Java SearchHit.sourceAsMap方法代码示例

import org.elasticsearch.search.SearchHit; //导入方法依赖的package包/类
public ComperableHitResult(SearchHit hit , String[] fieldsOrder ,String seperator) {
    this.hit = hit;
    Map<String, Object> hitAsMap = hit.sourceAsMap();
    this.flattenMap = new HashMap<>();
    List<String> results = new ArrayList<>();
    this.isAllNull = true;

    for(int i = 0 ; i< fieldsOrder.length ;i++){
        String field = fieldsOrder[i];
        Object result = Util.deepSearchInMap(hitAsMap,field);
        if(result == null){
            results.add("");
        }
        else {
            this.isAllNull = false;
            results.add(result.toString());
            this.flattenMap.put(field,result);
        }
    }
    this.comperator = Joiner.on(seperator).join(results);
}
 

import org.elasticsearch.search.SearchHit; //导入方法依赖的package包/类
private String getComparisonKey(List<Map.Entry<Field, Field>> t1ToT2FieldsComparison, SearchHit hit, boolean firstTable, Map<String, List<Object>> optimizationTermsFilterStructure) {
    String key = "";
    Map<String, Object> sourceAsMap = hit.sourceAsMap();
    for (Map.Entry<Field, Field> t1ToT2 : t1ToT2FieldsComparison) {
        //todo: change to our function find if key contains '.'
        String name;
        if (firstTable) name = t1ToT2.getKey().getName();
        else name = t1ToT2.getValue().getName();

        Object data = deepSearchInMap(sourceAsMap, name);
        if (firstTable && useQueryTermsFilterOptimization) {
            updateOptimizationData(optimizationTermsFilterStructure, data, t1ToT2.getValue().getName());
        }
        if (data == null)
            key += "|null|";
        else
            key += "|" + data.toString() + "|";
    }
    return key;
}
 

import org.elasticsearch.search.SearchHit; //导入方法依赖的package包/类
private boolean hitsContains(SearchHit[] hits, Map<String, Object> matchMap) {
    for(SearchHit hit : hits){
        Map<String, Object> hitMap = hit.sourceAsMap();
        boolean matchedHit = true;
        for(Map.Entry<String,Object> entry: hitMap.entrySet()){
            if(!matchMap.containsKey(entry.getKey())) {
                matchedHit = false;
                break;
            }
            if(!equalsWithNullCheck(matchMap.get(entry.getKey()), entry.getValue())){
                matchedHit = false;
                break;
            }
        }
        if(matchedHit) return true;
    }
    return false;
}
 

import org.elasticsearch.search.SearchHit; //导入方法依赖的package包/类
private Object getFieldValue(SearchHit hit, String fieldName) {
    Map<String,Object> sourceAsMap = hit.sourceAsMap();
    if(fieldName.contains(".")){
        String[] split = fieldName.split("\\.");
        return Util.searchPathInMap(sourceAsMap, split);
    }
    else if(sourceAsMap.containsKey(fieldName)){
        return sourceAsMap.get(fieldName);
    }
    return null;
}
 

import org.elasticsearch.search.SearchHit; //导入方法依赖的package包/类
private List<String> createHeadersAndFillDocsMap(boolean flat, SearchHit[] hits, List<Map<String, Object>> docsAsMap) {
    Set<String> csvHeaders = new HashSet<>();
    for(SearchHit hit : hits){
        Map<String, Object> doc = hit.sourceAsMap();
        Map<String, SearchHitField> fields = hit.getFields();
        for(SearchHitField searchHitField : fields.values()){
            doc.put(searchHitField.getName(),searchHitField.value());
        }
        mergeHeaders(csvHeaders, doc, flat);
        if(this.indcludeId){
            doc.put("_id", hit.id());
        }
        if(this.includeScore){
            doc.put("_score", hit.score());
        }
        if(this.includeType){
            doc.put("_type",hit.type());
        }
        docsAsMap.add(doc);
    }
    ArrayList<String> headersList = new ArrayList<>(csvHeaders);
    if (this.indcludeId){
        headersList.add("_id");
    }
    if (this.includeScore){
        headersList.add("_score");
    }
    if (this.includeType){
        headersList.add("_type");
    }
    return headersList;
}
 

import org.elasticsearch.search.SearchHit; //导入方法依赖的package包/类
private List<String> createHeadersAndFillDocsMap(boolean flat, SearchHit[] hits, List<Map<String, Object>> docsAsMap) {
    Set<String> csvHeaders = new HashSet<>();
    for (SearchHit hit : hits) {
        Map<String, Object> doc = hit.sourceAsMap();
        Map<String, SearchHitField> fields = hit.getFields();
        for (SearchHitField searchHitField : fields.values()) {
            doc.put(searchHitField.getName(), searchHitField.value());
        }
        mergeHeaders(csvHeaders, doc, flat);
        if (this.includeScore) {
            doc.put("_score", hit.score());
        }
        if (this.includeType) {
            doc.put("_type", hit.type());
        }
        if (this.includeId) {
            doc.put("_id", hit.id());
        }
        docsAsMap.add(doc);
    }
    ArrayList<String> headersList = new ArrayList<>(csvHeaders);
    if (this.includeScore) {
        headersList.add("_score");
    }
    if (this.includeType) {
        headersList.add("_type");
    }
    if (this.includeId) {
        headersList.add("_id");
    }
    return headersList;
}
 

import org.elasticsearch.search.SearchHit; //导入方法依赖的package包/类
@Test
public void notLikeTests() throws IOException, SqlParseException, SQLFeatureNotSupportedException{
    //cant use string.format cause of %d
    SearchHits response = query("SELECT name FROM " +TEST_INDEX + "/gotCharacters where name.firstname not like '%d' LIMIT 1000");
    Assert.assertEquals(3, response.getTotalHits());
    for(SearchHit hit : response.getHits()) {
        Map<String, Object> sourceAsMap = hit.sourceAsMap();
        String name = ((HashMap<String, Object>) sourceAsMap.get("name")).get("firstname").toString();
        Assert.assertFalse(name+" was in not like %d",name.startsWith("d"));
    }
}
 

import org.elasticsearch.search.SearchHit; //导入方法依赖的package包/类
@Override
public void onResponse(SearchResponse response) {
    long timeTaken = System.currentTimeMillis() - timeStartESQuery;

    Aggregations aggregs = response.getAggregations();

    Terms agg = aggregs.get("partition");

    int numhits = 0;
    int numBuckets = 0;
    int alreadyprocessed = 0;

    synchronized (buffer) {
        // For each entry
        for (Terms.Bucket entry : agg.getBuckets()) {
            String key = (String) entry.getKey(); // bucket key
            long docCount = entry.getDocCount(); // Doc count

            int hitsForThisBucket = 0;

            // filter results so that we don't include URLs we are already
            // being processed
            TopHits topHits = entry.getAggregations().get("docs");
            for (SearchHit hit : topHits.getHits().getHits()) {
                hitsForThisBucket++;

                Map<String, Object> keyValues = hit.sourceAsMap();
                String url = (String) keyValues.get("url");

                LOG.debug("{} -> id [{}], _source [{}]", logIdprefix,
                        hit.getId(), hit.getSourceAsString());

                // is already being processed - skip it!
                if (beingProcessed.containsKey(url)) {
                    alreadyprocessed++;
                    continue;
                }
                Metadata metadata = fromKeyValues(keyValues);
                buffer.add(new Values(url, metadata));
            }

            if (hitsForThisBucket > 0)
                numBuckets++;

            numhits += hitsForThisBucket;

            LOG.debug("{} key [{}], hits[{}], doc_count [{}]", logIdprefix,
                    key, hitsForThisBucket, docCount, alreadyprocessed);
        }

        // Shuffle the URLs so that we don't get blocks of URLs from the
        // same
        // host or domain
        Collections.shuffle((List) buffer);
    }

    LOG.info(
            "{} ES query returned {} hits from {} buckets in {} msec with {} already being processed",
            logIdprefix, numhits, numBuckets, timeTaken, alreadyprocessed);

    esQueryTimes.addMeasurement(timeTaken);
    eventCounter.scope("already_being_processed").incrBy(alreadyprocessed);
    eventCounter.scope("ES_queries").incrBy(1);
    eventCounter.scope("ES_docs").incrBy(numhits);

    // remove lock
    isInESQuery.set(false);
}