couchdb-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Adam Kocoloski <kocol...@apache.org>
Subject Re: Integers changed to float in Couch 0.11.0
Date Tue, 13 Apr 2010 01:25:29 GMT
It's a trivial change if the community decides we don't want the mochijson2 behavior.  Do we
know what their rationale was for conversion?

Adam

On Apr 12, 2010, at 5:22 PM, Zachary Zolton wrote:

> LOL... I just got bit by this as well, did anyone ever open an JIRA
> ticket on this, or shall I?
> 
> On Fri, Apr 9, 2010 at 1:54 PM, Paul Davis <paul.joseph.davis@gmail.com> wrote:
>> On Fri, Apr 9, 2010 at 1:13 PM, Filipe David Manana <fdmanana@gmail.com> wrote:
>>> On Fri, Apr 2, 2010 at 3:37 PM, Paul Davis <paul.joseph.davis@gmail.com>wrote:
>>> 
>>>> 
>>>> Well, there's a limit to what can be stored in an integer remember. I
>>>> wouldn't really hold my breath for anything larger than 2147483648
>>>> (2^31). The reason that Futon shows something different is because
>>>> JavaScript treats all numbers as floats and then at display time it'll
>>>> not display a fractional part if it's zero.
>>>> 
>>>> I'm suprised that it used to work. Can anyone else on 0.10 verify
>>>> that? I'm not sure if I remember anything changing in the number
>>>> handling in mochijson or not.
>>>> 
>>>> 
>>> It changed:
>>> 
>>> http://code.google.com/p/mochiweb/source/detail?r=43
>>> 
>>> JSON (and JavaScript) doesn't have a float or integer type. Everything is of
>>> type number. So anything greater than (2^31) - 1 can be represented without
>>> the decimal separator.
>>> 
>>> In JavaScript:
>>> 
>>>>>> n1 = 1270162450695
>>> 1270162450695
>>>>>> n2 = 1270162450695.0
>>> 1270162450695
>>>>>> n1 === n2
>>> true
>>> 
>>> I dunno why mochijson2 is making a distinction between integers vs floats.
>>> 
>> 
>> Oh how awesome. I have no idea why mochijson would do such a thing.
>> The only thing that's ever specified about numbers and limits is in
>> section 4 with the note "An implementation may set limits on the range
>> of numbers."
>> 
>> Blargh.
>> 
>>> 
>>> 
>>>> HTH,
>>>> Paul Davis
>>>> 
>>> 
>>> 
>>> 
>>> --
>>> Filipe David Manana,
>>> fdmanana@gmail.com
>>> 
>>> "Reasonable men adapt themselves to the world.
>>> Unreasonable men adapt the world to themselves.
>>> That's why all progress depends on unreasonable men."
>>> 
>> 


Mime
View raw message