I'm building a graphics-intensive site in HTML5 that makes heavy use of the canvas context's drawImage() function. In IE9 on Windows 7, performance is smooth, but in Firefox 4, things get choppy. I'm trying to isolate bottlenecks so I can start optimizing.
If I use the JavaScript performance profiling feature of Firebug 1.7.0, I can see statistics for my own functions, but I'd like to see calls to the built-in JavaScript functions as well. Is there any way to do this in Firebug or some other tool?
As a workaround, I suppose I could make the profiling granularity finer by breaking things down into lots of tiny functions, but I'd rather my design not be driven by how easy it is to profile.
Update: Looking back at this question, it occurs to me that the built-in functions in question are not truly JavaScript functions but functions provided by the host. In this case, they're DOM objects from the browser. Just thought I'd clarify that technical detail.
Last I used it Firebug doesn't give you the ability to profile native code.
What you can do is make your own function that just calls the native piece you want to call. As in turn the code:
ctx.fillRect(50,50,50,50);`
Into:
// wrap in function
function fillR() {
ctx.fillRect(50,50,50,50);
};
// call it immediately afterwards
fillR();
Then you can get the stats for fillR. Not the best fix, but it may useful enough.
You might try playing around with Venkman, Mozilla's JS debugger. At the moment the version on addons.mozilla.org is broken in Firefox 4.0.
You can get the most recent version via mercurial, which does work with Firefox 4.0:
hg clone http://hg.mozilla.org/venkman/
cd venkman/xpi
./makexpi.py
firefox venkman-0.9.88.1.xpi
After contemplating this on and off for a couple of days, I came up with a new solution and wrote a blog post about it:
http://andrewtwest.com/2011/03/26/profiling-built-in-javascript-functions-with-firebug/
This method does the following:
Checks to see if the Firebug console
is open.
Does a combination of overriding and
wrapping native functions with
user-defined functions.
These steps combined provide a way to profile DOM functions that doesn't affect your original script code unless you're debugging.
Related
Not a coding question, and the question applies to a wider tag of XYZ becoming obsolete, but asking as there seems to be tons of comments on the web about how there are so many apps out there that use this command.
Why is this now obsolete? (tried executing on vscode, would not work)
Based on the MDN entry for Document.execCommand, in most cases the APIs made available by execCommand can be achieved using other modern, purpose-built JavaScript APIs that behave the same way across browsers.
Each command in execCommand's toolbox was contentiously implemented in different ways across browsers meaning that testing on every single browser was a necessity of web development. Just skimming through the draft spec you linked in the comments shows all of the notes detailing some of the different ways they were implemented. Because of these differences there was a shift away from HTML-based rich-text document rendering to canvas-based rendering for Google Docs and img-based rendering for Office 365 Web Apps (at least for viewing them, editing is still HTML-based).
Additionally, the MDN entry notes:
Return value
A boolean value that is false if the command is unsupported or disabled.
Note: document.execCommand() only returns true if it is invoked as part of a user interaction. You can't use it to verify browser support before calling a command. From Firefox 82, nested document.execCommand() calls will always return false.
Because execCommand can only be invoked as part of user interaction, it might work in one bit of your code, and not in another - which can cause time-consuming headaches that involve learning the intricacies of JavaScript's event loop.
Importantly, with the advent of JavaScript running on mobile devices and modern JavaScript frameworks having their own "event->queue->render" loops, what even defines a user interaction event? A click, tap, keypress, etc? What about a screen rotation or resizing the window? What about asynchronous event changes where you make a server request and then show the user something based on the result? How would the engine know those two things are linked together with modern Promise-based APIs rather than callback-based APIs?
If you want to do some further digging, take a look at this thread on another question asking about what replacements are available to execCommand().
So I've been coding for a while now in school, and I'm having my first introduction into a real working code base at my new job. When it comes to debugging, like if I want to find out what a certain variable is equal to, I'll often just console.log() things out and hope that something useful comes out of it.
It seems really inefficient considering how many moving parts there are in the code. Most of the time, it works eventually, but it feels so barbaric just editing the code over and over and looking at the browser's console until something useful comes out, and it feels like there has to be a better way to debug in a production environment. I am specifically using Ember.JS, but this question would apply to any framework like React or Angular or Node.
There are many tools out there to debug JS, the most "basic" being the debugger in your browser. Some tools go further also, for example VSCode has a slew of extensions that open and control the browsers debugger.
Chrome: https://developers.google.com/web/tools/chrome-devtools/javascript/
Firefox: https://developer.mozilla.org/en-US/docs/Tools/Debugger
Edge: https://learn.microsoft.com/en-us/microsoft-edge/devtools-guide/debugger
Safari also has one in the develop menu but there is no site describing it in detail - it is located in the sources pane
If you use VS Code, you can run Node apps in debug mode then select Node.js.
This is kind of a two-part question.
Why does IE require so much special treatment when handling Javascript? And are there any tricks, resources, and/or systems you have picked up for making your js IE-compatible, besides Firebug lite?
Using standardised libraries like J-query so you don't have to jump through all the hoops yourself works on the javascript side!
Also Yahoo User Interface (YUI) is good for making websites look extremely similar over different browsers. Their Grids library works really well.
Could you please elaborate little what you're referring to?
JavaScript in it's core is have mainly been the same the last 10 years. If you're worried about older IE versions (IE6-7) you can remain calm. IE6 was released with JavaScript 1.5 support so all JS code should run fine. Mozilla has a very good JS ref document at their MDC site. At the bottom of each page there's usually a list of what version of each browser that supports that specific function.
However the difference between browser usually lies in the DOM implementation or event handling. Where properties may have different names, at the top of my head these properties mainly are related to element/scroll positions.
To find the correct property to use, check in the developer tool (Firebug in IE, Developer Tools in Webkit or Developer Toolbar in IE) for that browser to find what you are looking for. If you're unsure set at JavaScript breakpoint in your code using the debugger;keyword or send something to the console using console.log(). In IE Developer Toolbar is available from IE8+ (I think).
Most of these problems are already resolved in the major JavaScript frameworks like jQuery, MonoTools and so on.
The two main "special treatment" things that come to mind are:
Events. Including assignment of event handlers, the way the event object is made available to the handler function, and some of the properties of the event object. See this page for more info: http://www.quirksmode.org/js/introevents.html
Ajax. Use of XMLHttpRequest versus ActiveXObject("Microsoft.XMLHTTP").
Most everything else should be fine.
You can write yourself some (relatively) simple helper functions to get around these issues, or use a library like jQuery that normalises the differences for you. If you do write it yourself, be sure to test for feature support rather than try to test for which browser - see this (long) article for an explanation: http://jibbering.com/faq/notes/detect-browser/ (I'm sure there are shorter explanations around but I can't be bothered finding one.)
If you using jquery in your page then you can check for ie with this code
if ($.browser.msie && $.browser.version == '6.0') {
//do IE specific code
}
This code will only run when the user browser will be IE 6.0 or you leave this condition
I would like to see what the JavaScript interpreter is doing in real-time, which line it is reading and which function it is running, because I would like to make a full debug on every little piece of the JavaScript to make it faster. For this I need some tool to tell me what the interpreter is doing exactly, if it is defining a variable, if it's running a function, if it's in a loop, check the current intervals (defined in setInterval).
As far as I know Firebug can't do that.
Check out the javascript tab in Firebug, it's a full debugger. Set a break point and when your code hits that line you will have full debugging access to variables etc. Chrome has similar functionality in the developer tools which are included in a standard install.
If you're looking to do automated monitoring/analysis of your program, check out Chrome's Debugger Protocol. It provides a programatic API. I believe (but could be wrong) that this is what tools like Web Inspector and node-inspector are built on.
If you want something more than what the standard Firebug/Web Inspector interfaces are built on, you're going to have to either use something like this or start hacking on the internals of the V8 and Gecko JS interpreters.
As the other answer says,if you want to go step by step, setting a debug point is the way to go.
But since you seem interested in improving performance you might want to consider optimizing only the true bottlenecks in your application. To see which functions take the most to run, and other statistics you might want to take a look at console.profile() (available both in firebug and chrome).
Here is an example:
console.profile('myProfile');
//some code you want to know more about
console.profileEnd();
You can also start it manually by clicking the Profile button in the Console panel(in firebug)/ Profile panel (chrome)
In Chrome you might want to also take a look at Heap Snapshots (they tell you about memory usage).
I want to parse a html-page that unfortunately requires JavaScript to show any content. In order to do so I use a small python-script that pulls the html-code of the page, but after that I have to execute the JavaScript in a DOM-context which seems pretty hard.
To make it even harder I want to use it in a server environment that has no X11-server.
Note: I already read about http://code.google.com/p/pywebkitgtk/ but it seems to need a X-server.
You can simulate a browser environment using EnvJS. However, in order to make use of it, you will have to embed some kind of JavaScript runtime (e.g. Rhino) in your program (or spawn one as an external process).
You could try using Xvfb to have a fake frame buffer, so you won't need to run X11 (though it may be a dependency of Xvfb on your system). Most rendering engines don't have a headless mode, so something like Xvfb is necessary to run them. I used this technique successfully using XULRunner to navigate web pages, though not from python.
I'm still trying to figure this out myself, so take my answer with a grain of salt.
So far, I found http://blog.motane.lu/2009/06/18/pywebkitgtk-execute-javascript-from-python/, which describes the use and the quirks of Pywebkitgtk by someone who has similar needs to what we do.
Later, however, the writer of that blogpost discovered that he can't get it to work with Xvbf, so he hunted some more and found a Qt webkit (possibly in Qt itself, if I understand correctly) http://blog.motane.lu/2009/07/07/downloading-a-pages-content-with-python-and-webkit/. Apparently it's a much better solution than PywebkitGTK.
Naturally, I'll be looking into the other solutions offered here--but I wanted to bring up the Qt solution, because to me, it seems the most likely candidate for what I want to do...and if not, then perhaps it will be for someone else, looking for an answer to this question! :-)
I use VNC or Xvfb for this purpose, combined with Firefox. After experimenting with the two, I settled on XTightVNC. We use it to create screenshots on demand for various test purposes. It's nice to use one of these because you're executing it in an actual browser, same as a user would be (though most users probably won't be using the same OS as your server).
The handy thing about using VNC is that you can connect remotely to set up and test the browser when needed.
This might help: http://code.google.com/p/pyv8/