Re: [Mingw-users] mShell: mSys spawning system

[Julien L. <ju...@fa...>]

On 10/03/2006 01:30, Earnie Boyd wrote:
> Quoting Earnie Boyd <ea...@us...>:
> 
>> Quoting Julien Lecomte <ju...@fa...>:
>>>
>>> The current problem with msys ptys is that they haven't the maturity 
>>> of cygwin's ptys; more info towards the bottom of this email.
>>>
>>
>> Uhm, as of about six months or so ago the problems still existed in 
>> Cygwin.  I haven't given a try recently.  But ``rxvt -e 
>> /cygdrive/c/windows/system32/ftp.exe'' should give it a good test.
>>
> 
> And I just grabbed the cygwin1.dll an rxvt.exe and gave it a whurl.  I 
> don't know what you are talking about Julien but Cygwin still has the 
> same issues with pty's as MSYS does.  Perhaps something else is working 
> better.

Ooops... I was just saying that cygwin implemented additional functions 
to deal with pty's. I didn't (yet) look much further to see if they 
improved the current pty functions.

Sorry for the misunderstanding.

But I do believe that the pty functions that msys kept from the fork of 
cygwin have been improved in cygwin itself; when I implemented the 
missing pty functions in a test case msys dll; I couldn't achieve the 
same results when trying to port PuTTYcyg to mSys.

> The issue both Cygwin and MSYS have with pty's is that there is no way 
> to control the child process console handles when the child isn't using 
> their respect runtimes.  STDOUT always blocks until the buffer is full 
> and the buffer size isn't controllable either.

>  If you find a way to do 
> that, then we can fix the pty issue.

And will I get a cookie ? ;-)