optimize-selectors qSA before Sizzle

[mr21]

Hi :)

Here a first draft (who probably contain few mistakes) for the issue #666.

[agcolom]

@dmethvin @timmywil It would be truly wonderful if you could review this PR. Thank you :-)

[AurelioDeRosa]

I think this example might need a small introduction. Even 1-2 sentences should do it.

[AurelioDeRosa]

Hi @mr21. Are you still interested in this PR?

[mr21]

... I'm really sorry, yes I'm still interested.
I'll push ASAP your change

[kswedberg]

@mr21 Make sure you rebase from here before continuing your work. I merged a PR based on another issue and removed/rearranged a bunch of stuff, so you might have conflicts.

[mr21]

Understood :)

[mr21]

Hi :)

So, I have rebased and repush again my commit with the @AurelioDeRosa's modifications.
Except that I didn't add yet a small introduction in place of this removed sentence:
Beginning your selector with an ID is a safe bet.

[AurelioDeRosa]

Hi @mr21.

With "yet" you mean that you'll add another commit to update the PR?

[mr21]

Hi :)

For the moment I have replaced this:

---

ID-Based Selectors

Beginning your selector with an ID is a safe bet.

// Fast:
$( "#container div.robotarm" );

// Super-fast:
$( "#container" ).find( "div.robotarm" );

With the first approach, jQuery queries the DOM using document.querySelectorAll(). With the second, jQuery uses document.getElementById(), which is faster, although the speed improvement may be diminished by the subsequent call to .find().

---

by this:

Save calls to querySelectorAll

querySelectorAll is already really fast, if you want maintain this speed try to call it the least possible.

// If in your HTML there are 2 .container with 5 div in each,
// this line will call querySelectorAll 13 times (1 + 2 + 2*5).
$( ".container" ).children( "div" ).find( ".robotarm" );

// Against only 1 call with this:
$( ".container div .robotarm" );

---

But I don't know what I have to write to introduce my example.
If you have an idea, I will commit it quickly.

Thanks for your patience :)

[Krinkle]

I've confirmed from the previous CLA data that this PR was indeed signed as the bot originally indicated:

cb2c81b

/var/www/jquery-license/output/jquery/learn.jquery.com/cb/cb2c81b5816f112368c34090f4214bf613c1b881.json

{
  "data":{
    "commits":[{
      "hash":"cb2c81b5816f112368c34090f4214bf613c1b881",
      "email":"thomastortorini@gmail.com",
      "name":"Thomas Tortorini"
    }],
    "neglectedCommits":[],
    "neglectedAuthors":[]
  },
  "state":"success",
  "description":"All authors have signed the CLA"
}