Code Review: IronRails2

tfpt review “/shelveset:IronRails2;REDMOND\tomat”

Contains a simple change in DLR.

Ruby changes:
Improves IronRails - switches to IIS7 integrated mode, adds a log,
implements static file handling.
Updates Ruby HAPI.
Fixes writing to text streams (\n should be replaced by \r\n).
Ruby console host exposes the current Ruby engine via “iron_ruby”
variable int the console scope.

Tomas

Ruby and IronRails changes are good.
The call to Handler.HandleFile could go outside of the Rails lock as
there’s no Ruby code invoked.

Now wait just a minute…

:slight_smile:

You can’t casually mention “IronRails changes look good” on the list
without
any indication of what IronRails is. That sounds far too exciting to
pass
up. Can someone explain what this might be to us non MSofties?

-Eric

It’s a top secret experimental ASP.NET HttpHandler that dispatches to a
Rails application :slight_smile:

Tomas

From: [email protected]
[mailto:[email protected]] On Behalf Of Eric N.
Sent: Friday, October 17, 2008 11:48 AM
To: [email protected]
Subject: Re: [Ironruby-core] Code Review: IronRails2

Now wait just a minute…

:slight_smile:

You can’t casually mention “IronRails changes look good” on the list
without any indication of what IronRails is. That sounds far too
exciting to pass up. Can someone explain what this might be to us non
MSofties?

-Eric
On Fri, Oct 17, 2008 at 1:58 PM, Curt H.
<[email protected]mailto:[email protected]> wrote:
Ruby and IronRails changes are good.
The call to Handler.HandleFile could go outside of the Rails lock as
there’s no Ruby code invoked.

It’s an IIS handler to calls Rails directly without going through
Webrick or some other Ruby-based http server. It’s mostly experimental
at this point.

From: [email protected]
[mailto:[email protected]] On Behalf Of Eric N.
Sent: Friday, October 17, 2008 11:48 AM
To: [email protected]
Subject: Re: [Ironruby-core] Code Review: IronRails2

Now wait just a minute…

:slight_smile:

You can’t casually mention “IronRails changes look good” on the list
without any indication of what IronRails is. That sounds far too
exciting to pass up. Can someone explain what this might be to us non
MSofties?

-Eric
On Fri, Oct 17, 2008 at 1:58 PM, Curt H.
<[email protected]mailto:[email protected]> wrote:
Ruby and IronRails changes are good.
The call to Handler.HandleFile could go outside of the Rails lock as
there’s no Ruby code invoked.

Can you tell that we all just got back from lunch? :slight_smile:

From: [email protected]
[mailto:[email protected]] On Behalf Of Tomas M.
Sent: Friday, October 17, 2008 12:52 PM
To: [email protected]
Subject: Re: [Ironruby-core] Code Review: IronRails2

It’s a top secret experimental ASP.NET HttpHandler that dispatches to a
Rails application :slight_smile:

Tomas

From: [email protected]
[mailto:[email protected]] On Behalf Of Eric N.
Sent: Friday, October 17, 2008 11:48 AM
To: [email protected]
Subject: Re: [Ironruby-core] Code Review: IronRails2

Now wait just a minute…

:slight_smile:

You can’t casually mention “IronRails changes look good” on the list
without any indication of what IronRails is. That sounds far too
exciting to pass up. Can someone explain what this might be to us non
MSofties?

-Eric
On Fri, Oct 17, 2008 at 1:58 PM, Curt H.
<[email protected]mailto:[email protected]> wrote:
Ruby and IronRails changes are good.
The call to Handler.HandleFile could go outside of the Rails lock as
there’s no Ruby code invoked.

There’s quite a difference between a development environment like
ASP.NET that’s fully supported by Microsoft and one that we make
available on an as-is basis for the community to play with - and maybe
take ownership of. Our goal has always been to create a fully
conformant Ruby implementation, and by definition, that’s going to be
able to run Rails.

From: [email protected]
[mailto:[email protected]] On Behalf Of Mike M.
Sent: Friday, October 17, 2008 1:04 PM
To: [email protected]
Subject: Re: [Ironruby-core] Code Review: IronRails2

And Scott Guthrie said Microsoft had no interest in enabling Rails to
run in IronRuby… :slight_smile:
On Fri, Oct 17, 2008 at 1:52 PM, Tomas M.
<[email protected]mailto:[email protected]>
wrote:

It’s a top secret experimental ASP.NEThttp://ASP.NET HttpHandler that
dispatches to a Rails application :slight_smile:

Tomas

From:
[email protected]mailto:[email protected]
[mailto:[email protected]mailto:[email protected]]
On Behalf Of Eric N.
Sent: Friday, October 17, 2008 11:48 AM
To: [email protected]mailto:[email protected]
Subject: Re: [Ironruby-core] Code Review: IronRails2

Now wait just a minute…

:slight_smile:

You can’t casually mention “IronRails changes look good” on the list
without any indication of what IronRails is. That sounds far too
exciting to pass up. Can someone explain what this might be to us non
MSofties?

-Eric

On Fri, Oct 17, 2008 at 1:58 PM, Curt H.
<[email protected]mailto:[email protected]> wrote:

Ruby and IronRails changes are good.
The call to Handler.HandleFile could go outside of the Rails lock as
there’s no Ruby code invoked.

And Scott Guthrie said Microsoft had no interest in enabling Rails to
run in
IronRuby… :slight_smile:

On Fri, Oct 17, 2008 at 1:52 PM, Tomas M. <

On Fri, Oct 17, 2008 at 2:09 PM, Curt H.
[email protected]wrote:

There’s quite a difference between a development environment like ASP.NETthat’s fully supported by Microsoft and one that we make available on an
as-is basis for the community to play with – and maybe take ownership of.
Our goal has always been to create a fully conformant Ruby implementation,
and by definition, that’s going to be able to run Rails.

Sure, but the word before was that any ability to run Rails with
IronRuby
was not going to come from Microsoft and would have to be provided by
the
community. This looks to be coming directly from Microsoft. You could
have
left it alone and made us proxy from IIS to webrick/mongrel or use
FastCGI,
but this is evidence of the IronRuby team going above and beyond. You
are to
be commended.

I would like to hear more chatter about this though. What are the
results of
your experiments with IronRails so far?

That’s great! Thanks for the update!
-Eric

On Fri, Oct 17, 2008 at 5:53 PM, Tomas M. <

The result so far is that it is able to dispatch a request to Rails and
send output to client. The shelveset I’ve just submitted implements
static file handling (so that static file requests are not dispatched to
Rails).

Tomas

From: [email protected]
[mailto:[email protected]] On Behalf Of Mike M.
Sent: Friday, October 17, 2008 2:06 PM
To: [email protected]
Subject: Re: [Ironruby-core] Code Review: IronRails2

On Fri, Oct 17, 2008 at 2:09 PM, Curt H.
<[email protected]mailto:[email protected]> wrote:

There’s quite a difference between a development environment like
ASP.NEThttp://ASP.NET that’s fully supported by Microsoft and one that
we make available on an as-is basis for the community to play with - and
maybe take ownership of. Our goal has always been to create a fully
conformant Ruby implementation, and by definition, that’s going to be
able to run Rails.
Sure, but the word before was that any ability to run Rails with
IronRuby was not going to come from Microsoft and would have to be
provided by the community. This looks to be coming directly from
Microsoft. You could have left it alone and made us proxy from IIS to
webrick/mongrel or use FastCGI, but this is evidence of the IronRuby
team going above and beyond. You are to be commended.

I would like to hear more chatter about this though. What are the
results of your experiments with IronRails so far?