Hi Andy, Now, I have my web application runs under Tomcat and Fuseki2 also. I made tests to route requests from my webapp/servlet to Fuseki to do Graph Store Protocol operations. and it works well (Tests with tomcat with eclipse and Fuseki2 with command line).
When I want to query my TDB database with Fuseki2 under Tomcat I have 0 results and the storage directory containes files but these files are empty. I changed the permissions rights of etc/fuseki to allow Tomcat to write. Help me please and Thank you very much !!! 2014-08-07 11:49 GMT+02:00 Amira Sifaoui Ep Ghaddab <[email protected] >: > I m sorry Andy. > > It works well :) > > Thank you very much for your help :) > > > 2014-08-07 11:05 GMT+02:00 Andy Seaborne <[email protected]>: > >> On 06/08/14 16:01, Amira Sifaoui Ep Ghaddab wrote: >> >>> Hi Andy, >>> >>> Thank you very much for your reply. >>> >>> I want to tell you that when I want to execute query with Fuseki2 , it >>> works well with "Selection of classes" but with "Selection of triples" it >>> loads with no results. >>> >> >> Odd - because "Selection of triples" is a more general query than >> "Selection of classes". Is the "SPARQL endpoint" gets set correctly? >> >> Andy >> >> >> >>> >>> >>> >>> 2014-08-05 15:03 GMT+02:00 Andy Seaborne <[email protected]>: >>> >>> On 05/08/14 09:42, Amira Sifaoui Ep Ghaddab wrote: >>>> >>>> Hi Andy, >>>>> >>>>> Thank you very much for your help, my problem is resolved; >>>>> I have a cookies problem . >>>>> >>>>> >>>> Fuseki2 does not use cookies itself though IIRC jquery.dataTables does >>>> (for the query UI). >>>> >>>> (I did find that browsers were caching javascript aggressively). >>>> >>>> >>>> Now Fuseki2 run as a web application and it detects datasets. >>>> >>>>> >>>>> >>>> Good! >>>> >>>> >>>> I try to query my datasets , I have some problems I try to resolve >>>> them >>>> >>>>> :) >>>>> Just a question : I have my directory for storing ontologies and >>>>> datasets >>>>> with TDB, I used a servlet for that. >>>>> Should my directory be under Fuseki?? or Tomcat?? >>>>> >>>>> >>>> You can not share a TDB database between JVMs and sharing across webapps >>>> in Tomcat will not be reliable. >>>> >>>> Databases are managed by the system and it uses statics. When you open >>>> a >>>> database it must be the same database java objects in order to do >>>> caching >>>> and transaction management which in turn means coordinated disk access. >>>> >>>> The directory for your databases used by Fuseki can be where you want >>>> (given they can be large, disk volumes matter). >>>> >>>> You can have symbolic links within the Fuseki layout (e.g. under >>>> /etc/fuseki/databases or /etc/fuseki/databases itself can be a symbolic >>>> link) off to your preferred locations. >>>> >>>> >>>> In my case, I uplod files (ontologies+datasets) to store them with >>>> TDB. >>>> >>>>> The storing directory is under tomcat: I have used this code in my >>>>> servlet: getServletContext().getRealPath. >>>>> This is right?? >>>>> >>>>> >>>> Fuseki config files only know about file names, not the deployment >>>> structure of Tomcat. >>>> >>>> And, again, a warning that you can't have two running webapps using one >>>> TDB database. You can route requests from your webapp/servlet to >>>> Fuseki to >>>> do SPARQL Update or Graph Store Protocol operations. >>>> >>>> Andy >>>> >>>> It is possible to use the Fuseki service components inside your own >>>> webapp >>>> - this is not documented yet. Fuseki is a collection of servlets, a >>>> single >>>> database registry and runs using a servlet filter, but the filter is >>>> only >>>> to separate the dynamic structure of the databases and services from the >>>> fixed view of the world provided by web.xml. >>>> >>>> >>> >> >
