Re: [Sbcl-devel] [PATCH] sb-bsd-sockets: Add EWOULDBLOCK error to the list of recognized conditions.

[Mateusz B. <mat...@fa...>]

Like I said earlier, there is no need.

contrib/sb-bsd-sockets/win32-constants.lisp: (:integer EAGAIN 
"WSAEWOULDBLOCK")

Most UNIXES define those to be the same, but not all. Windows is special 
and handled differently.



On 31 May 2016, at 2:26, Stas Boukarev wrote:

> If it's supported, then add it to the windows constants.
>
> On Tue, May 31, 2016 at 5:58 AM, Mateusz Berezecki 
> <mat...@fa...> wrote:
>> On Sun, May 29, 2016, at 11:28, Stas Boukarev wrote:
>>> On Sun, May 29, 2016 at 9:27 PM, Mateusz Berezecki 
>>> <mat...@fa...>
>>> wrote:
>>>>
>>>>
>>>> On 29 May 2016, at 11:21, Stas Boukarev wrote:
>>>>
>>>>> On Sun, May 29, 2016 at 8:53 PM, Mateusz Berezecki 
>>>>> <mat...@fa...>
>>>>> wrote:
>>>>>>
>>>>>> ---
>>>>>>  contrib/sb-bsd-sockets/constants.lisp | 1 +
>>>>>>  contrib/sb-bsd-sockets/sockets.lisp   | 1 +
>>>>>>  2 files changed, 2 insertions(+)
>>>>>
>>>>> That's missing a windows definition.
>>>>
>>>>
>>>> Sorry, don’t have a single Windows machine laying around. Using
>>>> freebsd/linux/osx only at the moment.
>>>> What’s the corresponding Windows definition ?
>>> Probably the same, but it needs to be figured out.
>>
>> After thinking about it I don't think this needs to be supported.
>> Windows supports EWOULDBLOCK via BSD sockets and this is what this
>> package, sb-bsd-sockets, is about. If you want to use Windows 
>> specific
>> errors and API you use Winsock api and that uses WSAEWOULDBLOCK.
>> sb-bsd-sockets is not a winsock api. it is a BSD socket api which is
>> fairly well standardized and even under windows EWOULDBLOCK is 
>> returned.
>>
>> You can take a look at the code samples at Microsoft's own website. 
>> This
>> is what they say:
>> Source:
>> https://msdn.microsoft.com/en-us/library/windows/desktop/ms737828(v=vs.85).aspx
>>
>> Mateusz
>
>
>
> -- 
> With best regards, Stas.