6.5 System state and capabilities

6.5.1 The Navigator object

The navigator attribute of the Window interface must return an instance of the Navigator interface, which represents the identity and state of the user agent (the client), and allows Web pages to register themselves as potential protocol and content handlers:

interface Navigator {
  // objects implementing this interface also implement the interfaces given below
};
Navigator implements NavigatorID;
Navigator implements NavigatorOnLine;
Navigator implements NavigatorContentUtils;
Navigator implements NavigatorStorageUtils;

These interfaces are defined separately so that other specifications can re-use parts of the Navigator interface.

6.5.1.1 Client identification
[NoInterfaceObject]
interface NavigatorID {
  readonly attribute DOMString appName;
  readonly attribute DOMString appVersion;
  readonly attribute DOMString platform;
  readonly attribute DOMString userAgent;
};

In certain cases, despite the best efforts of the entire industry, Web browsers have bugs and limitations that Web authors are forced to work around.

This section defines a collection of attributes that can be used to determine, from script, the kind of user agent in use, in order to work around these issues.

Client detection should always be limited to detecting known current versions; future versions and unknown versions should always be assumed to be fully compliant.

window . navigator . appName

Returns the name of the browser.

window . navigator . appVersion

Returns the version of the browser.

window . navigator . platform

Returns the name of the platform.

window . navigator . userAgent

Returns the complete User-Agent header.

appName

Must return either the string "Netscape" or the full name of the browser, e.g. "Mellblom Browsernator".

appVersion

Must return either the string "4.0" or a string representing the version of the browser in detail, e.g. "1.0 (VMS; en-US) Mellblomenator/9000".

platform

Must return either the empty string or a string representing the platform on which the browser is executing, e.g. "MacIntel", "Win32", "FreeBSD i386", "WebTV OS".

userAgent

Must return the string used for the value of the "User-Agent" header in HTTP requests, or the empty string if no such header is ever sent.

Any information in this API that varies from user to user can be used to profile the user. In fact, if enough such information is available, a user can actually be uniquely identified. For this reason, user agent implementors are strongly urged to include as little information in this API as possible.

6.5.1.2 Custom scheme and content handlers
[NoInterfaceObject]
interface NavigatorContentUtils {
  // content handler registration
  void registerProtocolHandler(DOMString scheme, DOMString url, DOMString title);
  void registerContentHandler(DOMString mimeType, DOMString url, DOMString title);
  DOMString isProtocolHandlerRegistered(DOMString scheme, DOMString url);
  DOMString isContentHandlerRegistered(DOMString mimeType, DOMString url);
  void unregisterProtocolHandler(DOMString scheme, DOMString url);
  void unregisterContentHandler(DOMString mimeType, DOMString url);
};

The registerProtocolHandler() method allows Web sites to register themselves as possible handlers for particular schemes. For example, an online telephone messaging service could register itself as a handler of the sms: scheme, so that if the user clicks on such a link, he is given the opportunity to use that Web site. Analogously, the registerContentHandler() method allows Web sites to register themselves as possible handlers for content in a particular MIME type. For example, the same online telephone messaging service could register itself as a handler for text/vcard files, so that if the user has no native application capable of handling vCards, his Web browser can instead suggest he use that site to view contact information stored on vCards that he opens. [RFC5724] RFC6350

window . navigator . registerProtocolHandler(scheme, url, title)
window . navigator . registerContentHandler(mimeType, url, title)

Registers a handler for the given scheme or content type, at the given URL, with the given title.

The string "%s" in the URL is used as a placeholder for where to put the URL of the content to be handled.

Throws a SecurityError exception if the user agent blocks the registration (this might happen if trying to register as a handler for "http", for instance).

Throws a SyntaxError if the "%s" string is missing in the URL.

User agents may, within the constraints described in this section, do whatever they like when the methods are called. A UA could, for instance, prompt the user and offer the user the opportunity to add the site to a shortlist of handlers, or make the handlers his default, or cancel the request. UAs could provide such a UI through modal UI or through a non-modal transient notification interface. UAs could also simply silently collect the information, providing it only when relevant to the user.

User agents should keep track of which sites have registered handlers (even if the user has declined such registrations) so that the user is not repeatedly prompted with the same request.

The arguments to the methods have the following meanings and corresponding implementation requirements. The requirements that involve throwing exceptions must be processed in the order given below, stopping at the first exception thrown. (So the exceptions for the first argument take precedence over the exceptions for the second argument.)

scheme (registerProtocolHandler() only)

A scheme, such as mailto or web+auth. The scheme must be compared in an ASCII case-insensitive manner by user agents for the purposes of comparing with the scheme part of URLs that they consider against the list of registered handlers.

The scheme value, if it contains a colon (as in "mailto:"), will never match anything, since schemes don't contain colons.

If the registerProtocolHandler() method is invoked with a scheme that is neither a whitelisted scheme nor a scheme whose value starts with the substring "web+" and otherwise contains only characters in the range lowercase ASCII letters, the user agent must throw a SecurityError exception.

The following schemes are the whitelisted schemes:

  • irc
  • mailto
  • mms
  • news
  • nntp
  • sms
  • smsto
  • tel
  • urn
  • webcal
  • xmpp

This list can be changed. If there are schemes that should be added, please send feedback.

mimeType (registerContentHandler() only)

A MIME type, such as model/vnd.flatland.3dml or application/vnd.google-earth.kml+xml. The MIME type must be compared in an ASCII case-insensitive manner by user agents for the purposes of comparing with MIME types of documents that they consider against the list of registered handlers.

User agents must compare the given values only to the MIME type/subtype parts of content types, not to the complete type including parameters. Thus, if mimeType values passed to this method include characters such as commas or whitespace, or include MIME parameters, then the handler being registered will never be used.

The type is compared to the MIME type used by the user agent after the sniffing algorithms have been applied.

If the registerContentHandler() method is invoked with a MIME type that is in the type blacklist or that the user agent has deemed a privileged type, the user agent must throw a SecurityError exception.

The following MIME types are in the type blacklist:

This list can be changed. If there are schemes that should be added, please send feedback.

url

A string used to build the URL of the page that will handle the requests.

User agents must throw a SyntaxError exception if the url argument passed to one of these methods does not contain the exact literal string "%s".

User agents must throw a SyntaxError if resolving the url argument relative to the entry script's base URL, is not successful.

The resulting absolute URL would by definition not be a valid URL as it would include the string "%s" which is not a valid component in a URL.

User agents must throw a SecurityError exception if the resulting absolute URL has an origin that differs from the origin of the entry script.

This is forcibly the case if the %s placeholder is in the scheme, host, or port parts of the URL.

The resulting absolute URL is the proto-URL. It identifies the handler for the purposes of the methods described below.

When the user agent uses this handler, it must replace the first occurrence of the exact literal string "%s" in the url argument with an escaped version of the absolute URL of the content in question (as defined below), then resolve the resulting URL, relative to the base URL of the entry script at the time the registerContentHandler() or registerProtocolHandler() methods were invoked, and then navigate an appropriate browsing context to the resulting URL using the GET method (or equivalent for non-HTTP URLs).

To get the escaped version of the absolute URL of the content in question, the user agent must replace every character in that absolute URL that doesn't match the <query> production defined in RFC 3986 by the percent-encoded form of that character. [RFC3986]

If the user had visited a site at http://example.com/ that made the following call:

navigator.registerContentHandler('application/x-soup', 'soup?url=%s', 'SoupWeb™')

...and then, much later, while visiting http://www.example.net/, clicked on a link such as:

<a href="chickenkïwi.soup">Download our Chicken Kïwi soup!</a>

...then, assuming this chickenkïwi.soup file was served with the MIME type application/x-soup, the UA might navigate to the following URL:

http://example.com/soup?url=http://www.example.net/chickenk%C3%AFwi.soup

This site could then fetch the chickenkïwi.soup file and do whatever it is that it does with soup (synthesize it and ship it to the user, or whatever).

title

A descriptive title of the handler, which the UA might use to remind the user what the site in question is.

This section does not define how the pages registered by these methods are used, beyond the requirements on how to process the url value (see above). To some extent, the processing model for navigating across documents defines some cases where these methods are relevant, but in general UAs may use this information wherever they would otherwise consider handing content to native plugins or helper applications.

UAs must not use registered content handlers to handle content that was returned as part of a non-GET transaction (or rather, as part of any non-idempotent transaction), as the remote site would not be able to fetch the same data.


In addition to the registration methods, there are also methods for determining if particular handlers have been registered, and for unregistering handlers.

state = window . navigator . isProtocolHandlerRegistered(scheme, url)
state = window . navigator . isContentHandlerRegistered(mimeType, url)

Returns one of the following strings describing the state of the handler given by the arguments:

new
Indicates that no attempt has been made to register the given handler (or that the handler has been unregistered). It would be appropriate to promote the availability of the handler or to just automatically register the handler.
registered
Indicates that the given handler has been registered or that the site is blocked from registering the handler. Trying to register the handler again would have no effect.
declined
Indicates that the given handler has been offered but was rejected. Trying to register the handler again may prompt the user again.
state = window . navigator . unregisterProtocolHandler(scheme, url)
state = window . navigator . unregisterContentHandler(mimeType, url)

Unregisters the handler given by the arguments.

The isProtocolHandlerRegistered() method must return the handler state string that most closely describes the current state of the handler described by the two arguments to the method, where the first argument gives the scheme and the second gives the string used to build the URL of the page that will handle the requests.

The first argument must be compared to the schemes for which custom protocol handlers are registered in an ASCII case-insensitive manner to find the relevant handlers.

The second argument must be preprocessed as described below, and if that is successful, must then be matched against the proto-URLs of the relevant handlers to find the described handler.


The isContentHandlerRegistered() method must return the handler state string that most closely describes the current state of the handler described by the two arguments to the method, where the first argument gives the MIME type and the second gives the string used to build the URL of the page that will handle the requests.

The first argument must be compared to the MIME types for which custom content handlers are registered in an ASCII case-insensitive manner to find the relevant handlers.

The second argument must be preprocessed as described below, and if that is successful, must then be matched against the proto-URLs of the relevant handlers to find the described handler.


The handler state strings are the following strings. Each string describes several situations, as given by the following list.

new
The described handler has never been registered for the given scheme or type.
The described handler was once registered for the given scheme or type, but the site has since unregistered it. If the handler were to be reregistered, the user would be notified accordingly.
The described handler was once registered for the given scheme or type, but the site has since unregistered it, but the user has indicated that the site is to be blocked from registering the type again, so the user agent would ignore further registration attempts.
registered
An attempt was made to register the described handler for the given scheme or type, but the user has not yet been notified, and the user agent would ignore further registration attempts. (Maybe the user agent batches registration requests to display them when the user requests to be notified about them, and the user has not yet requested that the user agent notify it of the previous registration attempt.)
The described handler is registered for the given scheme or type (maybe, or maybe not, as the default handler).
The described handler is permanently blocked from being (re)registered. (Maybe the user marked the registration attempt as spam, or blocked the site for other reasons.)
declined
An attempt was made to register the described handler for the given scheme or type, but the user has not yet been notified; however, the user might be notified if another registration attempt were to be made. (Maybe the last registration attempt was made while the page was in the background and the user closed the page without looking at it, and the user agent requires confirmation for this registration attempt.)
An attempt was made to register the described handler for the given scheme or type, but the user has not yet responded.
An attempt was made to register the described handler for the given scheme or type, but the user declined the offer. The user has not indicated that the handler is to be permanently blocked, however, so another attempt to register the described handler might result in the user being prompted again.
The described handler was once registered for the given scheme or type, but the user has since removed it. The user has not indicated that the handler is to be permanently blocked, however, so another attempt to register the described handler might result in the user being prompted again.

The unregisterProtocolHandler() method must unregister the handler described by the two arguments to the method, where the first argument gives the scheme and the second gives the string used to build the URL of the page that will handle the requests.

The first argument must be compared to the schemes for which custom protocol handlers are registered in an ASCII case-insensitive manner to find the relevant handlers.

The second argument must be preprocessed as described below, and if that is successful, must then be matched against the proto-URLs of the relevant handlers to find the described handler.


The unregisterContentHandler() method must unregister the handler described by the two arguments to the method, where the first argument gives the MIME type and the second gives the string used to build the URL of the page that will handle the requests.

The first argument must be compared to the MIME types for which custom content handlers are registered in an ASCII case-insensitive manner to find the relevant handlers.

The second argument must be preprocessed as described below, and if that is successful, must then be matched against the proto-URLs of the relevant handlers to find the described handler.


The second argument of the four methods described above must be preprocessed as follows:

  1. If the string does not contain the substring "%s", abort these steps. There's no matching handler.

  2. Resolve the string relative to the base URL of the entry script.

  3. If this fails, then throw a SyntaxError exception, aborting the method.

  4. If the resoluting absolute URL's origin is not the same origin as that of the entry script throw a SecurityError exception, aborting the method.

  5. Return the resulting absolute URL as the result of preprocessing the argument.

6.5.1.3 Security and privacy

These mechanisms can introduce a number of concerns, in particular privacy concerns.

Hijacking all Web usage. User agents should not allow schemes that are key to its normal operation, such as http or https, to be rerouted through third-party sites. This would allow a user's activities to be trivially tracked, and would allow user information, even in secure connections, to be collected.

Hijacking defaults. It is strongly recommended that user agents do not automatically change any defaults, as this could lead the user to send data to remote hosts that the user is not expecting. New handlers registering themselves should never automatically cause those sites to be used.

Registration spamming. User agents should consider the possibility that a site will attempt to register a large number of handlers, possibly from multiple domains (e.g. by redirecting through a series of pages each on a different domain, and each registering a handler for video/mpeg — analogous practices abusing other Web browser features have been used by pornography Web sites for many years). User agents should gracefully handle such hostile attempts, protecting the user.

Misleading titles. User agents should not rely wholly on the title argument to the methods when presenting the registered handlers to the user, since sites could easily lie. For example, a site hostile.example.net could claim that it was registering the "Cuddly Bear Happy Content Handler". User agents should therefore use the handler's domain in any UI along with any title.

Hostile handler metadata. User agents should protect against typical attacks against strings embedded in their interface, for example ensuring that markup or escape characters in such strings are not executed, that null bytes are properly handled, that over-long strings do not cause crashes or buffer overruns, and so forth.

Leaking Intranet URLs. The mechanism described in this section can result in secret Intranet URLs being leaked, in the following manner:

  1. The user registers a third-party content handler as the default handler for a content type.
  2. The user then browses his corporate Intranet site and accesses a document that uses that content type.
  3. The user agent contacts the third party and hands the third party the URL to the Intranet content.

No actual confidential file data is leaked in this manner, but the URLs themselves could contain confidential information. For example, the URL could be http://www.corp.example.com/upcoming-aquisitions/the-sample-company.egf, which might tell the third party that Example Corporation is intending to merge with The Sample Company. Implementors might wish to consider allowing administrators to disable this feature for certain subdomains, content types, or schemes.

Leaking secure URLs. User agents should not send HTTPS URLs to third-party sites registered as content handlers without the user's informed consent, for the same reason that user agents sometimes avoid sending Referer (sic) HTTP headers from secure sites to third-party sites.

Leaking credentials. User agents must never send username or password information in the URLs that are escaped and included sent to the handler sites. User agents may even avoid attempting to pass to Web-based handlers the URLs of resources that are known to require authentication to access, as such sites would be unable to access the resources in question without prompting the user for credentials themselves (a practice that would require the user to know whether to trust the third-party handler, a decision many users are unable to make or even understand).

6.5.1.4 Sample user interface

This section is non-normative.

A simple implementation of this feature for a desktop Web browser might work as follows.

The registerContentHandler() method could display a modal dialog box:

The modal dialog box could have the title 'Content Handler Registration', and could say 'This Web page: Kittens at work http://kittens.example.org/ ...would like permission to handle files of type: application/x-meowmeow using the following Web-based application: Kittens-at-work displayer http://kittens.example.org/?show=%s Do you trust the administrators of the "kittens.example.org" domain?' with two buttons, 'Trust kittens.example.org' and 'Cancel'.

In this dialog box, "Kittens at work" is the title of the page that invoked the method, "http://kittens.example.org/" is the URL of that page, "application/x-meowmeow" is the string that was passed to the registerContentHandler() method as its first argument (mimeType), "http://kittens.example.org/?show=%s" was the second argument (url), and "Kittens-at-work displayer" was the third argument (title).

If the user clicks the Cancel button, then nothing further happens. If the user clicks the "Trust" button, then the handler is remembered.

When the user then attempts to fetch a URL that uses the "application/x-meowmeow" MIME type, then it might display a dialog as follows:

The dialog box could have the title 'Unknown File Type' and could say 'You have attempted to access:' followed by a URL, followed by a prompt such as 'How would you like FerretBrowser to handle this resource?' with three radio buttons, one saying 'Contact the FerretBrowser plugin registry to see if there is an official way to handle this resource.', one saying 'Pass this URL to a local application' with an application selector, and one saying 'Pass this URL to the "Kittens-at-work displayer" application at "kittens.example.org"', with a checkbox labeled 'Always do this for resources using the "application/x-meowmeow" type in future.', and with two buttons, 'Ok' and 'Cancel'.

In this dialog, the third option is the one that was primed by the site registering itself earlier.

If the user does select that option, then the browser, in accordance with the requirements described in the previous two sections, will redirect the user to "http://kittens.example.org/?show=data%3Aapplication/x-meowmeow;base64,S2l0dGVucyBhcmUgdGhlIGN1dGVzdCE%253D".

The registerProtocolHandler() method would work equivalently, but for schemes instead of unknown content types.

6.5.1.5 Manually releasing the storage mutex
[NoInterfaceObject]
interface NavigatorStorageUtils {
  void yieldForStorageUpdates();
};
window . navigator . yieldForStorageUpdates()

If a script uses the document.cookie API, or the localStorage API, the browser will block other scripts from accessing cookies or storage until the first script finishes. [WEBSTORAGE]

Calling the navigator.yieldForStorageUpdates() method tells the user agent to unblock any other scripts that may be blocked, even though the script hasn't returned.

Values of cookies and items in the Storage objects of localStorage attributes can change after calling this method, whence its name. [WEBSTORAGE]

The yieldForStorageUpdates() method, when invoked, must, if the storage mutex is owned by the event loop of the task that resulted in the method being called, release the storage mutex so that it is once again free. Otherwise, it must do nothing.

6.5.2 The External interface

The external attribute of the Window interface must return an instance of the External interface. The same object must be returned each time.

interface External {
  void AddSearchProvider(DOMString engineURL);
  unsigned long IsSearchProviderInstalled(DOMString engineURL);
};
window . external . AddSearchProvider( url )

Adds the search engine described by the OpenSearch description document at url. [OPENSEARCH]

The OpenSearch description document has to be on the same server as the script that calls this method.

installed = window . external . IsSearchProviderInstalled( url )

Returns a value based on comparing url to the URLs of the results pages of the installed search engines.

0
None of the installed search engines match url.
1
One or more installed search engines match url, but none are the user's default search engine.
2
The user's default search engine matches url.

The url is compared to the URLs of the results pages of the installed search engines using a prefix match. Only results pages on the same domain as the script that calls this method are checked.

Another way of exposing search engines using OpenSearch description documents is using a link element with the search link type.

The AddSearchProvider() method, when invoked, must run the following steps:

  1. Optionally, abort these steps. User agents may implement the method as a stub method that never does anything, or may arbitrarily ignore invocations with particular arguments for security, privacy, or usability reasons.

  2. Resolve the value of the method's first argument relative to the entry script's base URL.

  3. If this fails, abort these steps.

  4. Process the resulting absolute URL as the URL to an OpenSearch description document. [OPENSEARCH]

The IsSearchProviderInstalled() method, when invoked, must run the following steps:

  1. Optionally, return 0 and abort these steps. User agents may implement the method as a stub method that never returns a non-zero value, or may arbitrarily ignore invocations with particular arguments for security, privacy, or usability reasons.

  2. If the origin of the entry script is an opaque identifier (i.e. it has no host component), then return 0 and abort these steps.

  3. Let host1 be the host component of the origin of the entry script.

  4. Resolve the scriptURL argument relative to the entry script's base URL.

  5. If this fails, return 0 and abort these steps.

  6. Let host2 be be the <host> component of the resulting absolute URL.

  7. If the longest suffix in the Public Suffix List that matches the end of host1 is different than the longest suffix in the Public Suffix List that matches the end of host2, then return 0 and abort these steps. [PSL]

    If the next domain component of host1 and host2 after their common suffix are not the same, then return 0 and abort these steps.

    Domain labels must be compared after applying the IDNA ToASCII algorithm to them, with both the AllowUnassigned and UseSTD3ASCIIRules flags set, in an ASCII case-insensitive manner. [RFC3490]

  8. Let search engines be the list of search engines known by the user agent and made available to the user by the user agent for which the resulting absolute URL is a prefix match of the search engine's URL, if any. For search engines registered using OpenSearch description documents, the URL of the search engine corresponds to the URL given in a Url element whose rel attribute is "results" (the default). [OPENSEARCH]

  9. If search engines is empty, return 0 and abort these steps.

  10. If the user's default search engine (as determined by the user agent) is one of the search engines in search engines, then return 2 and abort these steps.

  11. Return 1.