Thanks for the PR! Please check the items below to help us merge this faster. See the contributing docs for more information.

• Uses the issue template (#7273)

If you need to make changes to your pull request, you can update the commit with git commit --amend.
Then, update the pull request with git push -f.

Thanks again for your help!

I'm getting a throw new Error('Spec patterns did not match any files.'); error when I try to run grunt protractor:travis. I'm not sure why, and I can't really fix the errors in the protractor file easily with it. I've updated my version of protractor, but otherwise I'm a bit lost here.

Before someone looks this over, you should clean the PR, i.e squash all the commits into one.

Okay, @caitp said it's not necessary to squash the commits into one, but I would at least put them into logical packages that remove all that c&p from existing code, WIP, logging etc.

I'm not 100% sure, but you shouldn't need to run travis locally, it's done when you send a PR on github.

Working on that now, didn't see the contribute docs before, was going off of the docs on the site. Travis is easier to run locally to spot the errors, rather than waiting for committing. I got it working now, though, I believe, I just need 0a6f332 to run. Thanks, @Narretz.

caitp said it's not necessary to squash the commits into one before review, but I would at least put them into logical packages that remove all that c&p from existing code, WIP, logging etc.

Renamed the PR, but it's probably not the same as in the dev docs.

I'm no nearer to collapsing this; the issue is I merged in the recent version, and I don't know how to collapse some commits, then undo a merge, then collapse more. Should I just delete the PR, reset to the beginning, add all the changes in again in one commit, and reopen a PR? Seems to be the easiest option for me.

How is this different from <p id="{{variable}}">foo</p>?

@RichardLitt , either create a new pull request or

• use git rebase -i <sha-before-merge-commit>
• remove the merge commit and squash / reorder related commits
• after this, rebase your whole branch onto upstream/master
• git push -f your branch to update the PR

Alright, got it. Thanks.

@Wizek Basically, it does a couple of things more. It refuses to deal with space delimited strings, as those are invalid. It allows an object to be passed (with limits: if there are multiple truthy values, it only uses the key of the first one, much like {{a && 'aValue' || b && 'bValue' || 'cValue'}} might do, which is a workaround). If the {{variable}} evaluates as falsy, it removes the empty id class, as according to the html spec id must be a string of more than one character, and cannot be a boolean.

@caitp raised a good point in that it is poor design to reflect state in an ID, as they should just represent nodes in the DOM. I agree. She also pointed out that this might be useful if there was an id generator for ngRepeat. I think that's a good point, and I think it's something that should be added onto this directive. Whether this is useful at the moment isn't clear, as it may equate the use of ngId as the same as that of ngClass, which shouldn't be done - ngClass should be used in almost all cases.

Finally, I was hoping to be able to use $animate the same way that ngClass does, for transitions, but I'm not sure it's strictly necessary, as the id is simply replaced and not added and removed like classes can be, since multiple classes are valid.

Travis failed due to cross browser tests not delineated in the documentation. I opened an issue here: #7303

Did you know that you can use ng-attr-id instead of ng-id? It's pretty cool:

<div ng-attr-id="someExpression">Happy Rain!</div>

+1 @judytuna for mentioning ngAttr. It is really cool as it is really hard to find in the docs :)

@judytuna Yeah, that's a possibility. It still has all of the flaws I mentioned above, though. I would recommend against using it for id.

@gkalpak Maybe you want to submit a PR to the docs? :D

With the possibility to use ng-attr-id, then this PR is no longer needed

@lgalfaso using ng-attr-id can lead to invalid HTML. Are we ok with that?

data-ng-attr-id should keep people relying on validators happy

Word.