Improve performance of inserting text into an HTML element

[anonymous]

In Firefox, I am inserting text into ~500 DIVs using this code:

$(".divs").text("default"); ~500 DIVs

On my computer, this consistently takes 9ms if the DIVs are empty. However, this same code consistently takes 18ms if the DIVs already contain text.

Can we improve the performance of replacing text in a DIV that already contains text?

[Rick Waldron]

Confirmed, 1.5 is mega slow.

​http://jsperf.com/insert-to-500

[danheberden]

More specific, it's an issue of $.fn.append - zomg is it slow now. like 80% slower.

​http://jsperf.com/dh-jquery-dom-manip

[addyosmani]

+1 on all the above - it's slow as hell. Do we have an idea yet of what change related to $.fn.append is causing this?.

[jitter]

The real culprit isn't append but domManip. This performance regression potentially affects a whole bunch of methods (maybe not always but: text, append, appendTo, wrap, wrapAll, wrapInner, unwrap, html, replaceWith, replaceAll, jQuery.load, prepend, prependTo, before, after, show, animate) which either directly or indirectly depend on domManip.

The regression was introduced with this ​commit.

Specifically the change on ​this line which before was fragment.cloneNode( true ) :

---

I also want to address that the reporter originally didn't complain about a performance drop for .text/append/domManip between 1.4.4 and 1.5, but that he wanted

$("<div>test test</div>").text("bar")

to be as fast as

$("<div></div>").text("bar")

not sure if there is something which can be done about that

[Rick Waldron]

Duplicate of #7992.

[brad@…]

So #7992 is really about remove() being slow and it was introduced starting in 1.4.3

[Rick Waldron]

Patched. For some reason the browser scope tests arent updating correctly, so I grabbed screenshots of the tests along the way

Chrome: ​http://gyazo.com/c40de84ba91c1602838ca532f32da0a1.png
Opera: ​http://gyazo.com/e40704f30e7798f2fa6bae79c4d65722.png
Safari: ​http://gyazo.com/8cb924594e5f74e428ebb8904c8b111a.png
Firefox 3.6.13: ​http://gyazo.com/275f394bbc9e60fd06dbed0556f0f51b.png
Firefox 4b9: ​http://gyazo.com/dfc05f1da4413d41c7650194619de116.png

The IE patch that introduced the issue is still in place, instead I'm only running it through that path if its absolutely nec. IE requires the run through clone, so its still slow.

[jitter]

​pull request

rwaldron can you please include a link to the jsperf test you made. Also wouldn't this be better suited on #8017 as this bug report originally was about something different (speed of .text in 1.4.4 not the perf drop from 1.4.4 to 1.5b1)

[Rick Waldron]

@jitter,

The perf test will show that this is the appropriate ticket for the patch - they are actually all related.

The perf test will append a string to 500 divs per iteration with 4 different versions of jQuery.

​http://jsperf.com/dommanip-is-a-sloth

Also, I should note that jsperf is not sending the correct values to browserscope.

[Rick Waldron]

More results shots...

Chrome 8 on Win7: ​http://goo.gl/tVaZn

[danheberden]

IE6 - which of course, is running the slow bug fix for events...

​http://gyazo.com/b780d57adb15b970d5ff66d09a73f2f9.png

[Rick Waldron]

Patch is being refactored - please hold.

[danheberden]

holding music: ​http://proj.jetless.org/AWESOME-TIME/

[john]

Landed this pull request, which should resolve this issue: ​https://github.com/jquery/jquery/pull/200