I found in the RDF4J documentation here http://docs.rdf4j.org/custom-sparql-functions/ that it supports creating custom functions with Java.
I attempted to implement this palindrome example and run the example query against GraphDB. I am using the RDF4J java libraries to execute my queries against GraphDB. When executing the query, I get no obvious errors but no results were returned.
Does GraphDB support running custom RDF4J functions? The documentation for RDF4J custom functions states you must place the JAR on your classpath and it will work. What classpath? Can I build it into my project that is executing queries via RDF4J libraries or do I place the custom function JAR on the classpath of GraphDB before I start the server?
In case the graphdb comes as a platform dependent embedded launcher, e.g. .exe, .deb etc. you'll need an additional step.
Under the installation folder (say GrapDBFree), there is a folder named app. First, place the jar with the custom function within app/lib folder and then, edit the app/GraphDBFree.cfg file by adding the jar to the app.classpath= entry declared there.
Related
I added the jena arq 2.9.3 jar file to my code to be able to run sparql queries for DBpedia but it says that the literal and resource factory classes are not present. Do I need to add another version of the jar file or do I need to make changes to package names or import class names to resolve this issue?
I am using Maven as a build tool. On selniumhq site I can see selenium-server-standalone.jar file but I could not found related Maven dependency. Is there any maven dependency for selenium standalone file?
latest selenium standalone file is : selenium-server-standalone-3.2.0.jar
Note : I want to execute code on Remote desktop machine using Selenium grid (which require aforementioned jar file)
I think you should refer to this manual first: http://www.seleniumhq.org/docs/07_selenium_grid.jsp#starting-selenium-grid
i.e. in order to use grid you need you need to create grid instance first by running commands
from command line, then you need to register node, again by using command line, and after that from your code you need to create instance of RemoteWebDriver (refer to this page for more details: http://www.seleniumhq.org/docs/04_webdriver_advanced.jsp#remotewebdriver).
If, for some reason, you want to to create grid\register nodes from your project maybe it makes sense to download it to your resources folder, execute it from there using Runtime (refer to this link for more details: http://docs.oracle.com/javase/7/docs/api/java/lang/Runtime.html) and then connect to it. But I would strongly recommend avoiding doing so.
I am using drools for processing rules. Web-service calls a method in a class which is in jar included in lib directory of web-service. And this method in turn uses drools. Now the problem is web-service is able to find jar that is using drools but not the drools-compiler jar which is residing in same lib directory. And it gives Unable to load dialect 'org.drools.rule.builder.dialect.mvel.MVELDialectConfiguration:mvel' error. It works if I copy all jars in web-service.aar/lib to axis2/WEB_INF/lib. I also tried to set classpath in a way to take web-service.aar/lib jars first then the one's in axis2/WEB_INF/lib by setting classpath in setenv.sh and catalina configurations. But that didn't help either.
What can be the reason/solution?
You probably need to add a newer version of the mvel jar. I added the mvel-1.3.3-java1.5.jar and it did the trick for me, but just remember to restart your IDE.
I'm using PMEase QuickBuild to perform automated builds of our Maven2 projects and a nightly sanity test to ensure nothing is broken.
The test needs to untar packages which are created by the automated Maven2 projects. The problem is that the package names change frequently due to project versions being incremented all the time.
Does anyone know how I can configure QuickBuild to pick up the version (ideally from the POM file of the individual components), if this is possible at all?
I don't know if this is an option for you but it looks like you can do it the other way around. Quoting Build with Maven:
Control build version
If you want to control the build
version from QuickBuild side, please
follow below steps:
Change the POM file and define the project version as
${buildVersion}. Do not forget to
commit the file into your SCM after
change.
Define a build property like below when define the Maven build
step:
buildVersion=${build.version}
There are maybe other options but I must admit that my knowledge (zero) of QuickBuild is very limited
I created a work around to this issue by having QuickBuild execute a shell script which did the untarring by using wildcards, similar to the following (to avoid computing the exact version):
tar xzf filename-*.tar.gz
I couldn't figure out how to do this in QuickBuild, so I offloaded the work to the shell script.
I would like to produce a binary zip distribution of my project that would contain an uber jar and a set of scripts. Right now, I am using two descritors, first one for the uber jar, and the second for the zip that contains the uberjar + extra scripts and documentation. The problem is that both of these get deployed to maven repo, while I don't actually need the uberjar, only the distro. Is there any way to either:
create the distro using one descriptor or
avoid the uberjar being deployed?
Edit: I found this question regarding part 2, but perhaps there is an easier way to do it...
In my opinion, the easiest way is to create the uberjar in one module (set the skip optional parameter in the maven-deploy-plugin configuration to bypass it during deploy) and to create the zip distribution in another module (using a dependency on the previously created assembly as described in Assembling Assemblies via Assembly Dependencies). And that's very close to what you already have which is good news.