com.synedge.oss.client.QueryClient Maven / Gradle / Ivy

Go to download

Show more of this group Show more artifacts with this name
Show all versions of synedge-java-client Show documentation

This client allows you to easily connect to the Synedge API with any language running on the JDK

The newest version!

package com.synedge.oss.client;

import com.synedge.oss.client.query.Target;
import com.synedge.oss.client.query.TargetNode;

import java.util.List;

/**
 * Interface for making queries to the Synedge backend
 * @author Paul van Assen
 *
 */
public interface QueryClient {
    /**
     * Query a backend metric, from start to end. Start and end are in Graphite notation. Granuality can be 5m, 1h, 1d, 1w, 1mon or 1y. 
     * Targets are metrics to be queried. Some simple functions are available:
     * 
     * min
     * max
     * sum
     * avg
     * 
     * 

     * One special function is alias. This allows you to alias a target. It will in the results then not use the target as identifier but the alias. 

     * Sample method defines how data points should be treated when packed together in a bigger time frame. By default http metrics are stored in a 5 minute interval. If a
     * granularity of 1 hour is requested
     * data will be aggregated/packed. This can be done by adding it up, or taking the average. The methods are the same as the previously mentioned simple functions. 

     * Possible values granularity:
     * 
     * finest
     * hour
     * day
     * week
     * month
     * year
     * 
     * 

     * Possibel values for agregation are:
     * 
     * min
     * max
     * sum
     * avg
     * 
     * 
     * @param customerId Customer id
     * @param start Start, this may be an absolute date (ISO 8601 date) or relative: -1w for one week ago
     * @param end End time. Same rules apply as for start time
     * @param query Query to send. 
     * @return List of targets received
     */
    List query(String customerId, String start, String end, String query);
 
    /**
     * Metrics are stored in paths similar to Graphite. Each part is separated by a '.'. This allows for easy adding of new metrics. You can query each of the nodes or request a list of child nodes:
     * 
     * <distributiongroup>.http will return a list of one TargetNode containing the name http and leaf will be false. 
     * <distributiongroup>.http.* will return a list of nodes containing all children of http, at this moment it will contain europe and/or north_america and possibly south_america or asia. Each of the nodes will have the leaf set to false 
     * 
     * 
     * If leaf is set to true, it means data can be queried on it. 
     * 
     * @param customerId Customer id
     * @param parentNode Parent node. 
     * @return List of target nodes
     */
    List find(String customerId, String parentNode);
}