Re: [boost] Interest in a Boost.JSON library?

Subject: Re: [boost] Interest in a Boost.JSON library?
From: Kirit Sælensminde (kirit.saelensminde_at_[hidden])
Date: 2008-12-21 21:57:49

Davide Bolcioni wrote:
> On Sunday 21 December 2008 11:49:05 am Kirit Sælensminde wrote:
> 
>> Judging by the documentation[1] the parser seems to target a different
>> use case. Mine includes a full JSON object, rather than the only
>> partially compatible property tree -- for example, mine won't lose null,
>>   true or false type information (so can write back out the exact
>> equivalent JSON after it has been parsed).
> 
> Not having downloaded the code ... would it preserve comments or could it
> be easily extended in order to preserve comments ?

It does not preserve comments, in fact it will throw a parse error if 
comments are encountered. I guess JavaScript style comments are an 
obvious extension, but I can't see that they are part of the standard.

I can't help but think that trying to store comments with the the rest 
of the representation is anything other than opening a can of worms. If 
the JSON is processed then output again what happens to comments?

It seems to me that the preservation of comments is going to be hard 
without a load of examples and use cases, and even then is it going to 
be worth it? Not sure, but for what I use JSON for (mostly web services) 
I can't see it.

That the parser should skip comments that it finds I guess there could 
be a argument for.

K