Examples with Filter org.elasticsearch.search.aggregations.bucket.filter.Filter used on opensource projects

Example 41 with Filter

use of org.elasticsearch.search.aggregations.bucket.filter.Filter in project graylog2-server by Graylog2.

the class Searches method terms.

public TermsResult terms(String field, int size, String query, String filter, TimeRange range, Sorting.Direction sorting) {
    Terms.Order termsOrder;
    if (size == 0) {
        size = 50;
    }
    if (sorting == Sorting.Direction.DESC) {
        termsOrder = Terms.Order.count(false);
    } else {
        termsOrder = Terms.Order.count(true);
    }
    SearchRequestBuilder srb;
    if (filter == null) {
        srb = standardSearchRequest(query, determineAffectedIndices(range, null), range);
    } else {
        srb = filteredSearchRequest(query, filter, determineAffectedIndices(range, filter), range);
    }
    FilterAggregationBuilder builder = AggregationBuilders.filter(AGG_FILTER).subAggregation(AggregationBuilders.terms(AGG_TERMS).field(field).size(size).order(termsOrder)).subAggregation(AggregationBuilders.missing("missing").field(field)).filter(standardAggregationFilters(range, filter));
    srb.addAggregation(builder);
    final SearchRequest request = srb.request();
    SearchResponse r = c.search(request).actionGet();
    recordEsMetrics(r, range);
    final Filter f = r.getAggregations().get(AGG_FILTER);
    return new TermsResult(f.getAggregations().get(AGG_TERMS), f.getAggregations().get("missing"), f.getDocCount(), query, request.source(), r.getTook());
}

Example 42 with Filter

use of org.elasticsearch.search.aggregations.bucket.filter.Filter in project graylog2-server by Graylog2.

the class Searches method histogram.

public HistogramResult histogram(String query, DateHistogramInterval interval, String filter, TimeRange range) {
    FilterAggregationBuilder builder = AggregationBuilders.filter(AGG_FILTER).subAggregation(AggregationBuilders.dateHistogram(AGG_HISTOGRAM).field("timestamp").interval(interval.toESInterval())).filter(standardAggregationFilters(range, filter));
    QueryStringQueryBuilder qs = queryStringQuery(query);
    qs.allowLeadingWildcard(configuration.isAllowLeadingWildcardSearches());
    final Set<String> affectedIndices = determineAffectedIndices(range, filter);
    final SearchRequestBuilder srb = c.prepareSearch(affectedIndices.toArray(new String[affectedIndices.size()])).setIndicesOptions(IndicesOptions.lenientExpandOpen()).setQuery(qs).addAggregation(builder);
    final SearchRequest request = srb.request();
    SearchResponse r = c.search(request).actionGet();
    recordEsMetrics(r, range);
    final Filter f = r.getAggregations().get(AGG_FILTER);
    return new DateHistogramResult(f.getAggregations().get(AGG_HISTOGRAM), query, request.source(), interval, r.getTook());
}

Example 43 with Filter

use of org.elasticsearch.search.aggregations.bucket.filter.Filter in project stash-codesearch-plugin by palantir.

the class SearchServlet method doGet.

@Override
protected void doGet(HttpServletRequest req, HttpServletResponse resp) throws ServletException, IOException {
    // Make sure user is logged in
    try {
        validationService.validateAuthenticated();
    } catch (AuthorisationException notLoggedInException) {
        try {
            resp.sendRedirect(propertiesService.getLoginUri(URI.create(req.getRequestURL() + (req.getQueryString() == null ? "" : "?" + req.getQueryString()))).toASCIIString());
        } catch (Exception e) {
            log.error("Unable to redirect unauthenticated user to login page", e);
        }
        return;
    }
    // Query and parse settings
    SearchParams params = SearchParams.getParams(req, DateTimeZone.forTimeZone(propertiesService.getDefaultTimeZone()));
    GlobalSettings globalSettings = settingsManager.getGlobalSettings();
    ImmutableSet.Builder<String> noHighlightBuilder = new ImmutableSet.Builder<String>();
    for (String extension : globalSettings.getNoHighlightExtensions().split(",")) {
        extension = extension.trim().toLowerCase();
        if (!extension.isEmpty()) {
            noHighlightBuilder.add(extension);
        }
    }
    ImmutableSet<String> noHighlight = noHighlightBuilder.build();
    int maxPreviewLines = globalSettings.getMaxPreviewLines();
    int maxMatchLines = globalSettings.getMaxMatchLines();
    int maxFragments = globalSettings.getMaxFragments();
    int pageSize = globalSettings.getPageSize();
    TimeValue searchTimeout = new TimeValue(globalSettings.getSearchTimeout());
    float commitHashBoost = (float) globalSettings.getCommitHashBoost();
    float commitSubjectBoost = (float) globalSettings.getCommitBodyBoost();
    float commitBodyBoost = (float) globalSettings.getCommitBodyBoost();
    float fileNameBoost = (float) globalSettings.getFileNameBoost();
    // Execute ES query
    int pages = 0;
    long totalHits = 0;
    long searchTime = 0;
    SearchHit[] currentHits = {};
    String error = "";
    ArrayList<ImmutableMap<String, Object>> hitArray = new ArrayList<ImmutableMap<String, Object>>(currentHits.length);
    ImmutableMap<String, Object> statistics = ImmutableMap.of();
    if (params.doSearch) {
        // Repo map is null iff user is a system administrator (don't need to validate permissions).
        ImmutableMap<String, Repository> repoMap;
        try {
            validationService.validateForGlobal(Permission.SYS_ADMIN);
            repoMap = null;
        } catch (AuthorisationException e) {
            repoMap = repositoryServiceManager.getRepositoryMap(validationService);
            if (repoMap.isEmpty()) {
                error = "You do not have permissions to access any repositories";
            }
        }
        int startIndex = params.page * pageSize;
        SearchRequestBuilder esReq = es.getClient().prepareSearch(ES_SEARCHALIAS).setFrom(startIndex).setSize(pageSize).setTimeout(searchTimeout).setFetchSource(true);
        if (error != null && !error.isEmpty()) {
            log.warn("Not performing search due to error {}", error);
        } else {
            // Build query source and perform query
            QueryBuilder query = matchAllQuery();
            if (params.searchString != null && !params.searchString.isEmpty()) {
                QueryStringQueryBuilder queryStringQuery = queryString(params.searchString).analyzeWildcard(true).lenient(true).defaultOperator(QueryStringQueryBuilder.Operator.AND);
                if (params.searchCommits) {
                    queryStringQuery.field("commit.subject", commitSubjectBoost).field("commit.hash", commitHashBoost).field("commit.body", commitBodyBoost);
                }
                if (params.searchFilenames) {
                    queryStringQuery.field("file.path", fileNameBoost);
                }
                if (params.searchCode) {
                    queryStringQuery.field("file.contents", 1);
                }
                query = queryStringQuery;
            }
            FilterBuilder filter = andFilter(boolFilter().must(repoMap == null ? matchAllFilter() : sf.aclFilter(repoMap), sf.refFilter(params.refNames.split(",")), sf.projectFilter(params.projectKeys.split(",")), sf.repositoryFilter(params.repoNames.split(",")), sf.extensionFilter(params.extensions.split(",")), sf.authorFilter(params.authorNames.split(","))), sf.dateRangeFilter(params.committedAfter, params.committedBefore));
            FilteredQueryBuilder finalQuery = filteredQuery(query, filter);
            esReq.setQuery(finalQuery).setHighlighterPreTags("\u0001").setHighlighterPostTags("\u0001").addHighlightedField("contents", 1, maxFragments);
            String[] typeArray = {};
            if (params.searchCommits) {
                if (params.searchFilenames || params.searchCode) {
                    typeArray = new String[] { "commit", "file" };
                } else {
                    typeArray = new String[] { "commit" };
                }
            } else if (params.searchFilenames || params.searchCode) {
                typeArray = new String[] { "file" };
            }
            esReq.setTypes(typeArray);
            // Build aggregations if statistics were requested
            if (params.showStatistics) {
                esReq.addAggregation(cardinality("authorCardinality").field("authoremail.untouched").precisionThreshold(1000)).addAggregation(terms("authorRanking").field("authoremail.untouched").size(25)).addAggregation(percentiles("charcountPercentiles").field("charcount").percentiles(PERCENTILES)).addAggregation(extendedStats("charcountStats").field("charcount")).addAggregation(filter("commitCount").filter(typeFilter("commit"))).addAggregation(cardinality("extensionCardinality").field("extension").precisionThreshold(1000)).addAggregation(terms("extensionRanking").field("extension").size(25)).addAggregation(percentiles("linecountPercentiles").field("linecount").percentiles(PERCENTILES)).addAggregation(extendedStats("linecountStats").field("linecount"));
            }
            SearchResponse esResp = null;
            try {
                esResp = esReq.get();
            } catch (SearchPhaseExecutionException e) {
                log.warn("Query failure", e);
                error = "Make sure your query conforms to the Lucene/Elasticsearch query string syntax.";
            }
            if (esResp != null) {
                SearchHits esHits = esResp.getHits();
                totalHits = esHits.getTotalHits();
                pages = (int) Math.min(Integer.MAX_VALUE, (totalHits + pageSize - 1) / pageSize);
                currentHits = esHits.getHits();
                searchTime = esResp.getTookInMillis();
                for (ShardSearchFailure failure : esResp.getShardFailures()) {
                    log.warn("Shard failure {}", failure.reason());
                    if (error == null || error.isEmpty()) {
                        error = "Shard failure: " + failure.reason();
                    }
                }
                Aggregations aggs = esResp.getAggregations();
                if (params.showStatistics && aggs != null && !aggs.asList().isEmpty()) {
                    Cardinality authorCardinality = aggs.get("authorCardinality");
                    Terms authorRanking = aggs.get("authorRanking");
                    Percentiles charcountPercentiles = aggs.get("charcountPercentiles");
                    Filter commitCount = aggs.get("commitCount");
                    ExtendedStats charcountStats = aggs.get("charcountStats");
                    Cardinality extensionCardinality = aggs.get("extensionCardinality");
                    Terms extensionRanking = aggs.get("extensionRanking");
                    Percentiles linecountPercentiles = aggs.get("linecountPercentiles");
                    ExtendedStats linecountStats = aggs.get("linecountStats");
                    statistics = new ImmutableMap.Builder<String, Object>().put("authorCardinality", authorCardinality.getValue()).put("authorRanking", getSoyRankingList(authorRanking, commitCount.getDocCount())).put("charcount", new ImmutableMap.Builder<String, Object>().put("average", charcountStats.getAvg()).put("max", Math.round(charcountStats.getMax())).put("min", Math.round(charcountStats.getMin())).put("percentiles", getSoyPercentileList(charcountPercentiles, PERCENTILES)).put("sum", Math.round(charcountStats.getSum())).build()).put("commitcount", commitCount.getDocCount()).put("extensionCardinality", extensionCardinality.getValue()).put("extensionRanking", getSoyRankingList(extensionRanking, charcountStats.getCount())).put("filecount", charcountStats.getCount()).put("linecount", new ImmutableMap.Builder<String, Object>().put("average", linecountStats.getAvg()).put("max", Math.round(linecountStats.getMax())).put("min", Math.round(linecountStats.getMin())).put("percentiles", getSoyPercentileList(linecountPercentiles, PERCENTILES)).put("sum", Math.round(linecountStats.getSum())).build()).build();
                }
            }
        }
        // Iterate through current page of search hits
        for (SearchHit hit : currentHits) {
            ImmutableMap<String, Object> hitData = searchHitToDataMap(hit, repoMap, maxPreviewLines, maxMatchLines, noHighlight);
            if (hitData != null) {
                hitArray.add(hitData);
            }
        }
    }
    // Render page
    pbs.assembler().resources().requireContext("com.atlassian.auiplugin:aui-date-picker");
    pbs.assembler().resources().requireContext("com.atlassian.auiplugin:aui-experimental-tooltips");
    pbs.assembler().resources().requireWebResource("com.palantir.stash.stash-code-search:scs-resources");
    resp.setContentType("text/html");
    try {
        String queryString = req.getQueryString();
        String fullUri = req.getRequestURI() + "?" + (queryString == null ? "" : queryString.replaceAll("&?page=\\d*", ""));
        ImmutableMap<String, Object> data = new ImmutableMap.Builder<String, Object>().put("pages", pages).put("currentPage", params.page).put("prevParams", params.soyParams).put("doSearch", params.doSearch).put("totalHits", totalHits).put("hitArray", hitArray).put("statistics", statistics).put("error", error).put("fullUri", fullUri).put("baseUrl", propertiesService.getBaseUrl().toASCIIString()).put("resultFrom", Math.min(totalHits, params.page * pageSize + 1)).put("resultTo", Math.min(totalHits, (params.page + 1) * pageSize)).put("searchTime", searchTime).build();
        soyTemplateRenderer.render(resp.getWriter(), "com.palantir.stash.stash-code-search:codesearch-soy", "plugin.page.codesearch.searchPage", data);
    } catch (Exception e) {
        log.error("Error rendering Soy template", e);
    }
}

Example 44 with Filter

use of org.elasticsearch.search.aggregations.bucket.filter.Filter in project graylog2-server by Graylog2.

the class IndicesAdapterES7 method indexRangeStatsOfIndex.

@Override
public IndexRangeStats indexRangeStatsOfIndex(String index) {
    final FilterAggregationBuilder builder = AggregationBuilders.filter("agg", QueryBuilders.existsQuery(Message.FIELD_TIMESTAMP)).subAggregation(AggregationBuilders.min("ts_min").field(Message.FIELD_TIMESTAMP)).subAggregation(AggregationBuilders.max("ts_max").field(Message.FIELD_TIMESTAMP)).subAggregation(AggregationBuilders.terms("streams").size(Integer.MAX_VALUE).field(Message.FIELD_STREAMS));
    final SearchSourceBuilder query = SearchSourceBuilder.searchSource().aggregation(builder).size(0);
    final SearchRequest request = new SearchRequest().source(query).indices(index).searchType(SearchType.DFS_QUERY_THEN_FETCH).indicesOptions(IndicesOptions.lenientExpandOpen());
    final SearchResponse result = client.execute((c, requestOptions) -> c.search(request, requestOptions), "Couldn't build index range of index " + index);
    if (result.getTotalShards() == 0 || result.getAggregations() == null) {
        throw new IndexNotFoundException("Couldn't build index range of index " + index + " because it doesn't exist.");
    }
    final Filter f = result.getAggregations().get("agg");
    if (f == null) {
        throw new IndexNotFoundException("Couldn't build index range of index " + index + " because it doesn't exist.");
    } else if (f.getDocCount() == 0L) {
        LOG.debug("No documents with attribute \"timestamp\" found in index <{}>", index);
        return IndexRangeStats.EMPTY;
    }
    final Min minAgg = f.getAggregations().get("ts_min");
    final long minUnixTime = new Double(minAgg.getValue()).longValue();
    final DateTime min = new DateTime(minUnixTime, DateTimeZone.UTC);
    final Max maxAgg = f.getAggregations().get("ts_max");
    final long maxUnixTime = new Double(maxAgg.getValue()).longValue();
    final DateTime max = new DateTime(maxUnixTime, DateTimeZone.UTC);
    // make sure we return an empty list, so we can differentiate between old indices that don't have this information
    // and newer ones that simply have no streams.
    final Terms streams = f.getAggregations().get("streams");
    final List<String> streamIds = streams.getBuckets().stream().map(MultiBucketsAggregation.Bucket::getKeyAsString).collect(toList());
    return IndexRangeStats.create(min, max, streamIds);
}

Example 45 with Filter

use of org.elasticsearch.search.aggregations.bucket.filter.Filter in project graylog2-server by Graylog2.

the class IndexToolsAdapterES7 method fieldHistogram.

@Override
public Map<DateTime, Map<String, Long>> fieldHistogram(String fieldName, Set<String> indices, Optional<Set<String>> includedStreams, long interval) {
    final BoolQueryBuilder queryBuilder = buildStreamIdFilter(includedStreams);
    final FilterAggregationBuilder the_filter = AggregationBuilders.filter(AGG_FILTER, queryBuilder).subAggregation(AggregationBuilders.dateHistogram(AGG_DATE_HISTOGRAM).field("timestamp").subAggregation(AggregationBuilders.terms(AGG_MESSAGE_FIELD).field(fieldName)).fixedInterval(new DateHistogramInterval(interval + "ms")).minDocCount(1L));
    final SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder().query(QueryBuilders.matchAllQuery()).aggregation(the_filter);
    final SearchRequest searchRequest = new SearchRequest().source(searchSourceBuilder).indices(indices.toArray(new String[0]));
    final SearchResponse searchResult = client.search(searchRequest, "Unable to retrieve field histogram.");
    final Filter filterAggregation = searchResult.getAggregations().get(AGG_FILTER);
    final ParsedDateHistogram dateHistogram = filterAggregation.getAggregations().get(AGG_DATE_HISTOGRAM);
    final List<ParsedDateHistogram.ParsedBucket> histogramBuckets = (List<ParsedDateHistogram.ParsedBucket>) dateHistogram.getBuckets();
    final Map<DateTime, Map<String, Long>> result = Maps.newHashMapWithExpectedSize(histogramBuckets.size());
    for (ParsedDateHistogram.ParsedBucket bucket : histogramBuckets) {
        final ZonedDateTime zonedDateTime = (ZonedDateTime) bucket.getKey();
        final DateTime date = new DateTime(zonedDateTime.toInstant().toEpochMilli()).toDateTime(DateTimeZone.UTC);
        final Terms sourceFieldAgg = bucket.getAggregations().get(AGG_MESSAGE_FIELD);
        final List<? extends Terms.Bucket> termBuckets = sourceFieldAgg.getBuckets();
        final HashMap<String, Long> termCounts = Maps.newHashMapWithExpectedSize(termBuckets.size());
        for (Terms.Bucket termBucket : termBuckets) {
            termCounts.put(termBucket.getKeyAsString(), termBucket.getDocCount());
        }
        result.put(date, termCounts);
    }
    return ImmutableMap.copyOf(result);
}