You can clone with
No one assigned
Let's warn people as they approach this limit. I'd suggest a warning at 3800 rules and an error at 4095.
I set them up as two rules so the user could configure them as error or warning as they see fit. I don't think that the message level got hard coded and in order to allow for both levels it needs to be two rules from my understanding.
@stubbornella Hmmm, I'm not sure I agree with that. Warning someone that they might hit a limit doesn't seem very useful to me. When I'm linting something, I want to know what would be an issue in the code I just wrote rather than a potential issue that might occur sometime in the future (but then again, it might not). There are a whole category of things that might end up being an issue if you continue down that path.
In any event, I think having two rules is overkill. If you think it's a best practice to, for instance, have less than 4000 rules in general, then I would make that a rule with the explanation that there's probably a lot of duplication and there's a potential issue once you reach 4095 in Internet Explorer.
And just an FYI that we don't currently support per rule configuration over and above specifying whether something is a warning or an error.
@nzakas It seems we disagree, so I thought the best thing to do was to ask the community what would be most useful to them. https://twitter.com/stubbornella/status/245952112243138561
ok, I let people chime in for around a half an hour. They prefer both an error at 4095 and warning a bit in advance of that. In addition, if we have both, anyone who doesn't like it can turn off the warning rule. Win-win.
7- error only,
3 - warning only,
15 - both
So, let's go ahead with two rules, one for the error and one for the warning.
One interesting comment came up. There may be different limits for IE7 versus 8 related to media queries. I'm waiting to hear back from @jaffathecake about that.
Well, my opinion hasn't changed, but go for it. Just make sure the two different rules are in two different files with two different sets of tests (the current pull request has both rules in the same file).
Pull request #304 has been updated
@stubbornella Both are not changed to only IE