license

375KB
8K SLoC

Agnes

Build Status Documentation Join the chat at https://gitter.im/agnes-rs/community

agnes is a data wrangling library for Rust.

Some useful links:

Overview

agnes is a statically-typed high-performance data processing library for the Rust programming language. It provides utilities for data loading, preprocessing, aggregation, annotation, and serialization. The primary goal of agnes is to to provide functionality to help in the development of robust, efficient, readable applications for your data preprocessing tasks.

Features

  • Data structures for handling heterogeneously-typed tabular data.
  • Extensible data source framework which currently supports the loading of local or web-based CSV files, with more data source types under development.
  • Data output through serde, allowing output to any serialization output serde supports.
  • Handling of missing data (NaNs) for all data types.
  • Data merging and joining to combine data from multiple sources.
  • Iterator- or index-based data access

Design Principles

agnes was designed with the following general principles in mind:

  • Minimal data duplication: a single data source can be shared by multiple data views or outputs.
  • Type safety -- agnes leverages Rust's typing system to provide the compile-time advantges that static typing provides.
  • Embracing of existing Rust paradigms and best practices. This includes the use of iterators, explicit memory control, and existing Rust libraries (such as serde).

Usage

To use, add this this to your Cargo.toml:

[dependencies]
agnes = "0.3"

and this to your crate root:

extern crate agnes;

Example

As an simple example, let's build an application that reads in a data set, and displays it. A more complete example illustrating much more agnes functionality can be found in the guide here.

This example loads specific fields from a country-by-country data file, and shows off the table definition format, source specification format, and loading-from-URI functionality. This example can also be found here.

#[macro_use]
extern crate agnes;

use agnes::source::csv::load_csv_from_uri;

// specify the GDP table (only the fields we are concerned about)
tablespace![
    table gdp {
        CountryName: String,
        CountryCode: String,
        Gdp2015: f64,
    }
];

fn main() {
    // specify the source location for our GDP fields
    let gdp_spec = spec![
        fieldname gdp::CountryName = "Country Name";
        fieldname gdp::CountryCode = "Country Code";
        fieldname gdp::Gdp2015 = "2015";
    ];

    // load the CSV file from a URI
    let gdp_view =
        load_csv_from_uri("https://wee.codes/data/gdp.csv", gdp_spec).expect("CSV loading failed.");

    // print the DataView
    println!("{}", gdp_view);
}

Changes and Future Plans

License

This work is licensed under the MIT Licence.

Dependencies

~19–31MB
~447K SLoC