How to use JenaText with FusekiMain?

[kinow]

Hi,

I have a dataset configured for Skosmos that's similar to what's described in this page: https://jena.apache.org/documentation/query/text-query.html#text-dataset-assembler

But when I start FusekiMain (in Docker Compose) with:

command: --update --desc=/fuseki/skosmos.ttl /skosmos

the container fails to initialize with the stack trace:

skosmos-web     | [Mon Jul 03 17:39:21.315012 2023] [core:notice] [pid 8] AH00094: Command line: '/usr/sbin/apache2 -D FOREGROUND'
skosmos-fuseki  | org.apache.jena.sparql.ARQException: Found two matches: var ?root -> http://base/#text_dataset, http://base/#tdb_dataset_readwrite
skosmos-fuseki  |   at org.apache.jena.sparql.util.QueryExecUtils.getAtMostOne(QueryExecUtils.java:395)
skosmos-fuseki  |   at org.apache.jena.sparql.util.graph.GraphUtils.findRootByType(GraphUtils.java:213)
skosmos-fuseki  |   at org.apache.jena.sparql.core.assembler.AssemblerUtils.build(AssemblerUtils.java:138)
skosmos-fuseki  |   at org.apache.jena.sparql.core.assembler.AssemblerUtils.build(AssemblerUtils.java:132)
skosmos-fuseki  |   at arq.cmdline.ModAssembler.create(ModAssembler.java:68)
skosmos-fuseki  |   at arq.cmdline.ModDatasetAssembler.createDataset(ModDatasetAssembler.java:41)
skosmos-fuseki  |   at org.apache.jena.fuseki.main.cmds.FusekiMain.processModulesAndArgs(FusekiMain.java:370)
skosmos-fuseki  |   at org.apache.jena.cmd.CmdArgModule.process(CmdArgModule.java:39)
skosmos-fuseki  |   at org.apache.jena.cmd.CmdMain.mainMethod(CmdMain.java:86)
skosmos-fuseki  |   at org.apache.jena.cmd.CmdMain.mainRun(CmdMain.java:56)
skosmos-fuseki  |   at org.apache.jena.cmd.CmdMain.mainRun(CmdMain.java:43)
skosmos-fuseki  |   at org.apache.jena.fuseki.main.cmds.FusekiMain.run(FusekiMain.java:134)
skosmos-fuseki  |   at org.apache.jena.fuseki.main.cmds.FusekiMainCmd.main(FusekiMainCmd.java:47)

Reading our Java code, it looks to me like the Assembler will try to load a single RDFDataset (or similar type) from the given configuration, but we have the text dataset and the tdb2 dataset, thus the error.

At the same page in Jena docs, there's a comment that's similar to what I explained above, I believe:

since the assembler contains two dataset definitions, one for the text dataset, one for the base data. Therefore, the application needs to identify the text dataset by it’s URI http://localhost/jena_example/#text_dataset.

Is there a way to initialize Fuseki Main with a assembler file for Jena Text and indicate what's the main dataset?

Thank you!

-Bruno

[afs]

Try using --config. That walks from the server node to find datasets or search for rdf:type fuseki:Service.

--desc is a shortcut - it is for publishing a single dataset at a single name with the default services (c.f. --file and --mem).
It ignores fuseki:Service.

It exists for other command e.g. tdbquery.

[kinow]

That worked! Thanks!