jackrabbit-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sridhar Raman" <sridhar.ra...@gmail.com>
Subject Re: Same-name siblings problem
Date Mon, 05 Feb 2007 04:45:57 GMT
I am using Jackrabbit 1.1.

I am using same name siblings for the following reason:
Every DIM has a VALUES node. This VALUES node is the root for a tree
structure of VALUE nodes. Hence, when I do a query for a particular value, I
can do it using a simple query.

I am sorry ... I didn't understand your question regarding the test case.
Let me try explaining the problem.
I have 13 VALUE nodes which are like buckets. And into these buckets, I
would be adding new VALUE nodes dynamically. eg. If the buckets were [A-D],
[E-H], and so on. I would look at the new value, find out its first letter
and create a VALUE node for it which would be added into the appropriate
bucket.
The problem I face happens when I add this new node. I am adding a node to a
particular bucket. I find it hard to understand how the indexing of a level
higher is affected.

One more new problem I face is that after a certain number of nodes have
been added, the structure is totally gone. In fact, my dump at the end of an
import looks somewhat like this:

/DIM
/DIM/VALS
/DIM/VALS/VALUE
/DIM/VALS/VALUE/name = [1]
/DIM/VALS/VALUE/VALUE/
/DIM/VALS/VALUE/VALUE/name = 1.1
/VALUE
...
/VALUE[2]
...

This is really bizarre!

I am running the code using version 1.2.1. I hope that solves this problem.
But any form of help, I would really appreciate it.

Thanks in advance,
Sridhar

On 2/4/07, Tobias Bocanegra <tobias.bocanegra@day.com> wrote:
>
> hi sridhar,
> can you provide a test case that reproduces your problem?
> what version of jackrabbit are you using?
>
> generally, i would avoid using same name siblings since it causes
> always a lot of trouble and is not very performant. use some useful
> node name or just a simple number.
>
> regards, toby
>
> On 2/4/07, Sridhar Raman <sridhar.raman@gmail.com> wrote:
> > I still haven't been able to tackle the problem. An additional thing
> that
> > seems to be happening is that after a certain point of time, the nodes
> get
> > added to the root (I have no idea how!).
> >
> > But can someone help me with the initial problem I had posted?
> >
> > Thanks,
> > Sridhar
> >
> > On 2/2/07, Sridhar Raman <sridhar.raman@gmail.com> wrote:
> > >
> > > I have the following node structure:
> > >
> > > /DIM
> > > /DIM/VALS
> > > /DIM/VALS/VALUE
> > > /DIM/VALS/VALUE/name = [1]
> > > /DIM/VALS/VALUE[2]
> > > /DIM/VALS/VALUE[2]/name = [2]
> > > /DIM/VALS/VALUE[3]
> > > /DIM/VALS/VALUE[3]/name = [3]
> > > /DIM/VALS/VALUE[4]
> > > /DIM/VALS/VALUE[4]/name = [4]
> > > ...
> > > ...
> > > ...
> > > /DIM/VALS/VALUE[13]
> > > /DIM/VALS/VALUE[13]/name = [13]
> > >
> > > There are multiple DIM nodes. Each DIM node has ONLY one VALS node.
> Each
> > > VALS node can have multiple VALUE nodes. Similarly, the VALUE nodes
> can have
> > > multiple children VALUE nodes.
> > >
> > > This is the problem I am facing:
> > >
> > > Initially, all the VALUE nodes (1 to 13) are empty. I add nodes into
> them
> > > as and when I get the data. But this is what happens when I add a
> child
> > > VALUE node to /DIM/VALS/VALUE.
> > > /DIM
> > > /DIM/VALS
> > > /DIM/VALS/VALUE
> > > /DIM/VALS/VALUE/name = [1]
> > > /DIM/VALS/VALUE/VALUE/
> > > /DIM/VALS/VALUE/VALUE/name = 1.1
> > > /DIM/VALS/VALUE
> > > /DIM/VALS/VALUE/name = [2]
> > > /DIM/VALS/VALUE[2]
> > > /DIM/VALS/VALUE[2]/name = [3]
> > > /DIM/VALS/VALUE[3]
> > > /DIM/VALS/VALUE[3]/name = [4]
> > > ...
> > > ...
> > > ...
> > > /DIM/VALS/VALUE[12]
> > > /DIM/VALS/VALUE[12]/name = [13]
> > >
> > > Is this a bug? Or is this how it is supposed to behave?
> > >
> > > The consequences of such a bug are immense.
> > > Firstly, I can't do any indexing based node retrieval, as the indexes
> keep
> > > changing as and when I add children to the nodes. Despite the children
> being
> > > added at a level below.
> > > Secondly, I am also facing some lucene query exceptions when I try to
> do a
> > > query to obtain the right node.(Note: I am forced to do an XPATH
> query, as
> > > the indexes have become inconsistent)
> > > Thirdly, I am curious to find out how there can be 2 nodes with the
> same
> > > name and same index. Shouldn't that be an automatic error?
> > >
> > > Can someone help me with this? I do a simple parent.addNode(nodename,
> > > nodetype) call. And I save the session. Why does that affect the
> indexing?
> > >
> > > Thanks in advance,
> > > Sridhar
> > >
> >
>
>
> --
> -----------------------------------------< tobias.bocanegra@day.com >---
> Tobias Bocanegra, Day Management AG, Barfuesserplatz 6, CH - 4001 Basel
> T +41 61 226 98 98, F +41 61 226 98 97
> -----------------------------------------------< http://www.day.com >---
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message