Re: Boost.Charconv review runs January 15th - 25th, 2024

From: Andrzej Krzemienski (akrzemi1_at_[hidden])
Date: 2024-01-26 06:57:54

śr., 24 sty 2024 o 22:27 Andrzej Krzemienski <akrzemi1_at_[hidden]>
napisał(a):

>
>
> pon., 15 sty 2024 o 11:31 Christopher Kormanyos via Boost <
> boost_at_[hidden]> napisał(a):
>
>> Dear all,
>>
>> The review of Boost.Charconv by Matt Borland runs
>> Monday, January 15th through January 25th, 2024.
>>
>> Code: https://github.com/cppalliance/charconvDocs:
>> https://master.charconv.cpp.al/Review Schedule:
>> https://www.boost.org/community/review_schedule.html
>>
>> Thank you for considering this review.
>>
>> Please optionally provide feedback on the followinggeneral topics:
>>
>> - What is your evaluation of the design?
>> - What is your evaluation of the implementation?
>> - What is your evaluation of the documentation?
>> - What is your evaluation of the potential usefulness
>> of the library?
>> - Did you try to use the library? With what compiler? Did
>> you have any problems?
>> - How much effort did you put into your evaluation?
>> A glance? A quick reading? In-depth study?
>> - Are you knowledgeable about the problem domain?
>>
>> Christopher KormanyosBoost.Charconv Review Manager
>>
>
> Not a review, but some observations about docs.
>
> It looks like `limits` is something that every responsible usage of
> to_chars should use. It should be introduced in all the initial examples of
> to_chars.
>
> https://master.charconv.cpp.al/#limits_examples
> -- these examples do not compile (missing closing square bracket).
>

Let me re-post this question, as I didn't get the answer, but I find it
important in the context of this review:

>
> Are there some requirements on how the numbers are converted to
> characters? How is it that 1.03 is rendered as "1.03" when there are surely
> longer sequences of characters that are closer to the numeric value really
> stored in the object of type `double`.
>

If we treated the requirement to represent the printed number as accurately
as possible, then surely "1.029999998" would be a closer approximation of
the value being stored, no?

Regards,
&rzej;

>
> The docs do not make it clear if a trailing zero is appended to the
> produced output.
> `std::to_chars` does not append the nul-terminator. If boost::to_chars
> does the same, the examples in the docs that use strcmp cannot be right.
> (You are not zeroing the buffer initially).
>
> Generally I feel Boost needs a library that does this (to_chars for
> C++11). I just do not feel qualified to review the implementation.
>
> However, using `malloc` in the corner cases (
> https://github.com/cppalliance/charconv/blob/master/include/boost/charconv/detail/from_chars_float_impl.hpp#L152)
> and using word  "non-allocating" in the docs doesn't seem right. You could
> say in the docs what corner cases these are. Maybe it makes sense to
> allocate in those cases. But you cannot say they are "non-allocating".
>
> Regards,
> &rzej;
>