Schedule.h

Go to the documentation of this file.

#ifndef FWCore_Framework_Schedule_h
#define FWCore_Framework_Schedule_h

/*
  Author: Jim Kowalkowski  28-01-06

  A class for creating a schedule based on paths in the configuration file.
  The schedule is maintained as a sequence of paths.
  After construction, events can be fed to the object and passed through
  all the modules in the schedule.  All accounting about processing
  of events by modules and paths is contained here or in object held
  by containment.

  The trigger results producer and product are generated and managed here.
  This class also manages endpaths and calls to endjob and beginjob.
  Endpaths are just treated as a simple list of modules that need to
  do processing of the event and do not participate in trigger path
  activities.

  This class requires the high-level process pset.  It uses @process_name.
  If the high-level pset contains an "options" pset, then the
  following optional parameter can be present:
  bool wantSummary = true/false   # default false

  wantSummary indicates whether or not the pass/fail/error stats
  for modules and paths should be printed at the end-of-job.

  A TriggerResults object will always be inserted into the event
  for any schedule.  The producer of the TriggerResults EDProduct
  is always the first module in the endpath.  The TriggerResultInserter
  is given a fixed label of "TriggerResults".

  Processing of an event happens by pushing the event through the Paths.
  The scheduler performs the reset() on each of the workers independent
  of the Path objects.

  ------------------------

  About Paths:
  Paths fit into two categories:
  1) trigger paths that contribute directly to saved trigger bits
  2) end paths
  The Schedule holds these paths in two data structures:
  1) main path list
  2) end path list

  Trigger path processing always precedes endpath processing.
  The order of the paths from the input configuration is
  preserved in the main paths list.

  ------------------------

  The Schedule uses the TriggerNamesService to get the names of the
  trigger paths and end paths. When a TriggerResults object is created
  the results are stored in the same order as the trigger names from
  TriggerNamesService.

*/

#include "DataFormats/Provenance/interface/ModuleDescription.h"
#include "FWCore/Framework/interface/ExceptionActions.h"
#include "FWCore/Framework/interface/EventPrincipal.h"
#include "FWCore/Framework/interface/ExceptionHelpers.h"
#include "FWCore/Framework/interface/Frameworkfwd.h"
#include "FWCore/Framework/interface/OccurrenceTraits.h"
#include "FWCore/Framework/interface/WorkerManager.h"
#include "FWCore/Framework/src/Worker.h"
#include "FWCore/Framework/src/WorkerRegistry.h"
#include "FWCore/Framework/src/GlobalSchedule.h"
#include "FWCore/Framework/src/StreamSchedule.h"
#include "FWCore/Framework/src/PreallocationConfiguration.h"
#include "FWCore/MessageLogger/interface/ExceptionMessages.h"
#include "FWCore/MessageLogger/interface/JobReport.h"
#include "FWCore/MessageLogger/interface/MessageLogger.h"
#include "FWCore/ServiceRegistry/interface/Service.h"
#include "FWCore/Utilities/interface/Algorithms.h"
#include "FWCore/Utilities/interface/BranchType.h"
#include "FWCore/Utilities/interface/ConvertException.h"
#include "FWCore/Utilities/interface/Exception.h"
#include "FWCore/Utilities/interface/StreamID.h"

#include "boost/shared_ptr.hpp"

#include <map>
#include <memory>
#include <set>
#include <string>
#include <vector>
#include <sstream>

namespace edm {

  namespace service {
    class TriggerNamesService;
  }
  class ActivityRegistry;
  class BranchIDListHelper;
  class EventSetup;
  class ExceptionCollector;
  class OutputModuleCommunicator;
  class ProcessContext;
  class PreallocationConfiguration;
  class StreamSchedule;
  class GlobalSchedule;
  class TriggerTimingReport;
  class ModuleRegistry;
  class TriggerResultInserter;
  
  class Schedule {
  public:
    typedef std::vector<std::string> vstring;
    typedef boost::shared_ptr<Worker> WorkerPtr;
    typedef std::vector<Worker*> AllWorkers;
    typedef std::vector<boost::shared_ptr<OutputModuleCommunicator>> AllOutputModuleCommunicators;

    typedef std::vector<Worker*> Workers;

    Schedule(ParameterSet& proc_pset,
             service::TriggerNamesService& tns,
             ProductRegistry& pregistry,
             BranchIDListHelper& branchIDListHelper,
             ExceptionToActionTable const& actions,
             boost::shared_ptr<ActivityRegistry> areg,
             boost::shared_ptr<ProcessConfiguration> processConfiguration,
             const ParameterSet* subProcPSet,
             PreallocationConfiguration const& config,
             ProcessContext const* processContext);

    template <typename T>
    void processOneEvent(unsigned int iStreamID,
                         typename T::MyPrincipal& principal,
                         EventSetup const& eventSetup,
                         bool cleaningUpAfterException = false);

    template <typename T>
    void processOneGlobal(typename T::MyPrincipal& principal,
                          EventSetup const& eventSetup,
                          bool cleaningUpAfterException = false);

    template <typename T>
    void processOneStream(unsigned int iStreamID,
                          typename T::MyPrincipal& principal,
                          EventSetup const& eventSetup,
                          bool cleaningUpAfterException = false);

    void beginJob(ProductRegistry const&);
    void endJob(ExceptionCollector & collector);
    
    void beginStream(unsigned int);
    void endStream(unsigned int);

    // Write the luminosity block
    void writeLumi(LuminosityBlockPrincipal const& lbp, ProcessContext const*);

    // Write the run
    void writeRun(RunPrincipal const& rp, ProcessContext const*);

    // Call closeFile() on all OutputModules.
    void closeOutputFiles();

    // Call openNewFileIfNeeded() on all OutputModules
    void openNewOutputFilesIfNeeded();

    // Call openFiles() on all OutputModules
    void openOutputFiles(FileBlock& fb);

    // Call respondToOpenInputFile() on all Modules
    void respondToOpenInputFile(FileBlock const& fb);

    // Call respondToCloseInputFile() on all Modules
    void respondToCloseInputFile(FileBlock const& fb);

    // Call shouldWeCloseFile() on all OutputModules.
    bool shouldWeCloseOutput() const;

    void preForkReleaseResources();
    void postForkReacquireResources(unsigned int iChildIndex, unsigned int iNumberOfChildren);


    std::vector<ModuleDescription const*> getAllModuleDescriptions() const;

    void availablePaths(std::vector<std::string>& oLabelsToFill) const;

    void modulesInPath(std::string const& iPathLabel,
                       std::vector<std::string>& oLabelsToFill) const;

    int totalEvents() const;

    int totalEventsPassed() const;

    int totalEventsFailed() const;

    void enableEndPaths(bool active);

    bool endPathsEnabled() const;

    void getTriggerReport(TriggerReport& rep) const;
    
    void getTriggerTimingReport(TriggerTimingReport& rep) const;

    bool terminate() const;

    void clearCounters();

    bool changeModule(std::string const& iLabel, ParameterSet const& iPSet);

    AllWorkers const& allWorkers() const;

  private:

    void limitOutput(ParameterSet const& proc_pset, BranchIDLists const& branchIDLists);

    std::shared_ptr<TriggerResultInserter> resultsInserter_;
    boost::shared_ptr<ModuleRegistry> moduleRegistry_;
    std::vector<std::shared_ptr<StreamSchedule>> streamSchedules_;
    //In the future, we will have one GlobalSchedule per simultaneous transition
    std::unique_ptr<GlobalSchedule> globalSchedule_;

    AllOutputModuleCommunicators         all_output_communicators_;
    PreallocationConfiguration           preallocConfig_;


    bool                           wantSummary_;

    volatile bool           endpathsAreActive_;
  };


  template <typename T>
  void Schedule::processOneEvent(unsigned int iStreamID,
                                 typename T::MyPrincipal& ep,
                                 EventSetup const& es,
                                 bool cleaningUpAfterException) {
    assert(iStreamID<streamSchedules_.size());
    streamSchedules_[iStreamID]->processOneEvent<T>(ep,es,cleaningUpAfterException);
  }

  template <typename T>
  void Schedule::processOneStream(unsigned int iStreamID,
                                  typename T::MyPrincipal& ep,
                                  EventSetup const& es,
                                  bool cleaningUpAfterException) {
    assert(iStreamID<streamSchedules_.size());
    streamSchedules_[iStreamID]->processOneStream<T>(ep,es,cleaningUpAfterException);
  }
  template <typename T>
  void
  Schedule::processOneGlobal(typename T::MyPrincipal& ep,
                                 EventSetup const& es,
                                 bool cleaningUpAfterException) {
    globalSchedule_->processOneGlobal<T>(ep,es,cleaningUpAfterException);
  }
}
#endif