Hi Andy,

We advanced a few steps but now we're having trouble using the (full)
geodetic EPSG dataset.

$ docker-compose stop fuseki; docker-compose rm -f fuseki; docker-compose
up fuseki
Stopping fuseki ... done
Going to remove fuseki
Removing fuseki ... done
Creating fuseki ... done
Attaching to fuseki
fuseki    | /opt/java-minimal/bin/java -Xmx24g -Xms24g
-Dderby.system.home=/fuseki/databases/SIS/apache-sis-1.1/data/Databases
-classpath
/fuseki/databases/SIS/apache-sis-1.1/lib/sis-referencing-1.1.jar:/fuseki/jena-fuseki-geosparql-4.4.0.jar
org.apache.jena.fuseki.geosparql.Main -t /fuseki/databases/DB2-4 --tdb2
--rdf_file /fuseki/store/ligplaats.trig --dataset dst --inference --validate
fuseki    | WARNING: sun.reflect.Reflection.getCallerClass is not
supported. This will impact performance.
fuseki    | [2022-02-03 19:15:18] INFO  Main            :: Arguments
Received: [-t, /fuseki/databases/DB2-4, --tdb2, --rdf_file,
/fuseki/store/ligplaats.trig, --dataset, dst, --inference, --validate]
fuseki    | [2022-02-03 19:15:18] INFO  DatasetOperations :: Server
Configuration: port=3030, datsetName=dst, loopbackOnly=true,
updateAllowed=false, inference=true, applyDefaultGeometry=false,
validateGeometryLiteral=true, convertGeoPredicates=false,
removeGeoPredicates=false, queryRewrite=true,
tdbFile=/fuseki/databases/DB2-4,
fileGraphFormats=[FileGraphFormat{rdfFile=/fuseki/store/ligplaats.trig,
graphName=, rdfFormat=Turtle/pretty}], fileGraphDelimiters=[],
indexEnabled=true, indexSizes=[-1, -1, -1], indexExpiries=[5000, 5000,
5000], spatialIndexFile=null, tdb2=true, help=false
fuseki    | [2022-02-03 19:15:18] INFO  DatasetOperations :: TDB Dataset:
/fuseki/databases/DB2-4, TDB2: true
fuseki    | [2022-02-03 19:15:19] WARN  system          :: The “SIS_DATA”
environment variable is not set.
fuseki    | [2022-02-03 19:15:19] INFO  DatasetOperations :: Reading RDF -
Started - File: /fuseki/store/ligplaats.trig, Graph Name: , RDF Format:
Turtle/pretty
fuseki    | [2022-02-03 19:15:20] WARN  SRSRegistry     :: SRS URI not
recognised so operation results may not be correct or accurate:
http://www.opengis.net/def/crs/EPSG/0/28992 - Unrecognised SRS URI code:
http://www.opengis.net/def/crs/EPSG/0/28992 - No
‘CoordinateReferenceSystem’ object found for code “28992”. However only a
subset of the EPSG geodetic dataset has been queried. See
http://sis.apache.org/epsg.html for instruction about how to install the
full EPSG database.
fuseki    | [2022-02-03 19:15:27] INFO  DatasetOperations :: Reading RDF -
Completed - File: /fuseki/store/ligplaats.trig, Graph Name: , RDF Format:
Turtle/pretty
fuseki    | [2022-02-03 19:15:27] INFO  GeoSPARQLOperations :: Applying
GeoSPARQL Schema - Started
fuseki    | [2022-02-03 19:15:32] INFO  GeoSPARQLOperations :: GeoSPARQL
schema applied to graph: default
fuseki    | [2022-02-03 19:15:32] INFO  GeoSPARQLOperations :: Applying
GeoSPARQL Schema - Completed
fuseki    | [2022-02-03 19:15:32] INFO  GeoSPARQLOperations :: Find Mode
SRS - Started
fuseki    | [2022-02-03 19:15:32] INFO  GeoSPARQLOperations :: Find Mode
SRS - Completed
fuseki    | [2022-02-03 19:15:32] ERROR Main            :: GeoSPARQL
Server:  Exiting - No SRS found. Check '
http://www.opengis.net/ont/geosparql#hasSerialization' or '
http://www.w3.org/2003/01/geo/wgs84_pos#lat'/'http://www.w3.org/2003/01/geo/wgs84_pos#lon'
predicates are present in the source data. Hint: Inferencing with GeoSPARQL
schema may be required.: dst

The was created manually with the CLI with the work around from :


   - cd /fuseki/databases/SIS/apache-sis-1.1/data
   - rm -rf Databases
   - wget
   
https://repo1.maven.org/maven2/org/apache/sis/non-free/sis-embedded-data/1.1/sis-embedded-data-1.1.jar
   - unzip sis-embedded-data-1.1.jar SIS_DATA/Databases/*
   - mv SIS_DATA/Databases .
   - rmdir SIS_DATA
   - cd ..
   - sis crs EPSG:28992

output:

ProjectedCRS["Amersfoort / RD New",
  BaseGeodCRS["Amersfoort",
    Datum["Amersfoort",
      Ellipsoid["Bessel 1841", 6377397.155, 299.1528128]],
    Unit["degree", 0.017453292519943295]],
  Conversion["RD New",
    Method["Oblique Stereographic"],
    Parameter["Latitude of natural origin", 52.15616055555556],
    Parameter["Longitude of natural origin", 5.387638888888891],
    Parameter["Scale factor at natural origin", 0.9999079],
    Parameter["False easting", 155000.0],
    Parameter["False northing", 463000.0]],
  CS[Cartesian, 2],
    Axis["Easting (X)", east],
    Axis["Northing (Y)", north],
    Unit["metre", 1],
  Scope["Large and medium scale topographic mapping and engineering
survey."],
  Area["Netherlands - onshore, including Waddenzee, Dutch Wadden Islands
and 12-mile offshore coastal zone."],
  BBox[50.75, 3.20, 53.70, 7.22],
  Id["EPSG", 28992, "9.9.1", URI["urn:ogc:def:crs:EPSG:9.9.1:28992"]],
  Remark["Replaces Amersfoort / RD Old (code 28991)."]]




Op wo 2 feb. 2022 om 12:11 schreef Andy Seaborne <[email protected]>:

>
>
> On 02/02/2022 10:17, Erik Bijsterbosch wrote:
> > Hi Andy,
> >
> > Good to know that the code loops through all the (named) graphs.
> > I noticed later in the logs that fuseki was still busy with the next
> > (named) graph, apparently.
> > It seems that that attempt failed on the type of geometry we have in this
> > graph:
> >
> > [2022-02-02 09:32:27] WARN  system          :: The “SIS_DATA” environment
> > variable is not set.
> > [2022-02-02 09:32:28] WARN  system          :: The “SIS_DATA” environment
> > variable is not set.
> > Exception in thread "main"
> > org.apache.jena.dboe.transaction.txn.TransactionException: Write
> > transaction with no commit() or abort() before end() - forced abort
> >          at
> >
> org.apache.jena.dboe.transaction.txn.Transaction.end(Transaction.java:203)
> >          at
> >
> org.apache.jena.dboe.transaction.txn.TransactionalBase._end(TransactionalBase.java:262)
> >          at
> >
> org.apache.jena.dboe.transaction.txn.TransactionalBase.end(TransactionalBase.java:168)
> >          at
> >
> org.apache.jena.dboe.storage.system.DatasetGraphStorage.end(DatasetGraphStorage.java:73)
> >          at
> >
> org.apache.jena.sparql.core.DatasetGraphWrapper.end(DatasetGraphWrapper.java:284)
> >          at
> org.apache.jena.sparql.core.DatasetImpl.end(DatasetImpl.java:162)
> >          at
> >
> org.apache.jena.geosparql.configuration.GeoSPARQLOperations.applyInferencing(GeoSPARQLOperations.java:297)
> >          at
> >
> org.apache.jena.geosparql.configuration.GeoSPARQLOperations.applyInferencing(GeoSPARQLOperations.java:264)
> >          at
> >
> org.apache.jena.fuseki.geosparql.DatasetOperations.setup(DatasetOperations.java:83)
> >          at org.apache.jena.fuseki.geosparql.Main.main(Main.java:64)
>
> Should be a log line: "Inferencing Error: ..." as well or there is a
> java Throwable/Error happening e.g. OutOfMemoryError.
>
> >
> > For that we'll checkout the recommendations here first:
> > https://sis.apache.org/epsg.html
> >
> >
> > Op wo 2 feb. 2022 om 09:33 schreef Andy Seaborne <[email protected]>:
> >
> >>
> >>
> >> On 01/02/2022 17:13, Erik Bijsterbosch wrote:
> >>> Hi all,
> >>>
> >>> I created a new jena tdb2 database and loaded it with 256 million
> triples
> >>> in a named graph, using a fuseki2/tdb2 4.3.2 docker setup.
> >>> The data also consist of geometries so I built a new docker image based
> >> on
> >>> jena-fuseki-geosparql 4.3.2
> >>> When I bring up the new fuseki container it seems I can't address the
> >> named
> >>> graph reusing the tdb2 database, it searches through the default graph.
> >>> According to the --help graph names can only be provided when loading
> >> from
> >>> files or is there another way?
> >>
> >>
> >> Hi Erik,
> >>
> >>> fuseki    | [2022-02-01 17:08:15] INFO  GeoSPARQLOperations ::
> GeoSPARQL
> >>> schema not applied to empty graph: default
> >>
> >> That message is for the attempt the default graph but it does not mean
> >> it is ignoring the named graphs. The code loops on the named graphs as
> >> well.
> >>
> >>
> >>
> https://github.com/apache/jena/blob/main/jena-geosparql/src/main/java/org/apache/jena/geosparql/configuration/GeoSPARQLOperations.java#L269
> >>
> >> Is that what you mean by "can't address the named graph"?
> >>
> >> Or from a SPARQL query you can't get expected results?
> >>
> >>       Andy
> >>
> >
>

Reply via email to