Last week, Microsoft officially launched the first preview of Internet Explorer 11 with Windows 8.1. In this way, it's time to rest on all the rumors about this controversial leaked version of the web browser. We now know some important details about Internet Explorer 11, including support for WebGL, prefetch (Translator's note: preread, such as uploading files, etc., and I don't know if the translation is correct), prerender (preprepresentation), flexbox, mutation observers and other web standards. Perhaps more interesting in IE11 is whether it is IE.
Microsoft has removed some features from the Internet Explorer browser for the first time in a long time. The user-agent string has also changed. It seems that Microsoft has abandoned its own way so that existing IE-detecting code branches will return false in Internet Explorer 11 browser, whether in JavaScript or on the server.
Changes to User-agent
Compared with previous versions, Internet Explorer 11's user-agent string is shorter, and there are some interesting changes:
The code copy is as follows:
Mozilla/5.0 (Windows NT 6.3; Trident/7.0; rv 11.0) like Gecko
For user-agent strings for Internet Explorer 10 on Windows 7
The code copy is as follows:
Mozilla/5.0 (compatible; MSIE 10.0; Windows NT 6.1; WOW64; Trident/6.0)
The most obvious difference is that the "MSIE" tag of the user-agent string in the Internet Explorer browser has been removed. Also compelling is the addition of "like Gecko" to the tail of the user-agent. This shows that Internet Explorer would rather be recognized as a Gecko browser than as an Internet Explorer itself. Safari is the first browser to add "like Gecko", so that anyone can sniff the "Gecko" character in the user-agent string to allow the browser to pass.
Now any sniffing code that looks for "MSIE" does not work under the new user-agent string. You can still search for the "Trident" character to determine that it is the IE browser (the "Trident" logo was introduced with Internet Explorer). The real version of Internet Explorer is identified by "rv".
In addition, the information of the navigator object has also changed, which further conceals the browser being used.
navigator.appName is set to "Netscape"
navigator.product is set to "Gecko"
This may seem like an attempt to sneakily trick the developer, but these are actually HTML5 specified. The navigator.product property must be "Gecko", and the navigator.appName should also be "Netscape" or more specific. But Internet Explorer 11 lacks follow this weird suggestion.
Side effects caused by changes in navigator information, browser detection based on Javascript logic in use may not be available, causing Internet Explorer 11 to recognize as a Gecko-based browser.
document.all and his friends
Since IE4, document.all has done everything in the IE browser. The document.getElementById() implemented previously is similar to document.all is the IE method of obtaining a DOM element reference. Although document.all maintains support for DOM from IE5 to IE10. However, in 11, the product left over from this era has been set to return false, which means that any code branch judgment based on document.all will fail in IE11, even if the code actually uses document.all to work properly.
Another legacy is the attachEvent() that adds the event function, similar to the detachEvent() method. This method has been removed from IE11. These methods are removed to avoid some logical judgments, such as:
The code copy is as follows:
function addEvent(element, type, handler) {
if (element.attachEvent) {
element.attachEvent("on" + type, handler);
} else if (element.addEventListener) {
element.addEventListener(type, handler, false);
}
}
Of course, we recommend that the version you are best used to test is always stable and meet the standards. On some level, removing the attachEvent method does not cause any discomfort. However, the Internet is full of feature detection logic codes for the Internet. The removal of the attachEvent method ensures that any code written in the above manner will use the standard version instead of IE-specific methods.
Some other removed features:
window.execScript() - IE version of eval() method
window.doScroll() - IE window scrolling method
script.onreadystatechange - state changes of the loading script in IE
script.readyState—the state of script loading in IE
document.selection--the currently selected text in IE
document.createStyleSheet--Create stylesheet document in IE
style.styleSheet--A style object in IE browser that references a style sheet
All of these IE browser methods are replaced by standard features. After removing these features and methods, the cross-browser code based on standard functional feature detection can still function normally without changing.
in conclusion
IE11 is probably the best IE browser for a long time. Microsoft is finally preparing to eliminate mistakes made in the past and is ready to start with browsers based on current standards. Deleting old features and modifying user agent strings that are not recognized as IE browsers is a unique practice, which ensures that all websites continue to work. If the web application uses feature detection instead of browser sniffing, the code should be run in IE11. For applications that use user-agent sniffing, users can still see a website with good functionality, because IE11 has excellent support for standard.
A future without IE branch code is coming, let's look forward to it together.
(Updated on July 2, 2013), the document.all mentioned in the revision was not actually deleted, but was changed to falsesy
I haven't translated the article for a long time. This time I translated Nicholas C. Zakas's Blog. The original text is here: http://www.nczonline.net/blog/2013/07/02/internet-explorer-11-dont-call-me-ie/