Crawly is an application framework for crawling web sites and extracting structured data which can be used for a wide range of useful applications, like data mining, information processing or historical archival.
- Elixir "~> 1.7"
- Works on Linux, Windows, OS X and BSD
-
Add Crawly as a dependencies:
# mix.exs defp deps do [ {:crawly, "~> 0.11.0"}, {:floki, "~> 0.26.0"} ] end
-
Fetch dependencies:
$ mix deps.get
-
Create a spider
# lib/crawly_example/esl_spider.ex defmodule EslSpider do use Crawly.Spider alias Crawly.Utils @impl Crawly.Spider def base_url(), do: "https://www.erlang-solutions.com" @impl Crawly.Spider def init(), do: [start_urls: ["https://www.erlang-solutions.com/blog.html"]] @impl Crawly.Spider def parse_item(response) do {:ok, document} = Floki.parse_document(response.body) hrefs = document |> Floki.find("a.more") |> Floki.attribute("href") requests = Utils.build_absolute_urls(hrefs, base_url()) |> Utils.requests_from_urls() title = document |> Floki.find("article.blog_post h1") |> Floki.text() %{ :requests => requests, :items => [%{title: title, url: response.request_url}] } end end
-
Configure Crawly
- By default, Crawly does not require any configuration. But obviously you will need a configuration for fine tuning the crawls:
# in config.exs config :crawly, closespider_timeout: 10, concurrent_requests_per_domain: 8, middlewares: [ Crawly.Middlewares.DomainFilter, Crawly.Middlewares.UniqueRequest, {Crawly.Middlewares.UserAgent, user_agents: ["Crawly Bot"]} ], pipelines: [ {Crawly.Pipelines.Validate, fields: [:url, :title]}, {Crawly.Pipelines.DuplicatesFilter, item_id: :title}, Crawly.Pipelines.JSONEncoder, {Crawly.Pipelines.WriteToFile, extension: "jl", folder: "/tmp"} ]
-
Start the Crawl:
$ iex -S mix
iex(1)> Crawly.Engine.start_spider(EslSpider)
-
Results can be seen with:
$ cat /tmp/EslSpider.jl
Crawly can be configured in the way that all fetched pages will be browser rendered, which can be very useful if you need to extract data from pages which has lots of asynchronous elements (for example parts loaded by AJAX).
You can read more here:
The CrawlyUI project is an add-on that aims to provide an interface for managing and rapidly developing spiders.
Checkout the code from GitHub or try it online CrawlyUIDemo
See more at Experimental UI
- Pluggable HTTP client
- Retries support
- Cookies support
- XPath support - can be actually done with meeseeks
- Project generators (spiders)
- UI for jobs management
- Blog post on Erlang Solutions website: https://www.erlang-solutions.com/blog/web-scraping-with-elixir.html
- Blog post about using Crawly inside a machine learning project with Tensorflow (Tensorflex): https://www.erlang-solutions.com/blog/how-to-build-a-machine-learning-project-in-elixir.html
- Web scraping with Crawly and Elixir. Browser rendering: https://medium.com/@oltarasenko/web-scraping-with-elixir-and-crawly-browser-rendering-afcaacf954e8
- Blog crawler: https://github.com/oltarasenko/crawly-spider-example
- E-commerce websites: https://github.com/oltarasenko/products-advisor
- Car shops: https://github.com/oltarasenko/crawly-cars
- JavaScript based website (Splash example): https://github.com/oltarasenko/autosites
We would gladly accept your contributions!
Please find documentation on the HexDocs
Using Crawly on production? Please let us know about your case!