POC Release Announcement for Guilder Build Project

[Mad Andy]

Hi Geeks

I am not quite sure if this is appropriate but I think it would be interesting for my fellow Groovy developers to see other projects in action. This project contains deferred GString evaluation because the data is not available when the GString is created and I did not want to force the user not to use double quotes.

As announced today at the LA-JUG I released the POC of the Guilder Project. You can read about it and also download the installer from:

https://madplanet.com'/trac/guilder

Basically Guilder indents to replace Maven 2 bringing back some of the great features of Maven 1 as well as adding new and hopefully cool features. The project is written in Groovy including the POM, the Plugins and the core code. Guilder builds itself using its own code.

This release is just a proof of concept and an invitation for you guys to give feedback to decide on the future of the project.

Have fun

Andreas Schaefer

CEO of Madplanet.com Inc.andreas.schaefer@...

andreas.schaefer@...

[Mike Dillon-4]

begin Andreas Schaefer (2) quotation:

> As announced today at the LA-JUG I released the POC of the Guilder  
> Project. You can read about it and also download the installer from:
>
> https://madplanet.com/trac/guilder
>
> Basically Guilder indents to replace Maven 2 bringing back some of the  
> great features of Maven 1 as well as adding new and hopefully cool  
> features. The project is written in Groovy including the POM, the  
> Plugins and the core code. Guilder builds itself using its own code.
>
> This release is just a proof of concept and an invitation for you guys  
> to give feedback to decide on the future of the project.

This reminds me of the Graven project:

    http://code.google.com/p/graven/

I heard about it at work from the project lead, Sam Pullara. I don't
know if it was ever announced on this list, but I know that Luke Daley
(the other listed member of the project) is on this list. You guys might
want to collaborate since you seem to be trying to do something similar.

-md

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email

[tugwilson]

Mad Andy wrote:

 This project contains deferred GString evaluation because the  
data is not available when the GString is created and I did not want  
to force the user not to use double quotes.

Hi!

Can you give us a bit more detail about this? It sounds a little like the use of Closures in GStrings which allows you to do things like this:

def a = 1
def s1 = "${a}"
def s2 = "${->a}"
a = 1
println s1
println s2

prints
1
2

i.e. if you embed zero parameter Closures in a GString the closure is evaluated every time the GString is evaluated and the result of the Closure is embedded in the resulting string.

John Wilson

[Bloois, Rene de]

John,

I'm sure you meant to write

a = 2

for the second assignment to a :)

Btw, thx, I didn't know that yet.

Greetz,
René

> -----Original Message-----
> From: tugwilson [mailto:tug@...]
> Sent: woensdag 5 maart 2008 08:26 AM
> To: user@...
> Subject: Re: [groovy-user] POC Release Announcement for
> Guilder Build Project
>
>
>
>
> Mad Andy wrote:
> >
> >  This project contains deferred GString evaluation because
> the data is
> > not available when the GString is created and I did not
> want to force
> > the user not to use double quotes.
> >
>
> Hi!
>
> Can you give us a bit more detail about this? It sounds a
> little like the use of Closures in GStrings which allows you
> to do things like this:
>
> def a = 1
> def s1 = "${a}"
> def s2 = "${->a}"
> a = 1
> println s1
> println s2
>
> prints
> 1
> 2
>
> i.e. if you embed zero parameter Closures in a GString the
> closure is evaluated every time the GString is evaluated and
> the result of the Closure is embedded in the resulting string.
>
> John Wilson
>
>
> --
> View this message in context:
> http://www.nabble.com/POC-Release-Announcement-for-Guilder-Bui

ld-Project-tp15843091p15844627.html

> Sent from the groovy - user mailing list archive at Nabble.com.
>
>
> ---------------------------------------------------------------------
> To unsubscribe from this list, please visit:
>
>     http://xircles.codehaus.org/manage_email
>
>
>
>

This e-mail and any attachment is for authorised use by the intended recipient(s) only. It may contain proprietary material, confidential information and/or be subject to legal privilege. It should not be copied, disclosed to, retained or used by, any other party. If you are not an intended recipient then please promptly delete this e-mail and any attachment and all copies and inform the sender. Thank you.



---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email

[tugwilson]

Bloois, Rene de wrote:

John,

I'm sure you meant to write

a = 2

for the second assignment to a :)

You are quite right, thanks for the correction (I had not had my first coffee when I posted that!)

John Wilson

PS

there is another use of Closures in GStrings which is slightly more specialised.

If you embed a closure with a single parameter then it is called with a Writer as the parameter every time the GSting is evaluated. Characters write to the Writer will be embedded in the resultant String.

This combines with the behaviour of GStings when they are passed to a Writer (the individual elements of the GString are written in order to the Writer rather than evaluating the GString to a String and writing that to a Writer).

The effect is that you can create GStrings which consist of very large amounts of data (gigabytes) but which can be written out without consuming excessive amounts of memory.

One use for this mechanism is to produce a GString which includes the contents of a file without having to read all the file into memory before writing it out.

[Bloois, Rene de]

John,

Cool, good to know that.

René

> -----Original Message-----
> From: tugwilson [mailto:tug@...]
> Sent: woensdag 5 maart 2008 10:49 AM
> To: user@...
> Subject: RE: [groovy-user] POC Release Announcement for
> Guilder Build Project
>
>
>
>
> Bloois, Rene de wrote:
> >
> >
> > John,
> >
> > I'm sure you meant to write
> >
> > a = 2
> >
> > for the second assignment to a :)
> >
>
>
> You are quite right, thanks for the correction (I had not had
> my first coffee when I posted that!)
>
> John Wilson
>
> PS
>
> there is another use of Closures in GStrings which is
> slightly more specialised.
>
> If you embed a closure with a single parameter then it is
> called with a Writer as the parameter every time the GSting
> is evaluated. Characters write to the Writer will be embedded
> in the resultant String.
>
> This combines with the behaviour of GStings when they are
> passed to a Writer (the individual elements of the GString
> are written in order to the Writer rather than evaluating the
> GString to a String and writing that to a Writer).
>
> The effect is that you can create GStrings which consist of
> very large amounts of data (gigabytes) but which can be
> written out without consuming excessive amounts of memory.
>
> One use for this mechanism is to produce a GString which
> includes the contents of a file without having to read all
> the file into memory before writing it out.
>
>
>
>
> --
> View this message in context:
> http://www.nabble.com/POC-Release-Announcement-for-Guilder-Bui
> ld-Project-tp15843091p15846658.html
> Sent from the groovy - user mailing list archive at Nabble.com.
>
>
> ---------------------------------------------------------------------
> To unsubscribe from this list, please visit:
>
>     http://xircles.codehaus.org/manage_email
>
>
>
>

This e-mail and any attachment is for authorised use by the intended recipient(s) only. It may contain proprietary material, confidential information and/or be subject to legal privilege. It should not be copied, disclosed to, retained or used by, any other party. If you are not an intended recipient then please promptly delete this e-mail and any attachment and all copies and inform the sender. Thank you.



---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email

[Mad Andy]

I am not quite sure if we have the same goals. They want to combine Groovy and Maven but I want to create a replacement for Maven 2. I was very critical of Maven 2 even before it was released and still prefer Maven 1 over 2. Now Jelly is not really my scripting language of choice, Maven 1 has no transitive dependency support  and its development has more or less seized. I selected Groovy as scripting language because I can mix Ant statements with scripting code. Later I decide to use Groovy for the entire project rather than just for the plugin scripting.

Nevertheless I will check out their code because my Groovy coding is still more like Java than Groovy.

Thanks

Andreas Schaefer

CEO of Madplanet.com Inc.

andreas.schaefer@...

On Mar 4, 2008, at 9:49 PM, Mike Dillon wrote:

begin Andreas Schaefer (2) quotation:

As announced today at the LA-JUG I released the POC of the Guilder  

Project. You can read about it and also download the installer from:

https://madplanet.com/trac/guilder

Basically Guilder indents to replace Maven 2 bringing back some of the  

great features of Maven 1 as well as adding new and hopefully cool  

features. The project is written in Groovy including the POM, the  

Plugins and the core code. Guilder builds itself using its own code.

This release is just a proof of concept and an invitation for you guys  

to give feedback to decide on the future of the project.

This reminds me of the Graven project:

   http://code.google.com/p/graven/

I heard about it at work from the project lead, Sam Pullara. I don't
know if it was ever announced on this list, but I know that Luke Daley
(the other listed member of the project) is on this list. You guys might
want to collaborate since you seem to be trying to do something similar.

-md

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

   http://xircles.codehaus.org/manage_email

[Russel Winder-4]

Andreas,

On Tue, 2008-03-04 at 20:31 -0800, Andreas Schaefer (2) wrote:

> https://madplanet.com'/trac/guilder
>
>
> Basically Guilder indents to replace Maven 2 bringing back some of the
> great features of Maven 1 as well as adding new and hopefully cool
> features. The project is written in Groovy including the POM, the
> Plugins and the core code. Guilder builds itself using its own code.
>
>
> This release is just a proof of concept and an invitation for you guys
> to give feedback to decide on the future of the project.

Some months ago -- at the Grails eXchange 2007 conference in fact --
Steven Devijver, Hans Dockter, Jochen Theoderou and myself discussed
where to take Gant.  After some thought after that meeting Hans decided
he needed to start afresh without the constraints of the Gant code base
and general philosophy.  The result is Gradle -- Hans (and I guess
Steven) have been hacking away now for a while.

So the question is whether Guilder and Gradle should slug it out in
traditional "survival of the fittest" style or whether it would be
better to collaborate and evolve a Maven/Buildr beater without the need
for blood on the floor?

Gant itself is of course evolving, though progress towards 2.0 is
significantly slower since Steven and Hans opted to not contribute to
Gant but to start Gradle instead :-(

--
Russel.
====================================================
Dr Russel Winder                 Partner

Concertant LLP                   t: [phone]
41 Buckmaster Road,              f: [phone]
London SW11 1EN, UK.             m: [phone]

[Mike Dillon-4]

begin Andreas Schaefer quotation:
> I am not quite sure if we have the same goals. They want to combine
> Groovy and Maven but I want to create a replacement for Maven 2. I was
> very critical of Maven 2 even before it was released and still prefer
> Maven 1 over 2. Now Jelly is not really my scripting language of
> choice, Maven 1 has no transitive dependency support  and its
> development has more or less seized. I selected Groovy as scripting
> language because I can mix Ant statements with scripting code. Later I
> decide to use Groovy for the entire project rather than just for the
> plugin scripting.

Yeah, I realized this once I looked at your code. I think it may still
be possible that there are some shared goals in the area of bringing
together the power of Groovy with a Maven-like system, but the
approaches are certainly different since Graven is limited by Maven 2's
architecture.

-md

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email

[Luke Daley]

On 05/03/2008, at 3:49 PM, Mike Dillon wrote:

> This reminds me of the Graven project:
>
>     http://code.google.com/p/graven/

"Project" is probably a bit strong.

This is just an attempt to allow maven users to define their pom in  
Groovy using builder syntax rather than XML. It doesn't aim to  
replace Maven in any way shape or form.

Another goal is to also allow templating of the pom. For example,  
googlecode projects have a lot of similarities. With Graven you could  
easily specify that this is a GoogleCode project and have the  
relevant pom sections declared for you.

This is not a high priority for myself (or Sam I think) so it's not  
being worked on that actively. That said, it really is quite trivial  
what it is trying to do. Sam spoke to the Maven guys about it and  
they were open to the idea of first class support for this if it  
worked well enough. Which basically would mean having a pom.groovy  
instead of pom.xml and everything would just work like normal.

LD.

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email

[hdockter]

Hi Russel,

On Mar 5, 2008, at 5:44 PM, Russel Winder wrote:

> Andreas,
>
> On Tue, 2008-03-04 at 20:31 -0800, Andreas Schaefer (2) wrote:
>
>> https://madplanet.com'/trac/guilder
>>
>>
>> Basically Guilder indents to replace Maven 2 bringing back some of  
>> the
>> great features of Maven 1 as well as adding new and hopefully cool
>> features. The project is written in Groovy including the POM, the
>> Plugins and the core code. Guilder builds itself using its own code.
>>
>>
>> This release is just a proof of concept and an invitation for you  
>> guys
>> to give feedback to decide on the future of the project.
>
> Some months ago -- at the Grails eXchange 2007 conference in fact --
> Steven Devijver, Hans Dockter, Jochen Theoderou and myself discussed
> where to take Gant.  After some thought after that meeting Hans  
> decided
> he needed to start afresh without the constraints of the Gant code  
> base
> and general philosophy.  The result is Gradle -- Hans (and I guess
> Steven) have been hacking away now for a while.
>
> So the question is whether Guilder and Gradle should slug it out in
> traditional "survival of the fittest" style or whether it would be
> better to collaborate and evolve a Maven/Buildr beater without the  
> need
> for blood on the floor?
>
> Gant itself is of course evolving, though progress towards 2.0 is
> significantly slower since Steven and Hans opted to not contribute to
> Gant but to start Gradle instead :-(

We are just before our first release, waiting desperately for http://
jira.codehaus.org/browse/GROOVY-2643 to be fixed.

I definitely going to have a look at the Guilder code.

As soon as we have released and the community had a first look on  
Gradle I'm looking forward to discussing on how to proceed. Who wants  
to join forces with whom, etc .... It is an exciting time for build  
tools :)

- Hans



---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email

[Russel Winder-4]

Hans,

On Thu, 2008-03-06 at 17:13 +0100, Hans Dockter wrote:

> We are just before our first release, waiting desperately for http://
> jira.codehaus.org/browse/GROOVY-2643 to be fixed.

So I spotted.  That has to be really frustrating.

> I definitely going to have a look at the Guilder code.
>
> As soon as we have released and the community had a first look on  
> Gradle I'm looking forward to discussing on how to proceed. Who wants  
> to join forces with whom, etc .... It is an exciting time for build  
> tools :)

In exactly the same way that Ant and Maven coexist -- if not happily :-)
-- there is room for more than one Groovy-based system to replace them.
Both Ant and Maven are used, have strong followings, and neither is
going to go away -- well not in the near future but maybe we can get rid
of them in the medium term.  I can envisage there being a couple of
standard Groovy-based build systems.  One low-level with great
flexibility, one high level with little or no control needed.  I am
hoping Gant fits the bill of the former now and in the future.  I guess
the question is whether there is room for more than one in the high
level role.  Clearly Buildr is trying to take that space as well, but I
suspect a Groovy-based system has a better chance than a JRuby-based
system.

I guess the question is whether to have lots of new systems and see what
people like, or whether to be more organized and have everyone line up
behind one.  There are pros and cons to both approaches.

With Sun now backing Jython as well as JRuby, the time will come when
SCons becomes a player in the JVM-centric build game.  Sorry, we are
going to loose the J in JVM of course!  However, that is another thread.

--
Russel.
====================================================
Dr Russel Winder                 Partner

Concertant LLP                   t: [phone]
41 Buckmaster Road,              f: [phone]
London SW11 1EN, UK.             m: [phone]

[hdockter]

Hi Russel,

On Mar 6, 2008, at 8:08 PM, Russel Winder wrote:

> Hans,
>
> On Thu, 2008-03-06 at 17:13 +0100, Hans Dockter wrote:
>
>> We are just before our first release, waiting desperately for http://
>> jira.codehaus.org/browse/GROOVY-2643 to be fixed.
>
> So I spotted.  That has to be really frustrating.
>
>> I definitely going to have a look at the Guilder code.
>>
>> As soon as we have released and the community had a first look on
>> Gradle I'm looking forward to discussing on how to proceed. Who wants
>> to join forces with whom, etc .... It is an exciting time for build
>> tools :)
>
> In exactly the same way that Ant and Maven coexist -- if not  
> happily :-)
> -- there is room for more than one Groovy-based system to replace  
> them.
> Both Ant and Maven are used, have strong followings, and neither is
> going to go away -- well not in the near future but maybe we can  
> get rid
> of them in the medium term.  I can envisage there being a couple of
> standard Groovy-based build systems.  One low-level with great
> flexibility, one high level with little or no control needed.  I am
> hoping Gant fits the bill of the former now and in the future.  I  
> guess
> the question is whether there is room for more than one in the high
> level role.  Clearly Buildr is trying to take that space as well,  
> but I
> suspect a Groovy-based system has a better chance than a JRuby-based
> system.

The problem with a build tool focusing only on the high level can be  
seen with ... We know where it can be seen ;)

I think a build tool which offers a build-by-convention approach with  
a Domain model modelling Java/Groovy or whatever projects is very  
important. Gradle offers this. But the build world is complex and  
there will always be situation where the domain model of the build  
tool does not fit and if you only offer the high level you lock  
people in. A major value that is driving Gradle is to offer maximum  
freedom and never to lock people in. We achieve this by having  
clearly separated layers. A low level layer with a general purpose  
language for dependency based programming (like Rake) and of top of  
that we build the higher levels. So our users can always fallback to  
the lower level where they have utmost flexibility. In the complex  
world of builds I think this is a necessity.

- Hans

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email

[Mad Andy]

Hi Russel

Sorry for the late reply but a flew rampaged through my family.

I started Guilder because I love to discuss issues over code than  
white papers because you never know if that works. Now that Guilder is  
out the door we have something to talk about. I will look closer into  
Gradle (just skimmed it for now) in the next few weeks.

I for sure don't want Guilder to be Maven 2 or even Maven 1 coded in  
Groovy. My time and the users' time are too valuable for that. On the  
other hand I am maybe tool blind and only see the Maven 1 way to build  
projects even though I like the mixture Maven 1 provided between  
structure and openness (even though I think the Maven 1 developer team  
never wanted to have it open like that). I want to accomplish a  
similar balance with the primary goal to make developing or just  
enhancing a plugin REALLY DEAD SIMPLE as promised by Maven 2 but they  
failed to do.

So I am always open for discussion even though for now I will keep on  
working on the project too see the pros and cons.

Have fun

Andreas Schaefer
CEO of Madplanet.com Inc.
andreas.schaefer@...


On Mar 5, 2008, at 8:44 AM, Russel Winder wrote:

> Andreas,
>
> On Tue, 2008-03-04 at 20:31 -0800, Andreas Schaefer (2) wrote:
>
>> https://madplanet.com'/trac/guilder
>>
>>
>> Basically Guilder indents to replace Maven 2 bringing back some of  
>> the
>> great features of Maven 1 as well as adding new and hopefully cool
>> features. The project is written in Groovy including the POM, the
>> Plugins and the core code. Guilder builds itself using its own code.
>>
>>
>> This release is just a proof of concept and an invitation for you  
>> guys
>> to give feedback to decide on the future of the project.
>
> Some months ago -- at the Grails eXchange 2007 conference in fact --
> Steven Devijver, Hans Dockter, Jochen Theoderou and myself discussed
> where to take Gant.  After some thought after that meeting Hans  
> decided
> he needed to start afresh without the constraints of the Gant code  
> base
> and general philosophy.  The result is Gradle -- Hans (and I guess
> Steven) have been hacking away now for a while.
>
> So the question is whether Guilder and Gradle should slug it out in
> traditional "survival of the fittest" style or whether it would be
> better to collaborate and evolve a Maven/Buildr beater without the  
> need
> for blood on the floor?
>
> Gant itself is of course evolving, though progress towards 2.0 is
> significantly slower since Steven and Hans opted to not contribute to
> Gant but to start Gradle instead :-(
>
> --
> Russel.
> ====================================================
> Dr Russel Winder                 Partner
>
> Concertant LLP                   t: [phone]
> 41 Buckmaster Road,              f: [phone]
> London SW11 1EN, UK.             m: [phone]

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email

[Luke Daley]

On 08/03/2008, at 4:39 PM, Andreas Schaefer wrote:

> I want to accomplish a similar balance with the primary goal to  
> make developing or just enhancing a plugin REALLY DEAD SIMPLE as  
> promised by Maven 2 but they failed to do.

Not trying to fan the Maven hate, but Jason's work on Groovy  
integration with Maven promises to significantly ease plugin  
development for Maven. FYI.

LD.

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email

[Raphaël]

Yup, i personally tried to create plugins in groovy, and that was
really simplier than java.
But i am biased as i made some jelly plugins as time of maven 1, and
also have made some java plugins for maven 2 :)

Regards,

Raphaël


2008/3/8, Luke Daley <ld@...>:

>
>  On 08/03/2008, at 4:39 PM, Andreas Schaefer wrote:
>
>  > I want to accomplish a similar balance with the primary goal to
>  > make developing or just enhancing a plugin REALLY DEAD SIMPLE as
>  > promised by Maven 2 but they failed to do.
>
>
> Not trying to fan the Maven hate, but Jason's work on Groovy
>  integration with Maven promises to significantly ease plugin
>  development for Maven. FYI.
>
>
>  LD.
>
>
>  ---------------------------------------------------------------------
>  To unsubscribe from this list, please visit:
>
>     http://xircles.codehaus.org/manage_email
>
>
>