Update status codes to official IANA list

[vanHoesel]

This updates the Status Codes to the official IANA Registry as described at:

https://www.iana.org/assignments/http-status-codes/http-status-codes.xhtml

I did remove a few that were not part of the official list, like 418: I'm a teapot. But then again include some existing unofficial ones that were already in the old version in a separate module.

Maybe, in the future we can use an import like:

use HTTP::Status ':nginx'

to include the 444, 495, 496, 497, and 499, for example.

[vanHoesel]

this would probably make it possible to close #17 (Add HTTP status codes 420, 421, 451, 508), considering the comments on that one

[karenetheridge]

this looks unintentional

[karenetheridge]

odd whitespace

[karenetheridge]

I don't think we should remove any pre-existing entries: someone might be using them.

[vanHoesel]

they were 'temporary' removed, to make the list IANA compliant ... intentionally
but I also planned to re add them.

I would not intentionally break people their stuff, quite aware that there is code out there in the wild.

But thanks for pointing out

[karenetheridge]

debugging code left behind?

[karenetheridge]

I'm not convinced that this is a good idea. It's easier to just leave the codes where they are in the main list.

[vanHoesel]

The idea I was toying with, was to make separate groups of codes in split files and that could be dynamically loaded.

Any new (proposed) RFC could than have it's own file, with it's own small documentation etc.

But ... lets leave that idea for now... me being ambitious

[karenetheridge]

this commit should have been squashed with the original that added this line, so we never saw it. Splitting it out into a separate change adds no value.

[vanHoesel]

:-( sorry, it was already done

[karenetheridge]

Grouping specific codes together makes more sense to do with an export tag, rather than with a separate namespace like ::Unofficial. Regardless, I think we should look at that in a separate PR rather than in this one, which should stick to just adding more codes from the RFCs.

[vanHoesel]

@karenetheridge , I have removed the 'unofficial' file and pulled the codes back into the main file again. But still in a separate section, rather then cluttering the list that now reflects 'official IANA'

Putting 'others' into the list of codes is something we can indeed keep for another discussion. Maybe I was a bit to eager here :-)

[skaji]

LGTM

Especially, thanks for pointing out the link
https://www.iana.org/assignments/http-status-codes/http-status-codes.xhtml.

[karenetheridge]

Ok, onward! Can you squash your changes down to the final set of commits? (i.e. "if it was all perfect right from the start, this is what the commit(s) would look like").

[vanHoesel]

@karenetheridge , on request, I've rewritten history ... as if all was perfect ... and with the small commits as if I knew what I was doing :-D

[oalders]

Did 425 get removed from the codes?

[vanHoesel]

@oalders 👍 well spotted!

yes ... 425 has been removed, it is not an official with IANA registered code!

but ... bad me 😞 for not having it added to the list of unofficial numbers at the bottom.

I did check again a few times but it seems to be okay now.

Added also the "RC_NO_CODE" alias to "RC_UNORDERED_COLLECTION", which is the most common interpretation these days.

[karenetheridge]

thanks Theo, it all looks good to me!

[oalders]

Thanks @vanHoesel, @skaji and @karenetheridge!

[Grinnz]

FYI the backcompat added here only added aliases for RC_NO_CODE, not HTTP_NO_CODE which was already in use, the RC_* constants seem to be legacy and not documented otherwise...

[Grinnz]

PR #145 includes the missing backcompat alias