Re: [ros-dev] Ros-dev Digest, Vol 50, Issue 14

[Olaf Siejka]

If i may say anything, I dont think its a good idea to force NET 3.5
dependency. It`ll reduce the software usability range as not everyone is
willing to upgrade from NET 2.0

Date: Mon, 13 Oct 2008 11:11:16 +0100
From: "gedmurphy" [e-mail]
Subject: Re: [ros-dev] [ros-diffs] [gschneider] 36737: RosDbg Part
       3/3: -  Named pipe implementation based on .net namespace IO.Pipes
   with
       support of threads - Previous win32 test version would  strip random
       characters from debug messages and crash on     entering kdbg
(related t
To: [e-mail]
Message-ID: 001901c92d1c$083d7770$18b86650$@com
Content-Type: text/plain;       charset="utf-8"

I'd originally refrained from doing it this way, using pinvoke instead to
keep .net 2.0 compatibility.
Using classes such as NamedPipeServerStream forces a .net 3.5 dependency. Is
this ok with everyone?

Ged.

[gedmurphy]

I'm all for using .NET 3.5. I think everyone should be using it now.

I only sent this email because I know some people are stubborn to change.

Rethinking it now though, only devs and testers are really gonna be using
this tool

If they haven't upgraded then I wonder what the reason could be. I can't
really think of a good reason not to.

Ged.

From: [e-mail] [mailto:[e-mail]] On
Behalf Of Olaf Siejka
Sent: 14 October 2008 12:01
To: [e-mail]
Subject: Re: [ros-dev] Ros-dev Digest, Vol 50, Issue 14

If i may say anything, I dont think its a good idea to force NET 3.5
dependency. It`ll reduce the software usability range as not everyone is
willing to upgrade from NET 2.0

Date: Mon, 13 Oct 2008 11:11:16 +0100

From: "gedmurphy" [e-mail]
Subject: Re: [ros-dev] [ros-diffs] [gschneider] 36737: RosDbg Part
       3/3: -  Named pipe implementation based on .net namespace IO.Pipes
with
       support of threads - Previous win32 test version would  strip random
       characters from debug messages and crash on     entering kdbg
(related t
To: [e-mail]
Message-ID: 001901c92d1c$083d7770$18b86650$@com
Content-Type: text/plain;       charset="utf-8"

I'd originally refrained from doing it this way, using pinvoke instead to
keep .net 2.0 compatibility.
Using classes such as NamedPipeServerStream forces a .net 3.5 dependency. Is
this ok with everyone?

Ged.

[Aleksey Bragin]

I would also support the idea of moving to 3.5, if it allows getting  
rid of "hacks" (pinvokes) in exchange to good code.

WBR,
Aleksey Bragin.

On Oct 14, 2008, at 3:31 PM, gedmurphy wrote:
> I’m all for using .NET 3.5. I think everyone should be using it now.
>
> I only sent this email because I know some people are stubborn to  
> change.
>
> Rethinking it now though, only devs and testers are really gonna be  
> using this tool
>
> If they haven’t upgraded then I wonder what the reason could be. I  
> can’t really think of a good reason not to.
>
> Ged.
>
> From: [e-mail] [mailto:ros-dev- 
> [e-mail]] On Behalf Of Olaf Siejka
> Sent: 14 October 2008 12:01
> To: [e-mail]
> Subject: Re: [ros-dev] Ros-dev Digest, Vol 50, Issue 14
>
> If i may say anything, I dont think its a good idea to force NET  
> 3.5 dependency. It`ll reduce the software usability range as not  
> everyone is willing to upgrade from NET 2.0
>
> Date: Mon, 13 Oct 2008 11:11:16 +0100
>
> From: "gedmurphy" [e-mail]
> Subject: Re: [ros-dev] [ros-diffs] [gschneider] 36737: RosDbg Part
>        3/3: -  Named pipe implementation based on .net namespace  
> IO.Pipes      with
>        support of threads - Previous win32 test version would   
> strip random
>        characters from debug messages and crash on     entering  
> kdbg (related t
> To: [e-mail]
> Message-ID: 001901c92d1c$083d7770$18b86650$@com
> Content-Type: text/plain;       charset="utf-8"
>
> I'd originally refrained from doing it this way, using pinvoke  
> instead to keep .net 2.0 compatibility.
> Using classes such as NamedPipeServerStream forces a .net 3.5  
> dependency. Is this ok with everyone?
>
> Ged.
>
> Ros-dev mailing list
> [e-mail]
> http://www.reactos.org/mailman/listinfo/ros-dev