gmane.comp.apache.jackrabbit.user

CorruptIndexException

pkrishna — 2008-12-04T23:30:35

I am luncene 2.2.0 with JackRabbit 1.4.6 and for some reason I am seeing the following exception: java.io.IOException: Directory was previously created with a different LockFactory instance; please pass null as the lockFactory instance and use setLockFactory to change it at org.apache.lucene.store.FSDirectory.getDirectory(FSDirectory.java:192) at org.apache.jackrabbit.core.query.lucene.PersistentIndex.(PersistentIndex.java:63) at org.apache.jackrabbit.core.query.lucene.MultiIndex.(MultiIndex.java:286) I debugged through the code core.query.lucene.PersistentIndex and the call to getDirectory always throws a CorruptIndexException. The directory folder points to temp folder under Documents and Settings. I was browsing through the forum to see if any others have seen this problem and there was one entry about this but the author mentioned that once he updated to latest version of lucene everything was working fine. My impression was JackRabbit 1.4.6 only works with lucene-c

Getting an exception when searching with the QueryManager

pkrishna — 2008-12-03T21:37:35

We are using JackRabbit to manage our repository. On some boxes, we are seeing an exception that is thrown by the lucene search engine while executing our integration tests. Would this be a timing issue? java.lang.AssertionError: Terms are out of order: field=_:PROPERTIES (number 4) lastField=_:PROPERTIES (number 4) text=3:frozenPrimaryType lastText=3:frozenPrimaryType at org.apache.lucene.index.TermInfosWriter.add(TermInfosWriter.java:154) at org.apache.lucene.index.DocumentsWriter.appendPostings(DocumentsWriter.java:2320) at org.apache.lucene.index.DocumentsWriter.writeSegment(DocumentsWriter.java:2015) at org.apache.lucene.index.DocumentsWriter.flush(DocumentsWriter.java:552) at org.apache.lucene.index.IndexWriter.doFlush(IndexWriter.java:2623) at org.apache.lucene.index.IndexWriter.flush(IndexWriter.java:2523) a

Re: [Fwd: Which RepositoryFactory for WebDAV?]

Kenneth Yue — 2008-12-03T21:07:24

Hmm, Jackrabbit specific coding, which is what I've been trying to avoid. I have nothing against Jackrabbit (or else I wouldn't be using it :-), but having Jackrabbit code (i.e import org.apache.jackrabbit.*) in my code makes the users of my software hard, if not impossible, to switch to another JSR170 implementation at deploy/run-time. For example, if my users want to use Day's CRX (pay Day money) instead of Jackrabbit, they can't because I'm using org.apache.jackrabbit.client.JCR2SPIRepositoryStub in my code. This completely defeats the purpose of having a JSR in the first place, IMHO. Is a javax.naming.spi.ObjectFactory implementation similar to org.apache.jackrabbit.rmi.client.ClientRepositoryFactory planned? Please consider doing it. That would make my life (and my users' lives) much easier. :-) I can't believe I'm the only one asking for it! Maybe you should, then you'll know what I meant above. :-) Thanks. Ken

Re: Repository Server setup on Jboss 4.2.3

Marcel Reutegger — 2008-12-03T14:41:32

Hi, the version store configured in repository.xml still uses derby: Repository/Versioning/PersistenceManager Christopher M. Logan wrote: afaik there isn't. regards marcel

Re: JackRabbit web explorers

Alexander Klimetschek — 2008-12-03T10:47:02

http://dev.day.com/microsling/content/blogs/main/jcrtools.html http://wiki.apache.org/jackrabbit/JcrLinks Regards, Alex

Re: JackRabbit web explorers

2008-12-03T09:55:57

Hi, BTW, I remember reading a few months ago in this mailing that Day Software was considering giving the CRX explorer to the Jackrabbit project. Is it still relevent today? BR, Stéphane Landelle Le Mer 3 décembre 2008 10:25, sbarriba a écrit :

JackRabbit web explorers

sbarriba — 2008-12-03T09:25:47

Hi all, What are people using as generic web explorers for JCR repositories? We're looking for web-app based solutions which can be collocated with other JCR modules (using JNDI). We've reviewed http://www.jcr-explorer.org/ but has been dormant for a while. All suggestions welcome. Regards, Shaun

Re: [Fwd: Which RepositoryFactory for WebDAV?]

Angela Schreiber — 2008-12-03T07:51:56

hi kenneth see http://svn.apache.org/repos/asf/jackrabbit/sandbox/spi/client/src/test/java/org/apache/jackrabbit/client/JCR2SPIRepositoryStub.java for an example. that used to work for me. if that doesn't it is simply broken. glad to hear. i didn't try that one ;) regards angela

about jackrabbit 1.5 clustered workspace creation

zoulunbing — 2008-12-03T07:44:25

i have use jackrabbit 1.5 for clustered workspace, who can give me the configuration instance? thanks!

Re: [Fwd: Which RepositoryFactory for WebDAV?]

Kenneth Yue — 2008-12-02T21:29:42

Thank you very much for checking this in. I tried it with and got Message: Could not create resource factory instance Cause: javax.naming.NamingException: Could not create resource factory instance [Root exception is java.lang.ClassCastException: org.apache.jackrabbit.client.RepositoryFactoryImpl cannot be cast to javax.naming.spi.ObjectFactory] So how do I use this? FYI, for a workaround, I've been using RMI and this works great: Thanks again. Ken

Re: Clustering and Index update

majohnst — 2008-12-02T16:41:10

I meant to add that if I delete the index directory of node2 and then restart tomcat, the index is rebuilt and I can search for my:node correctly in node2. So this is probably an issue of node1 alerting node2 of changes. majohnst wrote:

Clustering and Index update

majohnst — 2008-12-02T16:38:24

I have jackrabbit setup with a cluster of 2 nodes. Both nodes are running on my local computer but through different tomcat webapps. The clustering seems to be working correctly. My problem is when I try to search through the nodes. If I add a new jcr node to cluster node1, the node is added correctly and I can immediately search and find that node. But if I try to do the same node on cluster node2, I get no results. If I browse the contents of node2, I see my new node exists on the cluster. So somehow node2 is not indexing the new changes. Ex. - create new node, my:node and add to cluster node1 - search for //element(*,my:node) on node1 and I get the new node - search for //element(*,my:node) on node2 and I get no results

Re: OCM and custom_nodetypes.xml

Christophe Lombart — 2008-12-02T14:34:55

no, this is not mandatory in all cases. Furthermore, the usage of the custom_nodetypes is not depending on the OCM annotations or the OCM XML mapping file. Let me explain ... In some conditions, OCM needs a *discriminator ( = jcr property* of type String) that contains the class name to be used for instantiation. This node property must be named "ocm_classname". On loading objects from the jcr repository, OCM checks this attribute and instantiates objects of this type. This property is not intrusive because it is not necessary to map it into your java class attribute. It is only used internally by the Object Content Manager and again, it is not used in all cases (see below) Here is a summary of the different possibilities : 1. Mapping all java classes into the node type "nt:unstructured" : - The mixin "ocm:discriminator" is not necessary, so no need to add it into custom_nodetypes.xml. - OCM will add an extra property calle ocm_classname. 2. Mapping each java class to a different node type. - The mi

Re: Authentication for Repository

Jukka Zitting — 2008-12-02T14:11:50

Hi, On Tue, Dec 2, 2008 at 11:43 AM, hosseinakbari magfa.com> wrote: The Repository.login method takes an optional Credentials argument. You can use a SimpleCredentials instance with the username and password as this argument: import javax.jcr.*; Credentials credentials = new SimpleCredentials( "username", "password".toCharArray()); Session session = repository.login(credentials); BR, Jukka Zitting

Re: JackRabbit client / server

Alexander Klimetschek — 2008-12-02T12:42:08

You might want to take a look at this nice Sling introduction: http://www.unc.edu/home/adamc/sling-overview.html It describes the current status of the project very accurately: How Stable is this? From a technical perspective, Sling's fairly stable and has at least one full-fledged "release" under its belt. The main issue is that most of the developers work for the same company (Day Software). The Apache Software Foundation requires projects to establish that they have a sustainable, distributed community of committers, and that's why it's still in incubation. Since the project's getting more attention though, I don't expect it to be too long before it "graduates." Regards, Alex

Re: Jackrabbit JCA, glassfish (v2): shutdown repository

Edgar Merino — 2008-12-02T10:56:53

Anyone out there that knows the solution to this? it's very irritating to have to restart the application server each time I deploy my application... thanks in advance. Edgar Merino Edgar Merino escribió:

Re: [Fwd: Which RepositoryFactory for WebDAV?]

Angela Schreiber — 2008-12-02T10:54:03

hi kenneth i added an initial draft to the client module with rev. 722416 see https://issues.apache.org/jira/browse/JCR-1890 for a very rough summary. note: you have to update the jackrabbit trunk as well. the factory interface i used has just been added by marcel on friday (Revision: 721387, Issue JCR-1834). regards angela

Authentication for Repository

hosseinakbari — 2008-12-02T10:43:54

Hi, How can I put username and password for accessing a repository thanks

Re: [Fwd: Which RepositoryFactory for WebDAV?]

Alexander Klimetschek — 2008-12-02T10:22:58

Often it is all about the way you ask for something, and I personally found your wording inpolite. My response was also based on your previous mail where you wondered why you have to build the spi2dav project manually from svn - so I felt you had not much feeling for an OS project. But it's an inpolite term in the context of an OSS mailing list, at least to me. Regards, Alex

problem with Repository

hosseinakbari — 2008-12-02T10:20:40

Hi, I am new In Jackrabbit, I am trying to create 3 repository and register them with different names with RegistryHelper.registerRepository. my question is if I want to restart my application and because the repositories are already exist, how can I connect to the specific repository. thanks,

Re: [Fwd: Which RepositoryFactory for WebDAV?]

Kenneth Yue — 2008-12-01T23:36:35

It certainly wasn't my intention to treat Angela as my subordinate. If my request came across as a demand, I apologize. Look, I'm working on another FOSS project too. I'm familiar with the work I'm doing. However, I have not been involved in the development of Jackrabbit, so it will be awfully inefficient for me to go in and spend countless hours of time trying to figure out how the current Jackrabbit code works, how it's organized, what its architecture is, etc. in order to come up with a RepositoryFactory for spi2dav. I was really surprised to find out that the server side of JCR remoting over WebDAV is released but there is no client code for it (which makes JCR remoting over WebDAV pretty much useless). If I were to work on this today, this is the approach I'll make. Please let me know if this makes sense (all of these are just guessworks so I could be completely off-base): 1. Try to understand what the SPI architecture is. I suppose it's an implementation of JCR using SPI (what does