feature-detect for Document Fragments

[jokeyrhyme]

reports true as expected in IE 6-10, Chrome, Opera, probably in Firefox
reports false on a few old Nokia handsets in the office, probably on old Blackberries too

As using Document Fragments is considered a performance and stability best-practice for DOM manipulation, I think it's worth knowing when you can actually use this technique safely.

I have probably been a little too pedantic, so do let me know if some of the tests ought to be trimmed.

[paulirish]

Nice!

reading https://github.com/Modernizr/Modernizr/wiki/How-We-Test ... your tests are probably too exhaustive..

just checking !!document.createDocumentFragment may just be enough??

[jokeyrhyme]

Oh, you're probably right about simplifying the tests.

Whilst researching this feature, I came across notes that IE6 does not create a Fragment, but rather creates a new Document object. Much of my code is there to confirm that whatever is returned actually behaves the way it needs to. Even though IE6 was a little strange, it does complete the test.

I just wasn't sure if we'd be this lucky with every barely-conforming browser, but I'm happy to reduce this to a simple presence check if you like.

[paulirish]

Ah. Sounds like a bug in IE6 that we'd want to discover and then discard
its "support". Yeah?

[jokeyrhyme]

IE6's implementation does function as expected.

Hmm, just took a look at http://www.quirksmode.org/m/w3c_core.html#t112
Looks like a completely trusted result on this involves confirming that an element can start out life in the Document, be moved to the Fragment, and winds up in the Document again after the append. I personally use Fragment only for new elements, but I can't decide if this is something that is within the "90% use case".
The DOM manipulation here would definitely slow my test down even further. :S
It also looks like it was a DOM 1 feature, oops!

I don't think we need to check some of the other quirks IE 6 and 7 (http://reference.sitepoint.com/javascript/DocumentFragment) as that falls outside the "90% use case", or outside of mine, anyhow. :P Checking if Fragments have the "normalize" methods smells like unnecessary conformance testing to me.

So, do I make this more expensive to cover a known broken implementation? Or do I slim it down even more and trust that most people will never use the broken part?

[sindresorhus]

I would say; slim it down!

@paulirish what do you think?

[jokeyrhyme]

@paulirish @sindresorhus I've slimmed this down to just detecting that the methods required to use Document Fragments have been defined. Further, I've updated this for compatibility with Modernizr 3.

Please let me know if there are any further outstanding issues. <3

[paulirish]

nice! thanks for the update!

On Tue, Apr 2, 2013 at 10:35 PM, Ron Waldon notifications@github.comwrote:

@paulirish https://github.com/paulirish @sindresorhushttps://github.com/sindresorhusI've slimmed this down to just detecting that the methods required to use
Document Fragments have been defined. Further, I've updated this for
compatibility with Modernizr 3.

Please let me know if there are any further outstanding issues. <3

—
Reply to this email directly or view it on GitHubhttps://github.com//pull/641#issuecomment-15819236
.

[patrickkettner]

@jokeyrhyme it would be great ot get this merged in - could you squash all of the commits down into a single one?

[jokeyrhyme]

@patrickkettner done!

[patrickkettner]

tiny nit, then 👍

[jokeyrhyme]

@patrickkettner done!

[stucox]

LGTM, cheers Ron :-)