Splash Scripts Tutorial


Scripting support is an experimental feature for early adopters; API could change in future releases.


Splash can execute custom rendering scripts written in the Lua programming language. This allows us to use Splash as a browser automation tool similar to PhantomJS.

To execute a script and get the result back send it to the execute endpoint in a lua_source argument.


Most likely you’ll be able to follow Splash scripting examples even without knowing Lua; nevertheless, the language is worth learning. With Lua you can, for example, write Redis, Nginx, Apache, World of Warcraft scripts, create mobile apps using Moai or Corona SDK or use the state of the art Deep Learning framework Torch7. It is easy to get started and there are good online resources available like the tutorial Learn Lua in 15 minutes and the book Programming in Lua.

Let’s start with a basic example:

function main(splash)
    local title = splash:evaljs("document.title")
    return {title=title}

If we submit this script to the execute endpoint in a lua_source argument, Splash will go to the example.com website, wait until it loads, wait aother half-second, then get the page title (by evaluating a JavaScript snippet in page context), and then return the result as a JSON encoded object.


Splash UI provides an easy way to try scripts: there is a code editor for Lua and a button to submit a script to execute. Visit (or whatever host/port Splash is listening to).

Entry Point: the “main” Function

The script must provide a “main” function which is called by Splash. The result is returned as an HTTP response. The script could contain other helper functions and statements, but ‘main’ is required.

In the first example ‘main’ function returned a Lua table (an associative array similar to JavaScript Object or Python dict). Such results are returned as JSON.

The following will return the string {"hello":"world!"} as an HTTP response:

function main(splash)
    return {hello="world!"}

The script can also return a string:

function main(splash)
    return 'hello'

Strings are returned as-is (unlike tables they are not encoded to JSON). Let’s check it with curl:

$ curl ''

The “main” function receives an object that allows us to control the “browser tab”. All Splash features are exposed using this object. By convention, this argument is called “splash”, but you are not required to follow this convention:

function main(please)
    return "ok"

Where Are My Callbacks?

Here is a snippet from our first example:

local title = splash:evaljs("document.title")

The code looks like standard procedural code; there are no callbacks or fancy control-flow structures. It doesn’t mean Splash works in a synchronous way; under the hood it is still async. When you call splash.wait(0.5), Splash switches from the script to other tasks, and comes back after 0.5s.

It is possible to use loops, conditional statements, functions as usual in Splash scripts which enables more straightforward coding.

Let’s check an example PhantomJS script:

var users = ["PhantomJS", "ariyahidayat", /*...*/];

function followers(user, callback) {
    var page = require('webpage').create();
    page.open('http://mobile.twitter.com/' + user, function (status) {
        if (status === 'fail') {
            console.log(user + ': ?');
        } else {
            var data = page.evaluate(function () {
                return document.querySelector('div.profile td.stat.stat-last div.statnum').innerText;
            console.log(user + ': ' + data);
function process() {
    if (users.length > 0) {
        var user = users[0];
        users.splice(0, 1);
        followers(user, process);
    } else {

The code is (arguably) tricky: process function implements a loop by creating a chain of callbacks; followers function doesn’t return a value (it would be more complex to implement) - the result is logged to the console instead.

A similar Splash script:

users = {'PhantomJS', 'ariyahidayat'}

function followers(splash, user)
    local ok, msg = splash:go('http://mobile.twitter.com/' .. user)
    if not ok then
        return "?"
    return splash:evaljs([[
        document.querySelector('div.profile td.stat.stat-last div.statnum').innerText;

function process(splash, users)
    local result = {}
    for idx, user in ipairs(users) do
        result[user] = followers(splash, user)
    return result

function main(splash)
    local users = process(splash, users)
    return {users=users}


  • some Lua knowledge is helpful to be productive in Splash Scripts: ipairs, [[multi-line strings]] or string concatenation via .. could be unfamiliar;
  • in Splash variant followers function can return a result (a number of twitter followers); also, it doesn’t need a “callback” argument;
  • instead of a page.open callback which receives “status” argument there is a “blocking” splash:go call which returns “ok” flag;
  • error handling is different: in case of an HTTP 4xx or 5xx error PhantomJS doesn’t return an error code to page.open callback - example script will try to get the followers nevertheless because “status” won’t be “fail”; in Splash this error will be detected and ”?” will be returned;
  • process function can use a standard Lua for loop without a need to create a recursive callback chain;
  • instead of console messages we’ve got a JSON HTTP API;
  • apparently, PhantomJS allows to create multiple page objects and run several page.open requests in parallel (?); Splash only provides a single “browser tab” to a script via its splash parameter of main function (but you’re free to send multiple concurrent requests with Lua scripts to Splash).

There are great PhantomJS wrappers like CasperJS and NightmareJS which (among other things) bring a sync-looking syntax to PhantomJS scripts by providing custom control flow mini-languages. However, they all have their own gotchas and edge cases (loops? moving code to helper functions? error handling?). Splash scripts are standard Lua code.


PhantomJS itself and its wrappers are great, they deserve lots of respect; please don’t take this writeup as an attack on them. These tools are much more mature and feature complete than Splash. Splash tries to look at the problem from a different angle, but for each unique Splash feature there are ten unique PhantomJS features.

Living Without Callbacks


For the curious, Splash uses Lua coroutines under the hood.

Internally, “main” function is executed as a coroutine by Splash, and some of the splash:foo() methods use coroutine.yield. See http://www.lua.org/pil/9.html for Lua coroutines tutorial.

In Splash scripts it is not explicit which calls are async and which calls are blocking; this is a common criticism of coroutines/greenlets. Check this article for a good description of the problem.

However, these negatives have no real impact in Splash scripts which: are meant to be small, where shared state is minimized, and the API is designed to execute a single command at a time, so in most cases the control flow is linear.

If you want to be safe then think of all splash methods as async; consider that after you call splash:foo() a webpage being rendered can change. Often that’s the point of calling a method, e.g. splash:wait(time) or splash:go(url) only make sense because webpage changes after calling them, but still - keep it in mind.

There are async methods like splash:go, splash:wait, splash:wait_for_resume, etc.; most splash methods are currently not async, but thinking of them as of async will allow your scripts to work if we ever change that.

Calling Splash Methods

Unlike in many languages, methods in Lua are usually separated from an object using a colon :; to call “foo” method of “splash” object use splash:foo() syntax. See http://www.lua.org/pil/16.html for more details.

There are two main ways to call Lua methods in Splash scripts: using positional and named arguments. To call a method using positional arguments use parentheses splash:foo(val1, val2), to call it with named arguments use curly braces: splash:foo{name1=val1, name2=val2}:

-- Examples of positional arguments:
splash:wait(0.5, false)
local title = splash:evaljs("document.title")

-- The same using keyword arguments:
splash:wait{time=0.5, cancel_on_redirect=false}
local title = splash:evaljs{source="document.title"}

-- Mixed arguments example:
splash:wait{0.5, cancel_on_redirect=false}

For convenience all splash methods are designed to support both styles of calling: positional and named. But since there are no “real” named arguments in Lua most Lua functions (including the ones from the standard library) choose to support just positional arguments.

Error Handling

There are two ways to report errors in Lua: raise an exception and return an error flag. See http://www.lua.org/pil/8.3.html.

Splash uses the following convention:

  1. for developer errors (e.g. incorrect function arguments) exception is raised;
  2. for errors outside developer control (e.g. a non-responding remote website) status flag is returned: functions that can fail return ok, reason pairs which developer can either handle or ignore.

If main results in an unhandled exception then Splash returns HTTP 400 response with an error message.

It is possible to raise an exception manually using Lua error function:

error("A message to be returned in a HTTP 400 response")

To handle Lua exceptions (and prevent Splash from returning HTTP 400 response) use Lua pcall; see http://www.lua.org/pil/8.4.html.

To convert “status flag” errors to exceptions Lua assert function can be used. For example, if you expect a website to work and don’t want to handle errors manually, then assert allows to stop processing and return HTTP 400 if the assumption is wrong:

local ok, msg = splash:go("http://example.com")
if not ok then
    -- handle error somehow, e.g.

-- a shortcut for the code above: use assert


By default Splash scripts are executed in a restricted environment: not all standard Lua modules and functions are available, Lua require is restricted, and there are resource limits (quite loose though).

To disable the sandbox start Splash with --disable-lua-sandbox option:

$ python -m splash.server --disable-lua-sandbox

Custom Lua Modules

Splash provides a way to use custom Lua modules (stored on server) from scripts passed via HTTP API. This allows to

  1. reuse code without sending it over network again and again;
  2. use third-party Lua modules;
  3. implement features which need unsafe code and expose them safely in a sandbox.


To learn about Lua modules check e.g. http://lua-users.org/wiki/ModulesTutorial. Please prefer “the new way” of writing modules because it plays better with a sandbox. A good Lua modules style guide can be found here: http://hisham.hm/2014/01/02/how-to-write-lua-modules-in-a-post-module-world/

Setting Up

To use custom Lua modules, do the following steps:

  1. setup the path for Lua modules and add your modules there;
  2. tell Splash which modules are enabled in a sandbox;
  3. use Lua require function from a script to load a module.

To setup the path for Lua modules start Splash with --lua-package-path option. --lua-package-path value should be a semicolon-separated list of places where Lua looks for modules. Each entry should have a ? in it that’s replaced with the module name.


$ python -m splash.server --lua-package-path "/etc/splash/lua_modules/?.lua;/home/myuser/splash-modules/?.lua"


If you use Splash installed using Docker see Folders Sharing for more info on how to setup paths.


For the curious: --lua-package-path value is added to Lua package.path.

When you use a Lua sandbox (default) Lua require function is restricted when used in scripts: it only allows to load modules from a whitelist. This whitelist is empty by default, i.e. by default you can require nothing. To make your modules available for scripts start Splash with --lua-sandbox-allowed-modules option. It should contain a semicolon-separated list of Lua module names allowed in a sandbox:

$ python -m splash.server --lua-sandbox-allowed-modules "foo;bar" --lua-package-path "/etc/splash/lua_modules/?.lua"

After that it becomes possible to load these modules from Lua scripts using require:

local foo = require("foo")
function main(splash)
    return {result=foo.myfunc()}

Writing Modules

A basic module could look like the following:

-- mymodule.lua
local mymodule = {}

function mymodule.hello(name)
    return "Hello, " .. name

return mymodule

Usage in a script:

local mymodule = require("mymodule")

function main(splash)
    return mymodule.hello("world!")

Many real-world modules will likely want to use splash object. There are several ways to write such modules. The simplest way is to use functions that accept splash as an argument:

-- utils.lua
local utils = {}

-- wait until `condition` function returns true
function utils.wait_for(splash, condition)
    while not condition() do

return utils


local utils = require("utils")

function main(splash)

    -- wait until <h1> element is loaded
    utils.wait_for(splash, function()
       return splash:evaljs("document.querySelector('h1') != null")

    return splash:html()

Another way to write such module is to add a method to splash object. This can be done by adding a method to its Splash class - the approach is called “open classes” in Ruby or “monkey-patching” in Python.

-- wait_for.lua

-- Sandbox is not enforced in custom modules, so we can import
-- internal Splash class and change it - add a method.
local Splash = require("splash")

function Splash:wait_for(condition)
    while not condition() do

-- no need to return anything



function main(splash)

    -- wait until <h1> element is loaded
       return splash:evaljs("document.querySelector('h1') != null")

    return splash:html()

Which style to prefer is up to the developer. Functions are more explicit and composable, monkey patching enables a more compact code. Either way, require is explicit.

As seen in a previous example, sandbox restrictions for standard Lua modules and functions are not applied in custom Lua modules, i.e. you can use all the Lua powers. This makes it possible to import third-party Lua modules and implement advanced features, but requires developer to be careful. For example, let’s use os module:

-- evil.lua
local os = require("os")
local evil = {}

function evil.sleep()
    -- Don't do this! It blocks the event loop and has a startup cost.
    -- splash:wait is there for a reason.
    os.execute("sleep 2")

function evil.touch(filename)
    -- another bad idea
    os.execute("touch " .. filename)

-- todo: rm -rf /

return evil


By default Splash aborts script execution after a timeout (30s by default); it is a common problem for long scripts.

For more information see I’m getting lots of 504 Timeout errors, please help! and 2. Splash Lua script does too many things.