Log 404s to console

[edwardhorsford]

The kit currently swallows 404 errors.

We should either not swallow them, or else print them to the console.

Something like console.log('404:' + err) on line 161 of lib/utils.js could work.

[joelanman]

we can probably write a nicer error than the default

[mikeshawdev]

Adding a custom comment would be good! I wonder whether it would also be useful, to add an HTTP request logger, such as Morgan, to the prototype? It would output like below:

If folk feel this is, maybe, too much output we could add our own custom log statements in. I think it's useful to see exactly what requests are being served, and the response from the server

[joelanman]

I think the most useful default is to output errors and warnings. If we output everything, as in the screenshot above, it makes it harder to see those. We could make it configurable if people want to see more.

[mikeshawdev]

@joelanman I can add a condition to satisfy that. The log would look like below:

It only logs out on status codes that are greater than or equal to 400. Would this resolve the issue?

The other option I had was something like below, which may be easier for various users of the kit to understand but only logs for 404s:

[joelanman]

I think a custom message would be good. The first screenshot you have to know what a 404 is.

Not sure 'could not resolve path' is easy to understand either... but it's tricky. I was thinking 'Missing file' - but it's not always a file, could be a route in routes.js. How about

Error: 404 - couldn't find '/url/that/doesnt/work'

As that maps well to the technical meaning of 404 (Not found) but is a bit easier to read

[amyhupe]

Yes to @joelanman's suggestion. Though could you say 'could not' instead of 'couldn't'. I know our current position is that contractions are OK. but I think the proximity to the opening quotation here might make it easier to miss.

[mikeshawdev]

No problem, I'll get this added and create a PR! Will take @amyhupe's suggestion on the "could not" vs "couldn't"

[NickColley]

As referenced in the closed issue, this has now been fixed, thanks everyone.

#566