We're definitely using Torque in a production environment and love it. We currently use it to:
1) be able to rebuild our test database from scratch - allowing every developer their very own copy of the latest schema. (We also have import scripts to import a stable set of test data.) 2) enforce tracking schema changes (because the XML is stored in CVS - it was difficult in the past to find out when tables changed because you would have to compile the changes from the dba's alter scripts.) 3) get to spend almost all our time working on business logic rather than writing database accesses. I've been starting to put together some notes on how Torque best fits with a production environment. There are some tricks to it - who gets to modify the schema? How do you generate the alter table scripts from the modified XML? What if you have "future development" schema changes in your XML file on release? How you do prevent them from going into production? But overall, we love it. (But I've also heard that Hibernate is great, so...) -Peter On Mon, 2002-10-21 at 12:24, Jorge DeFlon wrote: > I was also impressed by Torque, but due to its beta status we have to change > to hibernate till the release of the non-beta decoupled version. > > Anyone using it in a production environment? > Jorge DeFlon > NetApps Mexico > > ----- Original Message ----- > From: "Joe McGlynn" <[EMAIL PROTECTED]> > To: "'Turbine Torque Users List'" <[EMAIL PROTECTED]> > Sent: Monday, October 21, 2002 10:59 AM > Subject: Inter-Tier data transfer (DAO/VO Patterns?) > > > > I downloaded Torque 3 this weekend and spent some time working with it, > I'm > > impressed -- this is nice package. > > > > I was surprised to see the tight coupling of data access objects and data > > objects though. What I observed is that if I set "torque.addSaveMethod = > > true" it does two things. First (in BaseXXX classes) is that Torque > > generates the setXXX() methods to include a test to see if the new value > is > > different from the current value, and if so calls setModified(). Second > > Torque generates a save() method in the BaseXXX classes. If I have > > addSaveMethod set to false then I don't get the "setModified()" > > functionality and I also don't get the "save()" methods. *BUT* the > > generated BaseXXX classes still have a private static reference to the > > XXXPeer object. > > > > Is there a way to generate the base classes _with_ the setModified() > > functionality, but no save() method and no reference to the peer object? > I > > want to be able to generate code that closely models the J2EE "Data Access > > Object" and "Value Object" patterns. > > > > Joe > > > > > -- > To unsubscribe, e-mail: <mailto:turbine-torque-user-unsubscribe@;jakarta.apache.org> > For additional commands, e-mail: <mailto:turbine-torque-user-help@;jakarta.apache.org> > -- To unsubscribe, e-mail: <mailto:turbine-torque-user-unsubscribe@;jakarta.apache.org> For additional commands, e-mail: <mailto:turbine-torque-user-help@;jakarta.apache.org>
