You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@velocity.apache.org by Geir Magnusson Jr <ge...@4quarters.com> on 2004/01/22 10:57:51 UTC
roadmap proposal
Here's a summary of release ideas that I thought I'd post for comment :
1) 1.4 : push out the door.
2) 1.5 : do a rc immediately of what we have already.
3) target 1.6 : Add the ability to have null in the context and let
RHS = null -> LHS = null in #set() as a switched on behavior, as well
as any other little tweaks and fixes we come across along the way
4) target 1.7 : flip the switch to go the other way, where LHS = null
is the default behavior
At this point, either the 1.6 or 1.7 branch remains as the 1.x branch
and HEAD is working to
5) target 2.0 : the <deep_voice>float patch</deep_voice>
I'd like to optionally call it "3.1" for historical humor value...
geir
--
Geir Magnusson Jr 203-247-1713(m)
geir@4quarters.com
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: roadmap proposal
Posted by Geir Magnusson Jr <ge...@4quarters.com>.
On Jan 23, 2004, at 5:36 PM, Charles N. Harvey III wrote:
> Sooo...
> Which version do I need to do:
>
> #set( $monkey.Say = ["Not",
> $my,
> "Fault"] )
>
>
>
1.5
--
Geir Magnusson Jr 203-247-1713(m)
geir@4quarters.com
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: roadmap proposal
Posted by "Charles N. Harvey III" <ch...@alloy.com>.
Sooo...
Which version do I need to do:
#set( $monkey.Say = ["Not",
$my,
"Fault"] )
Charlie
Geir Magnusson Jr wrote:
> Here's a summary of release ideas that I thought I'd post for comment :
>
> 1) 1.4 : push out the door.
>
> 2) 1.5 : do a rc immediately of what we have already.
>
> 3) target 1.6 : Add the ability to have null in the context and let
> RHS = null -> LHS = null in #set() as a switched on behavior, as well
> as any other little tweaks and fixes we come across along the way
>
> 4) target 1.7 : flip the switch to go the other way, where LHS = null
> is the default behavior
>
> At this point, either the 1.6 or 1.7 branch remains as the 1.x branch
> and HEAD is working to
>
> 5) target 2.0 : the <deep_voice>float patch</deep_voice>
>
> I'd like to optionally call it "3.1" for historical humor value...
>
> geir
>
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: roadmap proposal
Posted by Simon Christian <si...@stoutstick.com>.
I'd concur with that, the RHS=null change is a good thing IMO, and given
the relative levels of of support vs. opposition seems a shame not to
implement soon.
- simon
Christoph Reck wrote:
> Could we have the RHS=null in 1.5 since it is a small change (with a
> default BC switch)?
>
> The past has shown that releases are rare. I'm afraid if you do a 1.5
> and 1.6, you might lock some users to it that don't have the momentum
> to upgrade later (to be honest, I rarely update my vel jars, since the
> required workarounds are already implemeted they and thus work for
> me as they are). The result will be bug reports on 1.5 which will
> mostly receive the answer 'update to the latest velocity!'
>
> Cheers,
> Christoph
>
> Geir Magnusson Jr wrote:
>
>> Here's a summary of release ideas that I thought I'd post for comment :
>>
>> 1) 1.4 : push out the door.
>>
>> 2) 1.5 : do a rc immediately of what we have already.
>>
>> 3) target 1.6 : Add the ability to have null in the context and let
>> RHS = null -> LHS = null in #set() as a switched on behavior, as well
>> as any other little tweaks and fixes we come across along the way
>>
>> 4) target 1.7 : flip the switch to go the other way, where LHS = null
>> is the default behavior
>>
>> At this point, either the 1.6 or 1.7 branch remains as the 1.x branch
>> and HEAD is working to
>>
>> 5) target 2.0 : the <deep_voice>float patch</deep_voice>
>>
>> I'd like to optionally call it "3.1" for historical humor value...
>>
>> geir
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: velocity-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: velocity-dev-help@jakarta.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-dev-help@jakarta.apache.org
Re: roadmap proposal
Posted by Geir Magnusson Jr <ge...@4quarters.com>.
On Jan 22, 2004, at 6:50 AM, Christoph Reck wrote:
> Could we have the RHS=null in 1.5 since it is a small change (with a
> default BC switch)?
>
> The past has shown that releases are rare. I'm afraid if you do a 1.5
> and 1.6, you might lock some users to it that don't have the momentum
> to upgrade later (to be honest, I rarely update my vel jars, since the
> required workarounds are already implemeted they and thus work for
> me as they are). The result will be bug reports on 1.5 which will
> mostly receive the answer 'update to the latest velocity!'
I thought about that, but as this is a big change, I'm afraid it would
delay the 1.5 because we want to touch and test a few things.
I'm trying to fix that 'releases are rare' issue by stating now what
we'd do.
And I have no sympathy if we do the releases, making sure we're
careful, and then people don't want to upgrade... :)
geir
>
> Cheers,
> Christoph
>
> Geir Magnusson Jr wrote:
>> Here's a summary of release ideas that I thought I'd post for comment
>> :
>> 1) 1.4 : push out the door.
>> 2) 1.5 : do a rc immediately of what we have already.
>> 3) target 1.6 : Add the ability to have null in the context and let
>> RHS = null -> LHS = null in #set() as a switched on behavior, as well
>> as any other little tweaks and fixes we come across along the way
>> 4) target 1.7 : flip the switch to go the other way, where LHS = null
>> is the default behavior
>> At this point, either the 1.6 or 1.7 branch remains as the 1.x branch
>> and HEAD is working to
>> 5) target 2.0 : the <deep_voice>float patch</deep_voice>
>> I'd like to optionally call it "3.1" for historical humor value...
>> geir
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: velocity-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: velocity-dev-help@jakarta.apache.org
>
>
--
Geir Magnusson Jr 203-247-1713(m)
geir@4quarters.com
--
Geir Magnusson Jr 203-247-1713(m)
geir@4quarters.com
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-dev-help@jakarta.apache.org
Re: roadmap proposal
Posted by Christoph Reck <ap...@recks.org>.
Could we have the RHS=null in 1.5 since it is a small change (with a
default BC switch)?
The past has shown that releases are rare. I'm afraid if you do a 1.5
and 1.6, you might lock some users to it that don't have the momentum
to upgrade later (to be honest, I rarely update my vel jars, since the
required workarounds are already implemeted they and thus work for
me as they are). The result will be bug reports on 1.5 which will
mostly receive the answer 'update to the latest velocity!'
Cheers,
Christoph
Geir Magnusson Jr wrote:
> Here's a summary of release ideas that I thought I'd post for comment :
>
> 1) 1.4 : push out the door.
>
> 2) 1.5 : do a rc immediately of what we have already.
>
> 3) target 1.6 : Add the ability to have null in the context and let RHS
> = null -> LHS = null in #set() as a switched on behavior, as well as any
> other little tweaks and fixes we come across along the way
>
> 4) target 1.7 : flip the switch to go the other way, where LHS = null is
> the default behavior
>
> At this point, either the 1.6 or 1.7 branch remains as the 1.x branch
> and HEAD is working to
>
> 5) target 2.0 : the <deep_voice>float patch</deep_voice>
>
> I'd like to optionally call it "3.1" for historical humor value...
>
> geir
>
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-dev-help@jakarta.apache.org
Re: Separe Istance for path loader Templates
Posted by Geir Magnusson Jr <ge...@4quarters.com>.
On Apr 26, 2004, at 5:40 AM, Sara Frongillo wrote:
> Hi,
>
> I have a problem to load dynamic templates path, so I have a Velocity
> Servlet with method
> Properties loadConfiguration(ServletConfig config){
> .......................
> p.setProperty(Velocity.FILE_RESOURCE_LOADER_PATH, path);
>
> }
> I need to load different path for template about a object value in the
> session so I need to set path template in the handleRequest method so
> How can I do??
>
If I understand you correctly, you are going to get into *heaps* of
trouble with this once you have more than one concurrent user, because
the servlet is shared among multiple threads. So you don't want to try
and reconfigure velocity's resource management on a per-request basis,
because you'll be really surprised under load :)
One common solution is to have a root from which all can be seen by the
loader :
root/
foo/
index.vm
bar/
index.vm
and then based on request
Template t = vel.getTemplate( userPath + "/" + templateName);
for example.
There are other solutions, such as using the non-supported
URLResourceLoader or other like approaches.
geir
> Thanks in advance
>
> Sara
>
>
> -------------------------------
> Sara Frongillo
> Software Area Executive
>
> I.T.Sys srl
> C.so Garibaldi, 215
> 84125 Salerno - Italy
>
> Office +390892580977
> Mobile +393280087824
> Email s.frongillo@itsys.it
> Web http://www.nexsoft.it
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: velocity-user-help@jakarta.apache.org
>
>
--
Geir Magnusson Jr 203-247-1713(m)
geir@4quarters.com
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Separe Istance for path loader Templates
Posted by Sara Frongillo <s....@virgilio.it>.
Hi,
I have a problem to load dynamic templates path, so I have a Velocity
Servlet with method
Properties loadConfiguration(ServletConfig config){
.......................
p.setProperty(Velocity.FILE_RESOURCE_LOADER_PATH, path);
}
I need to load different path for template about a object value in the
session so I need to set path template in the handleRequest method so
How can I do??
Thanks in advance
Sara
-------------------------------
Sara Frongillo
Software Area Executive
I.T.Sys srl
C.so Garibaldi, 215
84125 Salerno - Italy
Office +390892580977
Mobile +393280087824
Email s.frongillo@itsys.it
Web http://www.nexsoft.it
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: roadmap proposal
Posted by Tim Pizey <ti...@paneris.org>.
On Thursday 22 January 2004 9:57 am, Geir Magnusson Jr wrote:
>
> I'd like to optionally call it "3.1" for historical humor value...
It'll never get better if you pick it!
TimP
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: roadmap proposal
Posted by Nathan Bubna <na...@esha.com>.
Will Glass-Husain said:
...
[plenty of good ideas]
...
> I don't know if it's worth debating the specifics right now, but I suggest
> resolving these issues by 2.0.
...
considering that there are a number of milestones on that roadmap prior to
2.0, i do not think it is worth debating the specifics yet. personally, i
suggest resolving these issues not only "by 2.0", but after 1.5 or 1.6 final
are released.
essentially, i see one of the primary benefits of roadmaps as being the
establishment of priorities. focus is a good thing. :) also, the earlier we
get into mailing list dicussions, the more likely it is that we will repeat
most of it once it comes time to actually make the decision. as such, i think
any 2.0 discussion would be better done in Wiki at this stage. there it
should be both more permanent and less of a distraction.
Will, since you've clearly got the itch, would you care to start the
scratching? :)
http://nagoya.apache.org/wiki/apachewiki.cgi?VelocityProjectPages
Nathan Bubna
nathan@esha.com
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-dev-help@jakarta.apache.org
Re: roadmap proposal
Posted by Mike Kienenberger <mk...@alaska.net>.
Will Glass-Husain <wg...@forio.com> wrote:
> (1) optional braces around directives, e.g. #{else}
>
> Which removes the problem of
> #if($test)something#elsesomethingdifferent#end
>
> patch available:
> http://nagoya.apache.org/bugzilla/show_bug.cgi?id=7189
And makes it possible to start supporting alternate separators.
Right now, trying to generate a velocity template of a velocity template is
painful.
(My use case is transforming a database entity into struts web page
components:
Entity + Velocity templates => InitializeEntityAction.java,
ProcessEntityAction.java, EntityPage.vm, Entity-struts-config.xml)
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-dev-help@jakarta.apache.org
Re: roadmap proposal
Posted by Will Glass-Husain <wg...@forio.com>.
Seems great. Glad to see some momentum.
How about considering other mostly backwards compatible changes to syntax
for 2.0.
(1) optional braces around directives, e.g. #{else}
Which removes the problem of
#if($test)something#elsesomethingdifferent#end
patch available:
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=7189
(2) commas in macro arguments ( spaces can be confusing )
(3) revision to #stop behavior. Current behavior is mentioned occasionally
in list as confusing and not useful. Patch available:
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=24189
(4) array.length
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=25007
There's also half a dozen minor patches in Bugzilla that look simple and
backwards compatible. For example: Anakia custom context (21417), absolute
file names in FileResourceLoader (17379), a new StringResourceLoader
(20677) plus a few bug fixes.
I don't know if it's worth debating the specifics right now, but I suggest
resolving these issues by 2.0.
Best, WILL
----- Original Message -----
From: "Geir Magnusson Jr" <ge...@4quarters.com>
To: "Velocity Developers List" <ve...@jakarta.apache.org>
Sent: Thursday, January 22, 2004 9:44 AM
Subject: Re: roadmap proposal
>
> On Jan 22, 2004, at 12:25 PM, Nathan Bubna wrote:
>
> > Geir Magnusson Jr said:
> >> Here's a summary of release ideas that I thought I'd post for comment
> >> :
> >>
> >> 1) 1.4 : push out the door.
> >>
> >> 2) 1.5 : do a rc immediately of what we have already.
> >>
> >> 3) target 1.6 : Add the ability to have null in the context and let
> >> RHS = null -> LHS = null in #set() as a switched on behavior, as well
> >> as any other little tweaks and fixes we come across along the way
> >
> > such as removing VelocityServlet? (it's already deprecated in 1.5) :)
>
> If you'd like, sure. That and maybe some of the event handler stuff
> suggested by Will.
>
> >
> >> 4) target 1.7 : flip the switch to go the other way, where LHS = null
> >> is the default behavior
> >
> > awful small change for minor version number. i'm inclined to agree
> > with
> > Christoph on this, but i don't care enough to make a fuss.
>
>
> That's not what he's disagreeing w/. He's disagreeing w/ not rolling
> 1.6 stuff into 1.5. My only concern is that we want the 1.6 changes
> (#set) to be foolproof and 100% compatible, and the stuff in 1.5 has
> been there long enough to know if we have any major problems. This
> addresses the rate o' release criticism.
> >
> >> At this point, either the 1.6 or 1.7 branch remains as the 1.x branch
> >> and HEAD is working to
> >>
> >> 5) target 2.0 : the <deep_voice>float patch</deep_voice>
> >>
> >> I'd like to optionally call it "3.1" for historical humor value...
> >
> > and i assume we will throw BC concerns to the wind for 2.0? :) i hope
> > so.
> > and if we do, maybe the proposed 1.7 switch would actually fit better
> > here.
>
> Well, lets not say "to the wind" :) But yes - if there are changes
> that make big breakage, we want to ensure that it's clear to people,
> and that we can continue the 1.x branch if we want to for people that
> can't change immediately.
>
> >
> > overall, i give the roadmap a +0.99
> >
> > :)
> >
>
> Cool beans! Thanks.
>
> geir
>
> > Nathan Bubna
> > nathan@esha.com
> >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: velocity-dev-unsubscribe@jakarta.apache.org
> > For additional commands, e-mail: velocity-dev-help@jakarta.apache.org
> >
> >
> --
> Geir Magnusson Jr 203-247-1713(m)
> geir@4quarters.com
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: velocity-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: velocity-dev-help@jakarta.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-dev-help@jakarta.apache.org
Re: roadmap proposal
Posted by Geir Magnusson Jr <ge...@4quarters.com>.
On Jan 22, 2004, at 12:25 PM, Nathan Bubna wrote:
> Geir Magnusson Jr said:
>> Here's a summary of release ideas that I thought I'd post for comment
>> :
>>
>> 1) 1.4 : push out the door.
>>
>> 2) 1.5 : do a rc immediately of what we have already.
>>
>> 3) target 1.6 : Add the ability to have null in the context and let
>> RHS = null -> LHS = null in #set() as a switched on behavior, as well
>> as any other little tweaks and fixes we come across along the way
>
> such as removing VelocityServlet? (it's already deprecated in 1.5) :)
If you'd like, sure. That and maybe some of the event handler stuff
suggested by Will.
>
>> 4) target 1.7 : flip the switch to go the other way, where LHS = null
>> is the default behavior
>
> awful small change for minor version number. i'm inclined to agree
> with
> Christoph on this, but i don't care enough to make a fuss.
That's not what he's disagreeing w/. He's disagreeing w/ not rolling
1.6 stuff into 1.5. My only concern is that we want the 1.6 changes
(#set) to be foolproof and 100% compatible, and the stuff in 1.5 has
been there long enough to know if we have any major problems. This
addresses the rate o' release criticism.
>
>> At this point, either the 1.6 or 1.7 branch remains as the 1.x branch
>> and HEAD is working to
>>
>> 5) target 2.0 : the <deep_voice>float patch</deep_voice>
>>
>> I'd like to optionally call it "3.1" for historical humor value...
>
> and i assume we will throw BC concerns to the wind for 2.0? :) i hope
> so.
> and if we do, maybe the proposed 1.7 switch would actually fit better
> here.
Well, lets not say "to the wind" :) But yes - if there are changes
that make big breakage, we want to ensure that it's clear to people,
and that we can continue the 1.x branch if we want to for people that
can't change immediately.
>
> overall, i give the roadmap a +0.99
>
> :)
>
Cool beans! Thanks.
geir
> Nathan Bubna
> nathan@esha.com
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: velocity-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: velocity-dev-help@jakarta.apache.org
>
>
--
Geir Magnusson Jr 203-247-1713(m)
geir@4quarters.com
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-dev-help@jakarta.apache.org
Re: roadmap proposal
Posted by Nathan Bubna <na...@esha.com>.
Geir Magnusson Jr said:
> Here's a summary of release ideas that I thought I'd post for comment :
>
> 1) 1.4 : push out the door.
>
> 2) 1.5 : do a rc immediately of what we have already.
>
> 3) target 1.6 : Add the ability to have null in the context and let
> RHS = null -> LHS = null in #set() as a switched on behavior, as well
> as any other little tweaks and fixes we come across along the way
such as removing VelocityServlet? (it's already deprecated in 1.5) :)
> 4) target 1.7 : flip the switch to go the other way, where LHS = null
> is the default behavior
awful small change for minor version number. i'm inclined to agree with
Christoph on this, but i don't care enough to make a fuss.
> At this point, either the 1.6 or 1.7 branch remains as the 1.x branch
> and HEAD is working to
>
> 5) target 2.0 : the <deep_voice>float patch</deep_voice>
>
> I'd like to optionally call it "3.1" for historical humor value...
and i assume we will throw BC concerns to the wind for 2.0? :) i hope so.
and if we do, maybe the proposed 1.7 switch would actually fit better here.
overall, i give the roadmap a +0.99
:)
Nathan Bubna
nathan@esha.com
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-dev-help@jakarta.apache.org
Re: roadmap proposal
Posted by "Charles N. Harvey III" <ch...@alloy.com>.
Sooo...
Which version do I need to do:
#set( $monkey.Say = ["Not",
$my,
"Fault"] )
Charlie
Geir Magnusson Jr wrote:
> Here's a summary of release ideas that I thought I'd post for comment :
>
> 1) 1.4 : push out the door.
>
> 2) 1.5 : do a rc immediately of what we have already.
>
> 3) target 1.6 : Add the ability to have null in the context and let
> RHS = null -> LHS = null in #set() as a switched on behavior, as well
> as any other little tweaks and fixes we come across along the way
>
> 4) target 1.7 : flip the switch to go the other way, where LHS = null
> is the default behavior
>
> At this point, either the 1.6 or 1.7 branch remains as the 1.x branch
> and HEAD is working to
>
> 5) target 2.0 : the <deep_voice>float patch</deep_voice>
>
> I'd like to optionally call it "3.1" for historical humor value...
>
> geir
>
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org