Re: json, binary data, and ceph config-key

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Mon, Apr 9, 2018 at 8:39 PM, Sage Weil <sweil@xxxxxxxxxx> wrote:
> Should we
>
> 1- Print something else there even though it is not strictly correct?
> That would mean that something can't parse the (JSON) dump output and
> get the right answer.
>
> 2- Disallow binary data.  This seems weak, but at least behavior would be
> consistent.  We'll probably break some random out-of-tree users out there.
>
> 3- Use XML as the output format for this command?
>
> 4- Remove the 'dump' command entirely and make you get each item.  There
> is still 'ceph config-key list' to get the keys you would need to
> individually get.
>
> All of these options kind of suck.  Is there a better option I'm missing?

5) (I don't know about "better" but:) Encode all strings with e.g.
base64? Then at least the user of the json output must deal with
decoding the strings in a consistent way. Obviously that has the
unfortunate side-effect of making the json output completely human
unreadable.

Or, (6) output values as

    "config-history/1/": {
      "type": "string",
      "value": "bar"
    },
    "config-history/1/": {
      "type": "blob",
      "value": "base64://ABCD=="
    }
    "config-history/2/": {
      "type": "integer",
      "value": 1
    }

(Perhaps a last minute change to add blobs or maybe just dynamically
recognize them.)

> I'm feeling pretty let down by the collective industry decision to use
> JSON for all the things.

Indeed.

-- 
Patrick Donnelly
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html



[Index of Archives]     [CEPH Users]     [Ceph Large]     [Information on CEPH]     [Linux BTRFS]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux