datasketches 0.3.1 → 0.4.0

Sign up to get free protection for your applications and to get access to all the features.
Files changed (245) hide show
  1. checksums.yaml +4 -4
  2. data/CHANGELOG.md +10 -0
  3. data/NOTICE +1 -1
  4. data/README.md +0 -2
  5. data/ext/datasketches/cpc_wrapper.cpp +3 -3
  6. data/ext/datasketches/kll_wrapper.cpp +0 -10
  7. data/lib/datasketches/version.rb +1 -1
  8. data/lib/datasketches.rb +1 -1
  9. data/vendor/datasketches-cpp/CMakeLists.txt +23 -20
  10. data/vendor/datasketches-cpp/CODE_OF_CONDUCT.md +3 -0
  11. data/vendor/datasketches-cpp/CONTRIBUTING.md +50 -0
  12. data/vendor/datasketches-cpp/Doxyfile +2827 -0
  13. data/vendor/datasketches-cpp/LICENSE +0 -76
  14. data/vendor/datasketches-cpp/README.md +1 -3
  15. data/vendor/datasketches-cpp/common/CMakeLists.txt +12 -11
  16. data/vendor/datasketches-cpp/common/include/MurmurHash3.h +25 -27
  17. data/vendor/datasketches-cpp/common/include/common_defs.hpp +15 -10
  18. data/vendor/datasketches-cpp/common/include/count_zeros.hpp +11 -2
  19. data/vendor/datasketches-cpp/common/include/kolmogorov_smirnov.hpp +9 -6
  20. data/vendor/datasketches-cpp/common/include/memory_operations.hpp +5 -4
  21. data/vendor/datasketches-cpp/common/include/optional.hpp +148 -0
  22. data/vendor/datasketches-cpp/common/include/quantiles_sorted_view.hpp +95 -2
  23. data/vendor/datasketches-cpp/common/include/quantiles_sorted_view_impl.hpp +1 -1
  24. data/vendor/datasketches-cpp/common/include/serde.hpp +69 -20
  25. data/vendor/datasketches-cpp/common/test/CMakeLists.txt +2 -2
  26. data/vendor/datasketches-cpp/common/test/integration_test.cpp +6 -0
  27. data/vendor/datasketches-cpp/common/test/optional_test.cpp +85 -0
  28. data/vendor/datasketches-cpp/common/test/test_allocator.hpp +14 -14
  29. data/vendor/datasketches-cpp/{python/src/__init__.py → count/CMakeLists.txt} +25 -1
  30. data/vendor/datasketches-cpp/count/include/count_min.hpp +405 -0
  31. data/vendor/datasketches-cpp/count/include/count_min_impl.hpp +497 -0
  32. data/vendor/datasketches-cpp/{MANIFEST.in → count/test/CMakeLists.txt} +23 -20
  33. data/vendor/datasketches-cpp/count/test/count_min_allocation_test.cpp +155 -0
  34. data/vendor/datasketches-cpp/count/test/count_min_test.cpp +303 -0
  35. data/vendor/datasketches-cpp/cpc/include/cpc_common.hpp +14 -20
  36. data/vendor/datasketches-cpp/cpc/include/cpc_compressor.hpp +7 -4
  37. data/vendor/datasketches-cpp/cpc/include/cpc_compressor_impl.hpp +17 -17
  38. data/vendor/datasketches-cpp/cpc/include/cpc_confidence.hpp +3 -3
  39. data/vendor/datasketches-cpp/cpc/include/cpc_sketch.hpp +40 -40
  40. data/vendor/datasketches-cpp/cpc/include/cpc_sketch_impl.hpp +14 -11
  41. data/vendor/datasketches-cpp/cpc/include/cpc_union.hpp +35 -11
  42. data/vendor/datasketches-cpp/cpc/include/cpc_union_impl.hpp +8 -8
  43. data/vendor/datasketches-cpp/cpc/include/cpc_util.hpp +16 -8
  44. data/vendor/datasketches-cpp/cpc/include/u32_table.hpp +3 -2
  45. data/vendor/datasketches-cpp/cpc/include/u32_table_impl.hpp +5 -5
  46. data/vendor/datasketches-cpp/cpc/test/CMakeLists.txt +20 -7
  47. data/vendor/datasketches-cpp/cpc/test/cpc_sketch_deserialize_from_java_test.cpp +60 -0
  48. data/vendor/datasketches-cpp/cpc/test/cpc_sketch_serialize_for_java.cpp +38 -0
  49. data/vendor/datasketches-cpp/cpc/test/cpc_sketch_test.cpp +4 -29
  50. data/vendor/datasketches-cpp/cpc/test/cpc_union_test.cpp +4 -4
  51. data/vendor/datasketches-cpp/{tox.ini → density/CMakeLists.txt} +24 -8
  52. data/vendor/datasketches-cpp/density/include/density_sketch.hpp +256 -0
  53. data/vendor/datasketches-cpp/density/include/density_sketch_impl.hpp +543 -0
  54. data/vendor/datasketches-cpp/{python/datasketches/__init__.py → density/test/CMakeLists.txt} +15 -3
  55. data/vendor/datasketches-cpp/density/test/density_sketch_test.cpp +244 -0
  56. data/vendor/datasketches-cpp/fi/include/frequent_items_sketch.hpp +21 -9
  57. data/vendor/datasketches-cpp/fi/include/frequent_items_sketch_impl.hpp +6 -4
  58. data/vendor/datasketches-cpp/fi/include/reverse_purge_hash_map.hpp +9 -3
  59. data/vendor/datasketches-cpp/fi/test/CMakeLists.txt +14 -1
  60. data/vendor/datasketches-cpp/fi/test/frequent_items_sketch_deserialize_from_java_test.cpp +95 -0
  61. data/vendor/datasketches-cpp/fi/test/frequent_items_sketch_serialize_for_java.cpp +83 -0
  62. data/vendor/datasketches-cpp/fi/test/frequent_items_sketch_test.cpp +3 -42
  63. data/vendor/datasketches-cpp/hll/include/CouponList-internal.hpp +2 -2
  64. data/vendor/datasketches-cpp/hll/include/CouponList.hpp +3 -1
  65. data/vendor/datasketches-cpp/hll/include/Hll4Array-internal.hpp +19 -11
  66. data/vendor/datasketches-cpp/hll/include/Hll4Array.hpp +2 -5
  67. data/vendor/datasketches-cpp/hll/include/Hll6Array-internal.hpp +19 -7
  68. data/vendor/datasketches-cpp/hll/include/Hll6Array.hpp +1 -1
  69. data/vendor/datasketches-cpp/hll/include/Hll8Array-internal.hpp +98 -42
  70. data/vendor/datasketches-cpp/hll/include/Hll8Array.hpp +2 -0
  71. data/vendor/datasketches-cpp/hll/include/HllArray-internal.hpp +94 -61
  72. data/vendor/datasketches-cpp/hll/include/HllArray.hpp +20 -8
  73. data/vendor/datasketches-cpp/hll/include/HllSketch-internal.hpp +4 -4
  74. data/vendor/datasketches-cpp/hll/include/HllSketchImpl.hpp +3 -1
  75. data/vendor/datasketches-cpp/hll/include/HllSketchImplFactory.hpp +3 -21
  76. data/vendor/datasketches-cpp/hll/include/HllUnion-internal.hpp +8 -0
  77. data/vendor/datasketches-cpp/hll/include/HllUtil.hpp +14 -18
  78. data/vendor/datasketches-cpp/hll/include/coupon_iterator-internal.hpp +1 -1
  79. data/vendor/datasketches-cpp/hll/include/coupon_iterator.hpp +8 -2
  80. data/vendor/datasketches-cpp/hll/include/hll.hpp +79 -65
  81. data/vendor/datasketches-cpp/hll/test/CMakeLists.txt +14 -1
  82. data/vendor/datasketches-cpp/hll/test/HllUnionTest.cpp +7 -1
  83. data/vendor/datasketches-cpp/hll/test/ToFromByteArrayTest.cpp +0 -68
  84. data/vendor/datasketches-cpp/hll/test/hll_sketch_deserialize_from_java_test.cpp +69 -0
  85. data/vendor/datasketches-cpp/hll/test/hll_sketch_serialize_for_java.cpp +52 -0
  86. data/vendor/datasketches-cpp/kll/include/kll_helper.hpp +0 -1
  87. data/vendor/datasketches-cpp/kll/include/kll_helper_impl.hpp +2 -2
  88. data/vendor/datasketches-cpp/kll/include/kll_sketch.hpp +79 -53
  89. data/vendor/datasketches-cpp/kll/include/kll_sketch_impl.hpp +61 -132
  90. data/vendor/datasketches-cpp/kll/test/CMakeLists.txt +14 -1
  91. data/vendor/datasketches-cpp/kll/test/kll_sketch_deserialize_from_java_test.cpp +103 -0
  92. data/vendor/datasketches-cpp/kll/test/kll_sketch_serialize_for_java.cpp +62 -0
  93. data/vendor/datasketches-cpp/kll/test/kll_sketch_test.cpp +5 -40
  94. data/vendor/datasketches-cpp/quantiles/include/quantiles_sketch.hpp +76 -54
  95. data/vendor/datasketches-cpp/quantiles/include/quantiles_sketch_impl.hpp +66 -136
  96. data/vendor/datasketches-cpp/quantiles/test/CMakeLists.txt +14 -1
  97. data/vendor/datasketches-cpp/quantiles/test/quantiles_sketch_deserialize_from_java_test.cpp +84 -0
  98. data/vendor/datasketches-cpp/quantiles/test/quantiles_sketch_serialize_for_java.cpp +52 -0
  99. data/vendor/datasketches-cpp/quantiles/test/quantiles_sketch_test.cpp +15 -39
  100. data/vendor/datasketches-cpp/req/include/req_common.hpp +7 -3
  101. data/vendor/datasketches-cpp/req/include/req_compactor_impl.hpp +2 -4
  102. data/vendor/datasketches-cpp/req/include/req_sketch.hpp +105 -26
  103. data/vendor/datasketches-cpp/req/include/req_sketch_impl.hpp +50 -111
  104. data/vendor/datasketches-cpp/req/test/CMakeLists.txt +14 -1
  105. data/vendor/datasketches-cpp/req/test/req_sketch_deserialize_from_java_test.cpp +55 -0
  106. data/vendor/datasketches-cpp/{tuple/include/array_of_doubles_intersection_impl.hpp → req/test/req_sketch_serialize_for_java.cpp} +12 -7
  107. data/vendor/datasketches-cpp/req/test/req_sketch_test.cpp +3 -89
  108. data/vendor/datasketches-cpp/sampling/CMakeLists.txt +4 -0
  109. data/vendor/datasketches-cpp/sampling/include/ebpps_sample.hpp +210 -0
  110. data/vendor/datasketches-cpp/sampling/include/ebpps_sample_impl.hpp +535 -0
  111. data/vendor/datasketches-cpp/sampling/include/ebpps_sketch.hpp +281 -0
  112. data/vendor/datasketches-cpp/sampling/include/ebpps_sketch_impl.hpp +531 -0
  113. data/vendor/datasketches-cpp/sampling/include/var_opt_sketch.hpp +89 -32
  114. data/vendor/datasketches-cpp/sampling/include/var_opt_sketch_impl.hpp +33 -19
  115. data/vendor/datasketches-cpp/sampling/include/var_opt_union.hpp +13 -10
  116. data/vendor/datasketches-cpp/sampling/include/var_opt_union_impl.hpp +23 -19
  117. data/vendor/datasketches-cpp/sampling/test/CMakeLists.txt +55 -8
  118. data/vendor/datasketches-cpp/sampling/test/ebpps_allocation_test.cpp +96 -0
  119. data/vendor/datasketches-cpp/sampling/test/ebpps_sample_test.cpp +137 -0
  120. data/vendor/datasketches-cpp/sampling/test/ebpps_sketch_test.cpp +266 -0
  121. data/vendor/datasketches-cpp/sampling/test/var_opt_sketch_deserialize_from_java_test.cpp +81 -0
  122. data/vendor/datasketches-cpp/sampling/test/var_opt_sketch_serialize_for_java.cpp +54 -0
  123. data/vendor/datasketches-cpp/sampling/test/var_opt_sketch_test.cpp +33 -51
  124. data/vendor/datasketches-cpp/sampling/test/var_opt_union_deserialize_from_java_test.cpp +50 -0
  125. data/vendor/datasketches-cpp/sampling/test/var_opt_union_serialize_for_java.cpp +56 -0
  126. data/vendor/datasketches-cpp/sampling/test/var_opt_union_test.cpp +0 -20
  127. data/vendor/datasketches-cpp/theta/CMakeLists.txt +1 -0
  128. data/vendor/datasketches-cpp/theta/include/bit_packing.hpp +6279 -0
  129. data/vendor/datasketches-cpp/theta/include/bounds_on_ratios_in_sampled_sets.hpp +1 -0
  130. data/vendor/datasketches-cpp/theta/include/bounds_on_ratios_in_theta_sketched_sets.hpp +7 -6
  131. data/vendor/datasketches-cpp/theta/include/compact_theta_sketch_parser.hpp +14 -8
  132. data/vendor/datasketches-cpp/theta/include/compact_theta_sketch_parser_impl.hpp +60 -46
  133. data/vendor/datasketches-cpp/theta/include/theta_a_not_b.hpp +20 -5
  134. data/vendor/datasketches-cpp/theta/include/theta_constants.hpp +10 -4
  135. data/vendor/datasketches-cpp/theta/include/theta_helpers.hpp +3 -1
  136. data/vendor/datasketches-cpp/theta/include/theta_intersection.hpp +13 -5
  137. data/vendor/datasketches-cpp/theta/include/theta_intersection_base_impl.hpp +5 -5
  138. data/vendor/datasketches-cpp/theta/include/theta_intersection_impl.hpp +3 -3
  139. data/vendor/datasketches-cpp/theta/include/theta_jaccard_similarity.hpp +2 -1
  140. data/vendor/datasketches-cpp/theta/include/theta_jaccard_similarity_base.hpp +1 -0
  141. data/vendor/datasketches-cpp/theta/include/theta_set_difference_base_impl.hpp +1 -1
  142. data/vendor/datasketches-cpp/theta/include/theta_sketch.hpp +180 -33
  143. data/vendor/datasketches-cpp/theta/include/theta_sketch_impl.hpp +430 -130
  144. data/vendor/datasketches-cpp/theta/include/theta_union.hpp +17 -10
  145. data/vendor/datasketches-cpp/theta/include/theta_union_base_impl.hpp +10 -10
  146. data/vendor/datasketches-cpp/theta/include/theta_union_impl.hpp +3 -3
  147. data/vendor/datasketches-cpp/theta/include/theta_update_sketch_base.hpp +21 -6
  148. data/vendor/datasketches-cpp/theta/include/theta_update_sketch_base_impl.hpp +13 -3
  149. data/vendor/datasketches-cpp/theta/test/CMakeLists.txt +15 -1
  150. data/vendor/datasketches-cpp/theta/test/bit_packing_test.cpp +80 -0
  151. data/vendor/datasketches-cpp/theta/test/theta_sketch_deserialize_from_java_test.cpp +57 -0
  152. data/vendor/datasketches-cpp/theta/test/theta_sketch_serialize_for_java.cpp +61 -0
  153. data/vendor/datasketches-cpp/theta/test/theta_sketch_test.cpp +39 -188
  154. data/vendor/datasketches-cpp/theta/test/theta_union_test.cpp +25 -0
  155. data/vendor/datasketches-cpp/tuple/CMakeLists.txt +8 -7
  156. data/vendor/datasketches-cpp/tuple/include/array_of_doubles_sketch.hpp +19 -144
  157. data/vendor/datasketches-cpp/tuple/include/{array_of_doubles_a_not_b.hpp → array_tuple_a_not_b.hpp} +24 -16
  158. data/vendor/datasketches-cpp/tuple/include/{array_of_doubles_a_not_b_impl.hpp → array_tuple_a_not_b_impl.hpp} +4 -4
  159. data/vendor/datasketches-cpp/tuple/include/array_tuple_intersection.hpp +65 -0
  160. data/vendor/datasketches-cpp/tuple/include/array_tuple_intersection_impl.hpp +31 -0
  161. data/vendor/datasketches-cpp/tuple/include/array_tuple_sketch.hpp +237 -0
  162. data/vendor/datasketches-cpp/tuple/include/{array_of_doubles_sketch_impl.hpp → array_tuple_sketch_impl.hpp} +40 -41
  163. data/vendor/datasketches-cpp/tuple/include/array_tuple_union.hpp +81 -0
  164. data/vendor/datasketches-cpp/tuple/include/array_tuple_union_impl.hpp +43 -0
  165. data/vendor/datasketches-cpp/tuple/include/tuple_a_not_b.hpp +11 -2
  166. data/vendor/datasketches-cpp/tuple/include/tuple_intersection.hpp +17 -10
  167. data/vendor/datasketches-cpp/tuple/include/tuple_jaccard_similarity.hpp +2 -1
  168. data/vendor/datasketches-cpp/tuple/include/tuple_sketch.hpp +95 -32
  169. data/vendor/datasketches-cpp/tuple/include/tuple_sketch_impl.hpp +2 -1
  170. data/vendor/datasketches-cpp/tuple/include/tuple_union.hpp +19 -11
  171. data/vendor/datasketches-cpp/tuple/test/CMakeLists.txt +16 -1
  172. data/vendor/datasketches-cpp/tuple/test/aod_sketch_deserialize_from_java_test.cpp +76 -0
  173. data/vendor/datasketches-cpp/tuple/test/aod_sketch_serialize_for_java.cpp +62 -0
  174. data/vendor/datasketches-cpp/tuple/test/array_of_doubles_sketch_test.cpp +5 -129
  175. data/vendor/datasketches-cpp/tuple/test/engagement_test.cpp +85 -89
  176. data/vendor/datasketches-cpp/tuple/test/tuple_jaccard_similarity_test.cpp +3 -1
  177. data/vendor/datasketches-cpp/tuple/test/tuple_sketch_deserialize_from_java_test.cpp +47 -0
  178. data/vendor/datasketches-cpp/{python/src/datasketches.cpp → tuple/test/tuple_sketch_serialize_for_java.cpp} +16 -30
  179. data/vendor/datasketches-cpp/tuple/test/tuple_sketch_test.cpp +1 -1
  180. data/vendor/datasketches-cpp/version.cfg.in +1 -1
  181. metadata +61 -79
  182. data/vendor/datasketches-cpp/fi/test/items_sketch_string_from_java.sk +0 -0
  183. data/vendor/datasketches-cpp/fi/test/items_sketch_string_utf8_from_java.sk +0 -0
  184. data/vendor/datasketches-cpp/fi/test/longs_sketch_from_java.sk +0 -0
  185. data/vendor/datasketches-cpp/hll/test/array6_from_java.sk +0 -0
  186. data/vendor/datasketches-cpp/hll/test/compact_array4_from_java.sk +0 -0
  187. data/vendor/datasketches-cpp/hll/test/compact_set_from_java.sk +0 -0
  188. data/vendor/datasketches-cpp/hll/test/list_from_java.sk +0 -0
  189. data/vendor/datasketches-cpp/hll/test/updatable_array4_from_java.sk +0 -0
  190. data/vendor/datasketches-cpp/hll/test/updatable_set_from_java.sk +0 -0
  191. data/vendor/datasketches-cpp/kll/test/kll_sketch_from_java.sk +0 -0
  192. data/vendor/datasketches-cpp/pyproject.toml +0 -23
  193. data/vendor/datasketches-cpp/python/CMakeLists.txt +0 -81
  194. data/vendor/datasketches-cpp/python/README.md +0 -85
  195. data/vendor/datasketches-cpp/python/datasketches/PySerDe.py +0 -104
  196. data/vendor/datasketches-cpp/python/include/py_serde.hpp +0 -113
  197. data/vendor/datasketches-cpp/python/jupyter/CPCSketch.ipynb +0 -345
  198. data/vendor/datasketches-cpp/python/jupyter/FrequentItemsSketch.ipynb +0 -354
  199. data/vendor/datasketches-cpp/python/jupyter/HLLSketch.ipynb +0 -346
  200. data/vendor/datasketches-cpp/python/jupyter/KLLSketch.ipynb +0 -463
  201. data/vendor/datasketches-cpp/python/jupyter/ThetaSketchNotebook.ipynb +0 -403
  202. data/vendor/datasketches-cpp/python/pybind11Path.cmd +0 -21
  203. data/vendor/datasketches-cpp/python/src/cpc_wrapper.cpp +0 -90
  204. data/vendor/datasketches-cpp/python/src/fi_wrapper.cpp +0 -128
  205. data/vendor/datasketches-cpp/python/src/hll_wrapper.cpp +0 -134
  206. data/vendor/datasketches-cpp/python/src/kll_wrapper.cpp +0 -210
  207. data/vendor/datasketches-cpp/python/src/ks_wrapper.cpp +0 -68
  208. data/vendor/datasketches-cpp/python/src/py_serde.cpp +0 -111
  209. data/vendor/datasketches-cpp/python/src/quantiles_wrapper.cpp +0 -204
  210. data/vendor/datasketches-cpp/python/src/req_wrapper.cpp +0 -215
  211. data/vendor/datasketches-cpp/python/src/theta_wrapper.cpp +0 -172
  212. data/vendor/datasketches-cpp/python/src/vector_of_kll.cpp +0 -490
  213. data/vendor/datasketches-cpp/python/src/vo_wrapper.cpp +0 -173
  214. data/vendor/datasketches-cpp/python/tests/__init__.py +0 -16
  215. data/vendor/datasketches-cpp/python/tests/cpc_test.py +0 -64
  216. data/vendor/datasketches-cpp/python/tests/fi_test.py +0 -110
  217. data/vendor/datasketches-cpp/python/tests/hll_test.py +0 -130
  218. data/vendor/datasketches-cpp/python/tests/kll_test.py +0 -125
  219. data/vendor/datasketches-cpp/python/tests/quantiles_test.py +0 -126
  220. data/vendor/datasketches-cpp/python/tests/req_test.py +0 -126
  221. data/vendor/datasketches-cpp/python/tests/theta_test.py +0 -146
  222. data/vendor/datasketches-cpp/python/tests/vector_of_kll_test.py +0 -148
  223. data/vendor/datasketches-cpp/python/tests/vo_test.py +0 -125
  224. data/vendor/datasketches-cpp/req/test/req_float_empty_from_java.sk +0 -0
  225. data/vendor/datasketches-cpp/req/test/req_float_estimation_from_java.sk +0 -0
  226. data/vendor/datasketches-cpp/req/test/req_float_exact_from_java.sk +0 -0
  227. data/vendor/datasketches-cpp/req/test/req_float_raw_items_from_java.sk +0 -0
  228. data/vendor/datasketches-cpp/req/test/req_float_single_item_from_java.sk +0 -0
  229. data/vendor/datasketches-cpp/sampling/test/binaries_from_java.txt +0 -67
  230. data/vendor/datasketches-cpp/sampling/test/varopt_sketch_long_sampling.sk +0 -0
  231. data/vendor/datasketches-cpp/sampling/test/varopt_sketch_string_exact.sk +0 -0
  232. data/vendor/datasketches-cpp/sampling/test/varopt_union_double_sampling.sk +0 -0
  233. data/vendor/datasketches-cpp/setup.py +0 -110
  234. data/vendor/datasketches-cpp/theta/test/theta_compact_empty_from_java.sk +0 -0
  235. data/vendor/datasketches-cpp/theta/test/theta_compact_estimation_from_java.sk +0 -0
  236. data/vendor/datasketches-cpp/theta/test/theta_compact_exact_from_java.sk +0 -0
  237. data/vendor/datasketches-cpp/theta/test/theta_compact_single_item_from_java.sk +0 -0
  238. data/vendor/datasketches-cpp/tuple/include/array_of_doubles_intersection.hpp +0 -52
  239. data/vendor/datasketches-cpp/tuple/include/array_of_doubles_union.hpp +0 -81
  240. data/vendor/datasketches-cpp/tuple/include/array_of_doubles_union_impl.hpp +0 -43
  241. data/vendor/datasketches-cpp/tuple/test/aod_1_compact_empty_from_java.sk +0 -1
  242. data/vendor/datasketches-cpp/tuple/test/aod_1_compact_estimation_from_java.sk +0 -0
  243. data/vendor/datasketches-cpp/tuple/test/aod_1_compact_non_empty_no_entries_from_java.sk +0 -0
  244. data/vendor/datasketches-cpp/tuple/test/aod_2_compact_exact_from_java.sk +0 -0
  245. data/vendor/datasketches-cpp/tuple/test/aod_3_compact_empty_from_java.sk +0 -1
@@ -0,0 +1,210 @@
1
+ /*
2
+ * Licensed to the Apache Software Foundation (ASF) under one
3
+ * or more contributor license agreements. See the NOTICE file
4
+ * distributed with this work for additional information
5
+ * regarding copyright ownership. The ASF licenses this file
6
+ * to you under the Apache License, Version 2.0 (the
7
+ * "License"); you may not use this file except in compliance
8
+ * with the License. You may obtain a copy of the License at
9
+ *
10
+ * http://www.apache.org/licenses/LICENSE-2.0
11
+ *
12
+ * Unless required by applicable law or agreed to in writing,
13
+ * software distributed under the License is distributed on an
14
+ * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
15
+ * KIND, either express or implied. See the License for the
16
+ * specific language governing permissions and limitations
17
+ * under the License.
18
+ */
19
+
20
+ #ifndef _EBPPS_SAMPLE_HPP_
21
+ #define _EBPPS_SAMPLE_HPP_
22
+
23
+ #include "common_defs.hpp"
24
+ #include "optional.hpp"
25
+ #include "serde.hpp"
26
+
27
+ #include <memory>
28
+ #include <vector>
29
+
30
+ namespace datasketches {
31
+
32
+ template<
33
+ typename T,
34
+ typename A = std::allocator<T>
35
+ >
36
+ class ebpps_sample {
37
+ public:
38
+ explicit ebpps_sample(uint32_t k, const A& allocator = A());
39
+
40
+ // constructor used to create a sample to merge one itme
41
+ template<typename TT>
42
+ ebpps_sample(TT&& item, double theta, const A& allocator = A());
43
+
44
+ // for deserialization
45
+ class items_deleter;
46
+ ebpps_sample(std::vector<T, A>&& data, optional<T>&& partial_item, double c, const A& allocator = A());
47
+
48
+ void reset();
49
+ void downsample(double theta);
50
+
51
+ template<typename FwdSample>
52
+ void merge(FwdSample&& other);
53
+
54
+ // standard way to query the sample
55
+ using result_type = std::vector<T, A>;
56
+ result_type get_sample() const;
57
+
58
+ double get_c() const;
59
+
60
+ // intended for internal use
61
+ // returns only full items
62
+ result_type get_full_items() const;
63
+
64
+ // intended for internal use
65
+ // handles only the partial item
66
+ bool has_partial_item() const;
67
+ T get_partial_item() const;
68
+
69
+ string<A> to_string() const;
70
+
71
+ /**
72
+ * @brief Returns the number of items contained in the sample
73
+ * Computes the number of items, full or partial, currently in the sample.
74
+ * The result should match ceiling(c);
75
+ * @return the number of items contained in the sample
76
+ */
77
+ inline uint32_t get_num_retained_items() const;
78
+
79
+ /**
80
+ * Computes size needed to serialize the current state of the sample.
81
+ * This version is for fixed-size arithmetic types (integral and floating point).
82
+ * @param sd instance of a SerDe
83
+ * @return size in bytes needed to serialize the items in this sample
84
+ */
85
+ template<typename TT = T, typename SerDe = serde<T>, typename std::enable_if<std::is_arithmetic<TT>::value, int>::type = 0>
86
+ inline size_t get_serialized_size_bytes(const SerDe& sd = SerDe()) const;
87
+
88
+ /**
89
+ * Computes size needed to serialize the items in the sample.
90
+ * This version is for all other types and can be expensive since every item needs to be looked at.
91
+ * @param sd instance of a SerDe
92
+ * @return size in bytes needed to serialize the items in this sample
93
+ */
94
+ template<typename TT = T, typename SerDe = serde<T>, typename std::enable_if<!std::is_arithmetic<TT>::value, int>::type = 0>
95
+ inline size_t get_serialized_size_bytes(const SerDe& sd = SerDe()) const;
96
+
97
+ // This is a convenience alias for users
98
+ // The type returned by the following serialize method
99
+ using vector_bytes = std::vector<uint8_t, typename std::allocator_traits<A>::template rebind_alloc<uint8_t>>;
100
+
101
+ /**
102
+ * This method serializes the sample as a vector of bytes.
103
+ * @param ptr pointer to start of pre-allocated memory to use
104
+ * @param end_ptr pointer to end of pre-allocated memory to use
105
+ * @param sd instance of a SerDe
106
+ * @return number of bytes written
107
+ */
108
+ template<typename SerDe = serde<T>>
109
+ size_t serialize(uint8_t* ptr, const uint8_t* end_ptr, const SerDe& sd = SerDe()) const;
110
+
111
+ /**
112
+ * This method serializes the sample into a given stream in a binary form
113
+ * @param os output stream
114
+ * @param sd instance of a SerDe
115
+ */
116
+ template<typename SerDe = serde<T>>
117
+ void serialize(std::ostream& os, const SerDe& sd = SerDe()) const;
118
+
119
+ /**
120
+ * This method deserializes a sample from a given array of bytes.
121
+ * @param ptr pointer to the array of bytes
122
+ * @param size the size of the array
123
+ * @param sd instance of a SerDe
124
+ * @param allocator instance of an allocator
125
+ * @return a std::pair with a sample and the number of bytes read
126
+ */
127
+ template<typename SerDe = serde<T>>
128
+ static std::pair<ebpps_sample, size_t> deserialize(const uint8_t* ptr, size_t size, const SerDe& sd = SerDe(), const A& allocator = A());
129
+
130
+ /**
131
+ * This method deserializes a sample from a given stream.
132
+ * @param is input stream
133
+ * @param sd instance of a SerDe
134
+ * @param allocator instance of an allocator
135
+ * @return an instance of a sample
136
+ */
137
+ template<typename SerDe = serde<T>>
138
+ static ebpps_sample deserialize(std::istream& is, const SerDe& sd = SerDe(), const A& allocator = A());
139
+
140
+ class const_iterator;
141
+
142
+ /**
143
+ * Iterator pointing to the first item in the sample.
144
+ * If the sample is empty, the returned iterator must not be dereferenced or incremented
145
+ * @param force_partial forces the inclusion of the partial item, if one exists
146
+ * @return iterator pointing to the first item in the sample
147
+ */
148
+ const_iterator begin() const;
149
+
150
+ /**
151
+ * Iterator pointing to the past-the-end item in the sample.
152
+ * The past-the-end item is the hypothetical item that would follow the last item.
153
+ * It does not point to any item, and must not be dereferenced or incremented.
154
+ * @return iterator pointing to the past-the-end item in the sample
155
+ */
156
+ const_iterator end() const;
157
+
158
+ private:
159
+ A allocator_;
160
+ double c_; // Current sample size, including fractional part
161
+ optional<T> partial_item_; // a sample item corresponding to a partial weight
162
+ std::vector<T, A> data_; // stored sampled items
163
+
164
+ template<typename FwdItem>
165
+ inline void set_partial(FwdItem&& item);
166
+ void swap_with_partial();
167
+ void move_one_to_partial();
168
+ void subsample(uint32_t num_samples);
169
+
170
+ static inline uint32_t random_idx(uint32_t max);
171
+ static inline double next_double();
172
+
173
+ friend class const_iterator;
174
+ };
175
+
176
+ template<typename T, typename A>
177
+ class ebpps_sample<T, A>::const_iterator {
178
+ public:
179
+ using iterator_category = std::input_iterator_tag;
180
+ using value_type = const T&;
181
+ using difference_type = void;
182
+ using pointer = const return_value_holder<value_type>;
183
+ using reference = value_type;
184
+
185
+ const_iterator(const const_iterator& other);
186
+ const_iterator& operator++();
187
+ const_iterator& operator++(int);
188
+ bool operator==(const const_iterator& other) const;
189
+ bool operator!=(const const_iterator& other) const;
190
+ reference operator*() const;
191
+ pointer operator->() const;
192
+
193
+ private:
194
+ static const size_t PARTIAL_IDX = static_cast<size_t>(-1);
195
+
196
+ // default iterator over sample
197
+ const_iterator(const ebpps_sample<T, A>* sample);
198
+
199
+ const ebpps_sample<T, A>* sample_;
200
+ size_t idx_;
201
+ bool use_partial_;
202
+
203
+ friend class ebpps_sample;
204
+ };
205
+
206
+ } // namespace datasketches
207
+
208
+ #include "ebpps_sample_impl.hpp"
209
+
210
+ #endif // _EBPPS_SAMPLE_HPP_