[kepler-dev] simple question with rebuilding kepler from sources
Christopher Brooks
cxh at eecs.berkeley.edu
Wed Apr 25 09:07:10 PDT 2007
Hi Dan,
Over here in Ptolemy-land, we try to tell people to put
update -P -d
in ~/.cvsrc. Maybe the Kepler CVS instructions should be updated
for this? The other common gotcha is to not set CVS_RSH to ssh.
Not setting $HOME can also sometimes cause issues under Cygwin, though
I have not seen that in awhile.
_Christopher
--------
Hi Unni,
It looks like you probably checked out Kepler without including the
'-P' switch to remove empty directories. This is a limitation of CVS -
empty directories never get deleted, so you have to use the CVS -P
switch to 'prune' those empty directories at checkout. 'ant buildkarlib'
check all directories in src/actors/ and has problems if empty
directories are there. [This is a recurring problem that we need to
figure out how to fix!]
Dan
---
Unni Pillai wrote:
> Okay it looks like the source of my problem is that something which
> was deprecated nearly 12 months ago is still being picked up by the
> build process. I'm going to continue debugging and see if I can fix
> the build locally, but appreciate if the domain experts have the
> correct build files to help out.
>
> the problem is as follows:
>
>
>
>
>
>
> On Apr 25, 2007, at 11:17 AM, Unni Pillai wrote:
>
>> Hi Folks
>>
>> I think my sources are matching the repositories for PTII and
>> Kepler. I went through the necessary hoops to get it.
>>
>> I was able to build my PTII correctly and invoke vergil editor and
>> test some demos. I went ahead and downloaded the kepler sources from
>> cvs account at ecoinformatics and went through steps outlined in the
>> install doc found
>> here http://cvs.ecoinformatics.org/cvs/cvsweb.cgi/~checkout~/kepler/INST
ALL?content-type=text/plain
>> <http://cvs.ecoinformatics.org/cvs/cvsweb.cgi/%7Echeckout%7E/kepler/INST
ALL?content-type=text/plain>
>>
>> The problem arises when I go to build kepler (ant buildkarlib)
>> missing MANIFEST.MF kepler/src/actors/ADNToHTMLConversion/MANIFEST.MF
>> does not exist.
>>
>> Can somebody verify if this file is missing from the repository or
>> suggest workaround.
>>
>> asterix:~/Documents/projects/ndiipp-ptv/kepler unni$ echo $KEPLER
>> /Users/unni/Documents/projects/ndiipp-ptv/kepler
>> asterix:~/Documents/projects/ndiipp-ptv/kepler unni$ echo $PTII
>> /Users/unni/Documents/projects/ndiipp-ptv/ptII
>> asterix:~/Documents/projects/ndiipp-ptv/kepler unni$ echo $JAVA_HOME
>> /Library/Java/Home
>> asterix:~/Documents/projects/ndiipp-ptv/kepler unni$ which ant
>> /usr/bin/ant
>> asterix:~/Documents/projects/ndiipp-ptv/kepler unni$ ant full-clean
>> ptolemy
>> Buildfile: build.xml
>>
>> BUILD SUCCESSFUL
>> Total time: 4 minutes 12 seconds
>>
>>
>>
>> asterix:~/Documents/projects/ndiipp-ptv/kepler unni$ ant buildkarlib
>> BUILD FAILED
>> /Users/unni/Documents/projects/ndiipp-ptv/kepler/build.xml:889: The
>> following error occurred while executing this line:
>> /Users/unni/Documents/projects/ndiipp-ptv/kepler/build.xml:909:
>> Manifest file:
>> /Users/unni/Documents/projects/ndiipp-ptv/kepler/src/actors/ADNToHTMLCon
version/MANIFEST.MF
>> does not exist.
>>
>> Total time: 0 seconds
>>
>>
>> Thanks
>>
>> Unni
>>
>> On Apr 24, 2007, at 4:24 PM, Christopher Brooks wrote:
>>
>>> Hi Unni,
>>>
>>> For information about read access to the ptII repository, see:
>>> http://chess.eecs.berkeley.edu/ptexternal/
>>>
>>> _Christopher
>>> --------
>>>
>>> Thanks for the response
>>>
>>> I guess I'd need access to the CHESS CVS server also. Since the =20
>>> PtolemyII I built is from sources obtained at PtolemyII
>>> website. Any =20=
>>>
>>> way to get access to these servers today if possible.
>>>
>>> Unni
>>>
>>>
>>> On Apr 24, 2007, at 4:08 PM, Chad Berkley wrote:
>>>
>>>> If you want to recompile stuff, you'll definitely want the CVS =20
>>>> version. Let me know if no one responds to your cvs account =20
>>>> request by tomorrow. You'll also need to check out ptolemy from =20
>>>> the CHESS CVS server. See the build instructions here: http://=20
>>>> cvs.ecoinformatics.org/cvs/cvsweb.cgi/~checkout~/kepler/INSTALL?=20
>>>> content-type=3Dtext/plain
>>>>
>>>> chad
>>>>
>>>> Unni Pillai wrote:
>>>>> Hi Dan
>>>>> So to work with nightly build we only need to set the KEPLER =20
>>>>> environment variable. I just need some clarification with how we =20
>>>>> go about rebuilding kepler if we touch any kepler java srcs from =20
>>>>> the nightly build or should I be working with kepler sources =20
>>>>> obtained from the cvs repository?
>>>>> I've requested pmc at ecoinformatics.org
>>>>> <mailto:pmc at ecoinformatics.org> =20
>>>>> <mailto:pmc at ecoinformatics.org> for read access to the cvs =20
>>>>> repository.
>>>>> Thanks
>>>>> Unni
>>>>> On Apr 23, 2007, at 4:29 PM, Dan Higgins wrote:
>>>>>> Hi,
>>>>>> If you are starting with the nightly zipped version of =20
>>>>>> Kepler, all of the needed Ptolemy code has been precompiled and =20
>>>>>> is in the kepler.jar file (both the source and the *.class =20
>>>>>> compiled files). You actually do not need the PTII env variable =20
>>>>>> for running kepler since the kepler.jar is in the class path. =20
>>>>>> Kepler source code is in the 'src/ subdirectory of the nightly =20
>>>>>> build. As Ilkay suggested, you probably should checkout fresh =20
>>>>>> CVS versions and start from there if you plan extensive changes.
>>>>>>
>>>>>> Ilkay Altintas wrote:
>>>>>>> Unni,
>>>>>>>
>>>>>>> The problem seems to be your PTII variable. To be able to build =20
>>>>>>> the nightly build from scratch, you need a version of Ptolemy in
>>>>>>> =20=
>>>
>>>>>>> your home directory (it probably already installs a version) and
>>>>>>> =20=
>>>
>>>>>>> point the PTII variable to that directory.
>>>>>>> You can also request a read-only kepler cvs account from =20
>>>>>>> "pmc at ecoinformatics.org <mailto:pmc at ecoinformatics.org>
>>>>>>> <mailto:pmc at ecoinformatics.org>" and =20
>>>>>>> check out both kepler and ptolemy to build everything from scratch.
>>>>>>>
>>>>>>> -ilkay
>>>>>>>
>>>>>>> On Apr 23, 2007, at 12:13 PM, Unni Pillai wrote:
>>>>>>>
>>>>>>>> Hi
>>>>>>>>
>>>>>>>> Following Lucas suggestion, I had downloaded the nightly build =20
>>>>>>>> (April 18th) and got it up and running using kepler.sh, but =20
>>>>>>>> I've run into some silly problems.
>>>>>>>>
>>>>>>>> Since I plan on doing some hacks locally I decided to be a bit =20
>>>>>>>> wild and did the following from a shell ant full-clean ptolemy.
>>>>>>>>
>>>>>>>> I'm testing on Mac OSX v 10.4.9
>>>>>>>> My environment variables are set as follows:
>>>>>>>> asterix:~/Documents/projects/ndiipp-ptv/kepler unni$ echo $KEPLER
>>>>>>>> /Users/unni/Documents/projects/ndiipp-ptv/kepler
>>>>>>>> *asterix:~/Documents/projects/ndiipp-ptv/kepler unni$ echo $PTII*
>>>>>>>> */Users/unni/Documents/projects/ndiipp-ptv/kepler/build*
>>>>>>>> asterix:~/Documents/projects/ndiipp-ptv/kepler unni$ echo =20
>>>>>>>> $JAVA_HOME
>>>>>>>> /Library/Java/Home
>>>>>>>>
>>>>>>>> My concern is did the nightly build bring in the appropriate =20
>>>>>>>> Ptolemy sources or am I missing some step since not sure if =20
>>>>>>>> Ptolemy came through or most likely I set my PTII environment =20
>>>>>>>> variable incorrectly.
>>>>>>>>
>>>>>>>> I'm getting whole slew of errors when I attempt to build
>>>>>>>> BUILD FAILED
>>>>>>>> /Users/unni/Documents/projects/ndiipp-ptv/kepler/build.xml:748:
>>>>>>>> =20=
>>>
>>>>>>>> The following error occurred while executing this line:
>>>>>>>> /Users/unni/Documents/projects/ndiipp-ptv/kepler/build-=20
>>>>>>>> ptolemy.xml:106: /Users/unni/Documents/projects/ndiipp-ptv/=20
>>>>>>>> kepler/build/ptolemy/distributed/config not found.
>>>>>>>>
>>>>>>>> Something is fishy with my PTII variable
>>>>>>>> Can someone point me to the correct environment variable setup =20
>>>>>>>> for working with nightly builds.
>>>>>>>>
>>>>>>>> Thanks
>>>>>>>>
>>>>>>>> Unni
>>>>>>>>
>>>>>>>> P.S. My previous postings to kepler dev list never show up, =20
>>>>>>>> hence decided to contact you folks who do development with the =20
>>>>>>>> nightly builds or build from source.
>>>>>>>>
>>>>>>>> Also as reference this is how my kepler directory looks after =20
>>>>>>>> ant clean
>>>>>>>>
>>>>>>>> asterix:~/Documents/projects/ndiipp-ptv/kepler unni$ ls -l
>>>>>>>> total 656
>>>>>>>> -rw-r--r-- 1 unni unni 6802 Apr 23 13:59 INSTALL
>>>>>>>> -rw-r--r-- 1 unni unni 7005 Apr 23 13:59 README.txt
>>>>>>>> -rw-r--r-- 1 unni unni 991 Apr 23 13:59 UDF_skolem.class
>>>>>>>> -rw-r--r-- 1 unni unni 736 Apr 23 13:59 UDF_split.class
>>>>>>>> -rw-r--r-- 1 unni unni 3497 Apr 23 13:59 build-doc.xml
>>>>>>>> -rw-r--r-- 1 unni unni 20897 Apr 23 13:59 build-ptolemy.xml
>>>>>>>> -rw-r--r-- 1 unni unni 5146 Apr 23 13:59 build-scia.xml
>>>>>>>> -rw-r--r-- 1 unni unni 66921 Apr 23 13:59 build.xml
>>>>>>>> drwxr-xr-x 6 unni unni 204 Apr 23 13:59 cog
>>>>>>>> drwxr-xr-x 4 unni unni 136 Apr 23 13:58 configs
>>>>>>>> -rw-r--r-- 1 unni unni 1230 Apr 23 13:59 copyright.txt
>>>>>>>> drwxr-xr-x 7 unni unni 238 Apr 23 13:58 demos
>>>>>>>> drwxr-xr-x 4 unni unni 136 Apr 23 13:59 dev
>>>>>>>> drwxr-xr-x 4 unni unni 136 Apr 23 13:58 doc
>>>>>>>> -rw-r--r-- 1 unni unni 3583 Apr 23 13:59 hsql-license.txt
>>>>>>>> -rw-r--r-- 1 unni unni 59 Apr 23 13:59 hsqlTool.bat
>>>>>>>> -rw-r--r-- 1 unni unni 4418 Apr 23 13:59 kepler.bat
>>>>>>>> -rw-r--r-- 1 unni unni 4394 Apr 23 13:59 kepler.sh
>>>>>>>> -rw-r--r-- 1 unni unni 121 Apr 23 13:59 keplerLog4J.log
>>>>>>>> drwxr-xr-x 50 unni unni 1700 Apr 23 13:58 lib
>>>>>>>> -rw-r--r-- 1 unni unni 143360 Apr 23 13:59 libexpat.dll
>>>>>>>> -rw-r--r-- 1 unni unni 2012 Apr 23 13:59 log4j.properties
>>>>>>>> -rw-r--r-- 1 unni unni 4397 Apr 23 13:59 momlexecute.sh
>>>>>>>> drwxr-xr-x 5 unni unni 170 Apr 23 13:57 plugins
>>>>>>>> drwxr-xr-x 11 unni unni 374 Apr 23 13:58 src
>>>>>>>> -rw-r--r-- 1 unni unni 14379 Apr 23 13:59 start.html
>>>>>>>> drwxr-xr-x 12 unni unni 408 Apr 23 13:58 webservice
>>>>>>>> drwxr-xr-x 4 unni unni 136 Apr 23 13:58 website
>>>>>>>> drwxr-xr-x 24 unni unni 816 Apr 23 13:58 workflows
>>>>>>>> drwxr-xr-x 3 unni unni 102 Apr 23 13:58 zipdir
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> *Unni Pillai* *|* Software Engineer / Analyst | Digital Library
>>>>>>>> =20=
>>>
>>>>>>>> <http://library.nyu.edu/diglib/> | NDIIPP-PDPTV <http://=20
>>>>>>>> www.ptvdigitalarchive.org/purpose/> =E2=9C=89
>>>>>>>> unni.pillai at nyu.edu <mailto:unni.pillai at nyu.edu> =20
>>> =
>>>
>>>>>>>> <mailto:unni.pillai at nyu.edu> | =E2=9C=86 (212) 992-9741 | AIM: =20
>>>>>>>> upillai at mac.com <mailto:upillai at mac.com>
>>>>>>>> <mailto:upillai at mac.com> | Jabber: =20
>>>>>>>> unni.pillainyc at gmail.com <mailto:unni.pillainyc at gmail.com>
>>>>>>>> <mailto:unni.pillainyc at gmail.com>
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>> --=20
>>>>>>> Ilkay ALTINTAS
>>>>>>> Lab Director, Scientific Workflow Automation Technologies (SWAT)
>>>>>>> San Diego Supercomputer Center(SDSC), UCSD
>>>>>>> 9500 Gilman Drive, MC: 0505 La Jolla, CA 92093-0505
>>>>>>> phone: (858) 822-5453 fax: (858) 534-8303
>>>>>>> web: http://users.sdsc.edu/~altintas
>>>>>>> <http://users.sdsc.edu/%7Ealtintas> <http://users.sdsc.edu/%=20
>>>>>>> 7Ealtintas>
>>>>>>>
>>>>>>>
>>>>>>> -------------------------------------------------------------------
=20
>>> =
>>>
>>>>>>> -----
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Kepler-dev mailing list
>>>>>>> Kepler-dev at ecoinformatics.org <mailto:Kepler-dev at ecoinformatics.org
>
>>>>>>> http://mercury.nceas.ucsb.edu/ecoinformatics/mailman/listinfo/=20
>>>>>>> kepler-dev
>>>>>>>
>>>>>>
>>>>>>
>>>>>> --=20
>>>>>> *******************************************************************
>>>>>> Dan Higgins
>>>>>> higgins at nceas.ucsb.edu <mailto:higgins at nceas.ucsb.edu>
>>>>>> http://www.nceas.ucsb.edu/ Ph: 805-893-5127
>>>>>> National Center for Ecological Analysis and Synthesis (NCEAS) =20
>>>>>> Marine Science Building - Room 3405
>>>>>> Santa Barbara, CA 93195
>>>>>> *******************************************************************
>>>>> ---------------------------------------------------------------------
=20
>>> =
>>>
>>>>> ---
>>>>> _______________________________________________
>>>>> Kepler-dev mailing list
>>>>> Kepler-dev at ecoinformatics.org <mailto:Kepler-dev at ecoinformatics.org>
>>>>> http://mercury.nceas.ucsb.edu/ecoinformatics/mailman/listinfo/=20
>>>>> kepler-dev
>>
>> _______________________________________________
>> Kepler-dev mailing list
>> Kepler-dev at ecoinformatics.org <mailto:Kepler-dev at ecoinformatics.org>
>> http://mercury.nceas.ucsb.edu/ecoinformatics/mailman/listinfo/kepler-dev
>
--
*******************************************************************
Dan Higgins higgins at nceas.ucsb.edu
http://www.nceas.ucsb.edu/ Ph: 805-893-5127
National Center for Ecological Analysis and Synthesis (NCEAS) Marine Scienc
e Building - Room 3405
Santa Barbara, CA 93195
*******************************************************************
_______________________________________________
Kepler-dev mailing list
Kepler-dev at ecoinformatics.org
http://mercury.nceas.ucsb.edu/ecoinformatics/mailman/listinfo/kepler-dev
--------
More information about the Kepler-dev
mailing list