Re: [Treebase-devel] Mesquite troubles

[Hilmar L. <hl...@ne...>]

Great!! Sounds like we're making progress wading through the mucky  
waters :-)

	-hilmar

On Dec 16, 2009, at 1:04 PM, Rutger Vos wrote:

> Hi all,
>
> Wayne's suggestion to pass mesquite's location programmatically works.
> I just managed to get it to parse a file through the web interface
> (and so this should also apply to the batch loader, which uses the
> same NexusServiceMesquite object). I still need to replicate the
> behaviour on the server, which means creating a bit of a Frankenstein
> system, unfortunately:
>
> * use the mesquite-2.01.tb.jar to compile against
>
> * place the class folder structure in treebase-core/lib in a location
> where tomcat sees it (e.g. in $CATALINA_HOME/shared/classes)
>
> * have the mesquite.folder_dir variable in jdbc.properties point to
> that location
>
> ...then the NexusServiceMesquite bean will be passed that location on
> instantiation, from where it can tell MesquiteModule where it lives.
>
> Phew.
>
> Rutger
>
> On Wed, Dec 16, 2009 at 10:02 AM, Rutger Vos <rut...@gm...>  
> wrote:
>> Yes, you are both right: the mesquite jar is "needed", but unless you
>> clean up your work environment (local repo) it will stick around and
>> so you won't notice.
>>
>> Unfortunately, leaving it in won't help us because the issue is that
>> the jar only works at compile time in that you can build against it,
>> but at runtime mesquite won't work inside a jar (because of its own
>> extension loading and bootstrapping implementation).
>>
>> What I'm doing instead is now committing the unpacked class folder  
>> for
>> mesquite into /treebase-core/lib, and setting my eclipse .project to
>> use that, which should amount to the same thing as having the jar  
>> file
>> in your classpath.
>>
>> From this point, we have to figure out how to correctly deploy this
>> class folder into our web application. I've tried placing it in
>> $CATALINA_HOME/shared/classes but it's not quite happy with that  
>> (same
>> error as before).
>>
>> Next I'll try and see what happens if I place it directly inside the
>> folder structure that is created when tomcat unpacks the
>> treebase-web.war. Not an ideal solution of course, but at this  
>> point I
>> just want to narrow down the variables.
>>
>> Rutger
>>
>> On Tue, Dec 15, 2009 at 10:57 PM, youjun guo <you...@ya...>  
>> wrote:
>>> Rutger,
>>>
>>> I Just noticed the mesquite jar, which was commented by you in the  
>>> pom.xml
>>> of treebase-core, is actually needed for the project. Sometimes  
>>> you can get
>>> the system compiled without it is because it already got  
>>> downloaded to the
>>> classpath,
>>>
>>> But for a newly check out project, it will not compile without it.
>>>
>>> Youjun
>>>
>>> On Tue, Dec 15, 2009 at 11:34 AM, Rutger Vos  
>>> <rut...@gm...> wrote:
>>>>
>>>> Hi all,
>>>>
>>>> I'm still having trouble getting mesquite to play nice, so this  
>>>> email
>>>> is a cry for help!
>>>>
>>>> I'm hoping that Vladimir and/or Youjun can give this some thought  
>>>> as
>>>> well, because maybe I'm just overlooking something obvious. What
>>>> continues to happen is that mesquite can't figure out where it is
>>>> located when I run it inside tomcat. It then tries to ask the user
>>>> where the mesquite manual is located (so that it can compute the
>>>> correct path relative to the manual file), but because in headless
>>>> mode we default to "OK" on all user queries everything breaks down
>>>> from then on.
>>>>
>>>> Here's what I've done recently:
>>>>
>>>> * I added anything that seems pertinent w.r.t. mesquite <=>  
>>>> treebase
>>>> interaction to the wiki page:
>>>>
>>>> https://sourceforge.net/apps/mediawiki/treebase/index.php?title=HeadlessMesquite
>>>>
>>>> * I committed the treebase class files to /treebase-core/lib
>>>>
>>>> * I removed all reference to mesquite from the pom.xml files
>>>> (apologies for adding them in the first place, I was hoping we  
>>>> could
>>>> integrate it in the maven system)
>>>>
>>>> * I emailed Jin Ruan, Peter Midford, Mark-Jason Dominus and Wayne
>>>> Maddison.
>>>>
>>>> Is Peter at nescent right now? Could someone knock on his door and
>>>> plead for his help? He knows a lot about mesquite!
>>>>
>>>> Thanks,
>>>>
>>>> Rutger
>>>>
>>>> --
>>>> Dr. Rutger A. Vos
>>>> School of Biological Sciences
>>>> Philip Lyle Building, Level 4
>>>> University of Reading
>>>> Reading
>>>> RG6 6BX
>>>> United Kingdom
>>>> Tel: +44 (0) 118 378 7535
>>>> http://www.nexml.org
>>>> http://rutgervos.blogspot.com
>>>>
>>>>
>>>> ------------------------------------------------------------------------------
>>>> Return on Information:
>>>> Google Enterprise Search pays you back
>>>> Get the facts.
>>>> http://p.sf.net/sfu/google-dev2dev
>>>> _______________________________________________
>>>> Treebase-devel mailing list
>>>> Tre...@li...
>>>> https://lists.sourceforge.net/lists/listinfo/treebase-devel
>>>
>>>
>>
>>
>>
>> --
>> Dr. Rutger A. Vos
>> School of Biological Sciences
>> Philip Lyle Building, Level 4
>> University of Reading
>> Reading
>> RG6 6BX
>> United Kingdom
>> Tel: +44 (0) 118 378 7535
>> http://www.nexml.org
>> http://rutgervos.blogspot.com
>>
>
>
>
> -- 
> Dr. Rutger A. Vos
> School of Biological Sciences
> Philip Lyle Building, Level 4
> University of Reading
> Reading
> RG6 6BX
> United Kingdom
> Tel: +44 (0) 118 378 7535
> http://www.nexml.org
> http://rutgervos.blogspot.com
>
> ------------------------------------------------------------------------------
> This SF.Net email is sponsored by the Verizon Developer Community
> Take advantage of Verizon's best-in-class app development support
> A streamlined, 14 day to market process makes app distribution fast  
> and easy
> Join now and get one step closer to millions of Verizon customers
> http://p.sf.net/sfu/verizon-dev2dev
> _______________________________________________
> Treebase-devel mailing list
> Tre...@li...
> https://lists.sourceforge.net/lists/listinfo/treebase-devel

-- 
===========================================================
: Hilmar Lapp  -:- Durham, NC -:- informatics.nescent.org :
===========================================================