datasketches 0.2.3 → 0.2.6
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/CHANGELOG.md +12 -0
- data/README.md +8 -8
- data/ext/datasketches/kll_wrapper.cpp +7 -3
- data/ext/datasketches/theta_wrapper.cpp +20 -4
- data/lib/datasketches/version.rb +1 -1
- data/vendor/datasketches-cpp/CMakeLists.txt +25 -5
- data/vendor/datasketches-cpp/MANIFEST.in +3 -0
- data/vendor/datasketches-cpp/NOTICE +6 -5
- data/vendor/datasketches-cpp/README.md +76 -9
- data/vendor/datasketches-cpp/cmake/DataSketchesConfig.cmake.in +10 -0
- data/vendor/datasketches-cpp/common/CMakeLists.txt +18 -13
- data/vendor/datasketches-cpp/common/include/binomial_bounds.hpp +1 -0
- data/vendor/datasketches-cpp/common/include/common_defs.hpp +14 -0
- data/vendor/datasketches-cpp/{kll → common}/include/kolmogorov_smirnov.hpp +5 -3
- data/vendor/datasketches-cpp/{kll → common}/include/kolmogorov_smirnov_impl.hpp +13 -16
- data/vendor/datasketches-cpp/common/include/quantile_sketch_sorted_view.hpp +121 -0
- data/vendor/datasketches-cpp/common/include/quantile_sketch_sorted_view_impl.hpp +91 -0
- data/vendor/datasketches-cpp/common/test/test_type.hpp +2 -0
- data/vendor/datasketches-cpp/cpc/CMakeLists.txt +15 -35
- data/vendor/datasketches-cpp/cpc/include/cpc_compressor_impl.hpp +3 -1
- data/vendor/datasketches-cpp/cpc/include/cpc_confidence.hpp +1 -0
- data/vendor/datasketches-cpp/cpc/include/cpc_sketch_impl.hpp +5 -3
- data/vendor/datasketches-cpp/cpc/include/cpc_union_impl.hpp +10 -6
- data/vendor/datasketches-cpp/cpc/test/cpc_sketch_allocation_test.cpp +17 -0
- data/vendor/datasketches-cpp/cpc/test/cpc_sketch_test.cpp +1 -0
- data/vendor/datasketches-cpp/cpc/test/cpc_union_test.cpp +2 -0
- data/vendor/datasketches-cpp/fi/CMakeLists.txt +5 -15
- data/vendor/datasketches-cpp/fi/include/frequent_items_sketch.hpp +37 -5
- data/vendor/datasketches-cpp/fi/include/frequent_items_sketch_impl.hpp +29 -11
- data/vendor/datasketches-cpp/fi/test/frequent_items_sketch_custom_type_test.cpp +2 -1
- data/vendor/datasketches-cpp/fi/test/frequent_items_sketch_test.cpp +1 -0
- data/vendor/datasketches-cpp/hll/CMakeLists.txt +33 -56
- data/vendor/datasketches-cpp/hll/include/AuxHashMap-internal.hpp +2 -0
- data/vendor/datasketches-cpp/hll/include/CompositeInterpolationXTable-internal.hpp +1 -0
- data/vendor/datasketches-cpp/hll/include/CouponHashSet-internal.hpp +2 -2
- data/vendor/datasketches-cpp/hll/include/CouponList-internal.hpp +1 -0
- data/vendor/datasketches-cpp/hll/include/HllSketch-internal.hpp +6 -4
- data/vendor/datasketches-cpp/hll/include/HllSketchImpl-internal.hpp +2 -0
- data/vendor/datasketches-cpp/hll/include/HllSketchImplFactory.hpp +2 -0
- data/vendor/datasketches-cpp/hll/test/AuxHashMapTest.cpp +1 -0
- data/vendor/datasketches-cpp/hll/test/CouponHashSetTest.cpp +1 -0
- data/vendor/datasketches-cpp/hll/test/CouponListTest.cpp +2 -0
- data/vendor/datasketches-cpp/hll/test/HllArrayTest.cpp +1 -0
- data/vendor/datasketches-cpp/hll/test/HllSketchTest.cpp +59 -0
- data/vendor/datasketches-cpp/hll/test/HllUnionTest.cpp +2 -0
- data/vendor/datasketches-cpp/hll/test/TablesTest.cpp +1 -0
- data/vendor/datasketches-cpp/kll/CMakeLists.txt +5 -19
- data/vendor/datasketches-cpp/kll/include/kll_helper.hpp +0 -4
- data/vendor/datasketches-cpp/kll/include/kll_helper_impl.hpp +5 -2
- data/vendor/datasketches-cpp/kll/include/kll_sketch.hpp +108 -41
- data/vendor/datasketches-cpp/kll/include/kll_sketch_impl.hpp +150 -132
- data/vendor/datasketches-cpp/kll/test/kll_sketch_test.cpp +165 -31
- data/vendor/datasketches-cpp/kll/test/kolmogorov_smirnov_test.cpp +1 -1
- data/vendor/datasketches-cpp/pyproject.toml +1 -1
- data/vendor/datasketches-cpp/python/CMakeLists.txt +8 -1
- data/vendor/datasketches-cpp/python/README.md +13 -9
- data/vendor/datasketches-cpp/python/src/datasketches.cpp +4 -0
- data/vendor/datasketches-cpp/python/src/fi_wrapper.cpp +6 -1
- data/vendor/datasketches-cpp/python/src/kll_wrapper.cpp +48 -13
- data/vendor/datasketches-cpp/python/src/ks_wrapper.cpp +68 -0
- data/vendor/datasketches-cpp/python/src/quantiles_wrapper.cpp +240 -0
- data/vendor/datasketches-cpp/python/src/req_wrapper.cpp +9 -2
- data/vendor/datasketches-cpp/python/src/vector_of_kll.cpp +1 -0
- data/vendor/datasketches-cpp/python/tests/kll_test.py +10 -4
- data/vendor/datasketches-cpp/python/tests/quantiles_test.py +126 -0
- data/vendor/datasketches-cpp/quantiles/CMakeLists.txt +42 -0
- data/vendor/datasketches-cpp/quantiles/include/quantiles_sketch.hpp +656 -0
- data/vendor/datasketches-cpp/quantiles/include/quantiles_sketch_impl.hpp +1373 -0
- data/vendor/datasketches-cpp/quantiles/test/CMakeLists.txt +44 -0
- data/vendor/datasketches-cpp/quantiles/test/Qk128_n1000_v0.3.0.sk +0 -0
- data/vendor/datasketches-cpp/quantiles/test/Qk128_n1000_v0.6.0.sk +0 -0
- data/vendor/datasketches-cpp/quantiles/test/Qk128_n1000_v0.8.0.sk +0 -0
- data/vendor/datasketches-cpp/quantiles/test/Qk128_n1000_v0.8.3.sk +0 -0
- data/vendor/datasketches-cpp/quantiles/test/Qk128_n50_v0.3.0.sk +0 -0
- data/vendor/datasketches-cpp/quantiles/test/Qk128_n50_v0.6.0.sk +0 -0
- data/vendor/datasketches-cpp/quantiles/test/Qk128_n50_v0.8.0.sk +0 -0
- data/vendor/datasketches-cpp/quantiles/test/Qk128_n50_v0.8.3.sk +0 -0
- data/vendor/datasketches-cpp/quantiles/test/kolmogorov_smirnov_test.cpp +110 -0
- data/vendor/datasketches-cpp/quantiles/test/quantiles_compatibility_test.cpp +129 -0
- data/vendor/datasketches-cpp/quantiles/test/quantiles_sketch_test.cpp +975 -0
- data/vendor/datasketches-cpp/req/CMakeLists.txt +6 -21
- data/vendor/datasketches-cpp/req/include/req_common.hpp +0 -5
- data/vendor/datasketches-cpp/req/include/req_compactor.hpp +6 -0
- data/vendor/datasketches-cpp/req/include/req_compactor_impl.hpp +30 -2
- data/vendor/datasketches-cpp/req/include/req_sketch.hpp +73 -23
- data/vendor/datasketches-cpp/req/include/req_sketch_impl.hpp +95 -63
- data/vendor/datasketches-cpp/req/test/req_sketch_test.cpp +74 -3
- data/vendor/datasketches-cpp/sampling/CMakeLists.txt +5 -9
- data/vendor/datasketches-cpp/sampling/include/var_opt_sketch.hpp +44 -7
- data/vendor/datasketches-cpp/sampling/include/var_opt_sketch_impl.hpp +44 -33
- data/vendor/datasketches-cpp/sampling/include/var_opt_union.hpp +41 -6
- data/vendor/datasketches-cpp/sampling/include/var_opt_union_impl.hpp +33 -15
- data/vendor/datasketches-cpp/sampling/test/var_opt_allocation_test.cpp +2 -2
- data/vendor/datasketches-cpp/sampling/test/var_opt_sketch_test.cpp +1 -0
- data/vendor/datasketches-cpp/sampling/test/var_opt_union_test.cpp +1 -0
- data/vendor/datasketches-cpp/setup.py +1 -1
- data/vendor/datasketches-cpp/theta/CMakeLists.txt +26 -45
- data/vendor/datasketches-cpp/theta/include/bounds_on_ratios_in_sampled_sets.hpp +1 -0
- data/vendor/datasketches-cpp/theta/include/compact_theta_sketch_parser_impl.hpp +92 -23
- data/vendor/datasketches-cpp/theta/include/theta_helpers.hpp +15 -0
- data/vendor/datasketches-cpp/theta/include/theta_intersection_base_impl.hpp +7 -6
- data/vendor/datasketches-cpp/theta/include/theta_set_difference_base_impl.hpp +3 -2
- data/vendor/datasketches-cpp/theta/include/theta_sketch.hpp +32 -15
- data/vendor/datasketches-cpp/theta/include/theta_sketch_impl.hpp +150 -93
- data/vendor/datasketches-cpp/theta/include/theta_union.hpp +6 -1
- data/vendor/datasketches-cpp/theta/include/theta_union_base.hpp +3 -1
- data/vendor/datasketches-cpp/theta/include/theta_union_base_impl.hpp +9 -2
- data/vendor/datasketches-cpp/theta/include/theta_union_impl.hpp +8 -5
- data/vendor/datasketches-cpp/theta/include/theta_update_sketch_base.hpp +5 -4
- data/vendor/datasketches-cpp/theta/include/theta_update_sketch_base_impl.hpp +34 -9
- data/vendor/datasketches-cpp/theta/test/CMakeLists.txt +1 -0
- data/vendor/datasketches-cpp/theta/test/theta_a_not_b_test.cpp +2 -0
- data/vendor/datasketches-cpp/theta/test/theta_compact_empty_from_java_v1.sk +0 -0
- data/vendor/datasketches-cpp/theta/test/theta_compact_empty_from_java_v2.sk +0 -0
- data/vendor/datasketches-cpp/theta/test/theta_compact_estimation_from_java_v1.sk +0 -0
- data/vendor/datasketches-cpp/theta/test/theta_compact_estimation_from_java_v2.sk +0 -0
- data/vendor/datasketches-cpp/theta/test/theta_compact_exact_from_java.sk +0 -0
- data/vendor/datasketches-cpp/theta/test/theta_intersection_test.cpp +2 -0
- data/vendor/datasketches-cpp/theta/test/theta_setop_test.cpp +446 -0
- data/vendor/datasketches-cpp/theta/test/theta_sketch_test.cpp +429 -1
- data/vendor/datasketches-cpp/theta/test/theta_union_test.cpp +25 -11
- data/vendor/datasketches-cpp/tuple/CMakeLists.txt +18 -33
- data/vendor/datasketches-cpp/tuple/include/array_of_doubles_sketch.hpp +1 -1
- data/vendor/datasketches-cpp/tuple/include/array_of_doubles_sketch_impl.hpp +3 -3
- data/vendor/datasketches-cpp/tuple/include/array_of_doubles_union.hpp +1 -1
- data/vendor/datasketches-cpp/tuple/include/array_of_doubles_union_impl.hpp +3 -3
- data/vendor/datasketches-cpp/tuple/include/tuple_sketch.hpp +29 -9
- data/vendor/datasketches-cpp/tuple/include/tuple_sketch_impl.hpp +34 -14
- data/vendor/datasketches-cpp/tuple/include/tuple_union.hpp +6 -1
- data/vendor/datasketches-cpp/tuple/include/tuple_union_impl.hpp +8 -3
- data/vendor/datasketches-cpp/tuple/test/array_of_doubles_sketch_test.cpp +16 -0
- data/vendor/datasketches-cpp/tuple/test/tuple_a_not_b_test.cpp +1 -0
- data/vendor/datasketches-cpp/tuple/test/tuple_intersection_test.cpp +1 -0
- data/vendor/datasketches-cpp/tuple/test/tuple_sketch_test.cpp +46 -8
- data/vendor/datasketches-cpp/tuple/test/tuple_union_test.cpp +8 -0
- metadata +33 -12
- data/vendor/datasketches-cpp/kll/include/kll_quantile_calculator.hpp +0 -75
- data/vendor/datasketches-cpp/kll/include/kll_quantile_calculator_impl.hpp +0 -184
- data/vendor/datasketches-cpp/req/include/req_quantile_calculator.hpp +0 -69
- data/vendor/datasketches-cpp/req/include/req_quantile_calculator_impl.hpp +0 -60
- data/vendor/datasketches-cpp/theta/test/theta_update_empty_from_java.sk +0 -0
- data/vendor/datasketches-cpp/theta/test/theta_update_estimation_from_java.sk +0 -0
@@ -32,64 +32,41 @@ target_include_directories(hll
|
|
32
32
|
target_link_libraries(hll INTERFACE common)
|
33
33
|
target_compile_features(hll INTERFACE cxx_std_11)
|
34
34
|
|
35
|
-
# TODO: would be useful if this didn't need to be reproduced in target_sources(), too
|
36
|
-
set(hll_HEADERS "")
|
37
|
-
list(APPEND hll_HEADERS "include/hll.hpp;include/AuxHashMap.hpp;include/CompositeInterpolationXTable.hpp")
|
38
|
-
list(APPEND hll_HEADERS "include/hll.private.hpp;include/HllSketchImplFactory.hpp")
|
39
|
-
list(APPEND hll_HEADERS "include/CouponHashSet.hpp;include/CouponList.hpp")
|
40
|
-
list(APPEND hll_HEADERS "include/CubicInterpolation.hpp;include/HarmonicNumbers.hpp;include/Hll4Array.hpp")
|
41
|
-
list(APPEND hll_HEADERS "include/Hll6Array.hpp;include/Hll8Array.hpp;include/HllArray.hpp")
|
42
|
-
list(APPEND hll_HEADERS "include/HllSketchImpl.hpp")
|
43
|
-
list(APPEND hll_HEADERS "include/HllUtil.hpp;include/coupon_iterator.hpp")
|
44
|
-
list(APPEND hll_HEADERS "include/RelativeErrorTables.hpp;include/AuxHashMap-internal.hpp")
|
45
|
-
list(APPEND hll_HEADERS "include/CompositeInterpolationXTable-internal.hpp")
|
46
|
-
list(APPEND hll_HEADERS "include/CouponHashSet-internal.hpp;include/CouponList-internal.hpp")
|
47
|
-
list(APPEND hll_HEADERS "include/CubicInterpolation-internal.hpp;include/HarmonicNumbers-internal.hpp")
|
48
|
-
list(APPEND hll_HEADERS "include/Hll4Array-internal.hpp;include/Hll6Array-internal.hpp")
|
49
|
-
list(APPEND hll_HEADERS "include/Hll8Array-internal.hpp;include/HllArray-internal.hpp")
|
50
|
-
list(APPEND hll_HEADERS "include/HllSketch-internal.hpp")
|
51
|
-
list(APPEND hll_HEADERS "include/HllSketchImpl-internal.hpp;include/HllUnion-internal.hpp")
|
52
|
-
list(APPEND hll_HEADERS "include/coupon_iterator-internal.hpp;include/RelativeErrorTables-internal.hpp")
|
53
|
-
|
54
35
|
install(TARGETS hll
|
55
36
|
EXPORT ${PROJECT_NAME}
|
56
37
|
)
|
57
38
|
|
58
|
-
install(FILES
|
39
|
+
install(FILES
|
40
|
+
include/hll.hpp
|
41
|
+
include/AuxHashMap.hpp
|
42
|
+
include/CompositeInterpolationXTable.hpp
|
43
|
+
include/hll.private.hpp
|
44
|
+
include/HllSketchImplFactory.hpp
|
45
|
+
include/CouponHashSet.hpp
|
46
|
+
include/CouponList.hpp
|
47
|
+
include/CubicInterpolation.hpp
|
48
|
+
include/HarmonicNumbers.hpp
|
49
|
+
include/Hll4Array.hpp
|
50
|
+
include/Hll6Array.hpp
|
51
|
+
include/Hll8Array.hpp
|
52
|
+
include/HllArray.hpp
|
53
|
+
include/HllSketchImpl.hpp
|
54
|
+
include/HllUtil.hpp
|
55
|
+
include/coupon_iterator.hpp
|
56
|
+
include/RelativeErrorTables.hpp
|
57
|
+
include/AuxHashMap-internal.hpp
|
58
|
+
include/CompositeInterpolationXTable-internal.hpp
|
59
|
+
include/CouponHashSet-internal.hpp
|
60
|
+
include/CouponList-internal.hpp
|
61
|
+
include/CubicInterpolation-internal.hpp
|
62
|
+
include/HarmonicNumbers-internal.hpp
|
63
|
+
include/Hll4Array-internal.hpp
|
64
|
+
include/Hll6Array-internal.hpp
|
65
|
+
include/Hll8Array-internal.hpp
|
66
|
+
include/HllArray-internal.hpp
|
67
|
+
include/HllSketch-internal.hpp
|
68
|
+
include/HllSketchImpl-internal.hpp
|
69
|
+
include/HllUnion-internal.hpp
|
70
|
+
include/coupon_iterator-internal.hpp
|
71
|
+
include/RelativeErrorTables-internal.hpp
|
59
72
|
DESTINATION "${CMAKE_INSTALL_INCLUDEDIR}/DataSketches")
|
60
|
-
|
61
|
-
target_sources(hll
|
62
|
-
INTERFACE
|
63
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/hll.hpp
|
64
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/hll.private.hpp
|
65
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/AuxHashMap.hpp
|
66
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/CompositeInterpolationXTable.hpp
|
67
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/CouponHashSet.hpp
|
68
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/CouponList.hpp
|
69
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/CubicInterpolation.hpp
|
70
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/HarmonicNumbers.hpp
|
71
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/Hll4Array.hpp
|
72
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/Hll6Array.hpp
|
73
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/Hll8Array.hpp
|
74
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/HllArray.hpp
|
75
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/HllSketchImpl.hpp
|
76
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/HllSketchImplFactory.hpp
|
77
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/HllUtil.hpp
|
78
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/RelativeErrorTables.hpp
|
79
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/coupon_iterator.hpp
|
80
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/AuxHashMap-internal.hpp
|
81
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/CompositeInterpolationXTable-internal.hpp
|
82
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/CouponHashSet-internal.hpp
|
83
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/CouponList-internal.hpp
|
84
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/CubicInterpolation-internal.hpp
|
85
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/HarmonicNumbers-internal.hpp
|
86
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/Hll4Array-internal.hpp
|
87
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/Hll6Array-internal.hpp
|
88
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/Hll8Array-internal.hpp
|
89
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/HllArray-internal.hpp
|
90
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/HllSketch-internal.hpp
|
91
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/HllSketchImpl-internal.hpp
|
92
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/HllUnion-internal.hpp
|
93
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/RelativeErrorTables-internal.hpp
|
94
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/coupon_iterator-internal.hpp
|
95
|
-
)
|
@@ -24,6 +24,7 @@
|
|
24
24
|
|
25
25
|
#include <cstring>
|
26
26
|
#include <exception>
|
27
|
+
#include <stdexcept>
|
27
28
|
|
28
29
|
namespace datasketches {
|
29
30
|
|
@@ -113,10 +114,9 @@ CouponHashSet<A>* CouponHashSet<A>::newSet(const void* bytes, size_t len, const
|
|
113
114
|
} else {
|
114
115
|
sketch->coupons_.resize(1ULL << lgArrInts);
|
115
116
|
sketch->couponCount_ = couponCount;
|
116
|
-
// only need to read valid coupons, unlike in stream case
|
117
117
|
std::memcpy(sketch->coupons_.data(),
|
118
118
|
data + hll_constants::HASH_SET_INT_ARR_START,
|
119
|
-
|
119
|
+
couponsInArray * sizeof(uint32_t));
|
120
120
|
}
|
121
121
|
|
122
122
|
return sketch;
|
@@ -246,10 +246,12 @@ string<A> hll_sketch_alloc<A>::to_string(const bool summary,
|
|
246
246
|
const bool detail,
|
247
247
|
const bool aux_detail,
|
248
248
|
const bool all) const {
|
249
|
-
|
249
|
+
// Using a temporary stream for implementation here does not comply with AllocatorAwareContainer requirements.
|
250
|
+
// The stream does not support passing an allocator instance, and alternatives are complicated.
|
251
|
+
std::stringstream os;
|
250
252
|
if (summary) {
|
251
253
|
os << "### HLL sketch summary:" << std::endl
|
252
|
-
<< " Log Config K : " << get_lg_config_k() << std::endl
|
254
|
+
<< " Log Config K : " << std::to_string(get_lg_config_k()) << std::endl
|
253
255
|
<< " Hll Target : " << type_as_string() << std::endl
|
254
256
|
<< " Current Mode : " << mode_as_string() << std::endl
|
255
257
|
<< " LB : " << get_lower_bound(1) << std::endl
|
@@ -258,7 +260,7 @@ string<A> hll_sketch_alloc<A>::to_string(const bool summary,
|
|
258
260
|
<< " OutOfOrder flag: " << (is_out_of_order_flag() ? "true" : "false") << std::endl;
|
259
261
|
if (get_current_mode() == HLL) {
|
260
262
|
HllArray<A>* hllArray = (HllArray<A>*) sketch_impl;
|
261
|
-
os << " CurMin : " << hllArray->getCurMin() << std::endl
|
263
|
+
os << " CurMin : " << std::to_string(hllArray->getCurMin()) << std::endl
|
262
264
|
<< " NumAtCurMin : " << hllArray->getNumAtCurMin() << std::endl
|
263
265
|
<< " HipAccum : " << hllArray->getHipAccum() << std::endl
|
264
266
|
<< " KxQ0 : " << hllArray->getKxQ0() << std::endl
|
@@ -338,7 +340,7 @@ string<A> hll_sketch_alloc<A>::to_string(const bool summary,
|
|
338
340
|
}
|
339
341
|
}
|
340
342
|
|
341
|
-
return os.str();
|
343
|
+
return string<A>(os.str().c_str(), sketch_impl->getAllocator());
|
342
344
|
}
|
343
345
|
|
344
346
|
template<typename A>
|
@@ -23,6 +23,7 @@
|
|
23
23
|
#include <cmath>
|
24
24
|
#include <string>
|
25
25
|
#include <exception>
|
26
|
+
#include <stdexcept>
|
26
27
|
|
27
28
|
#include "hll.hpp"
|
28
29
|
#include "CouponList.hpp"
|
@@ -31,6 +32,7 @@
|
|
31
32
|
namespace datasketches {
|
32
33
|
|
33
34
|
void println_string(std::string str) {
|
35
|
+
unused(str);
|
34
36
|
//std::cout << str << std::endl;
|
35
37
|
}
|
36
38
|
|
@@ -17,6 +17,8 @@
|
|
17
17
|
* under the License.
|
18
18
|
*/
|
19
19
|
|
20
|
+
#include <stdexcept>
|
21
|
+
|
20
22
|
#include "hll.hpp"
|
21
23
|
|
22
24
|
#include <catch.hpp>
|
@@ -382,4 +384,61 @@ TEST_CASE("hll sketch: deserialize HLL mode buffer overrun", "[hll_sketch]") {
|
|
382
384
|
REQUIRE(test_allocator_total_bytes == 0);
|
383
385
|
}
|
384
386
|
|
387
|
+
TEST_CASE("hll sketch: bytes serialize-deserialize-serialize list mode") {
|
388
|
+
test_allocator_total_bytes = 0;
|
389
|
+
{
|
390
|
+
hll_sketch_test_alloc s1(10, target_hll_type::HLL_4, false, 0);
|
391
|
+
s1.update(1);
|
392
|
+
s1.update(2);
|
393
|
+
s1.update(3);
|
394
|
+
std::cout << s1.to_string();
|
395
|
+
auto bytes1 = s1.serialize_compact();
|
396
|
+
auto s2 = hll_sketch_test_alloc::deserialize(bytes1.data(), bytes1.size(), 0);
|
397
|
+
auto bytes2 = s2.serialize_compact();
|
398
|
+
REQUIRE(bytes1 == bytes2);
|
399
|
+
}
|
400
|
+
REQUIRE(test_allocator_total_bytes == 0);
|
401
|
+
}
|
402
|
+
|
403
|
+
TEST_CASE("hll sketch: updatable bytes serialize-deserialize-serialize set mode") {
|
404
|
+
test_allocator_total_bytes = 0;
|
405
|
+
{
|
406
|
+
hll_sketch_test_alloc s1(10, target_hll_type::HLL_4, false, 0);
|
407
|
+
for (int i = 0; i < 10; ++i) s1.update(i);
|
408
|
+
std::cout << s1.to_string();
|
409
|
+
auto bytes1 = s1.serialize_updatable();
|
410
|
+
auto s2 = hll_sketch_test_alloc::deserialize(bytes1.data(), bytes1.size(), 0);
|
411
|
+
|
412
|
+
auto bytes2 = s2.serialize_updatable();
|
413
|
+
REQUIRE(bytes1 == bytes2);
|
414
|
+
}
|
415
|
+
REQUIRE(test_allocator_total_bytes == 0);
|
416
|
+
}
|
417
|
+
|
418
|
+
TEST_CASE("hll sketch: compact bytes serialize-deserialize-serialize set mode") {
|
419
|
+
test_allocator_total_bytes = 0;
|
420
|
+
{
|
421
|
+
hll_sketch_test_alloc s1(10, target_hll_type::HLL_4, false, 0);
|
422
|
+
for (int i = 0; i < 10; ++i) s1.update(i);
|
423
|
+
std::cout << s1.to_string();
|
424
|
+
auto bytes1 = s1.serialize_compact();
|
425
|
+
auto s2 = hll_sketch_test_alloc::deserialize(bytes1.data(), bytes1.size(), 0);
|
426
|
+
|
427
|
+
// cannot just compare bytes here
|
428
|
+
// hash set does not preserve the order after reconstruction in compact mode
|
429
|
+
// add more to push them to HLL mode
|
430
|
+
for (int i = 10; i < 100; ++i) {
|
431
|
+
s1.update(i);
|
432
|
+
s2.update(i);
|
433
|
+
}
|
434
|
+
std::cout << s1.to_string();
|
435
|
+
std::cout << s2.to_string();
|
436
|
+
|
437
|
+
auto bytes2 = s1.serialize_compact();
|
438
|
+
auto bytes3 = s2.serialize_compact();
|
439
|
+
REQUIRE(bytes2 == bytes3);
|
440
|
+
}
|
441
|
+
REQUIRE(test_allocator_total_bytes == 0);
|
442
|
+
}
|
443
|
+
|
385
444
|
} /* namespace datasketches */
|
@@ -32,27 +32,13 @@ target_include_directories(kll
|
|
32
32
|
target_link_libraries(kll INTERFACE common)
|
33
33
|
target_compile_features(kll INTERFACE cxx_std_11)
|
34
34
|
|
35
|
-
set(kll_HEADERS "")
|
36
|
-
list(APPEND kll_HEADERS "include/kll_sketch.hpp")
|
37
|
-
list(APPEND kll_HEADERS "include/kll_sketch_impl.hpp")
|
38
|
-
list(APPEND kll_HEADERS "include/kll_helper.hpp")
|
39
|
-
list(APPEND kll_HEADERS "include/kll_helper_impl.hpp")
|
40
|
-
list(APPEND kll_HEADERS "include/kll_quantile_calculator.hpp")
|
41
|
-
list(APPEND kll_HEADERS "include/kll_quantile_calculator_impl.hpp")
|
42
|
-
|
43
35
|
install(TARGETS kll
|
44
36
|
EXPORT ${PROJECT_NAME}
|
45
37
|
)
|
46
38
|
|
47
|
-
install(FILES
|
39
|
+
install(FILES
|
40
|
+
include/kll_sketch.hpp
|
41
|
+
include/kll_sketch_impl.hpp
|
42
|
+
include/kll_helper.hpp
|
43
|
+
include/kll_helper_impl.hpp
|
48
44
|
DESTINATION "${CMAKE_INSTALL_INCLUDEDIR}/DataSketches")
|
49
|
-
|
50
|
-
target_sources(kll
|
51
|
-
INTERFACE
|
52
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/kll_helper.hpp
|
53
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/kll_helper_impl.hpp
|
54
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/kll_sketch.hpp
|
55
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/kll_sketch_impl.hpp
|
56
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/kll_quantile_calculator.hpp
|
57
|
-
${CMAKE_CURRENT_SOURCE_DIR}/include/kll_quantile_calculator_impl.hpp
|
58
|
-
)
|
@@ -22,13 +22,9 @@
|
|
22
22
|
|
23
23
|
#include <random>
|
24
24
|
#include <stdexcept>
|
25
|
-
#include <chrono>
|
26
25
|
|
27
26
|
namespace datasketches {
|
28
27
|
|
29
|
-
static std::independent_bits_engine<std::mt19937, 1, uint32_t>
|
30
|
-
random_bit(static_cast<uint32_t>(std::chrono::system_clock::now().time_since_epoch().count()));
|
31
|
-
|
32
28
|
#ifdef KLL_VALIDATION
|
33
29
|
extern uint32_t kll_next_offset;
|
34
30
|
#endif
|
@@ -21,6 +21,9 @@
|
|
21
21
|
#define KLL_HELPER_IMPL_HPP_
|
22
22
|
|
23
23
|
#include <algorithm>
|
24
|
+
#include <stdexcept>
|
25
|
+
|
26
|
+
#include "common_defs.hpp"
|
24
27
|
|
25
28
|
namespace datasketches {
|
26
29
|
|
@@ -227,7 +230,7 @@ kll_helper::compress_result kll_helper::general_compress(uint16_t k, uint8_t m,
|
|
227
230
|
// move level over as is
|
228
231
|
// make sure we are not moving data upwards
|
229
232
|
if (raw_beg < out_levels[current_level]) throw std::logic_error("wrong move");
|
230
|
-
std::move(
|
233
|
+
std::move(items + raw_beg, items + raw_lim, items + out_levels[current_level]);
|
231
234
|
out_levels[current_level + 1] = out_levels[current_level] + raw_pop;
|
232
235
|
} else {
|
233
236
|
// The sketch is too full AND this level is too full, so we compact it
|
@@ -248,7 +251,7 @@ kll_helper::compress_result kll_helper::general_compress(uint16_t k, uint8_t m,
|
|
248
251
|
|
249
252
|
// level zero might not be sorted, so we must sort it if we wish to compact it
|
250
253
|
if ((current_level == 0) && !is_level_zero_sorted) {
|
251
|
-
std::sort(
|
254
|
+
std::sort(items + adj_beg, items + adj_beg + adj_pop, C());
|
252
255
|
}
|
253
256
|
|
254
257
|
if (pop_above == 0) { // Level above is empty, so halve up
|