Re: [boost] Phoenix 3 Review

Subject: Re: [boost] Phoenix 3 Review
From: Joel de Guzman (joel_at_[hidden])
Date: 2011-03-02 22:13:59

On 3/3/2011 1:49 AM, Thomas Heller wrote:
> On Wed, Mar 2, 2011 at 5:56 PM, Joel Falcou<joel.falcou_at_[hidden]>  wrote:
>> So here is my short review on phoenix 3.
>> I wont comment on the quality of the code and the implementation as I
>> followed Thomas durign his GSoC.
>>
>> On an user perspective, i'll place myself on the "advacned user" plate. i
>> see Phoenix as a great to enable lazy evalaution in a large number of
>> situation. its integration in proto is now making it a nice boilerplate code
>> for more advanced libraries. I've played with test adn examples and found
>> them to be good.
>>
>> My overall opinion is that YES, this version of Phoenix fulfills the
>> requirements of th eprevious review adn thomas addressed them in quite an
>> efficient way.
>
> Thanks.

Thank you, Joel!

>> Now, as an advanced nitpicker ;) i have a small set of request:
>>
>>   - the doc need more examples of proto interoperability either in use cases
>> or in various extension points. IIRC there was sone phoenix->  C cod ein a
>> string example that was never finished, it coudl be good if this could be
>> fleshed out. multistage programming is a Nice technique that need promotion
>> and exposure. Phoenix seems to be perfect for that.
>>
>>   - Could the various compile time / runtime performances tests be itnegrated
>> in the doc if they are a/ done and b/ give some insight. If needed i can run
>> some on non-trivial architecture/compiler and report results
>
> Yes, these topics will be addressed soon.

All these are reasonable requests.

>> So, I very welcome the addition of Phoenix 3 as a full citizen of boost.

Regards,
-- 
Joel de Guzman
http://www.boostpro.com
http://boost-spirit.com