aksw / erfurt

PHP5 / Zend based Semantic Web API for Social Semantic Software

Erfurt

Erfurt is a PHP/Zend based Semantic Web Framework for Social Semantic Software.

CI System	Status
Travis CI
owdev Jenkins

API Documentation

Features

• RDF Parser/serializer
• Storage abstraction/RDF Store
• Versioning
• SPARQL querying/update
• Caching
• Plugins/trigger
• ...

Developer Info

Repository Conventions

We use Vincent Driessen's branching model, Tom Preston-Werner's versioning specification as well as this note on good commit messages for this repository.

In addition to that, we suggest to use git-flow to keep naming conventions. Copy the following config section in you global ~/.gitconfig or the repository wide .git/config file.

# git flow default configuration for ~/.gitconfig
[gitflow "branch"]
    master = master
    develop = develop
[gitflow "prefix"]
    feature = feature/
    release = release/
    hotfix = hotfix/
    support = support/
    versiontag = v

Code Conventions

Currently, this library is developed using OntoWiki's coding standard.

erfurt's Issues

SysOnt redirect returns #404

http://ns.ontowiki.net/SysOnt/ is currently redirected to https://raw.github.com/AKSW/Erfurt/develop/Erfurt/include/SysOnt.rdf while it should be https://raw.github.com/AKSW/Erfurt/develop/library/Erfurt/include/SysOnt.rdf

(/library missing)

addMultipleStatements() only chunks on the first dimension in $statementsArray

What if we have one resource with many predicates? Looks like endless cycle while trying to increase chunk amount.

(Erfurt_Store_Adapter_Virtuoso)

RDF/XML Export does not work

Exporting a Knowledge Base as RDF/XML does not work. It only exports the prefixes, but no data.

All other formats work as expected AFIK.

Model creation failure with Virtuoso

Failed creating the model. Reason:
SPARQL Error: [OpenLink][Virtuoso iODBC Driver][Virtuoso Server]SR012:
Function aref needs a string or an array as argument 1, not an arg of type DB_NULL (204)
() in query: CREATE SILENT GRAPH <http://foo>.

Virtuoso info:

• Version: 06.01.3127
• Build: Apr 12 2012

Cache Invalidation throws warning when calling invalidateWithModelIri

When invalidateWithModelIri() is called a warning is generated as shown in this image:

This issue was discussed in 8731310 and 870b9ea

Turtle parser should support qnames without local part

According to [1] resp. [2] it is allowed to use just the prefix without local part in turtle.

@prefix rdf:  <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .
@prefix void: <http://rdfs.org/ns/void#> .
@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .
…
rdf:  rdfs:label "" ;
      void:exampleResource
rdf: .

This is currently parsed to the triples:

<http://www.w3.org/1999/02/22-rdf-syntax-ns#> <http://www.w3.org/2000/01/rdf-schema#label> "" .
<http://www.w3.org/1999/02/22-rdf-syntax-ns#> <http://rdfs.org/ns/void#exampleResource> <http://www.w3.org/1999/02/22-rdf-syntax-ns# .> .

Which is obviously wrong (Se the . before the last closing >).

[1] http://answers.semanticweb.com/questions/3239/can-qname-be-used-without-a-local-name-part
[2] http://www.w3.org/TeamSubmission/turtle/#sec-grammar-grammar (point 27)

Getting Started: Authentication should be loaded from config

When following https://github.com/AKSW/Erfurt/wiki/Getting-Started I got the impression that

# Authentification on OntoWiki (need for model access)
$app->authenticate( 'root', 'root' );

uses usernames and passwords that have been created with ontowiki, even if I don't have any ontowiki instance.

I could however use the store.zenddb.username and store.zenddb.password values from the config file without problems.

If that's wrong, please clarify that on the page. (https://github.com/AKSW/Erfurt/wiki/Internals specifically tells you to use those value). Also show how to use the settings from the config file without loading it myself, since erfurt should have loaded it automatically.

And why doesn't Erfurt use those settings automatically?

cache db backend: Invalid argument supplied for foreach()

From Jeff:

I downloaded the new archive, edited the config.ini-dist file, typed
localhost/ontowiki in my browser, and it seems to work! There was an
error message sort of hiding behind the contents of the page

Warning: Invalid argument supplied for foreach() in C:\xampp\htdocs
\ontowiki\libraries\Erfurt\Erfurt\Cache\Backend\Database.php on line 267

When I typed my login information, the error disappeared

addMultipleStatements() always guesses one chunk first

Maybe it should make a better first guess.

(Erfurt_Store_Adapter_Virtuoso)

Getting Started: Class 'Erfurt_Syntax_RdfParser' not found

When following https://github.com/AKSW/Erfurt/wiki/Getting-Started, I get an error

Fatal error: Class 'Erfurt_Syntax_RdfParser' not found in /home/cweiske/Dev/sabe/lib/Erfurt/Erfurt/Store.php on line 375

when running

$app->getStore()->getModel('..');

This is because Erfurt_Syntax_RdfParser::LOCATOR_FILE is used without that class being included yet.

Versioning failure by using deleteMatchingStatements function in a transaction

Prettier output from Erfurt RDF/Turtle export

The Turtle source as exported from
http://data.bbib.no/source/edit/r/Nordahl_Grieg
is not very readable for the part containing frbr:creatorOf. Never mind the frbr:creatorOf appearing twice. This is an error on my part whilst using OntoWiki, adding a separate instance of a property widget for frbr:creatorOf, instead of using the + button to expand the widget. This is an OntoWiki issue, I suppose.

The readability of the line

                frbr:creatorOf <http://data.bbib.no/data/17._mai_1914>, <http://data.bbib.no/data/Ung%20m%C3%A5%20verden%20ennu%20v%C3%A6re>, <http://data.bbib.no/data/Ung%20m%C3%A5%20verden%20ennu%20v%C3%A6re%20%3A%20roman>, <http://data.bbib.no/data/V%C3%A5r%20%C3%A6re%20og%20v%C3%A5r%20makt> ;

is what concerns this issue.

Exporting this as

                frbr:creatorOf <http://data.bbib.no/data/17._mai_1914>, 
                                <http://data.bbib.no/data/Ung%20m%C3%A5%20verden%20ennu%20v%C3%A6re>, 
                                <http://data.bbib.no/data/Ung%20m%C3%A5%20verden%20ennu%20v%C3%A6re%20%3A%20roman>, 
                                <http://data.bbib.no/data/V%C3%A5r%20%C3%A6re%20og%20v%C3%A5r%20makt> ;

would be more readable, and still valid Turtle syntax, I suppose.

;)Frode

Method to generate new URIs (like auto_increment)

When adding new statements to a store/model that e.g. are blank nodes, or just URIs you don't care about, one often uses a URL like

http://example.org/some/uri/$index

$index is being counted up here.

Currently, Erfurt offers no method to get a new URI based on the prefix (http://example.org/some/uri/), which makes it hard to continously generate such URIs among several PHP calls. It leads to code like this (from Syntax/RdfParser/Adapter/Turtle.php):

        while (true) {
           $id = self::BNODE_PREFIX . ++$this->_bnodeCounter;
           if (!isset($this->_usedBnodeIds[$id])) {
               break;
           }
       }    

This is also the same code I have to use to do that in my application. This code is very slow if you have to do that against a database.

Erfurt should offer a method to create such incrementing URIs. I imagine something like this:

var_dump($model->getNextFreeUri('http://example.org/some/uri/'));
string(xx) "http://example.org/some/uri/3721"
var_dump($model->getNextFreeUri('http://example.org/some/uri/'));
string(xx) "http://example.org/some/uri/3722"

Exception: Variable "" not selected

Hi,
is it Erfurt or OntoWiki related? I use Erfurt, because of the Exception from Erfurt.

I add some triples about our own extension ( forms.ontowiki). After that i wanted to get an overview about my new triples in the Resource view of OW ( http://localhost/dispedia/resource/instances ).

After click the item "M firstnameM lastname", i got follow error > Erfurt_Sparql_EngineDb_SqlGeneratorException: Variable "" not selected. > http://pastebin.com/44AmNHa4 (complete Stack trace)

Here is the faulty entry > http://pastebin.com/97cxJs0T

Maybe, are the white spaces in the subject prompt the error?

greeting

Please provide example Linked Data application

I think it is not really obvious, looking at the Wiki contents, how to quickly test a Linked Data application using Erfurt.

I've taken the liberty to add https://github.com/AKSW/Erfurt/wiki/Example-application which points to an example app that may be interesting (still very basic).

Hope this helps (and feel free to merge).

SPARQL answers with wrong dataformat

SPARQL answers with wrong dataformat.

Import this Resource into a virtuoso backed OntoWiki:

<http://localhost/activity/1/> a aair:Activity ;
    aair:activityActor <http://comiles.eu/natanael.rdf#me> ;
    ns3:published "2012-01-01T00:00:01+02:00"^^xsd:dateTime ;
    aair:activityVerb aair:Post ;
    aair:activityObject <http://localhost/object/1/> .

Do following query with the QueryEditor and with the Virtuoso Conductor

PREFIX atom: <http://www.w3.org/2005/Atom/>
PREFIX aair: <http://xmlns.notu.be/aair#>
SELECT ?activity ?date ?verb ?object
WHERE {
    ?activity a aair:Activity ;
              aair:activityActor  <http://comiles.eu/natanael.rdf#me> ;
              atom:published      ?date ;
              aair:activityVerb   ?verb ;
              aair:activityObject ?object .
}

correct answer by Virtuoso and OW with ZendDB

activity: http://localhost/1/
date: 2012-01-01T00:00:01+02:00
verb: http://xmlns.notu.be/aair#Post
object: http://localhost/object/1/

wrong answer by OW with Virtuoso

activity: http://localhost/1/
date: 2012-01-01 00:00:01
verb: http://xmlns.notu.be/aair#Post
object: http://localhost/object/1/

As you can see the date format is converted before the query result is returned.

Same with $model->sparqlQuery($query).

deleteMultipleStatements generates too big queries

SPARQL: Internal error: The length of generated SQL text has exceeded 10000 lines of code () In query

Virtuoso Adapter

Wrong handling of graph in SPARQL-execution

In Erfurt_Store_Adapter_Virtuoso there are multiple issues when executing a sparql-query.

Erfurt_Store_Adapter_Virtuoso::_execSparql():
When $graphUri = null, in Virtuoso 06.01.3127 odbc_exec() rises an exception "Fatal error: Out of memory (allocated 10485760) (tried to allocate 18446744073709551613 bytes)" (nearly 2^64 on 64bit-system). This is, because Virtuoso comes in trouble (also in ISQL console) when a graph named 'NULL' passed to DB.DBA.SPARQL_EVAL(). So it's only possible to quote the graph name when it's not null, otherwise real NULL must be passed as this parameter to virtuoso.

Erfurt_Store_Adapter_Virtuoso::_execSparqlUpdate():
Here we also have the param $graphUri, also used nowhere in the class like in Erfurt_Store_Adapter_Virtuoso::_execSparql(). The default graph definition is created in $graphSpec, but not in use. Is it possible to add a comment why to use DB.DBA.SPARQL_EVAL in _execSparql(), but a simple SPARQL in _execSparqlUpdate()? Furthermore the PHPDoc is missing for this method.

sparqlQuery changes 0 to empty string

When querying something that should result in a 0-value Erfurt instead returns an empty string "".

Virtuoso returns the correct value.

This was first filed here: AKSW/site.ontowiki/issues/31

Invalidation of Object & Query Cache results in a warning

When I select Object & Query Cache from the debug menu I get the following Warning (I'm using cache.backend.type = "database"):

( ! ) Warning: odbc_exec(): SQL error: [unixODBC][OpenLink][Virtuoso iODBC Driver][Virtuoso Server]SQ200: No table ef_cache_query_result, SQL state S0002 in SQLExecDirect in /home/…/public_html/OntoWiki/libraries/Erfurt/library/Erfurt/Store/Adapter/Virtuoso.php on line 1061
Call Stack
#   Time    Memory  Function    Location
1   0.0000  237920  {main}( )   ../index.php:0
2   4.3227  4303800 Zend_Application->run( )    ../index.php:188
3   4.3227  4303920 Zend_Application_Bootstrap_Bootstrap->run( )    ../Application.php:366
4   4.3228  4305776 Zend_Controller_Front->dispatch( )  ../Bootstrap.php:97
5   4.3256  4345760 Zend_Controller_Plugin_Broker->routeShutdown( ) ../Front.php:923
6   4.3257  4347104 OntoWiki_Controller_Plugin_SetupHelper->routeShutdown( )    ../Broker.php:260
7   4.3260  4347784 Erfurt_Store->getModel( )   ../SetupHelper.php:57
8   4.3260  4347832 Erfurt_Store->isModelAvailable( )   ../Store.php:810
9   4.3260  4347880 Erfurt_Store->_checkAc( )   ../Store.php:1130
10  4.3261  4348800 Erfurt_Ac_Default->isModelAllowed( )    ../Store.php:1813
11  4.3276  4355472 Erfurt_Ac_Default->_getUserModelRights( )   ../Default.php:285
12  4.3277  4366424 Erfurt_Ac_Default->_sparql( )   ../Default.php:551
13  4.3277  4367552 Erfurt_Store->sparqlQuery( )    ../Default.php:719
14  4.3281  4368984 Erfurt_Store->_prepareQuery( )  ../Store.php:1408
15  4.3291  4380536 Erfurt_Store->getImportsClosure( )  ../Store.php:1316
16  4.3291  4380768 Erfurt_Store->_getImportsClosure( ) ../Store.php:748
17  4.3291  4381016 Erfurt_Store_Adapter_Virtuoso->getImportsClosure( ) ../Store.php:770
18  4.3296  4417984 Erfurt_Cache_Frontend_QueryCache->load( )   ../Virtuoso.php:918
19  4.3297  4418208 Erfurt_Cache_Backend_QueryCache_Database->load( )   ../QueryCache.php:112
20  4.3297  4418648 Erfurt_Cache_Backend_QueryCache_Database->_query( ) ../Database.php:231
21  4.3297  4418792 Erfurt_Store->sqlQuery( )   ../Database.php:846
22  4.3297  4418840 Erfurt_Store_Adapter_Virtuoso->sqlQuery( )  ../Store.php:1503
23  4.3297  4419128 Erfurt_Store_Adapter_Virtuoso->_execSql( )  ../Virtuoso.php:777
24  4.3298  4419176 odbc_exec ( )   ../Virtuoso.php:1061

Some invalid queries lead to "Fatal error: Out of memory"

Some invalid queries lead to "Fatal error: Out of memory" rather than giving an exception e.g.

SELECT ?s ?p ?o ?u
WHERE {
    ?s ?p ?o
}

We should validate a query before we send it ti virtuoso.

Resource URI's with urlencoded characters are problematic

I , perhaps wrongly, filed this issue in AKSW/OntoWiki#151. Please look there for a detailed bug report.

Whenever there exists in the knowledge base resources with special characters in their names, things does not work as expected. Links are broken, and some views, such as the "Source" tab, does not work properly.

I had created a test resources to explain the problem. Theese were created using OntoWiki. As can be seen, the Turtle source is correct, but the OntoWiki interaction with them are not.

Regretfully the test resources are no longer available on my server, but the Turtle source can be used to recreate them fot testing purposes.

;)Frode

error running tests

PHPUnit 3.6.10 by Sebastian Bergmann.

Configuration read from /home/jonas/programming/php-workspace/ow/libraries/Erfurt/phpunit.xml.dist

.........PHP Fatal error: Call to a member function isVersioningEnabled() on a non-object in /home/jonas/programming/php-workspace/ow/libraries/Erfurt/tests/Erfurt/AppTest.php on line 176

PHP Stack trace:

Turtle parser returns wrong URI when base URI has no trailing slash

see testParseWithNoTrailingSlash

add support for blank nodes

Blank nodes are treated as strings currently, leading to duplicate blank node names when importing multiple RDF files (and thus data confusion).

The problem lies in Erfurt_Store_Adapter_Virtuoso::addStatement() which adds triples one by one, instead of storing them together so that real blank nodes can be used.

problem with writing quotes in zenddb store

please fix this asap ... :-)

update api documentation

Error when inserting triples in MySQL (on 1.1)

The following commit fixes missing string casts for model's URI when using MySQL backend in 1.1 : olberger@2c1eda9

array_split() is similar to array_chunk() built-in

And appears to use much more memory.

(Erfurt_Store_Adapter_Virtuoso)

Fix _decodeString function in turtle parser

... when Jonas detailed what the problem ("men leak") is...

Why re-authentication to create new model?

This issue occurs in relation to OntoWiki! I am in an action of a controller.

For instance:

class CollaboController extends OntoWiki_Controller_Component {

   /**
    * Import an ontology for given URI
    */
   public function importontologyAction () {

The plan is to create a new model via

$this->_erfurt->getStore()->getNewModel($ontologyUri, '', $newType);

But when iam doing this, it throws an exception that this operation is not allowed. For the record, i am logged in as Admin.

It runs perfectly when i authenticate myself again:

$backend = $this->_erfurt->getConfig()->store->backend;
$username = $this->_erfurt->getConfig()->store->{$backend}->username;
$password = $this->_erfurt->getConfig()->store->{$backend}->password;

$this->_erfurt->authenticate($username, $password);

After that i am able to create a new model. But was that your intention? I thought OntoWiki handle Erfurt authentication by itself.

Ping improvements

Additionally to #40 following things should be improved for the Erfurt_Ping class:

• receive() should return true or false
• an additional method getStatusCode()
• an additional method getTriples() which returns the new data from the source
• receive() should not always write the received triples to the store
• to make the pingback service usable for Erfurt standalone we have to move parts of linkeddataserver plugin to erfurt
• specify target graph for the ping to go in options
• add config option to also trigger onPingReceivedAgain event (AKSW/OntoWiki/issues/90)

Owl2Structured.php missing?

When I want to view an instance I get the following error:

Warning: include(Erfurt/Owl/Structured/Util/Owl2Structured.php): failed to open stream: No such file or directory in /usr/share/ontowiki/libraries/Zend/Loader.php on line 83

The file is not included?

Unexpected T_PAAMAYIM_NEKUDOTAYIM

A fresh install of OntoWiki fails to run on Windows (PHP 5.2) failing in Erfurt\Store\Adapter\Virtuoso.php with the above message.

Solution is to change line

if ($result == $queryCache::ERFURT_CACHE_NO_HIT) {

to

if ($result == Erfurt_Cache_Frontend_QueryCache::ERFURT_CACHE_NO_HIT) {

Rdf_Model->sparqlQuery() should use registered namespace prefixes

Currently sparqlQuery() on Rdf_Model (and I think also on Store) doesn't use the registered namespace prefixes.
This could be implemented as option or (what I would prefer) be the default behavior.

Request for comments!

Use Model for storing Pings

Use the default model to store incoming Pingback pings rather than the SQL database.

Template format pxml needs "short tags" to be disabled

While running tests I stumbled upon an error while including a template.
This template starts with a XML header which leads to a syntax error on include.
This error is avoidable by configuring PHP to NOT allow short tags.
For sure, your environment has been configured this way - however mine was not.

Solution:
Set PHP option "short_tags" to "off" in php.ini.
Do this for /etc/php5/cli/php.ini if you want to run unit tests.

It should be:

• either noted somewhere to turn short tags off
• or checked if short tags are off when using this template format using include

Serializer Interface lacks serializeQueryResultToString method

The Erfurt_Syntax_RdfSerializer_Adapter_Interface has no method serializeQueryResultToString, which is provided and called by Erfurt_Syntax_RdfSerializer.

Also Erfurt_Syntax_RdfSerializer_Adapter_RdfXml lacks this method

Should getNewModel return an existing Model?

Should the method Erfurt_Store->getNewModel() return the model if it already exists rather than throwing an Exception:

'Erfurt_Store_Exception' with message 'Failed creating the model.'

Or do we have an other method which creates a model if it doesn't exist or returns the existing one?

Integration of a worker queue backend

we need a backend abstraction for dropping worker tasks such as:

• do a semantic ping
• push new content
• email contribtutors
• create all resources for the site extension
• re-scale images

some tools for doing this are: gearman, beanstalkd, apache activeMQ and more

as a fallback, we need cron and instant worker queues.

• cron: save tasks somewhere, where a simle run script can start it
• instant: do not save tasks, just do them instantly (fallback if nothing is configured)

create knowledge base

During creating a knowledge base using e.g a turtle document it throws an error and a warning.

Warning: odbc_exec(): SQL error: [unixODBC][OpenLink][Virtuoso iODBC Driver][Virtuoso Server]SQ074: Line 171: syntax error at 'focus' before 'lies', SQL state 37000 in SQLExecDirect in /var/www/ontowiki/libraries/Erfurt/Erfurt/Store/Adapter/Virtuoso.php on line 975

The import error is thrown using the develop branch, master branch works fine.

current search in virtuoso broken

a search with the current develop head outputs this error message:

Erfurt_Store_Adapter_Exception: SPARQL Error: [unixODBC][OpenLink][Virtuoso iODBC Driver][Virtuoso Server]VD032: remote prepare: SQ200: No column t2.S. () in query: define output:format "JSON" SELECT DISTINCT ?resourceUri FROM http://localhost/OntoWiki/Config/ FROM http://ns.ontowiki.net/SysOnt/ WHERE { ?listresource ?p0 ?o0 . ?listresource ?resourceUri ?showPropsObj FILTER (!isBLANK(?resourceUri)) FILTER (bif:contains(?o0, '"Usergroup*"')) } ORDER BY ASC(?resourceUri)
/home/seebi/var/www/ow/tip/libraries/Erfurt/Erfurt/Store/Adapter/Virtuoso.php@957

This issue was introduces by revision 1e05aef of Erfurt/Store/Adapter/Virtuoso.php

Less-than comparison not working in Filters

As described in AKSW/OntoWiki#125 SparQL queries with a filter containing a less-than-char wouldn't work on ZendDB backend.

Max column length (255) of column 'DB.DBA.ef_cache_query_triple.subject' exceeded ()

After importing the dataset from the file into OntoWiki I've got the following error.

Dataset:
https://dl.dropboxusercontent.com/u/4882345/lodstats-complete.rdf

Error:
Something went wrong with the query cache: SQL Error: [unixODBC][OpenLink][Virtuoso iODBC Driver][Virtuoso Server]SR319: Max column length (255) of column 'DB.DBA.ef_cache_query_triple.subject' exceeded () in query: INSERT INTO ef_cache_query_triple ( subject, predicate, object) VALUES ( 'http://famu.eagle-i.net/sparqler/sparql?view=published-resources&format=text/turtle&query=PREFIX+rdfs%3A+%3Chttp%3A%2F%2Fwww.w3.org%2F2000%2F01%2Frdf-schema%23%3E+PREFIX+%3A+%3Chttp%3A%2F%2Feagle-i.org%2Font%2Frepo%2F1.0%2F%3E+construct+%7B%3Fs+%3Fp+%3Fo+.+%3Fp+rdfs%3Alabel+%3Fp_label+.+%3Fo+rdfs%3Alabel+%3Fo_label%7D+where+%7Bgraph+%3ANG_Published%7B%3Fs+%3Fp+%3Fo%7D+.+optional%7B%3Fp+rdfs%3Alabel+%3Fp_label+%7D+.+optional%7B+%3Fo+rdfs%3Alabel+%3Fo_label%7D%7D+', NULL, NULL) SQL:INSERT INTO ef_cache_query_triple ( subject, predicate, object) VALUES ( 'http://famu.eagle-i.net/sparqler/sparql?view=published-resources&format=text/turtle&query=PREFIX+rdfs%3A+%3Chttp%3A%2F%2Fwww.w3.org%2F2000%2F01%2Frdf-schema%23%3E+PREFIX+%3A+%3Chttp%3A%2F%2Feagle-i.org%2Font%2Frepo%2F1.0%2F%3E+construct+%7B%3Fs+%3Fp+%3Fo+.+%3Fp+rdfs%3Alabel+%3Fp_label+.+%3Fo+rdfs%3Alabel+%3Fo_label%7D+where+%7Bgraph+%3ANG_Published%7B%3Fs+%3Fp+%3Fo%7D+.+optional%7B%3Fp+rdfs%3Alabel+%3Fp_label+%7D+.+optional%7B+%3Fo+rdfs%3Alabel+%3Fo_label%7D%7D+', NULL, NULL)

Traceback:

Erfurt_Exception

/var/local/cubeviz.vhost.tld/libraries/Erfurt/library/Erfurt/Cache/Backend/QueryCache/Database.php@852 
#0: Erfurt_Cache_Backend_QueryCache_Database->_query@/var/local/cubeviz.vhost.tld/libraries/Erfurt/library/Erfurt/Cache/Backend/QueryCache/Database.php:750
#1: Erfurt_Cache_Backend_QueryCache_Database->_saveTriplePatterns@/var/local/cubeviz.vhost.tld/libraries/Erfurt/library/Erfurt/Cache/Backend/QueryCache/Database.php:196
#2: Erfurt_Cache_Backend_QueryCache_Database->save@/var/local/cubeviz.vhost.tld/libraries/Erfurt/library/Erfurt/Cache/Frontend/QueryCache.php:86
#3: Erfurt_Cache_Frontend_QueryCache->save@/var/local/cubeviz.vhost.tld/libraries/Erfurt/library/Erfurt/Store.php:1503
#4: Erfurt_Store->sparqlQuery@/var/local/cubeviz.vhost.tld/application/classes/OntoWiki/Model/TitleHelper.php:401
#5: OntoWiki_Model_TitleHelper->getTitleQueryResult@/var/local/cubeviz.vhost.tld/application/classes/OntoWiki/Model/TitleHelper.php:494
#6: OntoWiki_Model_TitleHelper->_fetchResourceTitlesFromQueryResult@/var/local/cubeviz.vhost.tld/application/classes/OntoWiki/Model/TitleHelper.php:321
#7: OntoWiki_Model_TitleHelper->getTitle@/var/local/cubeviz.vhost.tld/extensions/cubeviz/classes/DataCube/Query.php:95
#8: DataCube_Query->enrichResult@/var/local/cubeviz.vhost.tld/extensions/cubeviz/classes/DataCube/Query.php:319
#9: DataCube_Query->getComponentElements@/var/local/cubeviz.vhost.tld/extensions/cubeviz/classes/DataCube/Query.php:292
#10: DataCube_Query->getComponents@/var/local/cubeviz.vhost.tld/extensions/cubeviz/classes/CubeViz/ConfigurationLink.php:89
#11: CubeViz_ConfigurationLink->loadStandardConfigForData@/var/local/cubeviz.vhost.tld/extensions/cubeviz/classes/CubeViz/ConfigurationLink.php:342
#12: CubeViz_ConfigurationLink->read@/var/local/cubeviz.vhost.tld/extensions/cubeviz/classes/CubeViz/ViewHelper.php:106
#13: CubeViz_ViewHelper::initApp@/var/local/cubeviz.vhost.tld/extensions/cubeviz/DataselectionModule.php:137
#14: DataselectionModule->getContents@/var/local/cubeviz.vhost.tld/application/classes/OntoWiki/View.php:266
#15: OntoWiki_View->module@/var/local/cubeviz.vhost.tld/application/classes/OntoWiki/View.php:151
#16: OntoWiki_View->modules@/var/local/cubeviz.vhost.tld/application/classes/OntoWiki/Controller/Base.php:188
#17: OntoWiki_Controller_Base->preDispatch@/var/local/cubeviz.vhost.tld/libraries/Zend/Controller/Action.php:502
#18: Zend_Controller_Action->dispatch@/var/local/cubeviz.vhost.tld/libraries/Zend/Controller/Dispatcher/Standard.php:295
#19: Zend_Controller_Dispatcher_Standard->dispatch@/var/local/cubeviz.vhost.tld/libraries/Zend/Controller/Front.php:954
#20: Zend_Controller_Front->dispatch@/var/local/cubeviz.vhost.tld/libraries/Zend/Application/Bootstrap/Bootstrap.php:97
#21: Zend_Application_Bootstrap_Bootstrap->run@/var/local/cubeviz.vhost.tld/libraries/Zend/Application.php:366
#22: Zend_Application->run@/var/local/cubeviz.vhost.tld/index.php:174

Add a warning in case of non-writable logs dir in error_log

olberger@9a0748b proposes the addition of a message reported in error_log in case it isn't writable, which helps a great deal when starting debugging

Sparql Query2 creates incorrect count query

Code:

$query = new Erfurt_Sparql_Query2();
$query->addProjectionVar(new Erfurt_Sparql_Query2_Var('resourceUri'));
$query->setCountStar(true);
$query->setDistinct();
$query->addElements(NavigationHelper::getInstancesTriples($uri, $setup));

Expected output:

SELECT ?resourceUri COUNT(DISTINCT ?resourceUri) 
WHERE { ... } 

Actual output:

SELECT DISTINCT COUNT(*) 
WHERE { ... } 

SimpleQuery count() as doesn't work

The following query doesn't work (SimpleQuery just removes the count(…) as … part)

prefix ctag: <http://commontag.org/ns#> 
prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> 
select distinct ?tag ?label count(?tag) as ?weight {
  ?resource ctag:tagged ?tag . 
  optional { { ?tag ctag:label ?label } union { ?tag rdfs:label ?label } } 
} order by ?label

Query Cache Trouble if Table is Missing

The query cache gets the hiccups if one of its tables is somehow missing.
See also AKSW/OntoWiki/issues/285

Turtle parser returns wrong URIs for tel: URIs and other url encoded URIs

I think the urldecode call in the _decodeString function is the problem... Test cases were added to capture this issue.

see also:

commit 3c3a35a
Author: Jonas Brekle [email protected]
Date: Thu Oct 20 18:23:53 2011 +0200
use a new way to decode turtle strings (taken from arc) the old had a memory leak and i dont want to debug

strict standard problem antlr

Strict Standards: Declaration of AntlrLexer::matchAny() should be compatible with that of BaseRecognizer::matchAny() in /home/tr/projects/od.fmi/libraries/Erfurt/Erfurt/libraries/antlr/Php/Lexer.php on line 297

http://od.fmi.uni-leipzig.de/list?init=1&instancesconfig=%7B%22filter%22%3A%5B%7B%22mode%22%3A%22query%22%2C%22action%22%3A%22add%22%2C%22query%22%3A%22SELECT+DISTINCT+%3FresourceUri+++FROM+%3Chttp%3A%5C%2F%5C%2Fod.fmi.uni-leipzig.de%5C%2Fstudium%5C%2F%3E++WHERE+%7B++%3Chttp%3A%5C%2F%5C%2Fod.fmi.uni-leipzig.de%5C%2Fstudium%5C%2F07-203-4210%3E+%3Chttp%3A%5C%2F%5C%2Fod.fmi.uni-leipzig.de%5C%2Fmodel%5C%2FtoStudiengangSemester%3E+%3FresourceUri++FILTER+%28%21isBLANK%28%3FresourceUri%29%29++%7D++LIMIT+10++%22%7D%5D%7D