Error in icla-info.json format

[sebb <se...@gmail.com>]

I just had a look at icla-info.json, and noticed that the committers
section looks like this:

{
  "last_updated": "2016-02-15 19:49:41 UTC",
  "gem_version": "0.0.75",
  "committers": [
    [
      "a_horuzhenko",
      "Artyom Horuzhenko"
    ],
    [
      "aa",
      "Andrew Ahn"
    ],
...

This is not easy to use.
It was presumably intended to be a hash instead.

AFAICT, luckily the file is not being used except by phonebook, and
that assumes it is a hash so the contents are irrelevant.

I think this might be a good time to fix the output ...!

Perhaps create two new files for committers and non-committers; the
old format file can be deleted later?

Re: Error in icla-info.json format

[sebb <se...@gmail.com>]

On 15 February 2016 at 21:53, sebb <se...@gmail.com> wrote:
> I just had a look at icla-info.json, and noticed that the committers
> section looks like this:
>
> {
>   "last_updated": "2016-02-15 19:49:41 UTC",
>   "gem_version": "0.0.75",
>   "committers": [
>     [
>       "a_horuzhenko",
>       "Artyom Horuzhenko"
>     ],
>     [
>       "aa",
>       "Andrew Ahn"
>     ],
> ...
>
> This is not easy to use.
> It was presumably intended to be a hash instead.

Yes, in fact it was coded as such.
However a sort was added to the ids hash (mea culpa?) which silently
converted it into an array - one of the perils of a non strictly typed
language.

> AFAICT, luckily the file is not being used except by phonebook, and
> that assumes it is a hash so the contents are irrelevant.
>
> I think this might be a good time to fix the output ...!

Have fixed the hash, but not otherwise changed the output format yet.

> Perhaps create two new files for committers and non-committers; the
> old format file can be deleted later?

I still think it would be worth splitting the output.
Given that lack of issues with the broken file this should be safe enough