Hi all,
I finally managed to solve (part of) the problem.
As correctly said by Pablo, the command: mvn scala:run -Dlauncher=Indexer launches the Indexer. If you can live with the existing index, you can simply launch: mvn scala:run.

However, with the standard configuration the jersey server tries to listen on port 1111. Unfortunately, on my machine I already have Virtuoso installed, and isql listens on the same port. In order to solve this conflict, I had to manually edit the source file:
./src/main/scala/org/dbpedia/lookup/server/Server.scala
at the following line:
val serverURI = new URI("http://localhost:1111/";)
where I changed 1111 with 1112.

Then, I was able to launch the server. I would suggest to use a configuration file where to set such value, in order not to modify the source code.

Moreover, I don't really like the fact that the browser opens when the process starts, and it fires the standard query: http://localhost:1112/api/search.asmx/KeywordSearch?QueryString=Berlin

Is it really necessary? Could it simply be avoided?


I think now the next challenge will be recreate the new Index, using my own local DBpedia mirror. From the documentation I've read it should be done via the command:
mvn scala:run -Dlauncher=Indexer "-DaddArgs=indexDir|redirectsFile|data"
However it's not very clear to me the meaning of these parameters:

1) indexDir is the output directory that will hold the searchable structures --> OK
2) redirectsFile is the homonymous dbpedia data set --> what does it mean?
3) data is a collection of files with the properties you want to index --> in what format? Collection of files? How can I pass a collection of files?

If someone already managed to run the indexer from scratch, I'd be very pleased to have some hints or examples of the right command.


cheers,
roberto


Il 18/04/2012 10:07, Pablo Mendes ha scritto:

Well in your message you were trying to run Index, not Server. Can you please try again and copy+paste the command line and results?

On Apr 17, 2012 11:17 AM, "Roberto Mirizzi" <roberto.miri...@gmail.com <mailto:roberto.miri...@gmail.com>> wrote:

    Il 17/04/2012 1.57, Pablo Mendes ha scritto:


    Do you mean you are trying to run Serverand giving it the index
    you downloaded?


    Exactly, I think this is the easiest option. But I get the "build
    error" when try to launch the serve.
    If I don't succeed with this one, I can't figure out how to suceed
    with the other one. :-)


    Or are you trying to regenerate the index from DBpedia files?

    On Apr 16, 2012 6:35 PM, "Roberto Mirizzi"
    <roberto.miri...@gmail.com <mailto:roberto.miri...@gmail.com>> wrote:

        Hi all,
        one problem at a time, we're approaching the solution, I'm
        confident about this! :-)

        You were terribly right: I deleted the
        .m2/repository/com/sun/jersey folder, and tried to install
        the project again, and this time the build was successful.

        But today it's a rainy day, and if I try to launch scala with
        the command:

        mvn scala:run -Dlauncher=Indexer

        I obtain a BUILD ERROR.
        Attached is the output.

        I also tried with simply mvn scala:run, but the error is the
        same.

        I specified the index path in the file default_index_path with:
        
/media/Iomega500/dbpedia-dump/dbpedia-lookup/network/www5/dbpedia-lookup/lookup_index/

        cheers,
        roberto



        Il 16/04/2012 17:28, Jona Christopher Sahnwaldt ha scritto:

            Hi Roberto,

            I don't know what's going on, but this line at the end of
            your error
            message is weird:

            Not a v4.0.0 POM. for project
            com.sun.jersey:jersey-project at
            
/home/roberto/.m2/repository/com/sun/jersey/jersey-project/1.5/jersey-project-1.5.pom

            I would suggest you delete your local repository
            /home/roberto/.m2/repository/ and start again. Sometimes
            Maven
            downloads a broken pom from some repo, and deleting the
            local repo
            often helps. The next build will take a few minutes
            because it has to
            download all JARs again, but maybe that will fix the
            problem. If not,
            send us the maven output again and we'll see what we can do.

            Cheers,
            Christopher

            On Mon, Apr 16, 2012 at 17:06, Roberto Mirizzi
            <roberto.miri...@gmail.com
            <mailto:roberto.miri...@gmail.com>>  wrote:

                Thank you!
                I'm trying to install the DBpedia Lookup Service.
                However, when i try to install with maven:

                mvn clean install

                I get the following error:

                "Project ID: null:jersey-server:bundle:null

                Reason: Cannot find parent:
                com.sun.jersey:jersey-project for project:
                null:jersey-server:bundle:null for project
                null:jersey-server:bundle:null"

                Attached you can find the output produced by the
                commadn mvn -e clean
                install.

                Any ideas?


                thanks,
                roberto



                Il 15/04/2012 23:14, Jona Christopher Sahnwaldt ha
                scritto:

                    Hi there,

                    I just added a shell script to dbpedia lookup
                    that makes using mvn
                    more convenient. If you'e on UN*X, you can type

                    ./run Indexer

                    instead of

                    mvn scala:run -Dlauncher=Indexer

                    and

                    ./run Indexer indexDir redirectsFile data

                    instead of

                    mvn scala:run -Dlauncher=Indexer "-DaddArgs=
                    indexDir|redirectsFile|data"

                    Not a big deal, but quite nice. Tested with bash,
                    may also work with
                    other shells.

                    Sorry, no Windows version yet. If anyone is good
                    with Windows batch
                    files, such a script for Windows would be welcome!

                    Cheers,
                    JC

                    On Sun, Apr 15, 2012 at 22:56, Pablo
                    Mendes<pablomen...@gmail.com
                    <mailto:pablomen...@gmail.com>>    wrote:

                        Hi Roberto,
                        Thanks for your offer.

                        You can clone it with mercurial and then
                        build it with maven2

                        hg clone
                        
http://dbpedia.hg.sourceforge.net:8000/hgroot/dbpedia/lookup
                        mvn clean install

                        You can download the necessary index from here:
                        http://spotlight.dbpedia.org/download/lookup_index.tgz

                        Then you can just run the server:
                        mvn scala:run -Dlauncher=Indexer

                        To create the necessary index, you can run:
                        mvn scala:run -Dlauncher=Indexer "-DaddArgs=
                        indexDir|redirectsFile|data"

                        indexDir is the output directory that will
                        hold the searchable structures
                        redirectsFile is the homonymous dbpedia data set
                        data is a collection of files with the
                        properties you want to index.

                        I have never run this myself, but it seems
                        from reading source code [1]
                        that
                        it uses at least the lexicalization dataset,
                        categories, abstracts and
                        instance types. If you manage to run this,
                        would you be so kind to update
                        the documentation at the wiki?
                        http://wiki.dbpedia.org/lookup/install

                        Cheers,
                        Pablo

                        [1]

                        
http://dbpedia.hg.sourceforge.net/hgweb/dbpedia/lookup/file/475a32257232/src/main/scala/org/dbpedia/lookup/util/DBpedia2Lucene.scala


                        On Sun, Apr 15, 2012 at 8:39 PM, Roberto
                        Mirizzi<roberto.miri...@gmail.com
                        <mailto:roberto.miri...@gmail.com>>
                        wrote:

                            Il 15/04/2012 20.24, Pablo Mendes ha scritto:

                            It seems to have gotten too popular. :)

                            We have been trying to shift resources
                            around, as many of our servers
                            are
                            a bit under pressure. Hopefully we'll
                            manage to find a solution soon.

                            I'd recommend considering hosting your
                            own mirror if you're using it as
                            an
                            effective part of an application, rather
                            than for testing out some
                            prototype.


                            I'd be happy to host a mirror by my own,
                            we have already loaded the
                            DBpedia dumps on Virtuoso. Where could I
                            find the source code or some
                            hints
                            to set a mirror for the wonderful lookup
                            service? Here [1] there's not
                            so
                            much information about it.

                            [1] http://dbpedia.org/lookup

                            cheers,
                            roberto




                            Cheers,
                            Pablo

                            On Sun, Apr 15, 2012 at 8:06 PM, Roberto
                            Mirizzi
                            <roberto.miri...@gmail.com
                            <mailto:roberto.miri...@gmail.com>>    wrote:

                                Why is it so slow in providing
                                results (20, 30 seconds and more)? :-(
                                Last time I used it, it was great...

                                cheers,
                                roberto


                                Il 21/03/2012 16.53, Pablo Mendes ha
                                scritto:

                                Thanks for the report. It's restarting.

                                On Wed, Mar 21, 2012 at 4:31 PM,
                                Nemanja Vukosavljevic
                                <nemanja.vukosavlje...@gmail.com
                                <mailto:nemanja.vukosavlje...@gmail.com>>
                                   wrote:

                                    Hi guys,

                                    It looks like that Dbpedia lookup
                                    service is down but only the Keyword
                                    Search API. It gives back HTTP
                                    500 error. On the other hand, Prefix
                                    Search
                                    API works like a charm.

                                    Cheers,
                                    Nemanja



                                    
------------------------------------------------------------------------------
                                    This SF email is sponsosred by:
                                    Try Windows Azure free for 90
                                    days Click Here
                                    http://p.sf.net/sfu/sfd2d-msazure
                                    
_______________________________________________
                                    Dbpedia-discussion mailing list
                                    Dbpedia-discussion@lists.sourceforge.net
                                    
<mailto:Dbpedia-discussion@lists.sourceforge.net>
                                    
https://lists.sourceforge.net/lists/listinfo/dbpedia-discussion




                                
------------------------------------------------------------------------------
                                This SF email is sponsosred by:
                                Try Windows Azure free for 90 days
                                Click Here
                                http://p.sf.net/sfu/sfd2d-msazure



                                _______________________________________________
                                Dbpedia-discussion mailing list
                                Dbpedia-discussion@lists.sourceforge.net
                                
<mailto:Dbpedia-discussion@lists.sourceforge.net>
                                
https://lists.sourceforge.net/lists/listinfo/dbpedia-discussion



                        
------------------------------------------------------------------------------
                        For Developers, A Lot Can Happen In A Second.
                        Boundary is the first to Know...and Tell You.
                        Monitor Your Applications in Ultra-Fine
                        Resolution. Try it FREE!
                        http://p.sf.net/sfu/Boundary-d2dvs2
                        _______________________________________________
                        Dbpedia-discussion mailing list
                        Dbpedia-discussion@lists.sourceforge.net
                        <mailto:Dbpedia-discussion@lists.sourceforge.net>
                        
https://lists.sourceforge.net/lists/listinfo/dbpedia-discussion




------------------------------------------------------------------------------
Better than sec? Nothing is better than sec when it comes to
monitoring Big Data applications. Try Boundary one-second 
resolution app monitoring today. Free.
http://p.sf.net/sfu/Boundary-dev2dev
_______________________________________________
Dbpedia-discussion mailing list
Dbpedia-discussion@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/dbpedia-discussion

Reply via email to