Comments on: Introducing “mini”

By: cnlinkin

cnlinkin — Sat, 28 Nov 2009 13:53:56 +0000

hi,james:)
i’ve used your mini selector and it’s so cool~ but i found that in line 139-> var cacheIndex = elem[uid], nextCacheIndex = n++; will no work in IE when selector have the “select” tag,so would you like to take care of the poor IE in the next version?

By: Alex Robinson

Alex Robinson — Fri, 25 Sep 2009 11:37:11 +0000

David Hyatt’s response to this comment of mine on the Webkit blog

http://webkit.org/blog/156/queryselector-and-queryselectorall/#comment-23909

explains why querySelector is slower than getElementById

“Unfortunately, my understanding of the spec indicates that despite the invalidity of multiple elements having the same id, we’re required to return all elements with a given id”

By: Vasili

Vasili — Mon, 10 Aug 2009 22:47:25 +0000

You have inspired me to try out my JavaScript skills that I have been picking up from my book.

Mini looks very, very cool!

By: Graham B

Graham B — Sun, 09 Aug 2009 14:59:51 +0000

Nice work, James. While I don’t trust Slickspeed (many engines cache their returned nodesets which completely skews the results), it looks like your engine does pretty well. After building a CSS3-compliant engine I can certainly say that its a major task and that a slimline alternative such as this is far better in many situations.

By: James

James — Sun, 09 Aug 2009 14:28:26 +0000

@Elijah, During testing I found that context.getElementById(selector.substring(1)); was faster than context.querySelectorAll(selector) (selector being equal to '#id'). I think I'll take another look at it anyway, there may be a better way to optimise simple selectors.

By: Elijah Grey

Elijah Grey — Sun, 09 Aug 2009 13:46:20 +0000

Why only use document.querySelectorAll when the selector isn’t simple? Native is always faster so you should use it even if mini can handle it.