Browser detection using the user agent

Serving different Web pages or services to different browsers is usually a bad idea. The Web is meant to be accessible to everyone, regardless of which browser or device they're using. There are ways to develop your website to progressively enhance itself based on the availability of features rather than by targeting specific browsers.

But browsers and standards are not perfect, and there are still some edge cases where detecting the browser is needed. Using the user agent to detect the browser looks simple, but doing it well is, in fact, a very hard problem. This document will guide you in doing this as correctly as possible.

Considerations before using browser detection

When considering using the user agent string to detect which browser is being used, your first step is to try to avoid it if possible. Start by trying to identify why you want to do it.

Are you trying to work around a specific bug in some version of a browser?

Look, or ask, in specialized forums: you're unlikely to be the first to hit this problem. Also, experts, or simply people with another point of view, can give you ideas for working around the bug. If the problem seems uncommon, it's worth checking if this bug has been reported to the browser vendor via their bug tracking system (Mozilla; WebKit; Blink; Opera). Browser makers do pay attention to bug reports, and the analysis may hint about other workarounds for the bug.

Are you trying to check for the existence of a specific feature?

Your site needs to use a specific Web feature that some browsers don't yet support, and you want to send those users to an older Web site with fewer features but that you know will work. This is the worst reason to use user agent detection because odds are eventually all the other browsers will catch up. In addition, it is not practical to test every one of the less popular browsers and test for them. You should never do user agent sniffing. There is always the alternative of doing feature detection instead.

Do you want to provide different HTML depending on which browser is being used?

This is usually a bad practice, but there are some cases in which this is necessary. In these cases, you should first analyze your situation to be sure it's really necessary. Can you prevent it by adding some non-semantic <div> or <span> elements? The difficulty of successfully using user agent detection is worth a few disruptions to the purity of your HTML. Also, rethink your design: can you use progressive enhancement or fluid layouts to help remove the need to do this?

Avoiding user agent detection

If you want to try to avoid using user agent detection, there are options in some cases!

Feature detection

Feature detection is where you don't try to figure out which browser is rendering your page, but instead, you check to see if the specific feature you need is available. If it's not, you use a fallback. There are never any rare cases where you should use the user agent string, such as to detect a browser which implements the behavior of an API differently. Rather, in such rare cases, you should implement a test to detect how the browser implements the API and then determine when and how to use it instead. A good current example of feature detection is as follows. Recently, Chrome has added experimental look behind support to regular expressions, but no other browser currently supports this. So, you might incorrectly assume you should do this:

if (navigator.userAgent.indexOf("Chrome") !== -1){
    // YES, the user is suspected to support look-behind regexps 
} else { /*put your old fall back code here*/ }

However, the above code is absolutely terrible and ill-conceived. What if Chrome removes this look behind feature? What if another browser implements look behind regexps? What if another browser uses Chrome in their user agent string? The list goes on and on of things that could go terribly wrong. Thus, you should instead use feature detection like the following.

var isLookBehindSupported = false;
try { isLookBehindSupported = !!new RegExp("(?<=)"); } catch (e) {
/*In unsupported browsers, trying to create a lookbehind expression will simply error will simply error, which is caught here*/
}
if (isLookBehindSupported) {
    // Yay, lookbehind expressions are supported
} else {
    // Booo! Lookbehind not supported
}

As the above code demonstrates, there is always a way to test browser support on anything, no exceptions. There never any reason to check the user string for this.

Lastly, the above code snippets bring about a critical issue with cross-browser coding that must always be taken into account. Don't unintentionally use the API you are testing for in unsupported browsers. This may sound obvious and simple, but sometimes it is not. For example, in the above code snippets, using lookbehind in short-regexp notation (e.g. /reg/igm) will cause a parser error in unsupported browsers. Thus, in the above example, you would use new RegExp("(?<=look_behind_stuff)"); instead of /(?<=look_behind_stuff)/, even in the lookbehind supported section of your code.

Progressive enhancement

This design technique involves developing your Web site in 'layers', using a bottom-up approach, starting with a simpler layer and improving the capabilities of the site in successive layers, each using more features.

Graceful degradation

This is a top-down approach in which you build the best possible site using all the features you want, then tweak it to make it work on older browsers. This can be harder to do, and less effective, than progressive enhancement, but may be useful in some cases.

Mobile Device Detection

Arguably the most common use and misuse of user agent sniffing is to detect if the device is a mobile device. However, what is failed to be accountable is what they're really after. People use user agent sniffing to detect if the users' device is touch-friendly and has a small screen so they can optimize their website accordingly. While user agent sniffing can sometimes detect these, not all devices are the same. Some mobile devices have big screen sizes, some desktops have a small touchscreen, some people use smart TV's which are an entirely different ballgame altogether, some people can dynamically change the width and height of their screen by flipping their tablet on its side! So, user agent sniffing is definitely not the way to go. But, there are much better alternatives. Use Navigator.maxTouchPoints to detect if the user's device has a touchscreen. Then, default back to checking the user agent screen only if (!("maxTouchPoints" in Navigator)) { /*Code here*/}. Using this information of whether the device has a touchscreen, do not change the entire layout of the website just for touch devices: you will only create more work and maintenance for yourself. Rather, add in touch conveniences such as bigger, more easily clickable buttons (you can do this using CSS by simply increasing the font size). As for the screen size, simply use window.innerWidth and window.addEventListener("resize", function(){ /*refresh screen size dependent things*/ }). What you want to do for screen size is not slash off information on smaller screens. That will only annoy people because it will force them to use the desktop version. Rather, try to have fewer columns of information in a longer page on smaller screens while having more columns with a shorter page on larger screen sizes. This effect can be easily achieved using CSS flexboxes. Next, always make your code dynamic. The user can flip their mobile device on its side, changing the width and height of the page. Never be satisfied with your webpage until you can open up the dev tools side panel and resize the screen while the webpage looks smooth, fluid, and dynamically resized.

Which part of the user agent contains the information you are looking for

As there is no uniformity of the different part of the user agent string, this is the tricky part.

Browser Name

When people say they want "browser detection", often they actually want "rendering engine detection". Do you actually want to detect Firefox, as opposed to SeaMonkey, or Chrome as opposed to Chromium? Or do you actually simply want to see if the browser is using the Gecko or the WebKit rendering engine? If this is what you need, see further down the page.

Most browsers set the name and version in the format BrowserName/VersionNumber, with the notable exception of Internet Explorer. But as the name is not the only information in a user agent string that is in that format, you can not discover the name of the browser, you can only check if the name you are looking for. But note that some browsers are lying: Chrome for example reports both as Chrome and Safari. So to detect Safari you have to check for the Safari string and the absence of the Chrome string, Chromium often reports itself as Chrome too or Seamonkey sometimes reports itself as Firefox.

Also, pay attention not to use a simple regular expression on the BrowserName, user agents also contain strings outside the Keyword/Value syntax. Safari & Chrome contain the string 'like Gecko', for instance.

Of course, there is absolutely no guarantee that another browser will not hijack some of these things (like Chrome hijacked the Safari string in the past). That's why browser detection using the user agent string is unreliable and should be done only with the check of the version number (hijacking of past versions is less likely).

Browser version

The browser version is often, but not always, put in the value part of the BrowserName/VersionNumber token in the User Agent String. This is of course not the case for Internet Explorer (which puts the version number right after the MSIE token), and for Opera after version 10, which has added a Version/VersionNumber token.

Here again, be sure to take the right token for the browser you are looking for, as there is no guarantee that others will contain a valid number.

Rendering engine

As seen earlier, in most cases, looking for the rendering engine is a better way to go. This will help to not exclude lesser known browsers. Browsers sharing a common rendering engine will display a page in the same way: it is often a fair assumption that what will work in one will work in the other.

There are five major rendering engines: Trident, Gecko, Presto, Blink, and WebKit. As sniffing the rendering engines names is common, a lot of user agents added other rendering names to trigger detection. It is therefore important to pay attention not to trigger false-positives when detecting the rendering engine.

Rendering engine version

Most rendering engines put the version number in the RenderingEngine/VersionNumber token, with the notable exception of Gecko. Gecko puts the Gecko version number in the comment part of the User Agent after the rv: string. From Gecko 14 for the mobile version and Gecko 17 for the desktop version, it also puts this value in the Gecko/version token (previous version put there the build date, then a fixed date called the GeckoTrail).

OS

The Operating System is given in most User Agent strings (although not web-focussed platforms like Firefox OS), but the format varies a lot. It is a fixed string between two semi-colons, in the comment part of the User Agent. These strings are specific for each browser. They indicate the OS, but also often its version and information on the relying hardware (32 or 64 bits, or Intel/PPC for Mac).

Like in all cases, these strings may change in the future, one should use them only in conjunction with the detection of already released browsers. A technological survey must be in place to adapt the script when new browser versions are coming out.

Mobile, Tablet or Desktop

The most common reason to perform user agent sniffing is to determine which type of device the browser runs on. The goal is to serve different HTML to different device types.

• Never assume that a browser or a rendering engine only runs on one type of device. Especially don't make different defaults for different browsers or rendering engines.
• Never use the OS token to define if a browser is on mobile, tablet or desktop. The OS may run on more than one type of (for example, Android runs on tablets as well as phones).

The following table summarizes the way major browser vendors indicate that their browsers are running on a mobile device:

In summary, we recommend looking for the string “Mobi” anywhere in the User Agent to detect a mobile device.