Examples with StreamContext org.apache.solr.client.solrj.io.stream.StreamContext used on opensource projects

Example 11 with StreamContext

use of org.apache.solr.client.solrj.io.stream.StreamContext in project lucene-solr by apache.

the class GraphExpressionTest method testShortestPathStream.

@Test
public void testShortestPathStream() throws Exception {
    new UpdateRequest().add(id, "0", "from_s", "jim", "to_s", "mike", "predicate_s", "knows").add(id, "1", "from_s", "jim", "to_s", "dave", "predicate_s", "knows").add(id, "2", "from_s", "jim", "to_s", "stan", "predicate_s", "knows").add(id, "3", "from_s", "dave", "to_s", "stan", "predicate_s", "knows").add(id, "4", "from_s", "dave", "to_s", "bill", "predicate_s", "knows").add(id, "5", "from_s", "dave", "to_s", "mike", "predicate_s", "knows").add(id, "20", "from_s", "dave", "to_s", "alex", "predicate_s", "knows").add(id, "21", "from_s", "alex", "to_s", "steve", "predicate_s", "knows").add(id, "6", "from_s", "stan", "to_s", "alice", "predicate_s", "knows").add(id, "7", "from_s", "stan", "to_s", "mary", "predicate_s", "knows").add(id, "8", "from_s", "stan", "to_s", "dave", "predicate_s", "knows").add(id, "10", "from_s", "mary", "to_s", "mike", "predicate_s", "knows").add(id, "11", "from_s", "mary", "to_s", "max", "predicate_s", "knows").add(id, "12", "from_s", "mary", "to_s", "jim", "predicate_s", "knows").add(id, "13", "from_s", "mary", "to_s", "steve", "predicate_s", "knows").commit(cluster.getSolrClient(), COLLECTION);
    List<Tuple> tuples = null;
    Set<String> paths = null;
    ShortestPathStream stream = null;
    StreamContext context = new StreamContext();
    SolrClientCache cache = new SolrClientCache();
    context.setSolrClientCache(cache);
    StreamFactory factory = new StreamFactory().withCollectionZkHost("collection1", cluster.getZkServer().getZkAddress()).withFunctionName("shortestPath", ShortestPathStream.class);
    Map params = new HashMap();
    params.put("fq", "predicate_s:knows");
    stream = (ShortestPathStream) factory.constructStream("shortestPath(collection1, " + "from=\"jim\", " + "to=\"steve\"," + "edge=\"from_s=to_s\"," + "fq=\"predicate_s:knows\"," + "threads=\"3\"," + "partitionSize=\"3\"," + "maxDepth=\"6\")");
    stream.setStreamContext(context);
    paths = new HashSet();
    tuples = getTuples(stream);
    assertTrue(tuples.size() == 2);
    for (Tuple tuple : tuples) {
        paths.add(tuple.getStrings("path").toString());
    }
    assertTrue(paths.contains("[jim, dave, alex, steve]"));
    assertTrue(paths.contains("[jim, stan, mary, steve]"));
    //Test with batch size of 1
    params.put("fq", "predicate_s:knows");
    stream = (ShortestPathStream) factory.constructStream("shortestPath(collection1, " + "from=\"jim\", " + "to=\"steve\"," + "edge=\"from_s=to_s\"," + "fq=\"predicate_s:knows\"," + "threads=\"3\"," + "partitionSize=\"1\"," + "maxDepth=\"6\")");
    stream.setStreamContext(context);
    paths = new HashSet();
    tuples = getTuples(stream);
    assertTrue(tuples.size() == 2);
    for (Tuple tuple : tuples) {
        paths.add(tuple.getStrings("path").toString());
    }
    assertTrue(paths.contains("[jim, dave, alex, steve]"));
    assertTrue(paths.contains("[jim, stan, mary, steve]"));
    //Test with bad predicate
    stream = (ShortestPathStream) factory.constructStream("shortestPath(collection1, " + "from=\"jim\", " + "to=\"steve\"," + "edge=\"from_s=to_s\"," + "fq=\"predicate_s:crap\"," + "threads=\"3\"," + "partitionSize=\"3\"," + "maxDepth=\"6\")");
    stream.setStreamContext(context);
    paths = new HashSet();
    tuples = getTuples(stream);
    assertTrue(tuples.size() == 0);
    //Test with depth 2
    stream = (ShortestPathStream) factory.constructStream("shortestPath(collection1, " + "from=\"jim\", " + "to=\"steve\"," + "edge=\"from_s=to_s\"," + "fq=\"predicate_s:knows\"," + "threads=\"3\"," + "partitionSize=\"3\"," + "maxDepth=\"2\")");
    stream.setStreamContext(context);
    tuples = getTuples(stream);
    assertTrue(tuples.size() == 0);
    //Take out alex
    params.put("fq", "predicate_s:knows NOT to_s:alex");
    stream = (ShortestPathStream) factory.constructStream("shortestPath(collection1, " + "from=\"jim\", " + "to=\"steve\"," + "edge=\"from_s=to_s\"," + "fq=\" predicate_s:knows NOT to_s:alex\"," + "threads=\"3\"," + "partitionSize=\"3\"," + "maxDepth=\"6\")");
    stream.setStreamContext(context);
    paths = new HashSet();
    tuples = getTuples(stream);
    assertTrue(tuples.size() == 1);
    for (Tuple tuple : tuples) {
        paths.add(tuple.getStrings("path").toString());
    }
    assertTrue(paths.contains("[jim, stan, mary, steve]"));
    cache.close();
}

Example 12 with StreamContext

use of org.apache.solr.client.solrj.io.stream.StreamContext in project lucene-solr by apache.

the class GraphExpressionTest method testScoreNodesFacetStream.

@Test
public void testScoreNodesFacetStream() throws Exception {
    new UpdateRequest().add(id, "0", "basket_s", "basket1", "product_ss", "product1", "product_ss", "product3", "product_ss", "product5", "price_f", "1").add(id, "3", "basket_s", "basket2", "product_ss", "product1", "product_ss", "product6", "product_ss", "product7", "price_f", "1").add(id, "6", "basket_s", "basket3", "product_ss", "product4", "product_ss", "product3", "product_ss", "product1", "price_f", "1").add(id, "9", "basket_s", "basket4", "product_ss", "product4", "product_ss", "product3", "product_ss", "product1", "price_f", "1").commit(cluster.getSolrClient(), COLLECTION);
    List<Tuple> tuples = null;
    TupleStream stream = null;
    StreamContext context = new StreamContext();
    SolrClientCache cache = new SolrClientCache();
    context.setSolrClientCache(cache);
    StreamFactory factory = new StreamFactory().withCollectionZkHost("collection1", cluster.getZkServer().getZkAddress()).withDefaultZkHost(cluster.getZkServer().getZkAddress()).withFunctionName("gatherNodes", GatherNodesStream.class).withFunctionName("scoreNodes", ScoreNodesStream.class).withFunctionName("search", CloudSolrStream.class).withFunctionName("facet", FacetStream.class).withFunctionName("sort", SortStream.class).withFunctionName("count", CountMetric.class).withFunctionName("avg", MeanMetric.class).withFunctionName("sum", SumMetric.class).withFunctionName("min", MinMetric.class).withFunctionName("max", MaxMetric.class);
    String expr = "sort(by=\"nodeScore desc\",scoreNodes(facet(collection1, q=\"product_ss:product3\", buckets=\"product_ss\", bucketSorts=\"count(*) desc\", bucketSizeLimit=100, count(*))))";
    stream = factory.constructStream(expr);
    context = new StreamContext();
    context.setSolrClientCache(cache);
    stream.setStreamContext(context);
    tuples = getTuples(stream);
    Tuple tuple = tuples.get(0);
    assert (tuple.getString("node").equals("product3"));
    assert (tuple.getLong("docFreq") == 3);
    assert (tuple.getLong("count(*)") == 3);
    Tuple tuple0 = tuples.get(1);
    assert (tuple0.getString("node").equals("product4"));
    assert (tuple0.getLong("docFreq") == 2);
    assert (tuple0.getLong("count(*)") == 2);
    Tuple tuple1 = tuples.get(2);
    assert (tuple1.getString("node").equals("product1"));
    assert (tuple1.getLong("docFreq") == 4);
    assert (tuple1.getLong("count(*)") == 3);
    Tuple tuple2 = tuples.get(3);
    assert (tuple2.getString("node").equals("product5"));
    assert (tuple2.getLong("docFreq") == 1);
    assert (tuple2.getLong("count(*)") == 1);
    cache.close();
}

Example 13 with StreamContext

use of org.apache.solr.client.solrj.io.stream.StreamContext in project lucene-solr by apache.

the class GraphExpressionTest method testScoreNodesStream.

@Test
public void testScoreNodesStream() throws Exception {
    new UpdateRequest().add(id, "0", "basket_s", "basket1", "product_s", "product1", "price_f", "1").add(id, "1", "basket_s", "basket1", "product_s", "product3", "price_f", "1").add(id, "2", "basket_s", "basket1", "product_s", "product5", "price_f", "100").add(id, "3", "basket_s", "basket2", "product_s", "product1", "price_f", "1").add(id, "4", "basket_s", "basket2", "product_s", "product6", "price_f", "1").add(id, "5", "basket_s", "basket2", "product_s", "product7", "price_f", "1").add(id, "6", "basket_s", "basket3", "product_s", "product4", "price_f", "1").add(id, "7", "basket_s", "basket3", "product_s", "product3", "price_f", "1").add(id, "8", "basket_s", "basket3", "product_s", "product1", "price_f", "1").add(id, "9", "basket_s", "basket4", "product_s", "product4", "price_f", "1").add(id, "10", "basket_s", "basket4", "product_s", "product3", "price_f", "1").add(id, "11", "basket_s", "basket4", "product_s", "product1", "price_f", "1").add(id, "12", "basket_s", "basket5", "product_s", "product1", "price_f", "1").add(id, "13", "basket_s", "basket6", "product_s", "product1", "price_f", "1").add(id, "14", "basket_s", "basket7", "product_s", "product1", "price_f", "1").add(id, "15", "basket_s", "basket4", "product_s", "product1", "price_f", "1").commit(cluster.getSolrClient(), COLLECTION);
    List<Tuple> tuples = null;
    TupleStream stream = null;
    StreamContext context = new StreamContext();
    SolrClientCache cache = new SolrClientCache();
    context.setSolrClientCache(cache);
    StreamFactory factory = new StreamFactory().withCollectionZkHost("collection1", cluster.getZkServer().getZkAddress()).withDefaultZkHost(cluster.getZkServer().getZkAddress()).withFunctionName("gatherNodes", GatherNodesStream.class).withFunctionName("scoreNodes", ScoreNodesStream.class).withFunctionName("search", CloudSolrStream.class).withFunctionName("sort", SortStream.class).withFunctionName("count", CountMetric.class).withFunctionName("avg", MeanMetric.class).withFunctionName("sum", SumMetric.class).withFunctionName("min", MinMetric.class).withFunctionName("max", MaxMetric.class);
    String expr = "gatherNodes(collection1, " + "walk=\"product3->product_s\"," + "gather=\"basket_s\")";
    String expr2 = "sort(by=\"nodeScore desc\", " + "scoreNodes(gatherNodes(collection1, " + expr + "," + "walk=\"node->basket_s\"," + "gather=\"product_s\", " + "count(*), " + "avg(price_f), " + "sum(price_f), " + "min(price_f), " + "max(price_f))))";
    stream = factory.constructStream(expr2);
    context = new StreamContext();
    context.setSolrClientCache(cache);
    stream.setStreamContext(context);
    tuples = getTuples(stream);
    Tuple tuple0 = tuples.get(0);
    assert (tuple0.getString("node").equals("product4"));
    assert (tuple0.getLong("docFreq") == 2);
    assert (tuple0.getLong("count(*)") == 2);
    Tuple tuple1 = tuples.get(1);
    assert (tuple1.getString("node").equals("product1"));
    assert (tuple1.getLong("docFreq") == 8);
    assert (tuple1.getLong("count(*)") == 3);
    Tuple tuple2 = tuples.get(2);
    assert (tuple2.getString("node").equals("product5"));
    assert (tuple2.getLong("docFreq") == 1);
    assert (tuple2.getLong("count(*)") == 1);
    //Test using a different termFreq field then the default count(*)
    expr2 = "sort(by=\"nodeScore desc\", " + "scoreNodes(termFreq=\"avg(price_f)\",gatherNodes(collection1, " + expr + "," + "walk=\"node->basket_s\"," + "gather=\"product_s\", " + "count(*), " + "avg(price_f), " + "sum(price_f), " + "min(price_f), " + "max(price_f))))";
    stream = factory.constructStream(expr2);
    context = new StreamContext();
    context.setSolrClientCache(cache);
    stream.setStreamContext(context);
    tuples = getTuples(stream);
    tuple0 = tuples.get(0);
    assert (tuple0.getString("node").equals("product5"));
    assert (tuple0.getLong("docFreq") == 1);
    assert (tuple0.getDouble("avg(price_f)") == 100);
    tuple1 = tuples.get(1);
    assert (tuple1.getString("node").equals("product4"));
    assert (tuple1.getLong("docFreq") == 2);
    assert (tuple1.getDouble("avg(price_f)") == 1);
    tuple2 = tuples.get(2);
    assert (tuple2.getString("node").equals("product1"));
    assert (tuple2.getLong("docFreq") == 8);
    assert (tuple2.getDouble("avg(price_f)") == 1);
    cache.close();
}

Example 14 with StreamContext

use of org.apache.solr.client.solrj.io.stream.StreamContext in project lucene-solr by apache.

the class GraphTest method testShortestPathStream.

@Test
public void testShortestPathStream() throws Exception {
    new UpdateRequest().add(id, "0", "from_s", "jim", "to_s", "mike", "predicate_s", "knows").add(id, "1", "from_s", "jim", "to_s", "dave", "predicate_s", "knows").add(id, "2", "from_s", "jim", "to_s", "stan", "predicate_s", "knows").add(id, "3", "from_s", "dave", "to_s", "stan", "predicate_s", "knows").add(id, "4", "from_s", "dave", "to_s", "bill", "predicate_s", "knows").add(id, "5", "from_s", "dave", "to_s", "mike", "predicate_s", "knows").add(id, "20", "from_s", "dave", "to_s", "alex", "predicate_s", "knows").add(id, "21", "from_s", "alex", "to_s", "steve", "predicate_s", "knows").add(id, "6", "from_s", "stan", "to_s", "alice", "predicate_s", "knows").add(id, "7", "from_s", "stan", "to_s", "mary", "predicate_s", "knows").add(id, "8", "from_s", "stan", "to_s", "dave", "predicate_s", "knows").add(id, "10", "from_s", "mary", "to_s", "mike", "predicate_s", "knows").add(id, "11", "from_s", "mary", "to_s", "max", "predicate_s", "knows").add(id, "12", "from_s", "mary", "to_s", "jim", "predicate_s", "knows").add(id, "13", "from_s", "mary", "to_s", "steve", "predicate_s", "knows").commit(cluster.getSolrClient(), COLLECTION);
    List<Tuple> tuples = null;
    Set<String> paths = null;
    ShortestPathStream stream = null;
    String zkHost = cluster.getZkServer().getZkAddress();
    StreamContext context = new StreamContext();
    SolrClientCache cache = new SolrClientCache();
    context.setSolrClientCache(cache);
    SolrParams sParams = StreamingTest.mapParams("fq", "predicate_s:knows");
    stream = new ShortestPathStream(zkHost, "collection1", "jim", "steve", "from_s", "to_s", sParams, 20, 3, 6);
    stream.setStreamContext(context);
    paths = new HashSet();
    tuples = getTuples(stream);
    assertTrue(tuples.size() == 2);
    for (Tuple tuple : tuples) {
        paths.add(tuple.getStrings("path").toString());
    }
    assertTrue(paths.contains("[jim, dave, alex, steve]"));
    assertTrue(paths.contains("[jim, stan, mary, steve]"));
    //Test with batch size of 1
    sParams = StreamingTest.mapParams("fq", "predicate_s:knows");
    stream = new ShortestPathStream(zkHost, "collection1", "jim", "steve", "from_s", "to_s", sParams, 1, 3, 6);
    stream.setStreamContext(context);
    paths = new HashSet();
    tuples = getTuples(stream);
    assertTrue(tuples.size() == 2);
    for (Tuple tuple : tuples) {
        paths.add(tuple.getStrings("path").toString());
    }
    assertTrue(paths.contains("[jim, dave, alex, steve]"));
    assertTrue(paths.contains("[jim, stan, mary, steve]"));
    //Test with bad predicate
    sParams = StreamingTest.mapParams("fq", "predicate_s:crap");
    stream = new ShortestPathStream(zkHost, "collection1", "jim", "steve", "from_s", "to_s", sParams, 1, 3, 6);
    stream.setStreamContext(context);
    paths = new HashSet();
    tuples = getTuples(stream);
    assertTrue(tuples.size() == 0);
    //Test with depth 2
    sParams = StreamingTest.mapParams("fq", "predicate_s:knows");
    stream = new ShortestPathStream(zkHost, "collection1", "jim", "steve", "from_s", "to_s", sParams, 1, 3, 2);
    stream.setStreamContext(context);
    paths = new HashSet();
    tuples = getTuples(stream);
    assertTrue(tuples.size() == 0);
    //Take out alex
    sParams = StreamingTest.mapParams("fq", "predicate_s:knows NOT to_s:alex");
    stream = new ShortestPathStream(zkHost, "collection1", "jim", "steve", "from_s", "to_s", sParams, 10, 3, 6);
    stream.setStreamContext(context);
    paths = new HashSet();
    tuples = getTuples(stream);
    assertTrue(tuples.size() == 1);
    for (Tuple tuple : tuples) {
        paths.add(tuple.getStrings("path").toString());
    }
    assertTrue(paths.contains("[jim, stan, mary, steve]"));
    cache.close();
}