[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Chicken-hackers] [PATCH][for chicken-5] Remove srfi-13
From: |
Felix Winkelmann |
Subject: |
Re: [Chicken-hackers] [PATCH][for chicken-5] Remove srfi-13 |
Date: |
Thu, 11 Sep 2014 20:40:32 +0200 (CEST) |
From: Oleg Kolosov <address@hidden>
Subject: Re: [Chicken-hackers] [PATCH][for chicken-5] Remove srfi-13
Date: Thu, 11 Sep 2014 21:51:27 +0400
> On 09/11/14 20:45, Felix Winkelmann wrote:
>>> I don't see how providing faster alternatives to SRFI-13 procedures is
>>> mutually exclusive with making a cleaner API, unless you're planning on
>>> making completely unsafe versions.
>>
>> Yes (gulp!), I was also thinking about unsafe versions. But I think it
>> is more important to reduce the core API and reduce the number of
>> alternatives, which reduces the mental load on the programmer ("should
>> I use 'string-prefix?' or 'string-starts-with?', and so on...)
>>
>
> These could go to chicken.unsafe.strings or whatever: solves problem
> with naming, no additional mental load - it is easy to guess what you
> will get by using functions from this module.
>
> I prefer to avoid "for internal use only" APIs. It is better to
> modularize further using self descriptive names.
Well, they would be unexposed - no real "API"s. Things like
"##sys#delq", which is very fast (it could even be implemented as a
macro), and used in a sitation, where performance is critical (the
scheduler). Not having to achieve any sort of "usability" is a big win
here. Once such a procedure is "official", you lose any chance for
further tuning.
felix