exist-development — eXist Developer's List

Re: [Exist-development] [Exist-commits] SF.net SVN: exist:[13782] trunk/eXist/src/org/exist/storage/BrokerPool.java

[Dmitriy S. <sha...@gm...>]

settings files ... plugins one as example, but also it will be used for
others eXist settings files, like backup, serialize ...

On Wed, Feb 16, 2011 at 11:23 PM, Adam Retter <ad...@ex...> wrote:

> Whats it for?
>
> On 16 February 2011 16:33, Dmitriy Shabanov <sha...@gm...> wrote:
> > Wolfgang did ask to do not use '/db/system' because it have 'rwurwu---',
> but
> > 'etc' 'rwurwuR--'.
> >
> > On Wed, Feb 16, 2011 at 7:29 PM, Adam Retter <ad...@ex...> wrote:
> >>
> >> What is /db/etc? Should this not be /db/system/etc if anything?
> >>
> >>
> >> On 16 February 2011 14:13,  <sha...@us...> wrote:
> >> > Revision: 13782
> >> >          http://exist.svn.sourceforge.net/exist/?rev=13782&view=rev
> >> > Author:   shabanovd
> >> > Date:     2011-02-16 13:13:58 +0000 (Wed, 16 Feb 2011)
> >> >
> >> > Log Message:
> >> > -----------
> >> > [feature] create '/db/etc' collection.
> >> >
> >> > Owner: SYSTEM with permissions: rwu;
> >> > Group: DBA with permissions: rwu;
> >> > Other's permissions: r--;
> >> >
> >> > It will be monitored by configuration trigger and it must/should be
> used
> >> > for settings which must/should be readable by world.
> >> >
> >> > Modified Paths:
> >> > --------------
> >> >    trunk/eXist/src/org/exist/storage/BrokerPool.java
> >> >
> >> > Modified: trunk/eXist/src/org/exist/storage/BrokerPool.java
> >> > ===================================================================
> >> > --- trunk/eXist/src/org/exist/storage/BrokerPool.java   2011-02-16
> >> > 13:03:39 UTC (rev 13781)
> >> > +++ trunk/eXist/src/org/exist/storage/BrokerPool.java   2011-02-16
> >> > 13:13:58 UTC (rev 13782)
> >> > @@ -961,25 +961,23 @@
> >> >     }*/
> >> >
> >> >      /**
> >> > -     * Initialise system collections, if it doesnt exist yet
> >> > +     * Initialise system collections, if it doesn't exist yet
> >> >      *
> >> >      * @param sysBroker The system broker from before the brokerpool
> is
> >> > populated
> >> >      * @param sysCollectionUri XmldbURI of the collection to create
> >> > -     * @param permissions The persmissions to set on the created
> >> > collection
> >> > +     * @param permissions The permissions to set on the created
> >> > collection
> >> >      */
> >> > -    private void initialiseSystemCollection(DBBroker sysBroker,
> >> > XmldbURI sysCollectionUri, int permissions) throws EXistException {
> >> > -        //create /db/system
> >> > -        Collection sysCollection =
> >> > sysBroker.getCollection(sysCollectionUri);
> >> > -        if (sysCollection == null)
> >> > -        {
> >> > +    private Collection initialiseSystemCollection(DBBroker sysBroker,
> >> > XmldbURI sysCollectionUri, int permissions) throws EXistException {
> >> > +        Collection collection =
> >> > sysBroker.getCollection(sysCollectionUri);
> >> > +        if (collection == null) {
> >> >             TransactionManager transact = getTransactionManager();
> >> >             Txn txn = transact.beginTransaction();
> >> >             try {
> >> > -                sysCollection = sysBroker.getOrCreateCollection(txn,
> >> > sysCollectionUri);
> >> > -                if (sysCollection == null)
> >> > +                collection = sysBroker.getOrCreateCollection(txn,
> >> > sysCollectionUri);
> >> > +                if (collection == null)
> >> >                     throw new IOException("Could not create system
> >> > collection: " + sysCollectionUri);
> >> > -                sysCollection.setPermissions(permissions);
> >> > -                sysBroker.saveCollection(txn, sysCollection);
> >> > +                collection.setPermissions(permissions);
> >> > +                sysBroker.saveCollection(txn, collection);
> >> >
> >> >                 transact.commit(txn);
> >> >             } catch (Exception e) {
> >> > @@ -990,19 +988,36 @@
> >> >                 throw new EXistException(msg, e);
> >> >             }
> >> >         }
> >> > +
> >> > +        return collection;
> >> >     }
> >> >
> >> >     /**
> >> > -     * Initialise required system collections, if they dont exist yet
> >> > +     * Initialize required system collections, if they don't exist
> yet
> >> >      *
> >> >      * @param sysBroker - The system broker from before the brokerpool
> >> > is populated
> >> >      *
> >> >      * @throws EXistException If a system collection cannot be created
> >> >      */
> >> > -    private void initialiseSystemCollections(DBBroker sysBroker)
> throws
> >> > EXistException
> >> > +    private void initialiseSystemCollections(DBBroker broker) throws
> >> > EXistException
> >> >     {
> >> >         //create /db/system
> >> > -        initialiseSystemCollection(sysBroker,
> >> > XmldbURI.SYSTEM_COLLECTION_URI, 0770);
> >> > +        initialiseSystemCollection(broker,
> >> > XmldbURI.SYSTEM_COLLECTION_URI, 0770);
> >> > +        //create /db/etc
> >> > +        Collection collection = initialiseSystemCollection(broker,
> >> > XmldbURI.ETC_COLLECTION_URI, 0774);
> >> > +
> >> > +        //initialize configurations watcher trigger
> >> > +        if (collection != null) {
> >> > +               CollectionConfigurationManager manager =
> >> > broker.getBrokerPool().getConfigurationManager();
> >> > +               CollectionConfiguration collConf =
> >> > manager.getOrCreateCollectionConfiguration(broker, collection);
> >> > +               try {
> >> > +                               collConf.registerTrigger(broker,
> >> > +                                               "store,update,remove",
> >> > +
> >> > "org.exist.config.ConfigurationDocumentTrigger", null);
> >> > +                       } catch (CollectionConfigurationException e) {
> >> > +                               //LOG.error("Configuration changers
> >> > watcher could not the initialized.", e);
> >> > +                       }
> >> > +        }
> >> >     }
> >> >
> >> >     public long getReservedMem() {
> >> >
> >> >
> >> > This was sent by the SourceForge.net collaborative development
> platform,
> >> > the world's largest Open Source development site.
> >> >
> >> >
> >> >
> ------------------------------------------------------------------------------
> >> > The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio
> >> > XE:
> >> > Pinpoint memory and threading errors before they happen.
> >> > Find and fix more than 250 security defects in the development cycle.
> >> > Locate bottlenecks in serial and parallel code that limit performance.
> >> > http://p.sf.net/sfu/intel-dev2devfeb
> >> > _______________________________________________
> >> > Exist-commits mailing list
> >> > Exi...@li...
> >> > https://lists.sourceforge.net/lists/listinfo/exist-commits
> >> >
> >>
> >>
> >>
> >> --
> >> Adam Retter
> >>
> >> eXist Developer
> >> { United Kingdom }
> >> ad...@ex...
> >> irc://irc.freenode.net/existdb
> >
> >
> >
> > --
> > Dmitriy Shabanov
> >
>
>
>
> --
> Adam Retter
>
> eXist Developer
> { United Kingdom }
> ad...@ex...
> irc://irc.freenode.net/existdb
>



-- 
Dmitriy Shabanov

Re: [Exist-development] [Exist-commits] SF.net SVN: exist:[13782] trunk/eXist/src/org/exist/storage/BrokerPool.java

[Adam R. <ad...@ex...>]

Whats it for?

On 16 February 2011 16:33, Dmitriy Shabanov <sha...@gm...> wrote:
> Wolfgang did ask to do not use '/db/system' because it have 'rwurwu---', but
> 'etc' 'rwurwuR--'.
>
> On Wed, Feb 16, 2011 at 7:29 PM, Adam Retter <ad...@ex...> wrote:
>>
>> What is /db/etc? Should this not be /db/system/etc if anything?
>>
>>
>> On 16 February 2011 14:13,  <sha...@us...> wrote:
>> > Revision: 13782
>> >          http://exist.svn.sourceforge.net/exist/?rev=13782&view=rev
>> > Author:   shabanovd
>> > Date:     2011-02-16 13:13:58 +0000 (Wed, 16 Feb 2011)
>> >
>> > Log Message:
>> > -----------
>> > [feature] create '/db/etc' collection.
>> >
>> > Owner: SYSTEM with permissions: rwu;
>> > Group: DBA with permissions: rwu;
>> > Other's permissions: r--;
>> >
>> > It will be monitored by configuration trigger and it must/should be used
>> > for settings which must/should be readable by world.
>> >
>> > Modified Paths:
>> > --------------
>> >    trunk/eXist/src/org/exist/storage/BrokerPool.java
>> >
>> > Modified: trunk/eXist/src/org/exist/storage/BrokerPool.java
>> > ===================================================================
>> > --- trunk/eXist/src/org/exist/storage/BrokerPool.java   2011-02-16
>> > 13:03:39 UTC (rev 13781)
>> > +++ trunk/eXist/src/org/exist/storage/BrokerPool.java   2011-02-16
>> > 13:13:58 UTC (rev 13782)
>> > @@ -961,25 +961,23 @@
>> >     }*/
>> >
>> >      /**
>> > -     * Initialise system collections, if it doesnt exist yet
>> > +     * Initialise system collections, if it doesn't exist yet
>> >      *
>> >      * @param sysBroker The system broker from before the brokerpool is
>> > populated
>> >      * @param sysCollectionUri XmldbURI of the collection to create
>> > -     * @param permissions The persmissions to set on the created
>> > collection
>> > +     * @param permissions The permissions to set on the created
>> > collection
>> >      */
>> > -    private void initialiseSystemCollection(DBBroker sysBroker,
>> > XmldbURI sysCollectionUri, int permissions) throws EXistException {
>> > -        //create /db/system
>> > -        Collection sysCollection =
>> > sysBroker.getCollection(sysCollectionUri);
>> > -        if (sysCollection == null)
>> > -        {
>> > +    private Collection initialiseSystemCollection(DBBroker sysBroker,
>> > XmldbURI sysCollectionUri, int permissions) throws EXistException {
>> > +        Collection collection =
>> > sysBroker.getCollection(sysCollectionUri);
>> > +        if (collection == null) {
>> >             TransactionManager transact = getTransactionManager();
>> >             Txn txn = transact.beginTransaction();
>> >             try {
>> > -                sysCollection = sysBroker.getOrCreateCollection(txn,
>> > sysCollectionUri);
>> > -                if (sysCollection == null)
>> > +                collection = sysBroker.getOrCreateCollection(txn,
>> > sysCollectionUri);
>> > +                if (collection == null)
>> >                     throw new IOException("Could not create system
>> > collection: " + sysCollectionUri);
>> > -                sysCollection.setPermissions(permissions);
>> > -                sysBroker.saveCollection(txn, sysCollection);
>> > +                collection.setPermissions(permissions);
>> > +                sysBroker.saveCollection(txn, collection);
>> >
>> >                 transact.commit(txn);
>> >             } catch (Exception e) {
>> > @@ -990,19 +988,36 @@
>> >                 throw new EXistException(msg, e);
>> >             }
>> >         }
>> > +
>> > +        return collection;
>> >     }
>> >
>> >     /**
>> > -     * Initialise required system collections, if they dont exist yet
>> > +     * Initialize required system collections, if they don't exist yet
>> >      *
>> >      * @param sysBroker - The system broker from before the brokerpool
>> > is populated
>> >      *
>> >      * @throws EXistException If a system collection cannot be created
>> >      */
>> > -    private void initialiseSystemCollections(DBBroker sysBroker) throws
>> > EXistException
>> > +    private void initialiseSystemCollections(DBBroker broker) throws
>> > EXistException
>> >     {
>> >         //create /db/system
>> > -        initialiseSystemCollection(sysBroker,
>> > XmldbURI.SYSTEM_COLLECTION_URI, 0770);
>> > +        initialiseSystemCollection(broker,
>> > XmldbURI.SYSTEM_COLLECTION_URI, 0770);
>> > +        //create /db/etc
>> > +        Collection collection = initialiseSystemCollection(broker,
>> > XmldbURI.ETC_COLLECTION_URI, 0774);
>> > +
>> > +        //initialize configurations watcher trigger
>> > +        if (collection != null) {
>> > +               CollectionConfigurationManager manager =
>> > broker.getBrokerPool().getConfigurationManager();
>> > +               CollectionConfiguration collConf =
>> > manager.getOrCreateCollectionConfiguration(broker, collection);
>> > +               try {
>> > +                               collConf.registerTrigger(broker,
>> > +                                               "store,update,remove",
>> > +
>> > "org.exist.config.ConfigurationDocumentTrigger", null);
>> > +                       } catch (CollectionConfigurationException e) {
>> > +                               //LOG.error("Configuration changers
>> > watcher could not the initialized.", e);
>> > +                       }
>> > +        }
>> >     }
>> >
>> >     public long getReservedMem() {
>> >
>> >
>> > This was sent by the SourceForge.net collaborative development platform,
>> > the world's largest Open Source development site.
>> >
>> >
>> > ------------------------------------------------------------------------------
>> > The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio
>> > XE:
>> > Pinpoint memory and threading errors before they happen.
>> > Find and fix more than 250 security defects in the development cycle.
>> > Locate bottlenecks in serial and parallel code that limit performance.
>> > http://p.sf.net/sfu/intel-dev2devfeb
>> > _______________________________________________
>> > Exist-commits mailing list
>> > Exi...@li...
>> > https://lists.sourceforge.net/lists/listinfo/exist-commits
>> >
>>
>>
>>
>> --
>> Adam Retter
>>
>> eXist Developer
>> { United Kingdom }
>> ad...@ex...
>> irc://irc.freenode.net/existdb
>
>
>
> --
> Dmitriy Shabanov
>



-- 
Adam Retter

eXist Developer
{ United Kingdom }
ad...@ex...
irc://irc.freenode.net/existdb

Re: [Exist-development] [Exist-commits] SF.net SVN: exist:[13782] trunk/eXist/src/org/exist/storage/BrokerPool.java

[Dmitriy S. <sha...@gm...>]

Wolfgang did ask to do not use '/db/system' because it have 'rwurwu---', but
'etc' 'rwurwuR--'.

On Wed, Feb 16, 2011 at 7:29 PM, Adam Retter <ad...@ex...> wrote:

> What is /db/etc? Should this not be /db/system/etc if anything?
>
>
> On 16 February 2011 14:13,  <sha...@us...> wrote:
> > Revision: 13782
> >          http://exist.svn.sourceforge.net/exist/?rev=13782&view=rev
> > Author:   shabanovd
> > Date:     2011-02-16 13:13:58 +0000 (Wed, 16 Feb 2011)
> >
> > Log Message:
> > -----------
> > [feature] create '/db/etc' collection.
> >
> > Owner: SYSTEM with permissions: rwu;
> > Group: DBA with permissions: rwu;
> > Other's permissions: r--;
> >
> > It will be monitored by configuration trigger and it must/should be used
> for settings which must/should be readable by world.
> >
> > Modified Paths:
> > --------------
> >    trunk/eXist/src/org/exist/storage/BrokerPool.java
> >
> > Modified: trunk/eXist/src/org/exist/storage/BrokerPool.java
> > ===================================================================
> > --- trunk/eXist/src/org/exist/storage/BrokerPool.java   2011-02-16
> 13:03:39 UTC (rev 13781)
> > +++ trunk/eXist/src/org/exist/storage/BrokerPool.java   2011-02-16
> 13:13:58 UTC (rev 13782)
> > @@ -961,25 +961,23 @@
> >     }*/
> >
> >      /**
> > -     * Initialise system collections, if it doesnt exist yet
> > +     * Initialise system collections, if it doesn't exist yet
> >      *
> >      * @param sysBroker The system broker from before the brokerpool is
> populated
> >      * @param sysCollectionUri XmldbURI of the collection to create
> > -     * @param permissions The persmissions to set on the created
> collection
> > +     * @param permissions The permissions to set on the created
> collection
> >      */
> > -    private void initialiseSystemCollection(DBBroker sysBroker, XmldbURI
> sysCollectionUri, int permissions) throws EXistException {
> > -        //create /db/system
> > -        Collection sysCollection =
> sysBroker.getCollection(sysCollectionUri);
> > -        if (sysCollection == null)
> > -        {
> > +    private Collection initialiseSystemCollection(DBBroker sysBroker,
> XmldbURI sysCollectionUri, int permissions) throws EXistException {
> > +        Collection collection =
> sysBroker.getCollection(sysCollectionUri);
> > +        if (collection == null) {
> >             TransactionManager transact = getTransactionManager();
> >             Txn txn = transact.beginTransaction();
> >             try {
> > -                sysCollection = sysBroker.getOrCreateCollection(txn,
> sysCollectionUri);
> > -                if (sysCollection == null)
> > +                collection = sysBroker.getOrCreateCollection(txn,
> sysCollectionUri);
> > +                if (collection == null)
> >                     throw new IOException("Could not create system
> collection: " + sysCollectionUri);
> > -                sysCollection.setPermissions(permissions);
> > -                sysBroker.saveCollection(txn, sysCollection);
> > +                collection.setPermissions(permissions);
> > +                sysBroker.saveCollection(txn, collection);
> >
> >                 transact.commit(txn);
> >             } catch (Exception e) {
> > @@ -990,19 +988,36 @@
> >                 throw new EXistException(msg, e);
> >             }
> >         }
> > +
> > +        return collection;
> >     }
> >
> >     /**
> > -     * Initialise required system collections, if they dont exist yet
> > +     * Initialize required system collections, if they don't exist yet
> >      *
> >      * @param sysBroker - The system broker from before the brokerpool is
> populated
> >      *
> >      * @throws EXistException If a system collection cannot be created
> >      */
> > -    private void initialiseSystemCollections(DBBroker sysBroker) throws
> EXistException
> > +    private void initialiseSystemCollections(DBBroker broker) throws
> EXistException
> >     {
> >         //create /db/system
> > -        initialiseSystemCollection(sysBroker,
> XmldbURI.SYSTEM_COLLECTION_URI, 0770);
> > +        initialiseSystemCollection(broker,
> XmldbURI.SYSTEM_COLLECTION_URI, 0770);
> > +        //create /db/etc
> > +        Collection collection = initialiseSystemCollection(broker,
> XmldbURI.ETC_COLLECTION_URI, 0774);
> > +
> > +        //initialize configurations watcher trigger
> > +        if (collection != null) {
> > +               CollectionConfigurationManager manager =
> broker.getBrokerPool().getConfigurationManager();
> > +               CollectionConfiguration collConf =
> manager.getOrCreateCollectionConfiguration(broker, collection);
> > +               try {
> > +                               collConf.registerTrigger(broker,
> > +                                               "store,update,remove",
> > +
> "org.exist.config.ConfigurationDocumentTrigger", null);
> > +                       } catch (CollectionConfigurationException e) {
> > +                               //LOG.error("Configuration changers
> watcher could not the initialized.", e);
> > +                       }
> > +        }
> >     }
> >
> >     public long getReservedMem() {
> >
> >
> > This was sent by the SourceForge.net collaborative development platform,
> the world's largest Open Source development site.
> >
> >
> ------------------------------------------------------------------------------
> > The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
> > Pinpoint memory and threading errors before they happen.
> > Find and fix more than 250 security defects in the development cycle.
> > Locate bottlenecks in serial and parallel code that limit performance.
> > http://p.sf.net/sfu/intel-dev2devfeb
> > _______________________________________________
> > Exist-commits mailing list
> > Exi...@li...
> > https://lists.sourceforge.net/lists/listinfo/exist-commits
> >
>
>
>
> --
> Adam Retter
>
> eXist Developer
> { United Kingdom }
> ad...@ex...
> irc://irc.freenode.net/existdb
>



-- 
Dmitriy Shabanov

Re: [Exist-development] [Exist-commits] SF.net SVN: exist:[13782] trunk/eXist/src/org/exist/storage/BrokerPool.java

[Adam R. <ad...@ex...>]

What is /db/etc? Should this not be /db/system/etc if anything?


On 16 February 2011 14:13,  <sha...@us...> wrote:
> Revision: 13782
>          http://exist.svn.sourceforge.net/exist/?rev=13782&view=rev
> Author:   shabanovd
> Date:     2011-02-16 13:13:58 +0000 (Wed, 16 Feb 2011)
>
> Log Message:
> -----------
> [feature] create '/db/etc' collection.
>
> Owner: SYSTEM with permissions: rwu;
> Group: DBA with permissions: rwu;
> Other's permissions: r--;
>
> It will be monitored by configuration trigger and it must/should be used for settings which must/should be readable by world.
>
> Modified Paths:
> --------------
>    trunk/eXist/src/org/exist/storage/BrokerPool.java
>
> Modified: trunk/eXist/src/org/exist/storage/BrokerPool.java
> ===================================================================
> --- trunk/eXist/src/org/exist/storage/BrokerPool.java   2011-02-16 13:03:39 UTC (rev 13781)
> +++ trunk/eXist/src/org/exist/storage/BrokerPool.java   2011-02-16 13:13:58 UTC (rev 13782)
> @@ -961,25 +961,23 @@
>     }*/
>
>      /**
> -     * Initialise system collections, if it doesnt exist yet
> +     * Initialise system collections, if it doesn't exist yet
>      *
>      * @param sysBroker The system broker from before the brokerpool is populated
>      * @param sysCollectionUri XmldbURI of the collection to create
> -     * @param permissions The persmissions to set on the created collection
> +     * @param permissions The permissions to set on the created collection
>      */
> -    private void initialiseSystemCollection(DBBroker sysBroker, XmldbURI sysCollectionUri, int permissions) throws EXistException {
> -        //create /db/system
> -        Collection sysCollection = sysBroker.getCollection(sysCollectionUri);
> -        if (sysCollection == null)
> -        {
> +    private Collection initialiseSystemCollection(DBBroker sysBroker, XmldbURI sysCollectionUri, int permissions) throws EXistException {
> +        Collection collection = sysBroker.getCollection(sysCollectionUri);
> +        if (collection == null) {
>             TransactionManager transact = getTransactionManager();
>             Txn txn = transact.beginTransaction();
>             try {
> -                sysCollection = sysBroker.getOrCreateCollection(txn, sysCollectionUri);
> -                if (sysCollection == null)
> +                collection = sysBroker.getOrCreateCollection(txn, sysCollectionUri);
> +                if (collection == null)
>                     throw new IOException("Could not create system collection: " + sysCollectionUri);
> -                sysCollection.setPermissions(permissions);
> -                sysBroker.saveCollection(txn, sysCollection);
> +                collection.setPermissions(permissions);
> +                sysBroker.saveCollection(txn, collection);
>
>                 transact.commit(txn);
>             } catch (Exception e) {
> @@ -990,19 +988,36 @@
>                 throw new EXistException(msg, e);
>             }
>         }
> +
> +        return collection;
>     }
>
>     /**
> -     * Initialise required system collections, if they dont exist yet
> +     * Initialize required system collections, if they don't exist yet
>      *
>      * @param sysBroker - The system broker from before the brokerpool is populated
>      *
>      * @throws EXistException If a system collection cannot be created
>      */
> -    private void initialiseSystemCollections(DBBroker sysBroker) throws EXistException
> +    private void initialiseSystemCollections(DBBroker broker) throws EXistException
>     {
>         //create /db/system
> -        initialiseSystemCollection(sysBroker, XmldbURI.SYSTEM_COLLECTION_URI, 0770);
> +        initialiseSystemCollection(broker, XmldbURI.SYSTEM_COLLECTION_URI, 0770);
> +        //create /db/etc
> +        Collection collection = initialiseSystemCollection(broker, XmldbURI.ETC_COLLECTION_URI, 0774);
> +
> +        //initialize configurations watcher trigger
> +        if (collection != null) {
> +               CollectionConfigurationManager manager = broker.getBrokerPool().getConfigurationManager();
> +               CollectionConfiguration collConf = manager.getOrCreateCollectionConfiguration(broker, collection);
> +               try {
> +                               collConf.registerTrigger(broker,
> +                                               "store,update,remove",
> +                                               "org.exist.config.ConfigurationDocumentTrigger", null);
> +                       } catch (CollectionConfigurationException e) {
> +                               //LOG.error("Configuration changers watcher could not the initialized.", e);
> +                       }
> +        }
>     }
>
>     public long getReservedMem() {
>
>
> This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
>
> ------------------------------------------------------------------------------
> The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
> Pinpoint memory and threading errors before they happen.
> Find and fix more than 250 security defects in the development cycle.
> Locate bottlenecks in serial and parallel code that limit performance.
> http://p.sf.net/sfu/intel-dev2devfeb
> _______________________________________________
> Exist-commits mailing list
> Exi...@li...
> https://lists.sourceforge.net/lists/listinfo/exist-commits
>



-- 
Adam Retter

eXist Developer
{ United Kingdom }
ad...@ex...
irc://irc.freenode.net/existdb

Re: [Exist-development] XQuery: index causes empty result ?!

[Dannes W. <di...@ex...>]

This question should be posted on exist-open.......

Sent from my iPhone

On 15 feb. 2011, at 00:00, Lar...@ie... (Lars-Olof Krause) wrote:

> Hello,
> 
> I got a quit huge problem. I hope someone can help.
> I developed an XQuery for a study-project which takes 3.2s when
> executed on one XML-File using the Saxon-Processor of my Oxygen-Editor.
> Using eXist 1.5.0dev (eXist Build : 20110111) without
> index-configuration it takes 188.6s returning the same result as the
> Saxon-Processor of Oxygen.
> When defining an Index
> --------------------
> <collection xmlns="http://exist-db.org/collection-config/1.0">
>    <index>
>        <create qname="@start" type="xs:integer"/>
>        <create qname="@end" type="xs:integer"/>
>    </index>
> </collection>
> --------------------
> 
> eXist calculates 15.6s which is still slower than the Saxon-Processor
> of Oxygen but the main problem is: It returns an empty result!
> 
> Can someone please help me an give me an advice what to do?
> 
> Since the XQuery is quite long, I uploadet it to my server:
> http://loewe.fh.lok-soft.net/psc.xquery
> 
> You can find my test xml file here:
> http://loewe.fh.lok-soft.net/mueller_hpsg_2008-txsf.xml
> 
> 
> Please help me I really don't know what I could do now.
> 
> Lars-Olof Krause
> 
> 
> ------------------------------------------------------------------------------
> The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
> Pinpoint memory and threading errors before they happen.
> Find and fix more than 250 security defects in the development cycle.
> Locate bottlenecks in serial and parallel code that limit performance.
> http://p.sf.net/sfu/intel-dev2devfeb
> _______________________________________________
> Exist-development mailing list
> Exi...@li...
> https://lists.sourceforge.net/lists/listinfo/exist-development

Re: [Exist-development] XQuery: index causes empty result ?!

[Wolfgang M. <wol...@ex...>]

I actually had a look at your query yesterday. Creating indexes is certainly
the way to go, but I confirm I get an empty result when I create one on
start and end.

Unfortunately, the query is rather complex, so it's  hard to say where the
problem is. I do not understand it entirely yet. I'll have another look at
this later, though I won't have much time due to meetings.

Adding + 0 to both operands of a comparison expression may cause a problem
with index use and cost performance. You may try to factor this out, at
least for the left operand. But the problem remains that we get an empty
result with the integer index. This had to be investigated.

Wolfgang

Am 15.02.2011 00:00 schrieb "Lars-Olof Krause" <
Lar...@ie...>:
> Hello,
>
> I got a quit huge problem. I hope someone can help.
> I developed an XQuery for a study-project which takes 3.2s when
> executed on one XML-File using the Saxon-Processor of my Oxygen-Editor.
> Using eXist 1.5.0dev (eXist Build : 20110111) without
> index-configuration it takes 188.6s returning the same result as the
> Saxon-Processor of Oxygen.
> When defining an Index
> --------------------
> <collection xmlns="http://exist-db.org/collection-config/1.0">
> <index>
> <create qname="@start" type="xs:integer"/>
> <create qname="@end" type="xs:integer"/>
> </index>
> </collection>
> --------------------
>
> eXist calculates 15.6s which is still slower than the Saxon-Processor
> of Oxygen but the main problem is: It returns an empty result!
>
> Can someone please help me an give me an advice what to do?
>
> Since the XQuery is quite long, I uploadet it to my server:
> http://loewe.fh.lok-soft.net/psc.xquery
>
> You can find my test xml file here:
> http://loewe.fh.lok-soft.net/mueller_hpsg_2008-txsf.xml
>
>
> Please help me I really don't know what I could do now.
>
> Lars-Olof Krause
>
>
>
------------------------------------------------------------------------------
> The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
> Pinpoint memory and threading errors before they happen.
> Find and fix more than 250 security defects in the development cycle.
> Locate bottlenecks in serial and parallel code that limit performance.
> http://p.sf.net/sfu/intel-dev2devfeb
> _______________________________________________
> Exist-development mailing list
> Exi...@li...
> https://lists.sourceforge.net/lists/listinfo/exist-development

Re: [Exist-development] XQuery: index causes empty result ?!

[Dmitriy S. <sha...@gm...>]

let $doc :=
doc("/Users/LOK/Desktop/fh/Master/3Semester/Entwicklungsprojekt/korpus/loewe-a3/tocs/layers/txsf_u8/mueller_hpsg_2008-txsf.xml")

it will not use index, because it load in memory and do not use any
optimization then. Can you load it to db? What time you get if file stored
at db?

Also, you should rewrite 'where' to predicates.

Another thing is 'contains($selfRange, "Valenz")' or
'contains(lower-case($selfRange), "valenz")' and etc... can you use lucene
index for this? Only this will speed-up a lot.

On Tue, Feb 15, 2011 at 4:00 AM, Lars-Olof Krause <
Lar...@ie...> wrote:

> Hello,
>
> I got a quit huge problem. I hope someone can help.
> I developed an XQuery for a study-project which takes 3.2s when
> executed on one XML-File using the Saxon-Processor of my Oxygen-Editor.
> Using eXist 1.5.0dev (eXist Build : 20110111) without
> index-configuration it takes 188.6s returning the same result as the
> Saxon-Processor of Oxygen.
> When defining an Index
> --------------------
> <collection xmlns="http://exist-db.org/collection-config/1.0">
>    <index>
>        <create qname="@start" type="xs:integer"/>
>        <create qname="@end" type="xs:integer"/>
>    </index>
> </collection>
> --------------------
>
> eXist calculates 15.6s which is still slower than the Saxon-Processor
> of Oxygen but the main problem is: It returns an empty result!
>
> Can someone please help me an give me an advice what to do?
>
> Since the XQuery is quite long, I uploadet it to my server:
> http://loewe.fh.lok-soft.net/psc.xquery
>
> You can find my test xml file here:
> http://loewe.fh.lok-soft.net/mueller_hpsg_2008-txsf.xml
>
>

-- 
Dmitriy Shabanov

[Exist-development] XQuery: index causes empty result ?!

[<Lar...@ie...>]

Hello,

I got a quit huge problem. I hope someone can help.
I developed an XQuery for a study-project which takes 3.2s when
executed on one XML-File using the Saxon-Processor of my Oxygen-Editor.
Using eXist 1.5.0dev (eXist Build : 20110111) without
index-configuration it takes 188.6s returning the same result as the
Saxon-Processor of Oxygen.
When defining an Index
--------------------
<collection xmlns="http://exist-db.org/collection-config/1.0">
    <index>
        <create qname="@start" type="xs:integer"/>
        <create qname="@end" type="xs:integer"/>
    </index>
</collection>
--------------------

eXist calculates 15.6s which is still slower than the Saxon-Processor
of Oxygen but the main problem is: It returns an empty result!

Can someone please help me an give me an advice what to do?

Since the XQuery is quite long, I uploadet it to my server:
http://loewe.fh.lok-soft.net/psc.xquery

You can find my test xml file here:
http://loewe.fh.lok-soft.net/mueller_hpsg_2008-txsf.xml


Please help me I really don't know what I could do now.

Lars-Olof Krause

[Exist-development] Friday teleconference today

[Wolfgang M. <wol...@ex...>]

Hi guys,

Since we had already covered most of our topics last week, Dannes and I just
agreed that we may skip the Friday developers conference this week - unless
someone has an important topic to discuss. If so please shout.

Wolfgang

Re: [Exist-development] How do you get the exist.sourceforge site to be served be served by eXist?

[Chris M. <vol...@gm...>]

Thanks!
I'm wondering if it mightn't be possible to have sf.net reverse proxy
from my server machine ... but that's a question I'll direct at them.

On Thu, Feb 10, 2011 at 5:08 AM, Dannes Wessels <da...@ex...> wrote:
> Hi
>
> On Thu, Feb 10, 2011 at 5:59 AM, Dmitriy Shabanov <sha...@gm...>
> wrote:
>>
>> It hosted externally and redirected by sf.
>
> No, the 'static content' is hosted by sf.net ; some content is hosted at
> demo.exist-db.org .....
>
> cheers
>
> Dannes
>
> --
> eXist-db Native XML Database - http://exist-db.org
> Join us on linked-in: http://www.linkedin.com/groups?gid=35624
>

Re: [Exist-development] Compiling XQueries?

[Joe W. <jo...@gm...>]

> As far as I know there is still
> no XQuery beautifier available

Have you looked into xqpretty, a google code project created by Martin Probst?

  http://code.google.com/p/xqpretty/

See his blog post too:

  http://www.martin-probst.com/blog/2008/05/07/xquery-pretty-printer-and-grammar-open-source/

Joe

Re: [Exist-development] Compiling XQueries?

[Andrzej J. T. <an...@ch...>]

Thomas:

> I think this could be the easiest way to go. As far as I know there is still no XQuery beautifier
> available and if your code is of thousands of lines then the obfuscated code is a simple and
> efficient code protection which can be part of the build, that is good enough in most of the cases.

That is kinda what I'm thinking as well.  Just make it an Ant task/target and it becomes quite easy
to include in deployment processes.

-- 
Andrzej Taramina
Chaeron Corporation: Enterprise System Solutions
http://www.chaeron.com

Re: [Exist-development] Compiling XQueries?

[Thomas W. <tho...@gm...>]

Andrzej,


On 10 February 2011 13:31, Andrzej Jan Taramina <an...@ch...> wrote:

> The combination of obfuscating variable/function names along with
> serializing the query, would make
> the code pretty much useless for all but the most masochistic of hackers.
>

I think this could be the easiest way to go. As far as I know there is still
no XQuery beautifier available and if your code is of thousands of lines
then the obfuscated code is a simple and efficient code protection which can
be part of the build, that is good enough in most of the cases.

Thomas

------

Thomas White

Mobile:+44 7711 922 966
Skype: thomaswhite
gTalk: thomas.0007
http://www.linkedin.com/in/thomaswhite0007
http://www.facebook.com/thomas.0007

Re: [Exist-development] Compiling XQueries?

[Andrzej J. T. <an...@ch...>]

Dmitriy Shabanov wrote:

> The wrestling parser ( http://smartims.hg.sourceforge.net/hgweb/smartims/wrestling/ ) will be able
> to generate java byte code from xquery.

Not very compatible with eXist though, I assume, which is what I'm looking for.

> I don't see the big difference between serialized data and open xquery scripts, tool that generate
> xquery script from serialized data quite simple task.

It's not simple for casual users. Still better than open code.

>     The other alternative would be to write an obfuscator for XQuery.

> From my point of view, you will not find any good solution to solve your problem in this direction,
> try to look around for other options.

Why not?

The combination of obfuscating variable/function names along with serializing the query, would make
the code pretty much useless for all but the most masochistic of hackers.


-- 
Andrzej Taramina
Chaeron Corporation: Enterprise System Solutions
http://www.chaeron.com

Re: [Exist-development] How do you get the exist.sourceforge site to be served be served by eXist?

[Dannes W. <da...@ex...>]

Hi

On Thu, Feb 10, 2011 at 5:59 AM, Dmitriy Shabanov <sha...@gm...>wrote:

> It hosted externally and redirected by sf.


No, the 'static content' is hosted by sf.net ; some content is hosted at
demo.exist-db.org .....

cheers

Dannes

-- 
eXist-db Native XML Database - http://exist-db.org
Join us on linked-in: http://www.linkedin.com/groups?gid=35624

Re: [Exist-development] How do you get the exist.sourceforge site to be served be served by eXist?

[Dmitriy S. <sha...@gm...>]

It hosted externally and redirected by sf.

On Thu, Feb 10, 2011 at 8:26 AM, Chris Maloney <vol...@gm...> wrote:

> Hi, I just have a quick question I hope somebody can help me answer.  I'm
> running another Sourceforge project, and am also getting my feet wet with
> eXist, and I notice that exist.sourceforge.net is served by exist.  How do
> you set this up?  I've searched the Sourceforge help pages, but it doesn't
> seem to me that Java servlets are supported.  I've tried searching the email
> lists and blogs, but my search terms are too common -- can't get a hit.
>  Thanks!


-- 
Dmitriy Shabanov

Re: [Exist-development] Compiling XQueries?

[Dmitriy S. <sha...@gm...>]

On Thu, Feb 10, 2011 at 1:11 AM, Andrzej Jan Taramina
<an...@ch...>wrote:

> Wolfgang:
>
> > eXist compiles every xquery into a final representation, which is a tree
> of expression objects. Each
> > object is a subclass of Expression. I think it would be rather easy to
> generate java code which
> > recreates this expression tree. This way you could save the time needed
> for parsing and for
> > constructing the intermediate AST, but you would not save anything at
> runtime.
>
> Would that expression tree (or AST) be serializable, currently?
>
> If so, then it should be a pretty simple matter to reuse the existing
> compile code, grab the tree
> and serialize it out.  And on the other side, read the tree, re-instantiate
> it, and then execute it.
>
> > In a next step, one could also consider to directly generate the java
> code (or bytecode) for
> > handling certain types of expressions, eg. an if-then-else.
>
> I'm not interested in generating java or bytecode.
>
> The use case I'm thinking of is being able to distribute compiled XQueries,
> in the situation where
> you don't want to make the original source code available. Right now, if
> you deploy an
> eXist/XQuery-based app at a customer site, if they can get into eXist, they
> can get the source.
>

The wrestling parser (
http://smartims.hg.sourceforge.net/hgweb/smartims/wrestling/ ) will be able
to generate java byte code from xquery.

I don't see the big difference between serialized data and open xquery
scripts, tool that generate xquery script from serialized data quite simple
task.


> The other alternative would be to write an obfuscator for XQuery.
>

>From my point of view, you will not find any good solution to solve your
problem in this direction, try to look around for other options.

>
-- 
Dmitriy Shabanov

[Exist-development] How do you get the exist.sourceforge site to be served be served by eXist?

[Chris M. <vol...@gm...>]

Hi, I just have a quick question I hope somebody can help me answer.  I'm
running another Sourceforge project, and am also getting my feet wet with
eXist, and I notice that exist.sourceforge.net is served by exist.  How do
you set this up?  I've searched the Sourceforge help pages, but it doesn't
seem to me that Java servlets are supported.  I've tried searching the email
lists and blogs, but my search terms are too common -- can't get a hit.
Thanks!

Re: [Exist-development] Compiling XQueries?

[Andrzej J. T. <an...@ch...>]

> this remind me M$ seminars - same words about aspx files, is your code
> so ugly that you don`t want show it? :)

I was tempted to tell you to stick it in your aspx! ;-)

I'll stack up my code against anyone's....the issue is not that. Just don't want to release source
code for a proprietary application is all.

-- 
Andrzej Taramina
Chaeron Corporation: Enterprise System Solutions
http://www.chaeron.com

Re: [Exist-development] Compiling XQueries?

[Вячеслав С. <sch...@gm...>]

"The use case I'm thinking of is being able to distribute compiled
XQueries, in the situation where
you don't want to make the original source code available. Right now,
if you deploy an
eXist/XQuery-based app at a customer site, if they can get into eXist,
they can get the source."

this remind me M$ seminars - same words about aspx files, is your code
so ugly that you don`t want show it? :)

2011/2/9 Wolfgang Meier <wol...@ex...>:
>> Would that expression tree (or AST) be serializable, currently?
>
> Not yet. But since there are only maybe 20 or 30 core expressions, it
> would be feasible to make them serializable in one way or another (you
> could even use an ExpressionVisitor to generate the Java code needed
> to recreate the tree). The huge number of functions is not a problem
> since they would all be serialized in the same way.
>
> Wolfgang
>
> ------------------------------------------------------------------------------
> The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
> Pinpoint memory and threading errors before they happen.
> Find and fix more than 250 security defects in the development cycle.
> Locate bottlenecks in serial and parallel code that limit performance.
> http://p.sf.net/sfu/intel-dev2devfeb
> _______________________________________________
> Exist-development mailing list
> Exi...@li...
> https://lists.sourceforge.net/lists/listinfo/exist-development
>

Re: [Exist-development] Compiling XQueries?

[Wolfgang M. <wol...@ex...>]

> Would that expression tree (or AST) be serializable, currently?

Not yet. But since there are only maybe 20 or 30 core expressions, it
would be feasible to make them serializable in one way or another (you
could even use an ExpressionVisitor to generate the Java code needed
to recreate the tree). The huge number of functions is not a problem
since they would all be serialized in the same way.

Wolfgang

Re: [Exist-development] Compiling XQueries?

[Andrzej J. T. <an...@ch...>]

Wolfgang:

> eXist compiles every xquery into a final representation, which is a tree of expression objects. Each
> object is a subclass of Expression. I think it would be rather easy to generate java code which
> recreates this expression tree. This way you could save the time needed for parsing and for
> constructing the intermediate AST, but you would not save anything at runtime.

Would that expression tree (or AST) be serializable, currently?

If so, then it should be a pretty simple matter to reuse the existing compile code, grab the tree
and serialize it out.  And on the other side, read the tree, re-instantiate it, and then execute it.

> In a next step, one could also consider to directly generate the java code (or bytecode) for
> handling certain types of expressions, eg. an if-then-else.

I'm not interested in generating java or bytecode.

The use case I'm thinking of is being able to distribute compiled XQueries, in the situation where
you don't want to make the original source code available. Right now, if you deploy an
eXist/XQuery-based app at a customer site, if they can get into eXist, they can get the source.

The other alternative would be to write an obfuscator for XQuery.

-- 
Andrzej Taramina
Chaeron Corporation: Enterprise System Solutions
http://www.chaeron.com

Re: [Exist-development] Compiling XQueries?

[Wolfgang M. <wol...@ex...>]

eXist compiles every xquery into a final representation, which is a tree of
expression objects. Each object is a subclass of Expression. I think it
would be rather easy to generate java code which recreates this expression
tree. This way you could save the time needed for parsing and for
constructing the intermediate AST, but you would not save anything at
runtime.

In a next step, one could also consider to directly generate the java code
(or bytecode) for handling certain types of expressions, eg. an
if-then-else.

However, the big question is if you would win so much as to justify the
effort? My impression is that the raw language processing speed in eXist is
usually pretty good, with some exceptions where the processing has to be
further improved. Most real-world performance issues occur at the database
level. What normally helps here is improving index use, automatic query
rewriting and internal optimizations. I often experience that a certain
performance bottleneck is caused by the optimizer making bad decisions or
not using an index where it actually could. Working on those things often
results in huge performance gains.

Compared to this, the advantages of eg compiling xquery directly to java
bytecode would probably be rather small.

Wolfgang

Am 09.02.2011 19:49 schrieb "Andrzej Jan Taramina" <an...@ch...>:
> I know that internally, we compile XQueries and then cache them, prior to
execution.
>
> How difficult/involved would it be to expose the compile process, so that
I could compile XQueries
> ahead of time (say using an appropriate Ant task), and store/execute the
pre-compiled XQueries
> in/from the DB?
>
> Thanks!
>
> --
> Andrzej Taramina
> Chaeron Corporation: Enterprise System Solutions
> http://www.chaeron.com
>
>
------------------------------------------------------------------------------
> The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
> Pinpoint memory and threading errors before they happen.
> Find and fix more than 250 security defects in the development cycle.
> Locate bottlenecks in serial and parallel code that limit performance.
> http://p.sf.net/sfu/intel-dev2devfeb
> _______________________________________________
> Exist-development mailing list
> Exi...@li...
> https://lists.sourceforge.net/lists/listinfo/exist-development

Re: [Exist-development] Compiling XQueries?

[Joe W. <jo...@gm...>]

>> I know that internally, we compile XQueries and then cache them, prior to execution.
>>
>> How difficult/involved would it be to expose the compile process, so that I could compile XQueries
>> ahead of time (say using an appropriate Ant task), and store/execute the pre-compiled XQueries
>> in/from the DB?
>
> is it compiled to java?

Ah, module caching - I haven't heard that in a while.  The last
discussion that I recall was back in 2009:
http://markmail.org/message/lpmvwc7q4zw5zqw7.

>>> We were wondering if anything could be optimized where these are not having to be repeated loaded?
>>
>> The module caching is temporary disabled because of concurrency issues. We already agreed on refactoring after release to solve that problems.
>
> However, eXist does cache the main query, along with all the modules it imports.

This isn't an answer to your question, but this is where things still
remain unless I missed some other development.

Cheers,
Joe

Re: [Exist-development] Compiling XQueries?

[Вячеслав С. <sch...@gm...>]

is it compiled to java?

2011/2/9 Andrzej Jan Taramina <an...@ch...>:
> I know that internally, we compile XQueries and then cache them, prior to execution.
>
> How difficult/involved would it be to expose the compile process, so that I could compile XQueries
> ahead of time (say using an appropriate Ant task), and store/execute the pre-compiled XQueries
> in/from the DB?
>
> Thanks!
>
> --
> Andrzej Taramina
> Chaeron Corporation: Enterprise System Solutions
> http://www.chaeron.com
>
> ------------------------------------------------------------------------------
> The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
> Pinpoint memory and threading errors before they happen.
> Find and fix more than 250 security defects in the development cycle.
> Locate bottlenecks in serial and parallel code that limit performance.
> http://p.sf.net/sfu/intel-dev2devfeb
> _______________________________________________
> Exist-development mailing list
> Exi...@li...
> https://lists.sourceforge.net/lists/listinfo/exist-development
>