data.h File Reference

Program-level representation of streaming data. More...

#include "flowkey.h"
#include "logger.h"
#include "utils.h"
#include <boost/bimap.hpp>
#include <boost/bimap/unordered_set_of.hpp>
#include <boost/bimap/vector_of.hpp>
#include <filesystem>
#include <fmt/core.h>

Include dependency graph for data.h:

This graph shows which files directly or indirectly include this file:

Classes
struct	OmniSketch::Data::Record< key_len >
	Struct of a single record (i.e., a packet in a segment of streaming data) More...
class	OmniSketch::Data::DataFormat
	Parse the format in config file. More...
class	OmniSketch::Data::StreamData< key_len >
	Store the formatted streaming data. More...
class	OmniSketch::Data::GndTruth< key_len, T >
	Ground truth of the streaming data. More...
class	OmniSketch::Data::Estimation< key_len, T >
	Output of sketch as estimation of ground truth. More...

Namespaces
	OmniSketch
	OmniSketch::Data
	Miscellaneous tools for processing data.

Macros
#define	CHECK_CALLED_ONCE
#define	ASSERT_AND_TRUNCATE_MYSELF_TO_THE_FIRST_K_ELEMENTS
#define	ASSERT_AND_TRUNCATE_MYSELF_TO_ELEMENTS_WITH_GIVEN_VALUE

Enumerations
enum	OmniSketch::Data::CntMethod { OmniSketch::Data::InLength, OmniSketch::Data::InPacket }
	Specify the counting method. More...
enum	OmniSketch::Data::HXMethod { OmniSketch::Data::TopK, OmniSketch::Data::Percentile }
	Specify the defining rule of heavy X (X = [hitters|changers]) More...

Detailed Description

Program-level representation of streaming data.

Author

dromniscience (you@d.nosp@m.omai.nosp@m.n.com)

Copyright

Copyright (c) 2022

Macro Definition Documentation

ASSERT_AND_TRUNCATE_MYSELF_TO_ELEMENTS_WITH_GIVEN_VALUE

#define ASSERT_AND_TRUNCATE_MYSELF_TO_ELEMENTS_WITH_GIVEN_VALUE

Value:

  if (!(threshold >= 0.0 && threshold <= 1.0)) {                               \
    throw std::invalid_argument("Invalid Argument: Threshold should be in "    \
                                "[0,1] (Percentile), but got " +               \
                                std::to_string(threshold) + " intsead.");      \
  }                                                                            \
  T thres = threshold * save;                                                  \
  const auto end =                                                             \
      std::lower_bound(my_map.right.begin(), my_map.right.end(), thres,        \
                       [](const RightVal &p, const T &val) {                   \
                         return std::greater<T>()(p.first, val);               \
                       });                                                     \
  my_map.right.erase(end, my_map.right.end());                                 \
  for (auto ptr = my_map.begin(); ptr != my_map.end(); ptr++) {                \
    tot_value += ptr->get_right();                                             \
  }

ASSERT_AND_TRUNCATE_MYSELF_TO_THE_FIRST_K_ELEMENTS

#define ASSERT_AND_TRUNCATE_MYSELF_TO_THE_FIRST_K_ELEMENTS

Value:

  if (threshold < 1.0) {                                                       \
    throw std::invalid_argument(                                               \
        "Invalid Argument: Threshold should >= 1.0 (Top-K), but got " +        \
        std::to_string(threshold) + " intsead.");                              \
  }                                                                            \
  auto size = my_map.size();                                                   \
  size_t no = std::min(size, static_cast<size_t>(threshold));                  \
  my_map.right.erase(my_map.right.begin() + no, my_map.right.end());           \
  for (auto ptr = my_map.begin(); ptr != my_map.end(); ptr++) {                \
    tot_value += ptr->get_right();                                             \
  }

CHECK_CALLED_ONCE

#define CHECK_CALLED_ONCE

Value:

  called++;                                                                    \
  if (called > 1) {                                                            \
    LOG(WARNING,                                                               \
        fmt::format("getXXX(...) is called for the {:d}{} "                    \
                    "time. The instance is left unmodified.",                  \
                    called,                                                    \
                    (called % 10 == 2)                                         \
                        ? "nd"                                                 \
                        : ((called % 10 == 3)                                  \
                               ? "rd"                                          \
                               : ((called % 10 == 1) ? "st" : "th"))));        \
    return;                                                                    \
  }