feat(C++/regexp): add libraries submodules

.github/workflows/benchmark_on_change.yml

@@ -17,6 +17,7 @@ jobs:
       - name: Checkout
         uses: actions/checkout@v4
         with:
+          submodules: 'recursive'
           fetch-depth: 2
 
       - name: Pull Docker image

.gitmodules

@@ -0,0 +1,6 @@
+[submodule "c-plus-plus/lib/regex-boost"]
+	path = c-plus-plus/lib/regex-boost
+	url = https://github.com/boostorg/regex.git
+[submodule "c-plus-plus/lib/regex-hyperscan"]
+	path = c-plus-plus/lib/regex-hyperscan
+	url = https://github.com/intel/hyperscan

Dockerfile

@@ -10,6 +10,7 @@ RUN apt update \
         curl \
         wget \
         libc6 \
+        ragel \
         unzip \
         gnulib \
         gnupg2 \
@@ -24,6 +25,7 @@ RUN apt update \
         build-essential \
         ca-certificates \
         libreadline-dev \
+        libboost-all-dev \
         apt-transport-https \
         software-properties-common
 

benchmark.py

@@ -203,7 +203,7 @@ def writeResultsMarkdown(results):
         conf = loadConfiguration(configurationFilename)
 
         for run in conf['runs']:
-            if run['script']['title'] not in scripts and '*' not in scripts:
+            if run['script']['title'] not in scripts and run['script']['file'] not in scripts and '*' not in scripts:
                 continue
 
             split = run['command']['command'].split()

c-plus-plus/.gitignore

@@ -1,2 +1,8 @@
-*/*
-!**.cpp
+# Ignore everything in any subdirectory
+**/*
+
+# Un-ignore all directories
+!*/
+
+# Un-ignore .cpp files
+!**/*.cpp

c-plus-plus/Makefile

@@ -1,35 +1,39 @@
-CXXFLAGS = --std=c++17 -flto -s
+CXXFLAGS += -Wall -fPIC -std=c++17 -O3
+INCLUDES += -Ilib/regex-boost/include -Ilib/regex-hyperscan/src -Ilib/regex-hyperscan/bin -Ilib/regex-hyperscan/bin/lib/libhs.a
+OBJECTS += 
+LDFLAGS += -lstdc++ -lpthread
 
-SCRIPTS = $(wildcard **/*.cpp)
+SCRIPTS = $(wildcard ./src/*/*.cpp)
 
 BINARIES =  $(SCRIPTS:%.cpp=%-clang) \
 			$(SCRIPTS:%.cpp=%-clang-static) \
 			$(SCRIPTS:%.cpp=%-gcc) \
 			$(SCRIPTS:%.cpp=%-gcc-static) \
-			# $(SCRIPTS:%.cpp=%-graalvm) \
-			# $(SCRIPTS:%.cpp=%-graalvm-static) \
-
 
 %-clang: %.cpp
-	clang++ -O3 $(CXXFLAGS) -o $@ $<
+	clang++ $(CXXFLAGS) $(INCLUDES) -o $@ $< $(LDFLAGS)
 
 %-clang-static: %.cpp
-	clang++ -O3 $(CXXFLAGS) -static -o $@ $<
+	clang++ $(CXXFLAGS) $(INCLUDES) -static -o $@ $< $(LDFLAGS)
 
 %-gcc : %.cpp
-	g++ -O3 $(CXXFLAGS) -o $@ $<
+	g++ $(CXXFLAGS) $(INCLUDES) -o $@ $< $(LDFLAGS)
 
 %-gcc-static : %.cpp
-	g++ -O3 $(CXXFLAGS) -static -o $@ $<
-
-%-graalvm : %.cpp
-	$${GRAALVM_LLVM_TOOLCHAIN}/clang++ -O3 $(CXXFLAGS) -o $@ $<
+	g++ $(CXXFLAGS) $(INCLUDES) -static -o $@ $< $(LDFLAGS)
 
-%-graalvm-static : %.cpp
-	$${GRAALVM_LLVM_TOOLCHAIN}/clang++ -O3 $(CXXFLAGS) -static -o $@ $<
+all: libraries $(BINARIES)
 
-all: $(BINARIES)
+libraries:
+	cd ./lib/regex-boost && mkdir -p bin && cd bin && cmake -DCMAKE_POSITION_INDEPENDENT_CODE=ON .. && cmake --build .;
+	cd ./lib/regex-hyperscan && mkdir -p bin && cd bin && cmake -DCMAKE_POSITION_INDEPENDENT_CODE=ON .. && cmake --build .;
 
-.PHONY: clean
 clean:
-	rm -f $(BINARIES)
+	rm -f $(BINARIES)
+	cd ./lib/regex-boost && rm -rf bin;
+	cd ./lib/regex-hyperscan && rm -rf bin;
+
+list:
+	@echo $(BINARIES)
+
+.PHONY: clean list all libraries

c-plus-plus/benchmark.yml

@@ -15,16 +15,23 @@ strategy:
       #   command: 'graalvm.lli %s-graalvm'
 
     files:
-      - primes/Simple
-      - primes/Atkin
+      - title: 'primes/Simple'
+        file: 'src/primes/Simple'
 
-      - collatz/MaxSequence
+      - title: 'primes/Atkin'
+        file: 'src/primes/Atkin'
 
-      - mandelbrot/Simple
+      - title: 'collatz/MaxSequence'
+        file: 'src/collatz/MaxSequence'
 
-      - treap/Naive
+      - title: 'mandelbrot/Simple'
+        file: 'src/mandelbrot/Simple'
 
-      - recursion/Tak
+      - title: 'treap/Naive'
+        file: 'src/treap/Naive'
+
+      - title: 'recursion/Tak'
+        file: 'src/recursion/Tak'
 
     exclude:
       - command: 'gcc'

c-plus-plus/src/regexp/boost.cpp

@@ -0,0 +1,62 @@
+#include <iostream>
+#include <string>
+#include <chrono>
+#include <vector>
+#include <fstream>
+#include <filesystem>
+#include <boost/regex.hpp>
+
+const size_t N_REGEX_ITERATIONS = 1000;
+const size_t N_CHUNKS_ITERATIONS = 10;
+
+const std::vector<boost::regex> regexps = {
+    boost::regex("[Hh]ello [Ww]orld[!]?"),
+    boost::regex("[Hh]ello [Ww]orld[!]?$"),
+    boost::regex("[0-9][0-9][0-9][0-9][0-9]"),
+    boost::regex("[0-9][0-9][0-9][0-9][0-9]$"),
+    boost::regex("[CV]V[a-z]askjdvc[a-z0-9A-Z]"),
+    boost::regex("^[0-9][0-9]\\.[0-9][0-9]\\.[0-9][0-9]\\.[0-9][0-9]\\s[a-zA-Z]"),
+    boost::regex("^z.*$"),
+    boost::regex("^z.z.z.z.z.z.z.z.z.z.z.z.z.z*$"),
+    boost::regex("^[z][z][z][z][z][z][z][z][z][z][z][z][z][z][z]\\S*$"),
+    boost::regex("^[az]\\S\\D*$"),
+};
+
+const std::vector<std::string> chunks = {
+    "aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa",
+    "10.10.10.10 cjdhjhsdclkjhasl dflaskijd flkzsjd vlkszJ cvlsJKHD CVlaskjdvcl sdjvl ksDJv lkj lkj lzkjsdf lkj lzskdjv lkzj  66678",
+    "hello world !",
+    "bar!234ahem.. 'hello world !' ..c !hello worldzzz !zzzzzzzzzzzzzzzzzzz ksajf 874r hbsdfk i7r kjasdhf ikasuwhfia7234 kwaejhfkawehf7234h zzzzzzzzzzzzzzzzzzzzzzzzz dsssssssssssssssssssssssssssssssssssssssss aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa ssssssssssssssssssssssssssssssssssssssss ddddddddddddddddddddddddddddddddddddddddddddddddd ffffffffffffffffffffffffffffffffffffffffffff ggggggggggggggggggggggggggggggggggggggggggggggggggg hhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh jjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjj kkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkk llllllllllllllllllllllllllllllllllllllllllllllllllll qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee rrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr ttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttt yyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyy uuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo pppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppiauwdhfl,asdnfliehjrsgoirtjoersahjrouh waei7yi7q34f kwejfg hello world",
+    "zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz",
+};
+
+int main(int argc, char *argv[])
+{
+    const auto start_time = std::chrono::high_resolution_clock::now();
+
+    for (size_t i = 0; i < N_REGEX_ITERATIONS; i++)
+    {
+        for (const auto &re : regexps)
+        {
+            for (size_t j = 0; j < N_CHUNKS_ITERATIONS; j++)
+            {
+                for (auto chunk : chunks)
+                {
+                    for (boost::smatch sm; boost::regex_search(chunk, sm, re);)
+                    {
+                        // std::cout << sm.str() << std::endl;
+                        chunk = sm.suffix().str();
+                    }
+                }
+            }
+        }
+    }
+
+
+    const auto end_time = std::chrono::high_resolution_clock::now();
+    const auto duration = std::chrono::duration_cast<std::chrono::milliseconds>(end_time - start_time).count();
+
+    std::cout << "Execution time: " << duration << "ms" << std::endl;
+
+    return 0;
+}

c-plus-plus/src/regexp/hyperscan.cpp

@@ -0,0 +1,95 @@
+#include <iostream>
+#include <string>
+#include <chrono>
+#include <vector>
+#include <fstream>
+#include <filesystem>
+#include "hs.h"
+#include "database.h"
+#include "scratch.h"
+
+const size_t N_REGEX_ITERATIONS = 1000;
+const size_t N_CHUNKS_ITERATIONS = 10;
+
+const std::vector<std::string> rules = {
+    std::string("[Hh]ello [Ww]orld[!]?"),
+    std::string("[Hh]ello [Ww]orld[!]?$"),
+    std::string("[0-9][0-9][0-9][0-9][0-9]"),
+    std::string("[0-9][0-9][0-9][0-9][0-9]$"),
+    std::string("[CV]V[a-z]askjdvc[a-z0-9A-Z]"),
+    std::string("^[0-9][0-9]\\.[0-9][0-9]\\.[0-9][0-9]\\.[0-9][0-9]\\s[a-zA-Z]"),
+    std::string("^z.*$"),
+    std::string("^z.z.z.z.z.z.z.z.z.z.z.z.z.z*$"),
+    std::string("^[z][z][z][z][z][z][z][z][z][z][z][z][z][z][z]\\S*$"),
+    std::string("^[az]\\S\\D*$"),
+};
+
+std::vector<std::tuple<hs_database_t*, hs_scratch_t*>> regexps;
+
+const std::vector<std::string> chunks = {
+    "aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa",
+    "10.10.10.10 cjdhjhsdclkjhasl dflaskijd flkzsjd vlkszJ cvlsJKHD CVlaskjdvcl sdjvl ksDJv lkj lkj lzkjsdf lkj lzskdjv lkzj  66678",
+    "hello world !",
+    "bar!234ahem.. 'hello world !' ..c !hello worldzzz !zzzzzzzzzzzzzzzzzzz ksajf 874r hbsdfk i7r kjasdhf ikasuwhfia7234 kwaejhfkawehf7234h zzzzzzzzzzzzzzzzzzzzzzzzz dsssssssssssssssssssssssssssssssssssssssss aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa ssssssssssssssssssssssssssssssssssssssss ddddddddddddddddddddddddddddddddddddddddddddddddd ffffffffffffffffffffffffffffffffffffffffffff ggggggggggggggggggggggggggggggggggggggggggggggggggg hhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh jjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjj kkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkk llllllllllllllllllllllllllllllllllllllllllllllllllll qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee rrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr ttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttt yyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyy uuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo pppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppiauwdhfl,asdnfliehjrsgoirtjoersahjrouh waei7yi7q34f kwejfg hello world",
+    "zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz",
+};
+
+int main(int argc, char *argv[])
+{
+    regexps.resize(rules.size());
+
+    for (size_t i = 0; i < rules.size(); i++)
+    {
+        hs_compile_error_t *compile_err;
+        hs_database_t *db;
+        hs_scratch_t *scratch;
+
+        hs_error_t err = hs_compile(rules[i].c_str(), HS_FLAG_DOTALL | HS_FLAG_SINGLEMATCH, HS_MODE_BLOCK, NULL, &db, &compile_err);
+        if (err != HS_SUCCESS)
+        {
+            std::cerr << "ERROR: Unable to compile pattern \"" << rules[i] << "\": " << compile_err->message << std::endl;
+            hs_free_compile_error(compile_err);
+            return -1;
+        }
+
+        err = hs_alloc_scratch(db, &scratch);
+        if (err != HS_SUCCESS)
+        {
+            std::cerr << "ERROR: Unable to allocate scratch space. Exiting." << std::endl;
+            hs_free_database(db);
+            return -1;
+        }
+
+        regexps[i] = std::make_tuple(db, scratch);
+    }
+
+    const auto start_time = std::chrono::high_resolution_clock::now();
+
+    for (size_t i = 0; i < N_REGEX_ITERATIONS; i++)
+    {
+        for (const auto &re : regexps)
+        {
+            for (size_t j = 0; j < N_CHUNKS_ITERATIONS; j++)
+            {
+                for (auto chunk : chunks)
+                {
+                    hs_error_t err = hs_scan(static_cast<hs_database_t*>(std::get<0>(re)), chunk.c_str(), chunk.size(), 0, static_cast<hs_scratch_t*>(std::get<1>(re)), [](unsigned int id, unsigned long long from, unsigned long long to, unsigned int flags, void *ctx) -> int {
+                        return 0;
+                    }, nullptr);
+                    if (err != HS_SUCCESS)
+                    {
+                        std::cerr << "ERROR: Unable to scan chunk. Exiting." << std::endl;
+                        return -1;
+                    }
+                }
+            }
+        }
+    }
+
+    const auto end_time = std::chrono::high_resolution_clock::now();
+    const auto duration = std::chrono::duration_cast<std::chrono::milliseconds>(end_time - start_time).count();
+
+    std::cout << "Execution time: " << duration << "ms" << std::endl;
+
+    return 0;
+}

c-plus-plus/src/regexp/stdlib.cpp

@@ -0,0 +1,63 @@
+#include <iostream>
+#include <string>
+#include <chrono>
+#include <vector>
+#include <regex>
+#include <fstream>
+#include <filesystem>
+
+const size_t N_REGEX_ITERATIONS = 1000;
+const size_t N_CHUNKS_ITERATIONS = 10;
+
+// array of regexs
+const std::vector<std::regex> regexps = {
+    std::regex("[Hh]ello [Ww]orld[!]?"),
+    std::regex("[Hh]ello [Ww]orld[!]?$"),
+    std::regex("[0-9][0-9][0-9][0-9][0-9]"),
+    std::regex("[0-9][0-9][0-9][0-9][0-9]$"),
+    std::regex("[CV]V[a-z]askjdvc[a-z0-9A-Z]"),
+    std::regex("^[0-9][0-9]\\.[0-9][0-9]\\.[0-9][0-9]\\.[0-9][0-9]\\s[a-zA-Z]"),
+    std::regex("^z.*$"),
+    std::regex("^z.z.z.z.z.z.z.z.z.z.z.z.z.z*$"),
+    std::regex("^[z][z][z][z][z][z][z][z][z][z][z][z][z][z][z]\\S*$"),
+    std::regex("^[az]\\S\\D*$"),
+};
+
+const std::vector<std::string> chunks = {
+    "aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa",
+    "10.10.10.10 cjdhjhsdclkjhasl dflaskijd flkzsjd vlkszJ cvlsJKHD CVlaskjdvcl sdjvl ksDJv lkj lkj lzkjsdf lkj lzskdjv lkzj  66678",
+    "hello world !",
+    "bar!234ahem.. 'hello world !' ..c !hello worldzzz !zzzzzzzzzzzzzzzzzzz ksajf 874r hbsdfk i7r kjasdhf ikasuwhfia7234 kwaejhfkawehf7234h zzzzzzzzzzzzzzzzzzzzzzzzz dsssssssssssssssssssssssssssssssssssssssss aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa ssssssssssssssssssssssssssssssssssssssss ddddddddddddddddddddddddddddddddddddddddddddddddd ffffffffffffffffffffffffffffffffffffffffffff ggggggggggggggggggggggggggggggggggggggggggggggggggg hhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh jjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjj kkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkk llllllllllllllllllllllllllllllllllllllllllllllllllll qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee rrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr ttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttt yyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyy uuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo pppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppiauwdhfl,asdnfliehjrsgoirtjoersahjrouh waei7yi7q34f kwejfg hello world",
+    "zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz",
+};
+
+int main(int argc, char *argv[])
+{
+    const auto start_time = std::chrono::high_resolution_clock::now();
+
+    for (size_t i = 0; i < N_REGEX_ITERATIONS; i++)
+    {
+        for (const auto &re : regexps)
+        {
+            for (size_t j = 0; j < N_CHUNKS_ITERATIONS; j++)
+            {
+                for (auto chunk : chunks)
+                {
+                    for (std::smatch sm; std::regex_search(chunk, sm, re);)
+                    {
+                        // std::cout << sm.str() << std::endl;
+                        chunk = sm.suffix().str();
+                    }
+                }
+            }
+        }
+    }
+
+
+    const auto end_time = std::chrono::high_resolution_clock::now();
+    const auto duration = std::chrono::duration_cast<std::chrono::milliseconds>(end_time - start_time).count();
+
+    std::cout << "Execution time: " << duration << "ms" << std::endl;
+
+    return 0;
+}

run_on_changes.py

@@ -8,13 +8,13 @@
 # Parsing changed files to find unique language-script pairs
 benchmark_commands = set()
 for file_path in changed_files:
-    parts = file_path.split('/')
+    parts = file_path.split('/', 2)
     if len(parts) < 3:
         continue  # Skip invalid paths
     language, category, script = parts[0], parts[1], parts[2]
     script_name = script.split('.')[0]  # Extract script name without extension
     # Assuming the benchmark script supports language and script parameters
-    command = f"python3 ./benchmark.py --languages {language} --scripts {category}/{script_name} --times 1"
+    command = f"python3 ./benchmark.py run --languages {language} --scripts {category}/{script_name} --times 1"
     benchmark_commands.add(command)
 
 # Executing benchmark commands