Skip to content

A dead-simple crawler which focuses on ease of use and speed.

License

Notifications You must be signed in to change notification settings

manzanit0/spidey

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

84 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Spidey

A dead-simple, concurrent web crawler which focuses on ease of use and speed.

Installation

The package can be installed by adding spidey to your list of dependencies in mix.exs:

def deps do
  [
    {:spidey, "~> 0.3"}
  ]
end

The docs can be found at https://hexdocs.pm/spidey

Usage

Spidey has been thought with ease of usage in mind, so all you have to do to get started is:

iex> Spidey.crawl("https://manzanit0.github.io", :crawler_name, pool_size: 15)
[
  "https://https://manzanit0.github.io/foo",
  "https://https://manzanit0.github.io/bar-baz/#",
  ...
]

In a nutshell, the above line will:

  1. Spin up a new supervision tree under the Spidey OTP Application that will supervise a task supervisor and the queue of URLs.
  2. Create an ETS table to store crawled urls
  3. Crawl the website
  4. Return all the urls as a list
  5. Teardown the supervision tree and the ETS table

The function is blocking, but if you were to call it asynchronously multiple times, each invocation will spin up a new supervision trees with a new task supervisor and a new queue.

But why is it blocking?

The reason why it has been made blocking instead of non-blocking is because there are already multiple libraries which do async crawling out there... and I needed one that was blocking which allowed me to decide when to run it synchronously and when not to.

Specifying your own filter

Furthermore, if you would you want to specify your own filter for crawled URLs, you can do so by implementing the Spidey.Filter behaviour:

defmodule MyApp.RssFilter do
  @behaviour Spidey.Filter

  @impl true
  def filter_urls(urls, _opts) do
    urls
    |> Stream.reject(&String.ends_with?(&1, "feed/"))
    |> Stream.reject(&String.ends_with?(&1, "feed"))
  end
 end

And simply pass it down to the crawler as an option:

Spidey.crawl("https://manzanit0.github.io", :crawler_name, filter: MyApp.RssFilter)

It's encouraged to use the Stream module instead of the Enum since the code that handles the filtering uses streams.

Configuration

Currently Spidey supports the following configuration:

  • :log - the log level used when logging events with Elixir's Logger. If false, disables logging. Defaults to :debug
config :spidey, log: :info

Using the CLI

To be able to run the application make sure to have Elixir installed. Please check the official instructions: link

Once you have Elixir installed, to set up the application run:

git clone https://github.com/Manzanit0/spidey
cd spidey
mix deps.get
mix escript.build

To crawl websites, run the escript ./spidey:

./spidey --site https://manzanit0.github.io/

Escripts will run in any system which has Erlang/OTP installed, regardless if they have Elixir or not.

CLI options

Spidey provides two main functionalities – crawling a specific domain and saving it to a file according to the plain text site map protocol. For the latter, simply append --save to the execution.

About

A dead-simple crawler which focuses on ease of use and speed.

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages