[apps-discuss] Should this patch succeed?

Francis Galiegue <fgaliegue@gmail.com> Tue, 12 March 2013 20:03 UTC

MIME-Version: 1.0
Date: Tue, 12 Mar 2013 21:03:30 +0100
Message-ID: <CALcybBDsSpT24QeY_bE48cyp8dR5gGcBd32ev5--3EWh2RoNHw@mail.gmail.com>
From: Francis Galiegue <fgaliegue@gmail.com>
To: Mark Nottingham <mnot@mnot.net>
Content-Type: text/plain; charset="UTF-8"
Cc: apps-discuss@ietf.org
Subject: [apps-discuss] Should this patch succeed?
Precedence: list

On Tue, Mar 12, 2013 at 4:44 PM, Francis Galiegue <fgaliegue@gmail.com> wrote:
> On Tue, Mar 12, 2013 at 3:56 PM, Mark Nottingham <mnot@mnot.net> wrote:
>> Test cases are still more than welcome:
>>   https://github.com/json-patch/json-patch-tests
>>

One test reads something akin to this:

[ { "op": "add", "path": "/0", "value": 1 } ]

With input [], it is expected to give output [ 1 ].

However the spec says:

      [... ]The specified index MUST NOT be greater than the
      number of elements in the array.  If the "-" character is used to
      index the end of the array (see [JSON-Pointer]), this has the
      effect of appending the value to the array.

Strictly speaking, here the 0 in "/0" is not strictly greater than the
number of elements but it is an element which does not exist. And it
would seem that "/-" is the recommended approach here.

So, should this patch succeed? Was it the real intent of the text, or
was is meant to be "the specified index MUST refer to an existing
index in the array. If the character is" etc etc?

--
Francis Galiegue, fgaliegue@gmail.com
JSON Schema in Java: http://json-schema-validator.herokuapp.com

[apps-discuss] Should this patch succeed? Francis Galiegue
Re: [apps-discuss] Should this patch succeed? Paul C. Bryan