You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@roller.apache.org by Glen Mazza <gl...@gmail.com> on 2013/03/31 15:04:43 UTC
Planet source of record? (Was Re: Add planet-web to build process?)
Question below...
On 03/24/2013 09:19 AM, Dave wrote:
> On Sun, Mar 24, 2013 at 6:31 AM, Glen Mazza <gl...@gmail.com> wrote:
>
>> Hi Dave, no problem with separating out planet, but adding planet-web just
>> adds 5 seconds to the build time (roughly 2% longer), at least on my
>> machine, and is good for others looking at the code and for letting us know
>> if/when a planet-web dependency has fallen out of date and is no longer
>> available, so I'd like to reactivate it for the time being:
>
> Those are good points. I'm +0 on keeping the Planet Webapp in the build
> process.
>
>
>
>> But I noticed we have a more fundamental problem--we're presently
>> maintaining two sources of record for the planet source code, something I'd
>> like to rectify ASAP if I can:
>
>> 1.) http://svn.apache.org/viewvc/**roller/planet/core/trunk/<http://svn.apache.org/viewvc/roller/planet/core/trunk/>(Ant-based Planet)
>>
>> 2.) http://svn.apache.org/viewvc/**roller/trunk/<http://svn.apache.org/viewvc/roller/trunk/>(Maven-based Planet)
>>
>> It seems that when #1 was done the source code wasn't deleted in #2; OTOH,
>> if #2 was based on #1 I think I can go ahead and delete #1 to solve the
>> multiple source problem; later, we'll just need to move planet-web and
>> planet-business (and a new pom.xml) to #1's location and then we'll have
>> two separate web apps. (We might keep planet-business in its present
>> location for a longer term due to it being a dependency for Roller right
>> now.)
>>
> Yuck. I did not remember that little mess.
>
Hi Dave, which is the present most accurate source of record for Planet
Web? Neither of the planet WARs generated by the two branches work OOTB
(different errors though).
My read of the situation is that the never-formally-released
roller/trunk/planet-web wasn't fully Mavenized yet (which I can probably
take care of, leveraging what we already have in weblogger-war's pom)
which is why it was commented-out in the Roller root pom.xml, but what
is there (and *not* the Ant version) is nonetheless the most up-to-date
source for the Planet code.
That seems to be the case because planet-web/pom.xml was missing an
important dependency already in weblogger-web/pom.xml
(http://svn.apache.org/viewvc/roller/trunk/planet-web/pom.xml?r1=1462935&r2=1462934&pathrev=1462935),
meaning that couldn't have been working via Maven yet. (Even after
adding the dependency, the new bug report I get from the browser:
*"There is no Action mapped for namespace /planet-ui and action name
homepage. - [unknown location]*
com.opensymphony.xwork2.DefaultActionProxy.prepare(DefaultActionProxy.java:189)
org.apache.struts2.impl.StrutsActionProxy.prepare(StrutsActionProxy.java:61)
org.apache.struts2.impl.StrutsActionProxyFactory.createActionProxy(StrutsActionProxyFactory.java:39)
com.opensymphony.xwork2.DefaultActionProxyFactory.createActionProxy(DefaultActionProxyFactory.java:58)
org.apache.struts2.dispatcher.Dispatcher.serviceAction(Dispatcher.java:488)
) indicates some more hacking is needed to get it to work.
In contrast the war generated in the Ant version at
roller/planet/core/trunk hasn't been updated in 4 years, generates a WAR
called roller-planet-0.45.war, and generally has older dependencies than
what is in the Mavenized version: no Spring JARs, acegi-security
instead of Spring Security, Struts 2.0 vs. Struts 2.2, commons* JARs a
bit older, etc. The struts.xml in the Ant version is also substantively
more complex (I guess because it didn't switch to Spring-web yet) than
the one in planet-web ([1] vs. [2]). Then again, it could be that the
Ant version is the accurate one, it's just the planet-web pom has
dependencies copied over from weblogger-web pom without the underlying
Planet code incurring the architectural upgrade yet.
My goal is to get the Mavenized planet-web working, I'm just not sure
which source code I should be leveraging for that.
Thanks,
Glen
[1]
http://svn.apache.org/viewvc/roller/planet/core/trunk/web/WEB-INF/classes/struts.xml?view=markup
[2]
http://svn.apache.org/viewvc/roller/trunk/planet-web/src/main/resources/struts.xml?&view=markup
>
>> So I'd like to:
>> 1.) Reactivate planet-web in the pom.xml
>> 2.) Delete the source code in #1 (svn delete) so we're at one source of
>> record.
>> 3.) (at a later date) Move planet-web to 1's location, with a new pom.xml
>> and a dependency on planet-business (and possibly test-utils) in Roller.
>> 4.) (if I can sever the planet-business dependency from Roller), move
>> planet-business there too.
>>
> That all sounds good to me.
>
> - Dave
>
Re: Delete Planet-web?
Posted by Anil Gangolli <an...@busybuddha.org>.
+1
On 4/7/13 10:26 AM, Dave wrote:
> +1
>
>
> On Sat, Apr 6, 2013 at 5:48 PM, Glen Mazza <gl...@gmail.com> wrote:
>
>> Hi team, as part of Dave's Maven simplifications I'd like to see our rump
>> code that we have for planet-web (and the modified Ant version kept
>> separately at roller/planet/trunk) deleted from SVN so we can channel our
>> limited time to maintaining and modernizing the Roller application
>> exclusively. Dave has alerted me earlier that Roller already has a
>> Planet-type functionality built into it that we can continue to use and
>> enhanced as desired: http://rollerweblogger.org/**project/page/planet<http://rollerweblogger.org/project/page/planet>
>> .
>>
>> As I understand planet-web was never released, we have two partially
>> working and different versions of it and it's unclear which is the source
>> of record. Further, it is of little practical value to have a solid Planet
>> Web app unless the Roller it relies on is in very good shape, but the
>> amount of time it would take to make the Planet up-to-date would come at
>> the expense of Roller, making Planet irrelevant anyway. Planet can be
>> resurrected in the future should there be enough demand for it and the team
>> grows enough to have people willing to rebuild it.
>>
>> Thoughts?
>>
>> Regards,
>> Glen
>>
>>
>>
>> On 03/31/2013 09:04 AM, Glen Mazza wrote:
>>
>>> Question below...
>>>
>>> On 03/24/2013 09:19 AM, Dave wrote:
>>>
>>>> On Sun, Mar 24, 2013 at 6:31 AM, Glen Mazza<gl...@gmail.com>
>>>> wrote:
>>>>
>>>> Hi Dave, no problem with separating out planet, but adding planet-web
>>>>> just
>>>>> adds 5 seconds to the build time (roughly 2% longer), at least on my
>>>>> machine, and is good for others looking at the code and for letting us
>>>>> know
>>>>> if/when a planet-web dependency has fallen out of date and is no longer
>>>>> available, so I'd like to reactivate it for the time being:
>>>>>
>>>> Those are good points. I'm +0 on keeping the Planet Webapp in the build
>>>> process.
>>>>
>>>>
>>>>
>>>> But I noticed we have a more fundamental problem--we're presently
>>>>> maintaining two sources of record for the planet source code, something
>>>>> I'd
>>>>> like to rectify ASAP if I can:
>>>>> 1.)http://svn.apache.org/**viewvc/**roller/planet/core/**trunk/<http://svn.apache.org/viewvc/**roller/planet/core/trunk/>
>>>>> <http://svn.apache.org/**viewvc/roller/planet/core/**trunk/<http://svn.apache.org/viewvc/roller/planet/core/trunk/>>(Ant-based
>>>>> Planet)
>>>>>
>>>>> 2.)http://svn.apache.org/**viewvc/**roller/trunk/<http://svn.apache.org/viewvc/**roller/trunk/>
>>>>> <http://**svn.apache.org/viewvc/roller/**trunk/<http://svn.apache.org/viewvc/roller/trunk/>>(Maven-based
>>>>> Planet)
>>>>>
>>>>> It seems that when #1 was done the source code wasn't deleted in #2;
>>>>> OTOH,
>>>>> if #2 was based on #1 I think I can go ahead and delete #1 to solve the
>>>>> multiple source problem; later, we'll just need to move planet-web and
>>>>> planet-business (and a new pom.xml) to #1's location and then we'll have
>>>>> two separate web apps. (We might keep planet-business in its present
>>>>> location for a longer term due to it being a dependency for Roller right
>>>>> now.)
>>>>>
>>>>> Yuck. I did not remember that little mess.
>>>>
>>> Hi Dave, which is the present most accurate source of record for Planet
>>> Web? Neither of the planet WARs generated by the two branches work OOTB
>>> (different errors though).
>>>
>>> My read of the situation is that the never-formally-released
>>> roller/trunk/planet-web wasn't fully Mavenized yet (which I can probably
>>> take care of, leveraging what we already have in weblogger-war's pom) which
>>> is why it was commented-out in the Roller root pom.xml, but what is there
>>> (and *not* the Ant version) is nonetheless the most up-to-date source for
>>> the Planet code.
>>>
>>> That seems to be the case because planet-web/pom.xml was missing an
>>> important dependency already in weblogger-web/pom.xml (
>>> http://svn.apache.org/viewvc/**roller/trunk/planet-web/pom.**
>>> xml?r1=1462935&r2=1462934&**pathrev=1462935<http://svn.apache.org/viewvc/roller/trunk/planet-web/pom.xml?r1=1462935&r2=1462934&pathrev=1462935>),
>>> meaning that couldn't have been working via Maven yet. (Even after adding
>>> the dependency, the new bug report I get from the browser:
>>> *"There is no Action mapped for namespace /planet-ui and action name
>>> homepage. - [unknown location]*
>>> com.opensymphony.xwork2.**DefaultActionProxy.prepare(**
>>> DefaultActionProxy.java:189)
>>> org.apache.struts2.impl.**StrutsActionProxy.prepare(**
>>> StrutsActionProxy.java:61)
>>> org.apache.struts2.impl.**StrutsActionProxyFactory.**
>>> createActionProxy(**StrutsActionProxyFactory.java:**39)
>>> com.opensymphony.xwork2.**DefaultActionProxyFactory.**
>>> createActionProxy(**DefaultActionProxyFactory.**java:58)
>>> org.apache.struts2.dispatcher.**Dispatcher.serviceAction(**
>>> Dispatcher.java:488)
>>> ) indicates some more hacking is needed to get it to work.
>>>
>>> In contrast the war generated in the Ant version at
>>> roller/planet/core/trunk hasn't been updated in 4 years, generates a WAR
>>> called roller-planet-0.45.war, and generally has older dependencies than
>>> what is in the Mavenized version: no Spring JARs, acegi-security instead
>>> of Spring Security, Struts 2.0 vs. Struts 2.2, commons* JARs a bit older,
>>> etc. The struts.xml in the Ant version is also substantively more complex
>>> (I guess because it didn't switch to Spring-web yet) than the one in
>>> planet-web ([1] vs. [2]). Then again, it could be that the Ant version is
>>> the accurate one, it's just the planet-web pom has dependencies copied over
>>> from weblogger-web pom without the underlying Planet code incurring the
>>> architectural upgrade yet.
>>>
>>> My goal is to get the Mavenized planet-web working, I'm just not sure
>>> which source code I should be leveraging for that.
>>>
>>> Thanks,
>>> Glen
>>>
>>> [1] http://svn.apache.org/viewvc/**roller/planet/core/trunk/web/**
>>> WEB-INF/classes/struts.xml?**view=markup<http://svn.apache.org/viewvc/roller/planet/core/trunk/web/WEB-INF/classes/struts.xml?view=markup>
>>> [2] http://svn.apache.org/viewvc/**roller/trunk/planet-web/src/**
>>> main/resources/struts.xml?&**view=markup<http://svn.apache.org/viewvc/roller/trunk/planet-web/src/main/resources/struts.xml?&view=markup>
>>>
>>>
>>>
>>>
>>>
>>> So I'd like to:
>>>>> 1.) Reactivate planet-web in the pom.xml
>>>>> 2.) Delete the source code in #1 (svn delete) so we're at one source of
>>>>> record.
>>>>> 3.) (at a later date) Move planet-web to 1's location, with a new
>>>>> pom.xml
>>>>> and a dependency on planet-business (and possibly test-utils) in Roller.
>>>>> 4.) (if I can sever the planet-business dependency from Roller), move
>>>>> planet-business there too.
>>>>>
>>>>> That all sounds good to me.
>>>> - Dave
>>>>
>>>>
Re: Delete Planet-web?
Posted by Dave <sn...@gmail.com>.
+1
On Sat, Apr 6, 2013 at 5:48 PM, Glen Mazza <gl...@gmail.com> wrote:
> Hi team, as part of Dave's Maven simplifications I'd like to see our rump
> code that we have for planet-web (and the modified Ant version kept
> separately at roller/planet/trunk) deleted from SVN so we can channel our
> limited time to maintaining and modernizing the Roller application
> exclusively. Dave has alerted me earlier that Roller already has a
> Planet-type functionality built into it that we can continue to use and
> enhanced as desired: http://rollerweblogger.org/**project/page/planet<http://rollerweblogger.org/project/page/planet>
> .
>
> As I understand planet-web was never released, we have two partially
> working and different versions of it and it's unclear which is the source
> of record. Further, it is of little practical value to have a solid Planet
> Web app unless the Roller it relies on is in very good shape, but the
> amount of time it would take to make the Planet up-to-date would come at
> the expense of Roller, making Planet irrelevant anyway. Planet can be
> resurrected in the future should there be enough demand for it and the team
> grows enough to have people willing to rebuild it.
>
> Thoughts?
>
> Regards,
> Glen
>
>
>
> On 03/31/2013 09:04 AM, Glen Mazza wrote:
>
>> Question below...
>>
>> On 03/24/2013 09:19 AM, Dave wrote:
>>
>>> On Sun, Mar 24, 2013 at 6:31 AM, Glen Mazza<gl...@gmail.com>
>>> wrote:
>>>
>>> Hi Dave, no problem with separating out planet, but adding planet-web
>>>> just
>>>> adds 5 seconds to the build time (roughly 2% longer), at least on my
>>>> machine, and is good for others looking at the code and for letting us
>>>> know
>>>> if/when a planet-web dependency has fallen out of date and is no longer
>>>> available, so I'd like to reactivate it for the time being:
>>>>
>>> Those are good points. I'm +0 on keeping the Planet Webapp in the build
>>> process.
>>>
>>>
>>>
>>> But I noticed we have a more fundamental problem--we're presently
>>>> maintaining two sources of record for the planet source code, something
>>>> I'd
>>>> like to rectify ASAP if I can:
>>>> 1.)http://svn.apache.org/**viewvc/**roller/planet/core/**trunk/<http://svn.apache.org/viewvc/**roller/planet/core/trunk/>
>>>> <http://svn.apache.org/**viewvc/roller/planet/core/**trunk/<http://svn.apache.org/viewvc/roller/planet/core/trunk/>>(Ant-based
>>>> Planet)
>>>>
>>>> 2.)http://svn.apache.org/**viewvc/**roller/trunk/<http://svn.apache.org/viewvc/**roller/trunk/>
>>>> <http://**svn.apache.org/viewvc/roller/**trunk/<http://svn.apache.org/viewvc/roller/trunk/>>(Maven-based
>>>> Planet)
>>>>
>>>> It seems that when #1 was done the source code wasn't deleted in #2;
>>>> OTOH,
>>>> if #2 was based on #1 I think I can go ahead and delete #1 to solve the
>>>> multiple source problem; later, we'll just need to move planet-web and
>>>> planet-business (and a new pom.xml) to #1's location and then we'll have
>>>> two separate web apps. (We might keep planet-business in its present
>>>> location for a longer term due to it being a dependency for Roller right
>>>> now.)
>>>>
>>>> Yuck. I did not remember that little mess.
>>>
>>>
>> Hi Dave, which is the present most accurate source of record for Planet
>> Web? Neither of the planet WARs generated by the two branches work OOTB
>> (different errors though).
>>
>> My read of the situation is that the never-formally-released
>> roller/trunk/planet-web wasn't fully Mavenized yet (which I can probably
>> take care of, leveraging what we already have in weblogger-war's pom) which
>> is why it was commented-out in the Roller root pom.xml, but what is there
>> (and *not* the Ant version) is nonetheless the most up-to-date source for
>> the Planet code.
>>
>> That seems to be the case because planet-web/pom.xml was missing an
>> important dependency already in weblogger-web/pom.xml (
>> http://svn.apache.org/viewvc/**roller/trunk/planet-web/pom.**
>> xml?r1=1462935&r2=1462934&**pathrev=1462935<http://svn.apache.org/viewvc/roller/trunk/planet-web/pom.xml?r1=1462935&r2=1462934&pathrev=1462935>),
>> meaning that couldn't have been working via Maven yet. (Even after adding
>> the dependency, the new bug report I get from the browser:
>> *"There is no Action mapped for namespace /planet-ui and action name
>> homepage. - [unknown location]*
>> com.opensymphony.xwork2.**DefaultActionProxy.prepare(**
>> DefaultActionProxy.java:189)
>> org.apache.struts2.impl.**StrutsActionProxy.prepare(**
>> StrutsActionProxy.java:61)
>> org.apache.struts2.impl.**StrutsActionProxyFactory.**
>> createActionProxy(**StrutsActionProxyFactory.java:**39)
>> com.opensymphony.xwork2.**DefaultActionProxyFactory.**
>> createActionProxy(**DefaultActionProxyFactory.**java:58)
>> org.apache.struts2.dispatcher.**Dispatcher.serviceAction(**
>> Dispatcher.java:488)
>> ) indicates some more hacking is needed to get it to work.
>>
>> In contrast the war generated in the Ant version at
>> roller/planet/core/trunk hasn't been updated in 4 years, generates a WAR
>> called roller-planet-0.45.war, and generally has older dependencies than
>> what is in the Mavenized version: no Spring JARs, acegi-security instead
>> of Spring Security, Struts 2.0 vs. Struts 2.2, commons* JARs a bit older,
>> etc. The struts.xml in the Ant version is also substantively more complex
>> (I guess because it didn't switch to Spring-web yet) than the one in
>> planet-web ([1] vs. [2]). Then again, it could be that the Ant version is
>> the accurate one, it's just the planet-web pom has dependencies copied over
>> from weblogger-web pom without the underlying Planet code incurring the
>> architectural upgrade yet.
>>
>> My goal is to get the Mavenized planet-web working, I'm just not sure
>> which source code I should be leveraging for that.
>>
>> Thanks,
>> Glen
>>
>> [1] http://svn.apache.org/viewvc/**roller/planet/core/trunk/web/**
>> WEB-INF/classes/struts.xml?**view=markup<http://svn.apache.org/viewvc/roller/planet/core/trunk/web/WEB-INF/classes/struts.xml?view=markup>
>> [2] http://svn.apache.org/viewvc/**roller/trunk/planet-web/src/**
>> main/resources/struts.xml?&**view=markup<http://svn.apache.org/viewvc/roller/trunk/planet-web/src/main/resources/struts.xml?&view=markup>
>>
>>
>>
>>
>>
>> So I'd like to:
>>>> 1.) Reactivate planet-web in the pom.xml
>>>> 2.) Delete the source code in #1 (svn delete) so we're at one source of
>>>> record.
>>>> 3.) (at a later date) Move planet-web to 1's location, with a new
>>>> pom.xml
>>>> and a dependency on planet-business (and possibly test-utils) in Roller.
>>>> 4.) (if I can sever the planet-business dependency from Roller), move
>>>> planet-business there too.
>>>>
>>>> That all sounds good to me.
>>>
>>> - Dave
>>>
>>>
>>
>
Delete Planet-web?
Posted by Glen Mazza <gl...@gmail.com>.
Hi team, as part of Dave's Maven simplifications I'd like to see our
rump code that we have for planet-web (and the modified Ant version kept
separately at roller/planet/trunk) deleted from SVN so we can channel
our limited time to maintaining and modernizing the Roller application
exclusively. Dave has alerted me earlier that Roller already has a
Planet-type functionality built into it that we can continue to use and
enhanced as desired: http://rollerweblogger.org/project/page/planet.
As I understand planet-web was never released, we have two partially
working and different versions of it and it's unclear which is the
source of record. Further, it is of little practical value to have a
solid Planet Web app unless the Roller it relies on is in very good
shape, but the amount of time it would take to make the Planet
up-to-date would come at the expense of Roller, making Planet irrelevant
anyway. Planet can be resurrected in the future should there be enough
demand for it and the team grows enough to have people willing to
rebuild it.
Thoughts?
Regards,
Glen
On 03/31/2013 09:04 AM, Glen Mazza wrote:
> Question below...
>
> On 03/24/2013 09:19 AM, Dave wrote:
>> On Sun, Mar 24, 2013 at 6:31 AM, Glen Mazza<gl...@gmail.com> wrote:
>>
>>> Hi Dave, no problem with separating out planet, but adding planet-web just
>>> adds 5 seconds to the build time (roughly 2% longer), at least on my
>>> machine, and is good for others looking at the code and for letting us know
>>> if/when a planet-web dependency has fallen out of date and is no longer
>>> available, so I'd like to reactivate it for the time being:
>> Those are good points. I'm +0 on keeping the Planet Webapp in the build
>> process.
>>
>>
>>
>>> But I noticed we have a more fundamental problem--we're presently
>>> maintaining two sources of record for the planet source code, something I'd
>>> like to rectify ASAP if I can:
>>> 1.)http://svn.apache.org/viewvc/**roller/planet/core/trunk/<http://svn.apache.org/viewvc/roller/planet/core/trunk/>(Ant-based Planet)
>>>
>>> 2.)http://svn.apache.org/viewvc/**roller/trunk/<http://svn.apache.org/viewvc/roller/trunk/>(Maven-based Planet)
>>>
>>> It seems that when #1 was done the source code wasn't deleted in #2; OTOH,
>>> if #2 was based on #1 I think I can go ahead and delete #1 to solve the
>>> multiple source problem; later, we'll just need to move planet-web and
>>> planet-business (and a new pom.xml) to #1's location and then we'll have
>>> two separate web apps. (We might keep planet-business in its present
>>> location for a longer term due to it being a dependency for Roller right
>>> now.)
>>>
>> Yuck. I did not remember that little mess.
>>
>
> Hi Dave, which is the present most accurate source of record for
> Planet Web? Neither of the planet WARs generated by the two branches
> work OOTB (different errors though).
>
> My read of the situation is that the never-formally-released
> roller/trunk/planet-web wasn't fully Mavenized yet (which I can
> probably take care of, leveraging what we already have in
> weblogger-war's pom) which is why it was commented-out in the Roller
> root pom.xml, but what is there (and *not* the Ant version) is
> nonetheless the most up-to-date source for the Planet code.
>
> That seems to be the case because planet-web/pom.xml was missing an
> important dependency already in weblogger-web/pom.xml
> (http://svn.apache.org/viewvc/roller/trunk/planet-web/pom.xml?r1=1462935&r2=1462934&pathrev=1462935),
> meaning that couldn't have been working via Maven yet. (Even after
> adding the dependency, the new bug report I get from the browser:
> *"There is no Action mapped for namespace /planet-ui and action name
> homepage. - [unknown location]*
> com.opensymphony.xwork2.DefaultActionProxy.prepare(DefaultActionProxy.java:189)
> org.apache.struts2.impl.StrutsActionProxy.prepare(StrutsActionProxy.java:61)
> org.apache.struts2.impl.StrutsActionProxyFactory.createActionProxy(StrutsActionProxyFactory.java:39)
> com.opensymphony.xwork2.DefaultActionProxyFactory.createActionProxy(DefaultActionProxyFactory.java:58)
> org.apache.struts2.dispatcher.Dispatcher.serviceAction(Dispatcher.java:488)
> ) indicates some more hacking is needed to get it to work.
>
> In contrast the war generated in the Ant version at
> roller/planet/core/trunk hasn't been updated in 4 years, generates a
> WAR called roller-planet-0.45.war, and generally has older
> dependencies than what is in the Mavenized version: no Spring JARs,
> acegi-security instead of Spring Security, Struts 2.0 vs. Struts 2.2,
> commons* JARs a bit older, etc. The struts.xml in the Ant version is
> also substantively more complex (I guess because it didn't switch to
> Spring-web yet) than the one in planet-web ([1] vs. [2]). Then again,
> it could be that the Ant version is the accurate one, it's just the
> planet-web pom has dependencies copied over from weblogger-web pom
> without the underlying Planet code incurring the architectural upgrade
> yet.
>
> My goal is to get the Mavenized planet-web working, I'm just not sure
> which source code I should be leveraging for that.
>
> Thanks,
> Glen
>
> [1]
> http://svn.apache.org/viewvc/roller/planet/core/trunk/web/WEB-INF/classes/struts.xml?view=markup
> [2]
> http://svn.apache.org/viewvc/roller/trunk/planet-web/src/main/resources/struts.xml?&view=markup
>
>
>
>
>
>>> So I'd like to:
>>> 1.) Reactivate planet-web in the pom.xml
>>> 2.) Delete the source code in #1 (svn delete) so we're at one source of
>>> record.
>>> 3.) (at a later date) Move planet-web to 1's location, with a new pom.xml
>>> and a dependency on planet-business (and possibly test-utils) in Roller.
>>> 4.) (if I can sever the planet-business dependency from Roller), move
>>> planet-business there too.
>>>
>> That all sounds good to me.
>>
>> - Dave
>>
>