Change ignoreErrors to accept regular expressions

[jeremyhaile]

Currently - ignoreUrls accepts an array of regular expressions and ignoreErrors accepts an array of strings. Instead, each of these should accept an array of strings or regular expressions so that more flexible matching could be done. (or just regular expressions)

To give a specific example of when this is needed, I'm receiving errors such as "Uncaught Error: Error connecting to extension gclcmokkcfnjpghegbnebiokigholeli" I would like to ignore errors using a regex like /Error connecting to extension/ since those errors have nothing to do with my web application.

[acdha]

This was mentioned as an aside in #73. I'm 👍 on including it since it should be easy just to do a type-check on each entry in both arrays.

[jeremyhaile]

Exactly - we had agreed on it in #73 - but wanted to open a specific issue for tracking it, since it's a new enhancement.

[terinjokes]

Is there any work on this? I don't want to double efforts if someone else is already working on it.

[jeremyhaile]

@terinjokes did you do any work on this? branch? Can't wait to roll this out.

[mattrobenolt]

The only reason I haven't done this is because we either need to just assume all regex all the time, or type check. I dislike the idea of type checking, but now we've broken the API.

I like promoting just using all regex, and maybe this is an ok change for 1.1.

Does that sound reasonable?

[acdha]

Why is type checking such a bad idea? I wouldn't want to get carried away
with it but a simple if (foo.test) { … regexp … } else { … string … }
seems unlikely to cause much confusion.

[terinjokes]

I did some work for this, but haven't pushed to a GitHub branch yet. I'll push today.

[jeremyhaile]

@terinjokes great - thanks!

[jeremyhaile]

@terinjokes any update on that branch? and a pull request?