Annik,

I was only able to get jyve (version 1) up and running using 
a turbine build from Aug. 31st.  Since Aug 31st, several 
changes have been made to the Turbine code base that affect
jyve.  jyve was never updated to reflect these changes.


Here's what I propose:

   jyve (version 1.0)
   ------------------
   0. This should be strickly a version to run until a working
   version jyve2 can distributed. i.e. it will compile against
   the latest tdk, and the basic functionality of v1 works well.
   1. We distribute a 'static' version of jyve with the approprate
      turbine.jar file.  
   2. try to get it working with other dbase's other than mysql
      which means I need to strip out all the outer joins and
      provide a work around.  Hopefully the solution will be
      usable in jyve2.

      I'm not too familar with the peer model in Turbine, but
      if a dbase dosn't support outer joins, will it produce
      the right sql to emulate it?  If not, I propose that we
      add an attribute in the jyve.properties file:

      supportsOuterJoin:  [true|false]

      Then we can check that property before we generate any
      sql.  In any case, the above discussion is really
      targeted for jyve2.0.


   3. Don't use torque to provide the different schemas.  I can't 
   get torque to work (see point 1 below).  Also, it seems that
   torque isn't producing optimal sql for postgres.  Also, I
   can provide the foreign key support for postgres (for versions
   under 7.0).  It seems that providing the static schema's for
   Mysql and postgress should satisfy 80% of the people.
   I'm all for using Torque w/ jyve2.0 assuming we can get it
   to write the best sql for each dbase.


   Hopefully, 0-3 will allow people to download jyve and
   get it up and running painlessly and allow them to play
   with it while jyve2.0 becomes stable.

   I don't want to spend too much time with jyve1.0.  Just
   fix the stuff that's breaks inside it's code base.  We'll
   sol if problems come up in Turbine, etc.

   jyve (2.0)
   -----------

    1. I can't work on this until I can get jdk1.2 working on
    my freebsd box.  Luckly, I have enough to do with jyve (v1)
    to keep me busy.  I can get it to work against the old
    Turbine.jar file, but until 1.2 is available, I won't be
    able to compile it against the latest stuff in cvs. In
    any case, I can still start the process. 

    2. I would like to move to the peer model for the dbase stuff
    but do it in two phases:
       phase 1.  separate the sql/business logic from code base.
                 factor out some of the other common stuff I see.
                 get all the templating stuff to work right.
                 I vote to stick with webmacro for now.
       
       once phase 1 is done, compiles, works,  Then move to phase 2

       phase 2.   Use turbine's peer model to write the sql.
                  This is new territory for me.  But it seems
                  to be the right way to go. 

    3.  While doing all this keep in mind the 'wish' list and
    think about how we can integrate the new functionality into
    the existing model.


    mike

    I've got a majority of the stuff working for postgress with
    jyve1.0, currently working on the mother of all outer joins
    for the 'unreleased' stuff.  Once I am happy with how it works
    I will post my updates.
    

    ps. to answer your question, I did install it according to
    the INSTALL.txt file.  I had to use an old turbine.jar file.
    Once I get cvs rights, I can add it to the jyve distribution.





On Thu, Oct 12, 2000 at 08:32:56AM +0200, Annik Schoenmann wrote:
> Hi Mike I Hope than you will have time to answer me,
> 
> I try to use jyve with a turbine application made with TDK, it doesn't
> works.
> My boss say to me than the user managment isn't the same the I try to modify
> it.
> 
> Actually as you wrote <<I'm using a Turbine-2.0.jar file from Aug 31st, 2000
> that work's with
> jyve.  I'm getting an error when  saveToStorage() is called.
> BUT I deleted all the turbine src :( >>
> 
> I'm asking you how did you install jyve according to the install.txt file?
> What do you think than I should do stopping my installation a waiting for
> you new version or still trying to correct and install jyve2?
> 
> 
> Thanks for answering Annik
> 
> 
> 

--
--------------------------------------------------------------
To subscribe:        [EMAIL PROTECTED]
To unsubscribe:      [EMAIL PROTECTED]
Archives and Other:  <http://java.apache.org/main/mail.html>
Problems?:           [EMAIL PROTECTED]

Reply via email to