Craig L Russell wrote:
Hi Allen,
On Nov 14, 2006, at 9:06 AM, Allen Gilliland wrote:
Dave wrote:
On 11/13/06, Allen Gilliland <[EMAIL PROTECTED]> wrote:
One of the things that I am planning to do for the 3.2 release is do
some audit/cleanup of the current business layer code. There are a
variety of things which could use improving, but the main goal is to
fix
our Hibernate configuration so that we are 1) properly using the open
session in view pattern and 2) enabling lazy fetching on all objects
and
associations.
All good.
Right now our Hibernate config is pretty messy and doesn't take
advantage of many of Hibernate's performance features, so the main
reason to do this work is to improve the performance of the business
layer. The second big reason is just to reduce clutter and simplify
the
code as much as possible. There are plenty of places in the code where
we have methods that aren't used at all or methods which are
duplicated,
so those would all be cleaned up.
I have most of this work done already (but not checked in) and there
aren't really any surprise changes that I had to make except when it
came to the hierarchical objects. I tried for multiple days to get the
hierarchical objects to work with the updated hibernate config and the
current data model, but I kept running into problems. So to fix the
problem I had to make a small tweak to the way hierarchical objects are
persisted which fixed my issues and I believe drastically simplifies
the
problem overall. The basic change is that I have completely removed
the
HierarchicalPersistentObject class and Assoc and it's subclasses and
changed the data model so that we have a more normal hierarchical
model.
So, for weblog categories I added a simple 'parentid' column to the
weblogcategory table and that allows a category to manage relationships
between it's parent and children directly. Same goes for the
FolderData
class, but as it turns out that column already existed in the schema
but
wasn't being used. Upgrade path for both of these is fairly simple and
only requires populating these columns with the right value.
The hierarchical persistent object and the assoc tables were there for
one purpose: to make it possible to do a recursive query (i.e. query
for all entries under /pets and /pets/dogs, /pets/cats etc.) with only
one SQL statement. But that hasn't worked since we made the switch to
the Hibernate Criteria API back in 2004 and we've been doing fine
without it. And the code that started as complex has been moved around
so much that now it's almost unintelligible.
So I'm +1 with that change.
Actually, I have an idea that I think can accomplish what's described
above and actually kill a couple other birds with the same stone.
Basically, if we add another column for the full path of a node like
you did above (/pets, /pets/dogs, /pets/cats) then to get all entries
under the category /pets by getting all entries that have a category
with a path that starts with "/pets". So the sql query would just be ...
select * from weblogentry,weblogcategory where weblogentry.categoryid
= weblogcategory.id and weblogcategory.path like '/pets%';
This would also significantly improve the current performance of the
getPath() methods on our hierarchical objects because it would be a
simple attribute instead of needing to be calculated at runtime. And
this would also be much better for implementing equals() on these
objects since the true key for a hierarchical object is it's path.
The only down side is that the migration path for that would require
walking the category and folder trees for all weblogs and saving the
path for all of them, which is fairly complex :/
It seems to me that this is a tradeoff between simplicity and
performance. Simplicity is storing the local path name and a reference
to the parent; performance is storing the entire path.
Performance seems to be the better tradeoff here. Operations like moving
or renaming a path require recursively changing the entire path of all
children but I assume that the performance benefits outweigh the
complexity of these infrequent operations.
I definitely think it would be best to add this new path column and
setup our queries to run like I said above, but I didn't think it needed
to be done yet. I figured we can talk about it and agree on it first,
then look into what the migration path would be and see if we can
accomplish it in a reasonable fashion. My first step was just to
cleanup the actual object model and make it work in a more reliable
fashion, then we can look forward to performance improvements.
If everyone already agrees about adding this new column for the path
then I am happy to start looking at how to implement it as part of 3.2.
-- Allen
Craig
-- Allen
- Dave
Craig Russell
Architect, Sun Java Enterprise System http://java.sun.com/products/jdo
408 276-5638 mailto:[EMAIL PROTECTED]
P.S. A good JDO? O, Gasp!
