Re: core in value_core and reference_core

Subject: Re: core in value_core and reference_core
From: Mateusz Loskot (mateusz_at_[hidden])
Date: 2019-03-01 23:38:58

Thanks Christian, I'll take it.

I have ported the value_core and reference_core from the legacy tests
to new modular tests, renamed as pixel_value and pixel_reference
whth the old names as aliases to make it easier to find where it comes from:

https://github.com/boostorg/gil/blob/1af51300f41ffdc04273bf639dd8fc87a5a49674/test/pixel/test_fixture.hpp#L38-L89

Since the test/ directotry contains tests of the core features only,
the `core` preffix/suffix is superfluous. Thus, I renamed those xxx_core
with more descriptive names. (Yeah, I'm anal about clear naming :))

Thanks,
Matz

On 19-02-28 14:39:33, Christian Henning wrote:
>I think "core" just means gil only without any extensions. I might be
>wrong...
>
>On Thu, Feb 28, 2019 at 2:37 PM Mateusz Łoskot <mateusz_at_[hidden]> wrote:
>
>> Hi,
>>
>> In two legacy tests, channel.cpp [1] and pixel.cpp [2], there are so called
>> 'core' classes defined, with this comment in [1]:
>>
>> "// Different core classes depending on the different types of..." [1]
>>
>> I'm (as not a native speaker) not clear what is the semantic of the word
>> 'core'.
>> Is this the same as test fixture? Or, an archetype, prime example, a mold?
>>
>> I have a feeling that provenance of the 'core' name is historic like in
>> 'core
>> dump' [3]. There must be a good reason why the original GIL authors chose
>> to use
>> 'core' in those classes.
>>
>>
>> [1] https://github.com/boostorg/gil/blob/develop/test/legacy/channel.cpp
>> [2] https://github.com/boostorg/gil/blob/develop/test/legacy/pixel.cpp
>> [3] https://en.wikipedia.org/wiki/Core_dump
>>
>> Best regards,
>> --
>> Mateusz Loskot, http://mateusz.loskot.net
>> --
>> Boost-gil mailing list
>> Boost-gil_at_[hidden]
>> https://lists.boost.org/mailman/listinfo.cgi/boost-gil
>>
>-- 
>Boost-gil mailing list
>Boost-gil_at_[hidden]
>https://lists.boost.org/mailman/listinfo.cgi/boost-gil

-- 
Mateusz Loskot, http://mateusz.loskot.net