query command can execute SPARQL ASK, SELECT, and CONSTRUCT queries by using the
--query option with two arguments: a query file and an output file. The output file will only be written if the query returns some results. The output format will be inferred from the output file extension, or you can use the
ASK always produces
SELECT produces a table, if there are any results, defaulting to CSV format. For example:
robot query --input nucleus.owl \ --query cell_part.sparql results/cell_part.csv
CONSTRUCT produces RDF data, if there are any results, defaulting to Turtle format:
robot query --format ttl \ --input nucleus.owl \ --query part_of.sparql results/part_of.ttl
Instead of specifying one or more pairs (query file, output file), you can specify a single
--output-dir and use the
--queries option to provide one or more queries of any type. Each output file will be written to the output directory with the same base name as the query file that produced it. For example the
foo.sparql query file will produce the
foo.csv file. The output directory must exist.
robot query --input nucleus.owl \ --queries cell_part_ask.sparql \ --output-dir results/
query ignores import statements. To include all imports as named graphs, add
robot query --input imports.owl \ --use-graphs true --catalog catalog.xml \ --query named_graph.sparql results/named_graph.csv
The example above also uses the global
--catalog option to specify the catalog file for the import mapping. The default graph is the union of all graphs, which allows querying over an ontology and all its imports.
The names of the graphs correspond to the ontology IRIs of the imports. If the import does not have an ontology IRI, one will be automatically generated. Running
query with the
-vv flag will print the names of all graphs as they are added.
query command also supports SPARQL Update to insert and delete triples.
robot query --input nucleus.owl \ --update update.ru \ --output results/nucleus_update.owl
When using SPARQL update, you can either provide an
--output for the updated ontology, or chain it into another command.
You can perform multiple updates in one command to reduce time spent loading and saving the ontology. Updates are processed in the order that they are input.
robot query --input nucleus.owl \ --update update.ru \ --update revert.ru \ --output results/nucleus.owl
--update option only updates the ontology itself, not any of the imports.
Warning: The output of SPARQL updates will not include
xsd:string datatypes, because
xsd:string is considered implicit in RDF version 1.1. This behaviour differs from other ROBOT commands, where
xsd:string datatypes from the input are maintained in the output.
The file provided for
--update does not exist. Check the path and try again.
You must specify a query to execute with
The query was not able to be parsed. Often, this is as a result of an undefined prefix in the query. See the error message for more details.
Each SPARQL query should be a SELECT, ASK, DESCRIBE, or CONSTRUCT.