Table of contents
  1. 6 Web application APIs
    1. 6.1 Scripting
      1. 6.1.1 Introduction
      2. 6.1.2 Enabling and disabling scripting
      3. 6.1.3 Processing model
        1. 6.1.3.1 Definitions
        2. 6.1.3.2 Calling scripts
        3. 6.1.3.3 Creating scripts
        4. 6.1.3.4 Killing scripts
        5. 6.1.3.5 Runtime script errors
      4. 6.1.4 Event loops
        1. 6.1.4.1 Definitions
        2. 6.1.4.2 Processing model
        3. 6.1.4.3 Generic task sources
      5. 6.1.5 The javascript: URL scheme
      6. 6.1.6 Events
        1. 6.1.6.1 Event handlers
        2. 6.1.6.2 Event handlers on elements, Document objects, and Window objects
        3. 6.1.6.3 Event firing
        4. 6.1.6.4 Events and the Window object
    2. 6.2 Base64 utility methods

6 Web application APIs

6.1 Scripting

6.1.1 Introduction

Various mechanisms can cause author-provided executable code to run in the context of a document. These mechanisms include, but are probably not limited to:

6.1.2 Enabling and disabling scripting

Scripting is enabled in a browsing context when all of the following conditions are true:

Scripting is disabled in a browsing context when any of the above conditions are false (i.e. when scripting is not enabled).


Scripting is enabled for a node if the Document object of the node (the node itself, if it is itself a Document object) has an associated browsing context, and scripting is enabled in that browsing context.

Scripting is disabled for a node if there is no such browsing context, or if scripting is disabled in that browsing context.

6.1.3 Processing model

6.1.3.1 Definitions

A script has:

A script execution environment

The characteristics of the script execution environment depend on the language, and are not defined by this specification.

In JavaScript, the script execution environment consists of the interpreter, the stack of execution contexts, the global code and function code and the Function objects resulting, and so forth.

A list of code entry-points

Each code entry-point represents a block of executable code that the script exposes to other scripts and to the user agent.

Each Function object in a JavaScript script execution environment has a corresponding code entry-point, for instance.

The main program code of the script, if any, is the initial code entry-point. Typically, the code corresponding to this entry-point is executed immediately after the script is parsed.

In JavaScript, this corresponds to the execution context of the global code.

A relationship with the script's global object

An object that provides the APIs that the code can use.

This is typically a Window object. In JavaScript, this corresponds to the global object.

When a script's global object is an empty object, it can't do anything that interacts with the environment.

If the script's global object is a Window object, then in JavaScript, the ThisBinding of the global execution context for this script must be the Window object's WindowProxy object, rather than the global object. [ECMA262]

This is a willful violation of the JavaScript specification current at the time of writing (ECMAScript edition 5, as defined in section 10.4.1.1 Initial Global Execution Context, step 3). The JavaScript specification requires that the this keyword in the global scope return the global object, but this is not compatible with the security design prevalent in implementations as specified herein. [ECMA262]

A relationship with the script's browsing context

A browsing context that is assigned responsibility for actions taken by the script.

When a script creates and navigates a new top-level browsing context, the opener attribute of the new browsing context's Window object will be set to the script's browsing context's WindowProxy object.

A relationship with the script's document

A Document that is assigned responsibility for actions taken by the script.

When a script fetches a resource, the current address of the script's document will be used to set the Referer (sic) header.

A URL character encoding

A character encoding, set when the script is created, used to encode URLs. If the character encoding is set from another source, e.g. a document's character encoding, then the script's URL character encoding must follow the source, so that if the source's changes, so does the script's.

A base URL

A URL, set when the script is created, used to resolve relative URLs. If the base URL is set from another source, e.g. a document base URL, then the script's base URL must follow the source, so that if the source's changes, so does the script's.

6.1.3.2 Calling scripts

When a user agent is to jump to a code entry-point for a script, for example to invoke an event listener defined in that script, the user agent must run the following steps:

  1. If the script's global object is a Window object whose Document object is not fully active, then abort these steps without doing anything. The callback is not fired.

  2. Set the entry script to be the script being invoked.

  3. Make the script execution environment for the script execute the code for the given code entry-point.

  4. Set the entry script back to whatever it was when this algorithm started.

This algorithm is not invoked by one script calling another.

6.1.3.3 Creating scripts

When the specification says that a script is to be created, given some script source, its scripting language, a global object, a browsing context, a URL character encoding, and a base URL, the user agent must run the following steps:

  1. If scripting is disabled for browsing context passed to this algorithm, then abort these steps, as if the script did nothing but return void.

  2. Set up a script execution environment as appropriate for the scripting language.

  3. Parse/compile/initialize the source of the script using the script execution environment, as appropriate for the scripting language, and thus obtain the list of code entry-points for the script. If the semantics of the scripting language and the given source code are such that there is executable code to be immediately run, then the initial code entry-point is the entry-point for that code.

  4. Set up the script's global object, the script's browsing context, the script's document, the script's URL character encoding, and the script's base URL from the settings passed to this algorithm.

  5. If all the steps above succeeded (in particular, if the script was compiled successfully), Jump to the script's initial code entry-point.

    Otherwise, report the error using the onerror event handler of the script's global object. If the error is still not handled after this, then the error may be reported to the user.


When the user agent is to create an impotent script, given some script source, its scripting language, and a browsing context, the user agent must create a script, using the given script source and scripting language, using a new empty object as the global object, and using the given browsing context as the browsing context. The URL character encoding and base URL for the resulting script are not important as no APIs are exposed to the script.


When the specification says that a script is to be created from a node node, given some script source and its scripting language, the user agent must create a script, using the given script source and scripting language, and using the script settings determined from the node node.

The script settings determined from the node node are computed as follows:

  1. Let document be the Document of node (or node itself if it is a Document).

  2. The browsing context is the browsing context of document.

  3. The global object is the Window object of document.

  4. The URL character encoding is the character encoding of document. (This is a reference, not a copy.)

  5. The base URL is the base URL of document. (This is a reference, not a copy.)

6.1.3.4 Killing scripts

User agents may impose resource limitations on scripts, for example CPU quotas, memory limits, total execution time limits, or bandwidth limitations. When a script exceeds a limit, the user agent may either throw a QUOTA_EXCEEDED_ERR exception, abort the script without an exception, prompt the user, or throttle script execution.

For example, the following script never terminates. A user agent could, after waiting for a few seconds, prompt the user to either terminate the script or let it continue.

<script>
 while (true) { /* loop */ }
</script>

User agents are encouraged to allow users to disable scripting whenever the user is prompted either by a script (e.g. using the window.alert() API) or because of a script's actions (e.g. because it has exceeded a time limit).

If scripting is disabled while a script is executing, the script should be terminated immediately.

6.1.3.5 Runtime script errors

Whenever an uncaught runtime script error occurs in one of the scripts associated with a Document, the user agent must report the error using the onerror event handler of the script's global object. If the error is still not handled after this, then the error may be reported to the user.


When the user agent is required to report an error error using the event handler onerror, it must run these steps, after which the error is either handled or not handled:

If the value of onerror is a Function

The function must be invoked with three arguments. The three arguments passed to the function are all DOMStrings; the first must give the message that the UA is considering reporting, the second must give the absolute URL of the resource in which the error occurred, and the third must give the line number in that resource on which the error occurred.

If the function returns false, then the error is handled. Otherwise, the error is not handled.

Any uncaught exceptions thrown or errors caused by this function may be reported to the user immediately after the error that the function was called for; the report an error algorithm must not be used to handle exceptions thrown or errors caused by this function.

Otherwise

The error is not handled.

6.1.4 Event loops

6.1.4.1 Definitions

To coordinate events, user interaction, scripts, rendering, networking, and so forth, user agents must use event loops as described in this section.

There must be at least one event loop per user agent, and at most one event loop per unit of related similar-origin browsing contexts.

When there is more than one event loop for a unit of related browsing contexts, complications arise when a browsing context in that group is navigated such that it switches from one unit of related similar-origin browsing contexts to another. This specification does not currently describe how to handle these complications.

An event loop always has at least one browsing context. If an event loop's browsing contexts all go away, then the event loop goes away as well. A browsing context always has an event loop coordinating its activities.

Other specifications can define new kinds of event loops that aren't associated with browsing contexts; in particular, the Web Workers specification does so.

An event loop has one or more task queues. A task queue is an ordered list of tasks, which can be:

Events

Asynchronously dispatching an Event object at a particular EventTarget object is a task.

Not all events are dispatched using the task queue, many are dispatched synchronously during other tasks.

Parsing

The HTML parser tokenizing one or more bytes, and then processing any resulting tokens, is typically a task.

Callbacks

Calling a callback asynchronously is a task.

Using a resource

When an algorithm fetches a resource, if the fetching occurs asynchronously then the processing of the resource once some or all of the resource is available is a task.

Reacting to DOM manipulation

Some elements have tasks that trigger in response to DOM manipulation, e.g. when that element is inserted into the document.

When a user agent is to queue a task, it must add the given task to one of the task queues of the relevant event loop. All the tasks from one particular task source (e.g. the callbacks generated by timers, the events dispatched for mouse movements, the tasks queued for the parser) must always be added to the same task queue, but tasks from different task sources may be placed in different task queues.

For example, a user agent could have one task queue for mouse and key events (the user interaction task source), and another for everything else. The user agent could then give keyboard and mouse events preference over other tasks three quarters of the time, keeping the interface responsive but not starving other task queues, and never processing events from any one task source out of order.

Each task that is queued onto a task queue of an event loop defined by this specification is associated with a Document; if the task was queued in the context of an element, then it is the element's Document; if the task was queued in the context of a browsing context, then it is the browsing context's active document at the time the task was queued; if the task was queued by or for a script then the document is the script's document.

A user agent is required to have one storage mutex. This mutex is used to control access to shared state like cookies. At any one point, the storage mutex is either free, or owned by a particular event loop or instance of the fetching algorithm.

Whenever a script calls into a plugin, and whenever a plugin calls into a script, the user agent must release the storage mutex.

6.1.4.2 Processing model

An event loop must continually run through the following steps for as long as it exists:

  1. Run the oldest task on one of the event loop's task queues, ignoring tasks whose associated Documents are not fully active. The user agent may pick any task queue.

  2. If the storage mutex is now owned by the event loop, release it so that it is once again free.

  3. Remove that task from its task queue.

  4. Provide a stable state.

  5. If necessary, update the rendering or user interface of any Document or browsing context to reflect the current state.

  6. Return to the first step of the event loop.


When the user agent is to provide a stable state, if any asynchronously-running algorithms are awaiting a stable state, then the user agent must run their synchronous section and then resume running their asynchronous algorithm (if appropriate).

A synchronous section never mutates the DOM, runs any script, or has any other side-effects.

Steps in synchronous sections are marked with ⌛.


When an algorithm says to spin the event loop until a condition goal is met, the user agent must run the following steps:

  1. Let task source be the task source of the currently running task.

  2. Stop the currently running task, allowing the event loop to resume, but continue these steps asynchronously.

    This causes the event loop to move on to the second step of its processing model (defined above).

  3. Wait until the condition goal is met.

  4. Queue a task to continue running these steps, using the task source task source. Wait until this task runs before continuing these steps.

  5. Return to the caller.


Some of the algorithms in this specification, for historical reasons, require the user agent to pause while running a task until a condition goal is met. This means running the following steps:

  1. If any asynchronously-running algorithms are awaiting a stable state, then run their synchronous section and then resume running their asynchronous algorithm. (See the event loop processing model definition above for details.)

  2. If necessary, update the rendering or user interface of any Document or browsing context to reflect the current state.

  3. Wait until the condition goal is met. While a user agent has a paused task, the corresponding event loop must not run further tasks, and any script in the currently running task must block. User agents should remain responsive to user input while paused, however, albeit in a reduced capacity since the event loop will not be doing anything.


When a user agent is to obtain the storage mutex as part of running a task, it must run through the following steps:

  1. If the storage mutex is already owned by this task's event loop, then abort these steps.

  2. Otherwise, pause until the storage mutex can be taken by the event loop.

  3. Take ownership of the storage mutex.

6.1.4.3 Generic task sources

The following task sources are used by a number of mostly unrelated features in this and other specifications.

The DOM manipulation task source

This task source is used for features that react to DOM manipulations, such as things that happen asynchronously when an element is inserted into the document.

The user interaction task source

This task source is used for features that react to user interaction, for example keyboard or mouse input.

Asynchronous events sent in response to user input (e.g. click events) must be dispatched using tasks queued with the user interaction task source. [DOMEVENTS]

The networking task source

This task source is used for features that trigger in response to network activity.

The history traversal task source

This task source is used to queue calls to history.back() and similar APIs.

6.1.5 The javascript: URL scheme

When a URL using the javascript: scheme is dereferenced, the user agent must run the following steps:

  1. Let the script source be the string obtained using the content retrieval operation defined for javascript: URLs. [JSURL]

  2. Use the appropriate step from the following list:

    If a browsing context is being navigated to a javascript: URL, and the source browsing context for that navigation, if any, has scripting disabled

    Let result be void.

    If a browsing context is being navigated to a javascript: URL, and the active document of that browsing context has the same origin as the script given by that URL

    Let address be the address of the active document of the browsing context being navigated.

    If address is about:blank, and the browsing context being navigated has a creator browsing context, then let address be the address of the creator Document instead.

    Create a script from the Document node of the active document, using the aforementioned script source, and assuming the scripting language is JavaScript.

    Let result be the return value of the initial code entry-point of this script. If an exception was raised, let result be void instead. (The result will be void also if scripting is disabled.)

    When it comes time to set the document's address in the navigation algorithm, use address as the override URL.

    Otherwise

    Let result be void.

  3. If the result of executing the script is void (there is no return value), then the URL must be treated in a manner equivalent to an HTTP resource with an HTTP 204 No Content response.

    Otherwise, the URL must be treated in a manner equivalent to an HTTP resource with a 200 OK response whose Content-Type metadata is text/html and whose response body is the return value converted to a string value.

    Certain contexts, in particular img elements, ignore the Content-Type metadata.

So for example a javascript: URL for a src attribute of an img element would be evaluated in the context of an empty object as soon as the attribute is set; it would then be sniffed to determine the image type and decoded as an image.

A javascript: URL in an href attribute of an a element would only be evaluated when the link was followed.

The src attribute of an iframe element would be evaluated in the context of the iframe's own browsing context; once evaluated, its return value (if it was not void) would replace that browsing context's document, thus changing the variables visible in that browsing context.

6.1.6 Events

6.1.6.1 Event handlers

Many objects can have event handlers specified. These act as bubbling event listeners for the object on which they are specified.

An event handler can either have the value null or be set to a Function object. Initially, event handlers must be set to null.

Event handlers are exposed in one or two ways.

The first way, common to all event handlers, is as an event handler IDL attribute.

The second way is as an event handler content attribute. Event handlers on HTML elements and some of the event handlers on Window objects are exposed in this way.


Event handler IDL attributes, on setting, must set the corresponding event handler to their new value, and on getting, must return whatever the current value of the corresponding event handler is (possibly null).

If an event handler IDL attribute exposes an event handler of an object that doesn't exist, it must always return null on getting and must do nothing on setting.

This can happen in particular for event handler IDL attribute on body elements that do not have corresponding Window objects.

Certain event handler IDL attributes have additional requirements, in particular the onmessage attribute of MessagePort objects.


Event handler content attributes, when specified, must contain valid JavaScript code which, when parsed, would match the FunctionBody production after automatic semicolon insertion. [ECMA262]

When an event handler content attribute is set, if the element is owned by a Document that is in a browsing context, and scripting is enabled for that browsing context, the user agent must run the following steps to create a script after setting the content attribute to its new value:

  1. Set the corresponding event handler to null.

  2. Set up a script execution environment for JavaScript.

  3. Let body be the event handler content attribute's new value.

  4. If body is not parsable as FunctionBody or if parsing detects an early error then abort these steps.

    FunctionBody is defined in ECMAScript edition 5 section 13 Function Definition. Early error is defined in ECMAScript edition 5 section 16 Errors. [ECMA262]

  5. If body begins with a Directive Prologue that contains a Use Strict Directive then let strict be true, otherwise let strict be false.

    The terms "Directive Prologue" and "Use Strict Directive" are defined in ECMAScript edition 5 section 14.1 Directive Prologues and the Use Strict Directive. [ECMA262]

  6. Using the script execution environment created above, create a function object (as defined in ECMAScript edition 5 section 13.2 Creating Function Objects), with:

    Parameter list FormalParameterList
    If the attribute is the onerror attribute of the Window object
    Let the function have three arguments, named event, source, and fileno.
    Otherwise
    Let the function have a single argument called event.
    Function body FunctionBody
    The result of parsing body above.
    Lexical Environment Scope
    1. Let Scope be the result of NewObjectEnvironment(the element's Document, the global environment).
    2. If the element has a form owner, let Scope be the result of NewObjectEnvironment(the element's form owner, Scope).
    3. Let Scope be the result of NewObjectEnvironment(the element's object, Scope).

    NewObjectEnvironment() is defined in ECMAScript edition 5 section 10.2.2.3 NewObjectEnvironment (O, E). [ECMA262]

    Boolean flag Strict
    The value of strict.

    Let this new function be the only entry in the script's list of code entry-points.

  7. Set up the script's global object, the script's browsing context, the script's document, the script's URL character encoding, and the script's base URL from the script settings determined from the node on which the attribute is being set.

  8. Set the corresponding event handler to the aforementioned function.

When an event handler content attribute is removed, the user agent must set the corresponding event handler to null.

When an event handler content attribute is set on an element owned by a Document that is not in a browsing context, the corresponding event handler is not changed.


All event handlers on an object, whether an element or some other object, and whether set to null or to a Function object, must be registered as event listeners on the object when it is created, as if the addEventListener() method on the object's EventTarget interface had been invoked, with the event type (type argument) equal to the type corresponding to the event handler (the event handler event type), the listener set to be a target and bubbling phase listener (useCapture argument set to false), and the event listener itself (listener argument) set to do nothing while the event handler's value is not a Function object, and set to invoke the call() callback of the Function object associated with the event handler otherwise.

Event handlers therefore always fire before event listeners attached using addEventListener().

The listener argument is emphatically not the event handler itself.

The interfaces implemented by the event object do not influence whether an event handler is triggered or not.

When an event handler's Function object is invoked, its call() callback must be invoked with one argument, set to the Event object of the event in question.

The handler's return value must then be processed as follows:

If the event type is mouseover

If the return value is a boolean with the value true, then the event must be canceled.

If the event object is a BeforeUnloadEvent object

If the return value is a string, and the event object's returnValue attribute's value is the empty string, then set the returnValue attribute's value to the return value.

Otherwise

If the return value is a boolean with the value false, then the event must be canceled.


The Function interface represents a function in the scripting language being used. It is represented in IDL as follows:

[Callback=FunctionOnly, NoInterfaceObject]
interface Function {
  any call(in any... arguments);
};

The call(...) method is the object's callback.

In JavaScript, any Function object implements this interface.

If the Function object is a JavaScript Function, then when it is invoked by the user agent, the user agent must set the thisArg (as defined by ECMAScript edition 5 section 10.4.3 Entering Function Code) to the event handler's object. [ECMA262]

For example, the following document fragment:

<body onload="alert(this)" onclick="alert(this)">

...leads to an alert saying "[object Window]" when the document is loaded, and an alert saying "[object HTMLBodyElement]" whenever the user clicks something in the page.

The return value of the function affects whether the event is canceled or not: as described above, if the return value is false, the event is canceled (except for mouseover events, where the return value has to be true to cancel the event). With beforeunload events, the value is instead used to determine the message to show the user.

6.1.6.2 Event handlers on elements, Document objects, and Window objects

The following are the event handlers (and their corresponding event handler event types) that must be supported by all HTML elements, as both content attributes and IDL attributes, and on Document and Window objects, as IDL attributes.

Event handler Event handler event type
onabort abort
oncanplay canplay
oncanplaythrough canplaythrough
onchange change
onclick click
oncontextmenu contextmenu
oncuechange cuechange
ondblclick dblclick
ondrag drag
ondragend dragend
ondragenter dragenter
ondragleave dragleave
ondragover dragover
ondragstart dragstart
ondrop drop
ondurationchange durationchange
onemptied emptied
onended ended
oninput input
oninvalid invalid
onkeydown keydown
onkeypress keypress
onkeyup keyup
onloadeddata loadeddata
onloadedmetadata loadedmetadata
onloadstart loadstart
onmousedown mousedown
onmousemove mousemove
onmouseout mouseout
onmouseover mouseover
onmouseup mouseup
onmousewheel mousewheel
onpause pause
onplay play
onplaying playing
onprogress progress
onratechange ratechange
onreadystatechange readystatechange
onreset reset
onseeked seeked
onseeking seeking
onselect select
onshow show
onstalled stalled
onsubmit submit
onsuspend suspend
ontimeupdate timeupdate
onvolumechange volumechange
onwaiting waiting

The following are the event handlers (and their corresponding event handler event types) that must be supported by all HTML elements other than body, as both content attributes and IDL attributes, and on Document objects, as IDL attributes:

Event handler Event handler event type
onblur blur
onerror error
onfocus focus
onload load
onscroll scroll

The following are the event handlers (and their corresponding event handler event types) that must be supported by Window objects, as IDL attributes on the Window object, and with corresponding content attributes and IDL attributes exposed on the body and frameset elements:

Event handler Event handler event type
onafterprint afterprint
onbeforeprint beforeprint
onbeforeunload beforeunload
onblur blur
onerror error
onfocus focus
onhashchange hashchange
onload load
onmessage message
onoffline offline
ononline online
onpagehide pagehide
onpageshow pageshow
onpopstate popstate
onredo redo
onresize resize
onscroll scroll
onstorage storage
onundo undo
onunload unload

The onerror handler is also used for reporting script errors.

6.1.6.3 Event firing

Certain operations and methods are defined as firing events on elements. For example, the click() method on the HTMLElement interface is defined as firing a click event on the element. [DOMEVENTS]

Firing a simple event named e means that an event with the name e, which does not bubble (except where otherwise stated) and is not cancelable (except where otherwise stated), and which uses the Event interface, must be dispatched at the given target.

Firing a synthetic mouse event named e means that an event with the name e, which does not bubble (except where otherwise stated) and is not cancelable (except where otherwise stated), and which uses the MouseEvent interface, must be dispatched at the given target. The event object must have its screenX, screenY, clientX, clientY, and button attributes set to 0, its ctrlKey, shiftKey, altKey, and metaKey attributes set according to the current state of the key input device, if any (false for any keys that are not available), its detail attribute set to 1, and its relatedTarget attribute set to null. The getModifierState() method on the object must return values appropriately describing the state of the key input device at the time the event is created.

Firing a click event means firing a synthetic mouse event named click, which bubbles and is cancelable.

The default action of these events is to do nothing except where otherwise stated.

6.1.6.4 Events and the Window object

When an event is dispatched at a DOM node in a Document in a browsing context, if the event is not a load event, the user agent must also dispatch the event to the Window, as follows:

  1. In the capture phase, the event must propagate to the Window object before propagating to any of the nodes, as if the Window object was the parent of the Document in the dispatch chain.
  2. In the bubble phase, the event must propagate up to the Window object at the end of the phase, unless bubbling has been prevented, again as if the Window object was the parent of the Document in the dispatch chain.

6.2 Base64 utility methods

The atob() and btoa() methods allow authors to transform content to and from the base64 encoding.

[Supplemental, NoInterfaceObject]
interface WindowBase64 {
  DOMString btoa(in DOMString btoa);
  DOMString atob(in DOMString atob);
};
Window implements WindowBase64;

In these APIs, for mnemonic purposes, the "b" can be considered to stand for "binary", and the "a" for "ASCII". In practice, though, for primarily historical reasons, both the input and output of these functions are Unicode strings.

result = window . btoa( data )

Takes the input data, in the form of a Unicode string containing only characters in the range U+0000 to U+00FF, each representing a binary byte with values 0x00 to 0xFF respectively, and converts it to its base64 representation, which it returns.

Throws an INVALID_CHARACTER_ERR exception if the input string contains any out-of-range characters.

result = window . atob( data )

Takes the input data, in the form of a Unicode string containing base64-encoded binary data, decodes it, and returns a string consisting of characters in the range U+0000 to U+00FF, each representing a binary byte with values 0x00 to 0xFF respectively, corresponding to that binary data.

Throws an INVALID_CHARACTER_ERR exception if the input string is not valid base64 data.

The WindowBase64 interface adds to the Window interface and the WorkerUtils interface (part of Web Workers).

The btoa() method must throw an INVALID_CHARACTER_ERR exception if the method's first argument contains any character whose code point is greater than U+00FF. Otherwise, the user agent must convert that argument to a sequence of octets whose nth octet is the eight-bit representation of the code point of the nth character of the argument, and then must apply the base64 algorithm to that sequence of octets, and return the result. [RFC4648]

The atob() method must run the following steps to parse the string passed in the method's first argument:

  1. Let input be the string being parsed.

  2. Let position be a pointer into input, initially pointing at the start of the string.

  3. If the length of input divides by 4 leaving no remainder, then: if input ends with one or two U+003D EQUALS SIGN (=) characters, remove them from input.

  4. If the length of input divides by 4 leaving a remainder of 1, throw an INVALID_CHARACTER_ERR exception and abort these steps.

  5. If input contains a character that is not in the following list of characters and character ranges, throw an INVALID_CHARACTER_ERR exception and abort these steps:

    • U+002B PLUS SIGN (+)
    • U+002F SOLIDUS (/)
    • U+0030 DIGIT ZERO (0) to U+0039 DIGIT NINE (9)
    • U+0041 LATIN CAPITAL LETTER A to U+005A LATIN CAPITAL LETTER Z
    • U+0061 LATIN SMALL LETTER A to U+007A LATIN SMALL LETTER Z
  6. Let output be a string, initially empty.

  7. Let buffer be a buffer that can have bits appended to it, initially empty.

  8. While position does not point past the end of input, run these substeps:

    1. Find the character pointed to by position in the first column of the following table. Let n be the number given in the second cell of the same row.

      Character Number
      A0
      B1
      C2
      D3
      E4
      F5
      G6
      H7
      I8
      J9
      K10
      L11
      M12
      N13
      O14
      P15
      Q16
      R17
      S18
      T19
      U20
      V21
      W22
      X23
      Y24
      Z25
      a26
      b27
      c28
      d29
      e30
      f31
      g32
      h33
      i34
      j35
      k36
      l37
      m38
      n39
      o40
      p41
      q42
      r43
      s44
      t45
      u46
      v47
      w48
      x49
      y50
      z51
      052
      153
      254
      355
      456
      557
      658
      759
      860
      961
      +62
      /63
    2. Append to buffer the six bits corresponding to number, most significant bit first.

    3. If buffer has accumulated 24 bits, interpret them as three 8-bit big-endian numbers. Append the three characters with code points equal to those numbers to output, in the same order, and then empty buffer.

    4. Advance position by one character.

  9. If buffer is not empty, it contains either 12 or 18 bits. If it contains 12 bits, discard the last four and interpret the remaining eight as an 8-bit big-endian number. If it contains 18 bits, discard the last two and interpret the remaining 16 as two 8-bit big-endian numbers. Append the one or two characters with code points equal to those one or two numbers to output, in the same order.

    The discarded bits mean that, for instance, atob("YQ") and atob("YR") both return "a".

  10. Return output.

Some base64 encoders add newlines or other whitespace to their output. The atob() method throws an exception if its input contains characters other than those described by the regular expression bracket expression [+/=0-9A-Za-z], so other characters need to be removed before atob() is used for decoding.