query Command Group¤
List, execute, get status or open SPARQL queries.
Queries are identified either by a file path, a URI from the query catalog or a shortened URI (qname, using a default namespace).
One or more queries can be executed one after the other with the execute command. With open command you can jump to the query editor in your browser.
Queries can use a mustache like syntax to specify placeholder for parameter values (e.g. {{resourceUri}}). These parameter values need to be given as well, before the query can be executed (use the-p option).
Note
In order to get a list of queries from the query catalog, execute the query list command or use tab-completion.
query execute¤
Execute queries which are loaded from files or the query catalog.
Queries are identified either by a file path, a URI from the query catalog, or a shortened URI (qname, using a default namespace).
If multiple queries are executed one after the other, the first failing query stops the whole execution chain.
Limitations: All optional parameters (e.g. accept, base64, …) are provided for ALL queries in an execution chain. If you need different parameters for each query in a chain, run cmemc multiple times and use the logical operators && and || of your shell instead.
Options
--accept TEXT Accept header for the HTTP request(s).
Setting this to 'default' means that cmemc
uses an appropriate accept header for
terminal output (text/csv for tables,
text/turtle for graphs, * otherwise). Please
refer to the Corporate Memory system manual
for a list of accepted mime types.
[default: default]
--no-imports Graphs which include other graphs (using
owl:imports) will be queried as merged
overall-graph. This flag disables this
default behaviour. The flag has no effect on
update queries.
--base64 Enables base64 encoding of the query
parameter for the SPARQL requests (the
response is not touched). This can be useful
in case there is an aggressive firewall
between cmemc and Corporate Memory.
-p, --parameter <TEXT TEXT>... In case of a parameterized query
(placeholders with the '{{key}}' syntax),
this option fills all placeholder with a
given value before the query is
executed.Pairs of placeholder/value need to
be given as a tuple 'KEY VALUE'. A key can
be used only once.
--limit INTEGER Override or set the LIMIT in the executed
SELECT query. Note that this option will
never give you more results than the LIMIT
given in the query itself.
--offset INTEGER Override or set the OFFSET in the executed
SELECT query.
--distinct Override the SELECT query by make the result
set DISTINCT.
--timeout INTEGER Set max execution time for query evaluation
(in milliseconds).
query list¤
List available queries from the catalog.
Outputs a list of query URIs which can be used as reference for the query execute command.
Options
query open¤
Open queries in the editor of the query catalog in your browser.
With this command, you can open (remote) queries from the query catalog in the query editor in your browser (e.g. in order to change them). You can also load local query files into the query editor, in order to import them into the query catalog.
The command accepts multiple query URIs or files which results in opening multiple browser tabs.
query status¤
Get status information of executed and running queries.
With this command, you can access the latest executed SPARQL queries on the DataPlatform. These queries are identified by UUIDs and listed ordered by starting timestamp.
You can filter queries based on status and runtime in order to investigate slow queries. In addition to that, you can get the details of a specific query by using the ID as a parameter.
Options
--id-only Lists only query identifier and no labels or other
metadata. This is useful for piping the ids into
other cmemc commands.
--raw Outputs raw JSON response of the query status API.
--filter <TEXT TEXT>... Filter queries by one of the following filter names
and a corresponding value: status, type, trace-id,
user, graph, slower-than, regex.
query replay¤
Re-execute queries from a replay file.
This command reads a REPLAY_FILE and re-executes the logged queries. A REPLAY_FILE is a JSON document which is an array of JSON objects with at least a key queryString holding the query text OR a key iri holding the IRI of the query in the query catalog. It can be created with the query status command.
The output of this command shows basic query execution statistics.
The queries are executed one after another in the order given in the input REPLAY_FILE. Query placeholders / parameters are ignored. If a query results in an error, the duration is not counted.
The optional output file is the same JSON document which is used as input, but each query object is annotated with an additional replays object, which is an array of JSON objects which hold values for the replay|loop|run IDs, start and end time as well as duration and other data.
Options
--raw Output the execution statistic as raw JSON.
--loops INTEGER Number of loops to run the replay file. [default: 1]
--wait INTEGER Number of seconds to wait between query executions.
[default: 0]
--output-file FILE Save the optional output to this file. Input and output
of the command can be the same file. The output is
written at the end of a successful command execution.
The output can be stdout ('-') - in this case, the
execution statistic output is oppressed.
--run-label TEXT Optional label of this replay run.
query cancel¤
Cancel a running query.
With this command, you can cancel a running query. Depending on the backend triple store, this will result in a broken result stream (stardog, neptune and virtuoso) or a valid result stream with incomplete results (graphdb)