Add Five Benchmark Time Series Datasets

.github/actions/generate-dataset/action.yml

@@ -21,9 +21,11 @@ runs:
     # 2️⃣  Remove any .venv that might have been created earlier
     #     with Python 3.13, so we always recreate it with 3.12.
     # ─────────────────────────────────────────────────────────────
-    - name: Remove stale virtual-env
+    - name: Recreate virtual-env with the correct Python
       shell: bash
-      run: rm -rf "$GITHUB_WORKSPACE/.venv"
+      run: |
+        rm -rf "$GITHUB_WORKSPACE/.venv"
+        python3 -m venv "$GITHUB_WORKSPACE/.venv"
 
     # ─────────────────────────────────────────────────────────────
     # 3️⃣  Generate the synthetic data

benchmark/bench_compression_ratio.cpp

@@ -43,6 +43,11 @@ BenchmarkCase tpch_case {Tpch::dataset,
                          std::string(FLS_CMAKE_SOURCE_DIR) +
                              "/benchmark/result/compression_ratio/tpch/fastlanes_detailed.csv"};
 
+BenchmarkCase time_series_case {
+    TimeSeries::dataset,
+    std::string(FLS_CMAKE_SOURCE_DIR) + "/benchmark/result/compression_ratio/time_series/fastlanes.csv",
+    std::string(FLS_CMAKE_SOURCE_DIR) + "/benchmark/result/compression_ratio/time_series/fastlanes_detailed.csv"};
+
 void run_compression_ratio_benchmark(const BenchmarkCase& benchmark_case) {
 	const auto& result_file_path          = benchmark_case.result_file_path;
 	const auto  detailed_result_file_path = benchmark_case.detailed_result_file_path;

benchmark/result/compression_ratio/time_series/compression_summary.csv

@@ -0,0 +1,6 @@
+table_name,version,csv_size,fls_size,compression_ratio
+AMPds,0.1.4,2540627,363719,6.99
+Computer_Monitor,0.1.4,917528,57529,15.95
+Linear_Road,0.1.4,1290067,253407,5.09
+Smart_Gird,0.1.4,918847,250929,3.66
+Weather_Forcast,0.1.4,2595630,237159,10.94

benchmark/result/compression_ratio/time_series/fastlanes.csv

@@ -0,0 +1,6 @@
+table_name,version,file_size
+AMPds,0.1.4,363719
+Computer_Monitor,0.1.4,57529
+Linear_Road,0.1.4,253407
+Smart_Gird,0.1.4,250929
+Weather_Forcast,0.1.4,237159

benchmark/result/compression_ratio/time_series/fastlanes_detailed.csv

@@ -0,0 +1,35 @@
+compression,version,table_name,id,name,data_type,size(bytes),expression,bytes_per_value,bits_per_value
+FastLanes,0.1.4,AMPds,0,Column_1,DOUBLE,51665,"{[EXP_DICT_DBL_FFOR_SLPATCH_U08][0, 1, 2, 3, 4, 5, 6]}",0.79,6.31
+FastLanes,0.1.4,AMPds,1,Column_2,DOUBLE,28726,"{[EXP_DICT_DBL_FFOR_SLPATCH_U08][0, 1, 2, 3, 4, 5, 6]}",0.44,3.51
+FastLanes,0.1.4,AMPds,2,Column_3,DOUBLE,16744,"{[EXP_DICT_DBL_FFOR_SLPATCH_U08][0, 1, 2, 3, 4, 5, 6]}",0.26,2.04
+FastLanes,0.1.4,AMPds,3,Column_4,DOUBLE,42400,"{[EXP_DICT_DBL_FFOR_SLPATCH_U08][0, 1, 2, 3, 4, 5, 6]}",0.65,5.18
+FastLanes,0.1.4,AMPds,4,Column_5,DOUBLE,42542,"{[EXP_DICT_DBL_FFOR_SLPATCH_U08][0, 1, 2, 3, 4, 5, 6]}",0.65,5.19
+FastLanes,0.1.4,AMPds,5,Column_6,INT16,52348,"{[EXP_DICT_I16_FFOR_SLPATCH_U08][0, 1, 2, 3, 4, 5, 6]}",0.80,6.39
+FastLanes,0.1.4,AMPds,6,Column_7,INT8,17976,"{[EXP_FFOR_SLPATCH_I08][0, 1, 2, 3, 4, 5]}",0.27,2.19
+FastLanes,0.1.4,AMPds,7,Column_8,INT16,31412,"{[EXP_FFOR_SLPATCH_I16][0, 1, 2, 3, 4, 5]}",0.48,3.83
+FastLanes,0.1.4,AMPds,8,Column_9,INT8,8906,"{[EXP_DICT_I08_FFOR_SLPATCH_U08][0, 1, 2, 3, 4, 5, 6]}",0.14,1.09
+FastLanes,0.1.4,AMPds,9,Column_10,INT16,52976,"{[EXP_DICT_I16_FFOR_SLPATCH_U08][0, 1, 2, 3, 4, 5, 6]}",0.81,6.47
+FastLanes,0.1.4,AMPds,10,Column_11,INT8,17976,"{[EXP_FFOR_SLPATCH_I08][0, 1, 2, 3, 4, 5]}",0.27,2.19
+FastLanes,0.1.4,Computer_Monitor,0,Column_1,INT8,9476,"{[EXP_DICT_I08_FFOR_SLPATCH_U08][0, 1, 2, 3, 4, 5, 6]}",0.14,1.16
+FastLanes,0.1.4,Computer_Monitor,1,Column_2,DOUBLE,23493,"{[EXP_DICT_DBL_FFOR_SLPATCH_U08][0, 1, 2, 3, 4, 5, 6]}",0.36,2.87
+FastLanes,0.1.4,Computer_Monitor,2,Column_3,INT8,15046,"{[EXP_DICT_I08_FFOR_SLPATCH_U08][0, 1, 2, 3, 4, 5, 6]}",0.23,1.84
+FastLanes,0.1.4,Computer_Monitor,3,Column_4,INT8,9466,"{[EXP_DICT_I08_FFOR_SLPATCH_U08][0, 1, 2, 3, 4, 5, 6]}",0.14,1.16
+FastLanes,0.1.4,Linear_Road,0,Column_1,INT16,3748,"{[EXP_CROSS_RLE_I16][0, 1]}",0.06,0.46
+FastLanes,0.1.4,Linear_Road,1,Column_2,INT8,57728,"{[EXP_FFOR_I08][0, 1, 2]}",0.88,7.05
+FastLanes,0.1.4,Linear_Road,2,Column_3,INT8,1467,"{[EXP_FREQUENCY_I08][0, 1, 2, 3]}",0.02,0.18
+FastLanes,0.1.4,Linear_Road,3,Column_4,INT8,21614,"{[EXP_FFOR_SLPATCH_I08][0, 1, 2, 3, 4, 5]}",0.33,2.64
+FastLanes,0.1.4,Linear_Road,4,Column_5,INT8,11104,"{[EXP_DICT_I08_FFOR_SLPATCH_U08][0, 1, 2, 3, 4, 5, 6]}",0.17,1.36
+FastLanes,0.1.4,Linear_Road,5,Column_6,INT32,157698,"{[EXP_FFOR_SLPATCH_I32][0, 1, 2, 3, 4, 5]}",2.41,19.25
+FastLanes,0.1.4,Smart_Gird,0,Column_1,DOUBLE,174196,"{[EXP_DICT_DBL_FFOR_U16][0, 1, 2, 3]}",2.66,21.26
+FastLanes,0.1.4,Smart_Gird,1,Column_2,INT8,8579,"{[EXP_DICT_I08_FFOR_U08][0, 1, 2, 3]}",0.13,1.05
+FastLanes,0.1.4,Smart_Gird,2,Column_3,INT8,33167,"{[EXP_DICT_I08_FFOR_U08][0, 1, 2, 3]}",0.51,4.05
+FastLanes,0.1.4,Smart_Gird,3,Column_4,INT8,27100,"{[EXP_RLE_I08_U16][0, 1, 2, 3, 4]}",0.41,3.31
+FastLanes,0.1.4,Smart_Gird,4,Column_5,INT8,7839,"{[EXP_CROSS_RLE_I08][0, 1]}",0.12,0.96
+FastLanes,0.1.4,Weather_Forcast,0,Column_1,INT8,22,"{[EXP_CROSS_RLE_I08][0, 1]}",0.00,0.00
+FastLanes,0.1.4,Weather_Forcast,1,Column_2,INT64,95312,"{[EXP_RLE_I64_U16][0, 1, 2, 3, 4]}",1.45,11.63
+FastLanes,0.1.4,Weather_Forcast,2,Column_3,INT8,6788,"{[EXP_DICT_I08_FFOR_U08][0, 1, 2, 3]}",0.10,0.83
+FastLanes,0.1.4,Weather_Forcast,3,Column_4,INT8,31633,"{[EXP_DICT_I08_FFOR_U08][0, 1, 2, 3]}",0.48,3.86
+FastLanes,0.1.4,Weather_Forcast,4,Column_5,INT8,38912,"{[EXP_FFOR_I08][0, 1, 2]}",0.59,4.75
+FastLanes,0.1.4,Weather_Forcast,5,Column_6,INT8,37888,"{[EXP_FFOR_I08][0, 1, 2]}",0.58,4.62
+FastLanes,0.1.4,Weather_Forcast,6,Column_7,DOUBLE,12463,"{[EXP_DICT_DBL_FFOR_SLPATCH_U08][0, 1, 2, 3, 4, 5, 6]}",0.19,1.52
+FastLanes,0.1.4,Weather_Forcast,7,Column_7,INT8,14093,"{[EXP_DICT_I08_FFOR_SLPATCH_U08][0, 1, 2, 3, 4, 5, 6]}",0.22,1.72

benchmark/result/compression_time/time_series/fastlanes.csv

@@ -0,0 +1,6 @@
+table_name,version,compression_time_ms,n_repetition
+AMPds,0.1.4,64.601,1
+Computer_Monitor,0.1.4,21.5683,1
+Linear_Road,0.1.4,1232.85,1
+Smart_Gird,0.1.4,32.1722,1
+Weather_Forcast,0.1.4,41.9882,1

benchmark/result/decompression_time/time_series/fastlanes.csv

@@ -0,0 +1,6 @@
+table_name,version,decompression_time_ms,n_repetition
+AMPds,0.1.4,71.3487,1000
+Computer_Monitor,0.1.4,18.7312,1000
+Linear_Road,0.1.4,24.1296,1000
+Smart_Gird,0.1.4,17.0261,1000
+Weather_Forcast,0.1.4,28.9171,1000

data/include/data/TimeSeries.hpp

@@ -0,0 +1,35 @@
+// ────────────────────────────────────────────────────────
+// |                      FastLanes                       |
+// ────────────────────────────────────────────────────────
+// data/include/data/TimeSeries.hpp
+// ────────────────────────────────────────────────────────
+#ifndef DATA_TIMESERIES_HPP
+#define DATA_TIMESERIES_HPP
+
+#include <array>
+#include <string_view>
+
+namespace fastlanes {
+
+using timeseries_dataset_t = std::array<std::pair<std::string_view, std::string_view>, 5>;
+
+// clang-format off
+class TimeSeries {
+public:   
+    static constexpr std::string_view AMPds                             { FASTLANES_DATA_DIR "/time_series/tables/AMPds" };                    
+    static constexpr std::string_view WEATHER_FORCAST                   { FASTLANES_DATA_DIR "/time_series/tables/Weather_Forcast" };           
+    static constexpr std::string_view SMART_GRID                        { FASTLANES_DATA_DIR "/time_series/tables/Smart_Grid" };               
+    static constexpr std::string_view LINEAR_ROEAD                      { FASTLANES_DATA_DIR "/time_series/tables/Linear_Road" };               
+    static constexpr std::string_view COMPUTER_MONITOR                  { FASTLANES_DATA_DIR "/time_series/tables/Computer_Monitor" };  
+    static constexpr timeseries_dataset_t dataset = {{
+        { "AMPds",                           AMPds },
+        { "Weather_Forcast",                 WEATHER_FORCAST },
+        { "Smart_Grid",                      SMART_GRID },
+        { "Linear_Road",                     LINEAR_ROEAD },
+        { "Computer_Monitor",                COMPUTER_MONITOR }
+    }};
+};
+
+} // namespace fastlanes
+
+#endif // DATA_TIMESERIES_HPP

data/include/data/fastlanes_data.hpp

@@ -7,6 +7,7 @@
 #define DATA_FASTLANES_DATA_HPP
 
 #include "data/NextiaJD.hpp"
+#include "data/TimeSeries.hpp"
 #include "data/clickbench.hpp"
 #include "data/embedding.hpp"
 #include "data/example.hpp"

examples/rust_example/Cargo.toml

@@ -7,7 +7,7 @@ edition = "2021"
 
 [dependencies]
 anyhow = "1.0"
-fls-rs = "0.1.3-alpha.7"
+fls-rs = { path = "../../rust" }
 
 [[bin]]
 name = "rust_example"

mk/data.mk

@@ -10,8 +10,8 @@ PROJECT_ROOT := $(abspath $(dir $(lastword $(MAKEFILE_LIST)))/..)
 -include $(PROJECT_ROOT)/mk/python.mk           # non-fatal if missing
 
 VENV   ?= $(PROJECT_ROOT)/.venv
-PYTHON ?= $(VENV)/bin/python3                   # python.exe on Win via python.mk
-PIP    ?= $(PYTHON) -m pip
+PYTHON := $(VENV)/bin/python3                   # python.exe on Win via python.mk
+PIP    := $(PYTHON) -m pip
 
 DATA_DIR    := $(PROJECT_ROOT)/data
 SCRIPTS_DIR := $(PROJECT_ROOT)/scripts

rust/Cargo.toml

@@ -36,7 +36,7 @@ cxx = "1"
 anyhow = "1"
 
 [build-dependencies]
-vergen = { version = "9.0.5", default-features = false, features = ["build"] }
-vergen-gix = { version = "1.0.9", default-features = false, features = ["build"] }
+vergen = { version = "9.1.0", default-features = false, features = ["build"] }
+vergen-gix = { version = "9.1.0", default-features = false, features = ["build"] }
 cmake = "0.1"
 cxx-build = "1"

test/src/dataset_tests/CMakeLists.txt

@@ -5,6 +5,7 @@ add_executable(
         generated_data.cpp
         issues.cpp
         nextiajd_test.cpp
+        timeseries_test.cpp
         public_bi.cpp
         test_fc_bench.cpp
         tpch.cpp

test/src/dataset_tests/timeseries_test.cpp

@@ -0,0 +1,28 @@
+// ────────────────────────────────────────────────────────
+// |                      FastLanes                       |
+// ────────────────────────────────────────────────────────
+// test/src/dataset_tests/timeseries_test.cpp
+// ────────────────────────────────────────────────────────
+#include "data/TimeSeries.hpp"
+#include "fastlanes.hpp"
+#include "fls_tester.hpp"
+
+namespace fastlanes {
+
+#define TIMESERIES_TEST(DATASET_VAR)                                                                                   \
+	TEST_F(FastLanesReaderTester, TimeSeries_##DATASET_VAR) {                                                          \
+		const std::vector<n_t> constant_cols {};                                                                       \
+		const std::vector<n_t> equal_cols {};                                                                          \
+		const std::vector<n_t> one_to_one_mapped_col_indexes {};                                                       \
+		AllTest(TimeSeries::DATASET_VAR, constant_cols, equal_cols, one_to_one_mapped_col_indexes);                    \
+	}
+
+TIMESERIES_TEST(SMART_GRID)
+TIMESERIES_TEST(LINEAR_ROEAD)
+TIMESERIES_TEST(COMPUTER_MONITOR)
+TIMESERIES_TEST(AMPds)
+TIMESERIES_TEST(WEATHER_FORCAST)
+
+#undef TIMESERIES_TEST
+
+} // namespace fastlanes

test/src/quick_fuzz_tests/fuzz_config.json

@@ -1,6 +1,6 @@
 {
   "num_cases": 10,
-  "base_seed": 12,
+  "base_seed": 13,
   "delimiter": "|",
   "min_cols": 1,
   "max_cols": 2,