[Seaside-dev] When is Symbol>>#= not Symbol>>#==?

[John O'Keefe]

I guess the question is 'What do consumers of the JSON parser expect when
parsing names?'  I really don't know enough about JSON to argue one way or
the other on which is correct, but if consumers expect Symbols, then
returning Strings will break some code elsewhere.

I will need to retract the fix for Issue 240 to match the new behavior.  But
I guess a new fix for 240 is needed to ensure the JSON parser is really
returning Strings everywhere.

John O'Keefe [|], Principal Smalltalk Architect, Instantiations Inc.


On Fri, Nov 14, 2008 at 1:37 AM, Lukas Renggli <renggli at gmail.com> wrote:

> On Fri, Nov 14, 2008 at 7:00 AM, Philippe Marschall
> <philippe.marschall at gmail.com> wrote:
> > 2008/11/14 Lukas Renggli <renggli at gmail.com>:
> >> Thank you for fixing the test. I was not paying attention to that when
> >> writing this.
> >>
> >> Wouldn't it be simpler if the JSON parser would parse everything as a
> >> string and not convert certain values to a symbol?
> >
> > +1
>
> Name: Javascript-Core-lr.13
> Author: lr
> Time: 14 November 2008, 7:36:49 am
> UUID: fba167ec-7162-421a-85f4-a8019b30ae89
> Ancestors: Javascript-Core-lr.12
>
> - don't parse JSON names to symbols (please check if this passes
> agains Javascript-Tests-Core-jf.15.mcz)
>
> --
> Lukas Renggli
> http://www.lukas-renggli.ch
> _______________________________________________
>  seaside-dev mailing list
> seaside-dev at lists.squeakfoundation.org
> http://lists.squeakfoundation.org/mailman/listinfo/seaside-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.squeakfoundation.org/pipermail/seaside-dev/attachments/20081114/b122cffa/attachment.htm