Page tree
Skip to end of metadata
Go to start of metadata

You can access the installed queries of the server with the queries endpoint. If e.g. the would be a query with id 0, you could access the query with: http://localhost:8888/queries/0

The result will look similar to the following:

The important part for accessing the results can be found in the websockets section. For each root operator (i.e. the sink of a plan) you could find information how to access the results via a web socket connection.

Different kinds of data can be access. E.g. in der example above, the endpoint "/queries/5/d05c2648-2605-46be-a2e1-a88f5538aad3/0/CSV/ijqi2jrd9m0i170540h0dv0lss" will delivers the results of the query via the CSV protocol handler. (Binary will be the Odysseus Protocol Handler)

Reading in a Browser

You can access the result e.g. in a browser with java script. (see for an example)

Attention: In JSON the result contains the metadata appended after JSON-Document with an ";" as separator.

Reading in Odysseus

You could also use this output as an input to another odysseus operator as in the following

in = ACCESS({
	transport = 'WebsocketClient',	
	wrapper = 'GenericPush',
	datahandler = 'Tuple',
	protocol = 'CSV',
	source = 'access',
	readMetadata = 'true',
	options = [
	schema = [
		['person','id', 'INTEGER'],
		['person','name', 'STRING'],
		['person','mail', 'STRING']

The same information can be found for all other operators in the queryOperators section.

  • No labels