Google Image Search

A crate designed to search Google Images based on provided arguments. Due to the limitations of using only a single request to fetch images, only a max of about 100 images can be found per request. These images may be protected under copyright, and you shouldn't do anything punishable with them, like using them for commercial use.

Arguments

There are 2 required parameters, along with a variety of different arguments.

Search Arguments

These are optional arguments that Google can use to filter images, useful for narrowing your search. They are used via the various methods on the Arguments struct. Each argument is contained in an enum which contains all possible options.

Examples

Using the asynchronous API requires some sort of async runtime, usually tokio, which can be added to your Cargo.toml like so:

[dependencies]
image_search = "0.4"
tokio = { version = "1", features = ["full"] }

It can be used like this:

extern crate tokio;
extern crate image_search;

use std::path::PathBuf;
use image_search::{Arguments, Color, urls, search, download};
 
#[tokio::main]
async fn main() -> Result<(), image_search::Error> {
    let args = Arguments::new("example", 10)
        .color(Color::Gray)
        .directory(PathBuf::from("downloads")); // Only affects the download function
     
    let _image_urls = urls(args.clone()).await?;
    let _images = search(args.clone()).await?;
    let _paths = download(args).await?;
 
    Ok(())
}

Blocking

There is an optional "blocking" API that can be enabled:

[dependencies]
image_search = { version = "0.4", features = ["blocking"] }

This is called like so:

extern crate image_search;

use std::path::PathBuf;
use image_search::{Arguments, Time, blocking::{urls, search, download}};

fn main() -> Result<(), image_search::Error> {
    let args = Arguments::new("example", 10)
        .time(Time::Month)
        .directory(PathBuf::from("downloads")); // Only affects the download function
    
    let _image_urls = urls(args.clone())?;
    let _images = search(args.clone())?;
    let _paths = download(args)?;

    Ok(())
}

Clients

This crate uses surf for HTTP requests in order to allow for the customization of the client used for HTTP requests. This can allow programs to interface with C via CURL, pure Rust via hyper or async-h1, or even WASM. As with surf, the client used can be customized via features. In order to change you will have to set default-features=false in your Cargo.toml, since curl is used by default. The possible backends are listed here:

• curl: Uses CURL through isahc as the HTTP backend.
• hyper (default): Uses hyper as the HTTP backend.
• wasm: Uses window.fetch as the HTTP backend.
• h1: Uses async-h1 as the HTTP backend with native TLS for HTTPS.
• rustls: Uses async-h1 as the HTTP backend with rustls for HTTPS.