/**
 *    Copyright (C) 2018-present MongoDB, Inc.
 *
 *    This program is free software: you can redistribute it and/or modify
 *    it under the terms of the Server Side Public License, version 1,
 *    as published by MongoDB, Inc.
 *
 *    This program is distributed in the hope that it will be useful,
 *    but WITHOUT ANY WARRANTY; without even the implied warranty of
 *    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 *    Server Side Public License for more details.
 *
 *    You should have received a copy of the Server Side Public License
 *    along with this program. If not, see
 *    <http://www.mongodb.com/licensing/server-side-public-license>.
 *
 *    As a special exception, the copyright holders give permission to link the
 *    code of portions of this program with the OpenSSL library under certain
 *    conditions as described in each individual source file and distribute
 *    linked combinations including the program with the OpenSSL library. You
 *    must comply with the Server Side Public License in all respects for
 *    all of the code used other than as permitted herein. If you modify file(s)
 *    with this exception, you may extend this exception to your version of the
 *    file(s), but you are not obligated to do so. If you do not wish to do so,
 *    delete this exception statement from your version. If you delete this
 *    exception statement from all source files in the program, then also delete
 *    it in the license file.
 */


#include "mongo/platform/basic.h"

#include <memory>
#include <string>
#include <vector>

#include "mongo/db/auth/authorization_checks.h"
#include "mongo/db/auth/authorization_session.h"
#include "mongo/db/bson/dotted_path_support.h"
#include "mongo/db/client.h"
#include "mongo/db/clientcursor.h"
#include "mongo/db/commands.h"
#include "mongo/db/commands/run_aggregate.h"
#include "mongo/db/db_raii.h"
#include "mongo/db/exec/working_set_common.h"
#include "mongo/db/jsobj.h"
#include "mongo/db/matcher/extensions_callback_real.h"
#include "mongo/db/namespace_string.h"
#include "mongo/db/pipeline/aggregation_request_helper.h"
#include "mongo/db/query/collection_query_info.h"
#include "mongo/db/query/cursor_response.h"
#include "mongo/db/query/explain.h"
#include "mongo/db/query/find_common.h"
#include "mongo/db/query/get_executor.h"
#include "mongo/db/query/parsed_distinct.h"
#include "mongo/db/query/plan_summary_stats.h"
#include "mongo/db/query/query_planner_common.h"
#include "mongo/db/query/view_response_formatter.h"
#include "mongo/db/s/collection_sharding_state.h"
#include "mongo/db/s/query_analysis_writer.h"
#include "mongo/db/views/resolved_view.h"
#include "mongo/logv2/log.h"
#include "mongo/util/database_name_util.h"

#define MONGO_LOGV2_DEFAULT_COMPONENT ::mongo::logv2::LogComponent::kQuery


namespace mongo {
namespace {

namespace dps = dotted_path_support;

class DistinctCommand : public BasicCommand {
public:
    DistinctCommand() : BasicCommand("distinct") {}

    std::string help() const override {
        return "{ distinct : 'collection name' , key : 'a.b' , query : {} }";
    }

    AllowedOnSecondary secondaryAllowed(ServiceContext*) const override {
        return AllowedOnSecondary::kOptIn;
    }

    bool maintenanceOk() const override {
        return false;
    }

    bool supportsWriteConcern(const BSONObj& cmd) const override {
        return false;
    }

    bool collectsResourceConsumptionMetrics() const override {
        return true;
    }

    bool canIgnorePrepareConflicts() const override {
        return true;
    }

    ReadConcernSupportResult supportsReadConcern(const BSONObj& cmdObj,
                                                 repl::ReadConcernLevel level,
                                                 bool isImplicitDefault) const override {
        return ReadConcernSupportResult::allSupportedAndDefaultPermitted();
    }

    bool shouldAffectReadConcernCounter() const override {
        return true;
    }

    bool supportsReadMirroring(const BSONObj&) const override {
        return true;
    }

    ReadWriteType getReadWriteType() const override {
        return ReadWriteType::kRead;
    }

    std::size_t reserveBytesForReply() const override {
        return FindCommon::kInitReplyBufferSize;
    }

    Status checkAuthForOperation(OperationContext* opCtx,
                                 const DatabaseName& dbname,
                                 const BSONObj& cmdObj) const override {
        AuthorizationSession* authSession = AuthorizationSession::get(opCtx->getClient());

        if (!authSession->isAuthorizedToParseNamespaceElement(cmdObj.firstElement())) {
            return Status(ErrorCodes::Unauthorized, "Unauthorized");
        }

        const auto hasTerm = false;
        return auth::checkAuthForFind(authSession,
                                      CollectionCatalog::get(opCtx)->resolveNamespaceStringOrUUID(
                                          opCtx, CommandHelpers::parseNsOrUUID(dbname, cmdObj)),
                                      hasTerm);
    }

    bool allowedInTransactions() const final {
        return true;
    }

    bool allowedWithSecurityToken() const final {
        return true;
    }

    Status explain(OperationContext* opCtx,
                   const OpMsgRequest& request,
                   ExplainOptions::Verbosity verbosity,
                   rpc::ReplyBuilderInterface* result) const override {
        const DatabaseName dbName =
            DatabaseNameUtil::deserialize(request.getValidatedTenantId(), request.getDatabase());
        const BSONObj& cmdObj = request.body;
        // Acquire locks. The RAII object is optional, because in the case of a view, the locks
        // need to be released.
        boost::optional<AutoGetCollectionForReadCommandMaybeLockFree> ctx;
        ctx.emplace(
            opCtx,
            CommandHelpers::parseNsCollectionRequired(dbName, cmdObj),
            AutoGetCollection::Options{}.viewMode(auto_get_collection::ViewMode::kViewsPermitted));
        const auto nss = ctx->getNss();

        const ExtensionsCallbackReal extensionsCallback(opCtx, &nss);
        auto defaultCollator =
            ctx->getCollection() ? ctx->getCollection()->getDefaultCollator() : nullptr;
        auto parsedDistinct = uassertStatusOK(
            ParsedDistinct::parse(opCtx, nss, cmdObj, extensionsCallback, true, defaultCollator));

        if (ctx->getView()) {
            // Relinquish locks. The aggregation command will re-acquire them.
            ctx.reset();

            auto viewAggregation = parsedDistinct.asAggregationCommand();
            if (!viewAggregation.isOK()) {
                return viewAggregation.getStatus();
            }

            auto viewAggCmd =
                OpMsgRequest::fromDBAndBody(nss.db(), viewAggregation.getValue()).body;
            auto viewAggRequest = aggregation_request_helper::parseFromBSON(
                opCtx,
                nss,
                viewAggCmd,
                verbosity,
                APIParameters::get(opCtx).getAPIStrict().value_or(false));

            // An empty PrivilegeVector is acceptable because these privileges are only checked on
            // getMore and explain will not open a cursor.
            return runAggregate(
                opCtx, nss, viewAggRequest, viewAggregation.getValue(), PrivilegeVector(), result);
        }

        const auto& collection = ctx->getCollection();

        auto executor = uassertStatusOK(
            getExecutorDistinct(&collection, QueryPlannerParams::DEFAULT, &parsedDistinct));

        auto bodyBuilder = result->getBodyBuilder();
        Explain::explainStages(
            executor.get(), collection, verbosity, BSONObj(), cmdObj, &bodyBuilder);
        return Status::OK();
    }

    bool run(OperationContext* opCtx,
             const DatabaseName& dbName,
             const BSONObj& cmdObj,
             BSONObjBuilder& result) override {
        CommandHelpers::handleMarkKillOnClientDisconnect(opCtx);
        // Acquire locks and resolve possible UUID. The RAII object is optional, because in the case
        // of a view, the locks need to be released.
        boost::optional<AutoGetCollectionForReadCommandMaybeLockFree> ctx;
        ctx.emplace(
            opCtx,
            CommandHelpers::parseNsOrUUID(dbName, cmdObj),
            AutoGetCollection::Options{}.viewMode(auto_get_collection::ViewMode::kViewsPermitted));
        const auto& nss = ctx->getNss();

        if (!ctx->getView()) {
            // Distinct doesn't filter orphan documents so it is not allowed to run on sharded
            // collections in multi-document transactions.
            uassert(
                ErrorCodes::OperationNotSupportedInTransaction,
                "Cannot run 'distinct' on a sharded collection in a multi-document transaction. "
                "Please see http://dochub.mongodb.org/core/transaction-distinct for a recommended "
                "alternative.",
                !opCtx->inMultiDocumentTransaction() || !ctx->getCollection().isSharded());

            // Similarly, we ban readConcern level snapshot for sharded collections.
            uassert(
                ErrorCodes::InvalidOptions,
                "Cannot run 'distinct' on a sharded collection with readConcern level 'snapshot'",
                repl::ReadConcernArgs::get(opCtx).getLevel() !=
                        repl::ReadConcernLevel::kSnapshotReadConcern ||
                    !ctx->getCollection().isSharded());
        }

        const ExtensionsCallbackReal extensionsCallback(opCtx, &nss);
        auto defaultCollation =
            ctx->getCollection() ? ctx->getCollection()->getDefaultCollator() : nullptr;
        auto parsedDistinct = uassertStatusOK(
            ParsedDistinct::parse(opCtx, nss, cmdObj, extensionsCallback, false, defaultCollation));

        if (parsedDistinct.isMirrored()) {
            const auto& invocation = CommandInvocation::get(opCtx);
            invocation->markMirrored();
        }

        if (analyze_shard_key::supportsPersistingSampledQueries() && parsedDistinct.getSampleId()) {
            auto cq = parsedDistinct.getQuery();
            analyze_shard_key::QueryAnalysisWriter::get(opCtx)
                .addDistinctQuery(*parsedDistinct.getSampleId(),
                                  nss,
                                  cq->getQueryObj(),
                                  cq->getFindCommandRequest().getCollation())
                .getAsync([](auto) {});
        }

        if (ctx->getView()) {
            // Relinquish locks. The aggregation command will re-acquire them.
            ctx.reset();

            auto viewAggregation = parsedDistinct.asAggregationCommand();
            uassertStatusOK(viewAggregation.getStatus());

            BSONObj aggResult = CommandHelpers::runCommandDirectly(
                opCtx,
                OpMsgRequest::fromDBAndBody(dbName.db(), std::move(viewAggregation.getValue())));
            uassertStatusOK(ViewResponseFormatter(aggResult).appendAsDistinctResponse(&result));
            return true;
        }

        // Check whether we are allowed to read from this node after acquiring our locks.
        auto replCoord = repl::ReplicationCoordinator::get(opCtx);
        uassertStatusOK(replCoord->checkCanServeReadsFor(
            opCtx, nss, ReadPreferenceSetting::get(opCtx).canRunOnSecondary()));

        const auto& collection = ctx->getCollection();

        auto executor =
            getExecutorDistinct(&collection, QueryPlannerParams::DEFAULT, &parsedDistinct);
        uassertStatusOK(executor.getStatus());

        {
            stdx::lock_guard<Client> lk(*opCtx->getClient());
            CurOp::get(opCtx)->setPlanSummary_inlock(
                executor.getValue()->getPlanExplainer().getPlanSummary());
        }

        const auto key = cmdObj.getStringField(ParsedDistinct::kKeyField);

        std::vector<BSONObj> distinctValueHolder;
        BSONElementSet values(executor.getValue()->getCanonicalQuery()->getCollator());

        const int kMaxResponseSize = BSONObjMaxUserSize - 4096;

        try {
            size_t listApproxBytes = 0;
            BSONObj obj;
            while (PlanExecutor::ADVANCED == executor.getValue()->getNext(&obj, nullptr)) {
                // Distinct expands arrays.
                //
                // If our query is covered, each value of the key should be in the index key and
                // available to us without this.  If a collection scan is providing the data, we may
                // have to expand an array.
                BSONElementSet elts;
                dps::extractAllElementsAlongPath(obj, key, elts);

                for (BSONElementSet::iterator it = elts.begin(); it != elts.end(); ++it) {
                    BSONElement elt = *it;
                    if (values.count(elt)) {
                        continue;
                    }

                    // This is an approximate size check which safeguards against use of unbounded
                    // memory by the distinct command. We perform a more precise check at the end of
                    // this method to confirm that the response size is less than 16MB.
                    listApproxBytes += elt.size();
                    uassert(
                        17217, "distinct too big, 16mb cap", listApproxBytes < kMaxResponseSize);

                    auto distinctObj = elt.wrap();
                    values.insert(distinctObj.firstElement());
                    distinctValueHolder.push_back(std::move(distinctObj));
                }
            }
        } catch (DBException& exception) {
            auto&& explainer = executor.getValue()->getPlanExplainer();
            auto&& [stats, _] =
                explainer.getWinningPlanStats(ExplainOptions::Verbosity::kExecStats);
            LOGV2_WARNING(23797,
                          "Plan executor error during distinct command: {error}, "
                          "stats: {stats}, cmd: {cmd}",
                          "Plan executor error during distinct command",
                          "error"_attr = exception.toStatus(),
                          "stats"_attr = redact(stats),
                          "cmd"_attr = cmdObj);

            exception.addContext("Executor error during distinct command");
            throw;
        }

        auto curOp = CurOp::get(opCtx);

        // Get summary information about the plan.
        PlanSummaryStats stats;
        auto&& explainer = executor.getValue()->getPlanExplainer();
        explainer.getSummaryStats(&stats);
        if (collection) {
            CollectionQueryInfo::get(collection).notifyOfQuery(opCtx, collection, stats);
        }
        curOp->debug().setPlanSummaryMetrics(stats);

        if (curOp->shouldDBProfile(opCtx)) {
            auto&& [stats, _] =
                explainer.getWinningPlanStats(ExplainOptions::Verbosity::kExecStats);
            curOp->debug().execStats = std::move(stats);
        }

        BSONArrayBuilder valueListBuilder(result.subarrayStart("values"));
        for (const auto& value : values) {
            valueListBuilder.append(value);
        }
        valueListBuilder.doneFast();

        if (!opCtx->inMultiDocumentTransaction() &&
            repl::ReadConcernArgs::get(opCtx).getArgsAtClusterTime()) {
            result.append("atClusterTime"_sd,
                          repl::ReadConcernArgs::get(opCtx).getArgsAtClusterTime()->asTimestamp());
        }

        uassert(31299, "distinct too big, 16mb cap", result.len() < kMaxResponseSize);
        return true;
    }

    void appendMirrorableRequest(BSONObjBuilder* bob, const BSONObj& cmdObj) const override {
        static const auto kMirrorableKeys = [] {
            BSONObjBuilder keyBob;
            keyBob.append("distinct", 1);
            keyBob.append("key", 1);
            keyBob.append("query", 1);
            keyBob.append("collation", 1);
            keyBob.append("shardVersion", 1);
            return keyBob.obj();
        }();

        // Filter the keys that can be mirrored
        cmdObj.filterFieldsUndotted(bob, kMirrorableKeys, true);
    }


} distinctCmd;

}  // namespace
}  // namespace mongo