Make: Formatting and docs

ashvardanian · Oct 9, 2023 · b0a280d · b0a280d
1 parent eadad4e
commit b0a280d
Show file tree

Hide file tree

Showing 5 changed files with 100 additions and 87 deletions.
diff --git a/.vscode/settings.json b/.vscode/settings.json
@@ -118,7 +118,8 @@
     "strstream": "cpp",
     "filesystem": "cpp",
     "stringzilla.h": "c",
-    "__memory": "c"
+    "__memory": "c",
+    "charconv": "c"
   },
   "C_Cpp.default.configurationProvider": "ms-vscode.cmake-tools",
   "cSpell.words": [

diff --git a/CMakeLists.txt b/CMakeLists.txt
@@ -1,105 +1,109 @@
 # This CMake file is heavily inspired by following `stringzilla` CMake:
 # https://github.com/nlohmann/json/blob/develop/CMakeLists.txt
 cmake_minimum_required(VERSION 3.1)
-project(stringzilla VERSION 0.1.0 LANGUAGES C CXX)
+project(
+  stringzilla
+  VERSION 0.1.0
+  LANGUAGES C CXX)
 
-set (CMAKE_C_STANDARD 11)
-set (CMAKE_CXX_STANDARD 17)
+set(CMAKE_C_STANDARD 11)
+set(CMAKE_CXX_STANDARD 17)
 
-# Determine if USearch is built as a subproject (using `add_subdirectory`) or if it is the main project
+# Determine if USearch is built as a subproject (using `add_subdirectory`) or if
+# it is the main project
 set(STRINGZILLA_IS_MAIN_PROJECT OFF)
-if (CMAKE_CURRENT_SOURCE_DIR STREQUAL CMAKE_SOURCE_DIR)
-    set(STRINGZILLA_IS_MAIN_PROJECT ON)
+if(CMAKE_CURRENT_SOURCE_DIR STREQUAL CMAKE_SOURCE_DIR)
+  set(STRINGZILLA_IS_MAIN_PROJECT ON)
 endif()
 
 # Options
 option(STRINGZILLA_INSTALL "Install CMake targets" OFF)
-option(STRINGZILLA_BUILD_TEST "Compile a native unit test in C++" ${STRINGZILLA_IS_MAIN_PROJECT})
-option(STRINGZILLA_BUILD_BENCHMARK "Compile a native benchmark in C++" ${STRINGZILLA_IS_MAIN_PROJECT})
+option(STRINGZILLA_BUILD_TEST "Compile a native unit test in C++"
+       ${STRINGZILLA_IS_MAIN_PROJECT})
+option(STRINGZILLA_BUILD_BENCHMARK "Compile a native benchmark in C++"
+       ${STRINGZILLA_IS_MAIN_PROJECT})
 option(STRINGZILLA_BUILD_WOLFRAM "Compile Wolfram Language bindings" OFF)
 
 # Includes
 set(CMAKE_MODULE_PATH ${CMAKE_CURRENT_SOURCE_DIR}/cmake ${CMAKE_MODULE_PATH})
 include(ExternalProject)
 
-# Allow CMake 3.13+ to override options when using FetchContent / add_subdirectory
-if (POLICY CMP0077)
-    cmake_policy(SET CMP0077 NEW)
-endif ()
+# Allow CMake 3.13+ to override options when using FetchContent /
+# add_subdirectory
+if(POLICY CMP0077)
+  cmake_policy(SET CMP0077 NEW)
+endif()
 
 # Configuration
 include(GNUInstallDirs)
-set(STRINGZILLA_TARGET_NAME               ${PROJECT_NAME})
-set(STRINGZILLA_CONFIG_INSTALL_DIR        "${CMAKE_INSTALL_DATADIR}/cmake/${PROJECT_NAME}" CACHE INTERNAL "")
-set(STRINGZILLA_INCLUDE_INSTALL_DIR       "${CMAKE_INSTALL_INCLUDEDIR}")
-set(STRINGZILLA_TARGETS_EXPORT_NAME       "${PROJECT_NAME}Targets")
-set(STRINGZILLA_CMAKE_CONFIG_TEMPLATE     "cmake/config.cmake.in")
-set(STRINGZILLA_CMAKE_CONFIG_DIR          "${CMAKE_CURRENT_BINARY_DIR}")
-set(STRINGZILLA_CMAKE_VERSION_CONFIG_FILE "${STRINGZILLA_CMAKE_CONFIG_DIR}/${PROJECT_NAME}ConfigVersion.cmake")
-set(STRINGZILLA_CMAKE_PROJECT_CONFIG_FILE "${STRINGZILLA_CMAKE_CONFIG_DIR}/${PROJECT_NAME}Config.cmake")
-set(STRINGZILLA_CMAKE_PROJECT_TARGETS_FILE "${STRINGZILLA_CMAKE_CONFIG_DIR}/${PROJECT_NAME}Targets.cmake")
-set(STRINGZILLA_PKGCONFIG_INSTALL_DIR      "${CMAKE_INSTALL_DATADIR}/pkgconfig")
-
+set(STRINGZILLA_TARGET_NAME ${PROJECT_NAME})
+set(STRINGZILLA_CONFIG_INSTALL_DIR
+    "${CMAKE_INSTALL_DATADIR}/cmake/${PROJECT_NAME}"
+    CACHE INTERNAL "")
+set(STRINGZILLA_INCLUDE_INSTALL_DIR "${CMAKE_INSTALL_INCLUDEDIR}")
+set(STRINGZILLA_TARGETS_EXPORT_NAME "${PROJECT_NAME}Targets")
+set(STRINGZILLA_CMAKE_CONFIG_TEMPLATE "cmake/config.cmake.in")
+set(STRINGZILLA_CMAKE_CONFIG_DIR "${CMAKE_CURRENT_BINARY_DIR}")
+set(STRINGZILLA_CMAKE_VERSION_CONFIG_FILE
+    "${STRINGZILLA_CMAKE_CONFIG_DIR}/${PROJECT_NAME}ConfigVersion.cmake")
+set(STRINGZILLA_CMAKE_PROJECT_CONFIG_FILE
+    "${STRINGZILLA_CMAKE_CONFIG_DIR}/${PROJECT_NAME}Config.cmake")
+set(STRINGZILLA_CMAKE_PROJECT_TARGETS_FILE
+    "${STRINGZILLA_CMAKE_CONFIG_DIR}/${PROJECT_NAME}Targets.cmake")
+set(STRINGZILLA_PKGCONFIG_INSTALL_DIR "${CMAKE_INSTALL_DATADIR}/pkgconfig")
 
 # Define our header-only library
 add_library(${STRINGZILLA_TARGET_NAME} INTERFACE)
-add_library(${PROJECT_NAME}::${STRINGZILLA_TARGET_NAME} ALIAS ${STRINGZILLA_TARGET_NAME})
+add_library(${PROJECT_NAME}::${STRINGZILLA_TARGET_NAME} ALIAS
+            ${STRINGZILLA_TARGET_NAME})
 set(STRINGZILLA_INCLUDE_BUILD_DIR "${PROJECT_SOURCE_DIR}/include/")
 
 target_compile_definitions(
-    ${STRINGZILLA_TARGET_NAME}
-    INTERFACE
-    $<$<NOT:$<BOOL:${JSON_GlobalUDLs}>>:STRINGZILLA_USE_OPENMP=0>
-)
+  ${STRINGZILLA_TARGET_NAME}
+  INTERFACE $<$<NOT:$<BOOL:${JSON_GlobalUDLs}>>:STRINGZILLA_USE_OPENMP=0>)
 target_include_directories(
-    ${STRINGZILLA_TARGET_NAME}
-    ${STRINGZILLA_SYSTEM_INCLUDE} INTERFACE
-    $<BUILD_INTERFACE:${STRINGZILLA_INCLUDE_BUILD_DIR}>
-    $<INSTALL_INTERFACE:include>
-)
+  ${STRINGZILLA_TARGET_NAME} ${STRINGZILLA_SYSTEM_INCLUDE}
+  INTERFACE $<BUILD_INTERFACE:${STRINGZILLA_INCLUDE_BUILD_DIR}>
+            $<INSTALL_INTERFACE:include>)
 
 if(STRINGZILLA_INSTALL)
-    install(
-        DIRECTORY ${STRINGZILLA_INCLUDE_BUILD_DIR}
-        DESTINATION ${STRINGZILLA_INCLUDE_INSTALL_DIR}
-    )
-    install(
-        FILES ${STRINGZILLA_CMAKE_PROJECT_CONFIG_FILE} ${STRINGZILLA_CMAKE_VERSION_CONFIG_FILE}
-        DESTINATION ${STRINGZILLA_CONFIG_INSTALL_DIR}
-    )
-    export(
-        TARGETS ${STRINGZILLA_TARGET_NAME}
-        NAMESPACE ${PROJECT_NAME}::
-        FILE ${STRINGZILLA_CMAKE_PROJECT_TARGETS_FILE}
-    )
-    install(
-        TARGETS ${STRINGZILLA_TARGET_NAME}
-        EXPORT ${STRINGZILLA_TARGETS_EXPORT_NAME}
-        INCLUDES DESTINATION ${STRINGZILLA_INCLUDE_INSTALL_DIR}
-    )
-    install(
-        EXPORT ${STRINGZILLA_TARGETS_EXPORT_NAME}
-        NAMESPACE ${PROJECT_NAME}::
-        DESTINATION ${STRINGZILLA_CONFIG_INSTALL_DIR}
-    )
-    install(
-        FILES "${CMAKE_CURRENT_BINARY_DIR}/${PROJECT_NAME}.pc"
-        DESTINATION ${STRINGZILLA_PKGCONFIG_INSTALL_DIR}
-    )
+  install(DIRECTORY ${STRINGZILLA_INCLUDE_BUILD_DIR}
+          DESTINATION ${STRINGZILLA_INCLUDE_INSTALL_DIR})
+  install(FILES ${STRINGZILLA_CMAKE_PROJECT_CONFIG_FILE}
+                ${STRINGZILLA_CMAKE_VERSION_CONFIG_FILE}
+          DESTINATION ${STRINGZILLA_CONFIG_INSTALL_DIR})
+  export(
+    TARGETS ${STRINGZILLA_TARGET_NAME}
+    NAMESPACE ${PROJECT_NAME}::
+    FILE ${STRINGZILLA_CMAKE_PROJECT_TARGETS_FILE})
+  install(
+    TARGETS ${STRINGZILLA_TARGET_NAME}
+    EXPORT ${STRINGZILLA_TARGETS_EXPORT_NAME}
+    INCLUDES
+    DESTINATION ${STRINGZILLA_INCLUDE_INSTALL_DIR})
+  install(
+    EXPORT ${STRINGZILLA_TARGETS_EXPORT_NAME}
+    NAMESPACE ${PROJECT_NAME}::
+    DESTINATION ${STRINGZILLA_CONFIG_INSTALL_DIR})
+  install(FILES "${CMAKE_CURRENT_BINARY_DIR}/${PROJECT_NAME}.pc"
+          DESTINATION ${STRINGZILLA_PKGCONFIG_INSTALL_DIR})
 endif()
 
 if(${STRINGZILLA_BUILD_TEST} OR ${STRINGZILLA_BUILD_BENCHMARK})
-  add_executable(stringzilla_test scripts/test.c)
+  add_executable(stringzilla_test scripts/test.cpp)
   set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -g")
-  set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -O3")
+  set(CMAKE_CXX_FLAGS
+      "${CMAKE_CXX_FLAGS} -O3 -flto -march=native -finline-functions -funroll-loops"
+  )
 
   target_include_directories(stringzilla_test PRIVATE stringzilla)
-  set_target_properties(stringzilla_test PROPERTIES RUNTIME_OUTPUT_DIRECTORY ${CMAKE_BINARY_DIR})
+  set_target_properties(stringzilla_test PROPERTIES RUNTIME_OUTPUT_DIRECTORY
+                                                    ${CMAKE_BINARY_DIR})
 
-  if(${CMAKE_VERSION} VERSION_EQUAL 3.13 OR ${CMAKE_VERSION} VERSION_GREATER 3.13)
-      include(CTest)
-      enable_testing()
-      add_test(NAME stringzilla_test COMMAND stringzilla_test)
+  if(${CMAKE_VERSION} VERSION_EQUAL 3.13 OR ${CMAKE_VERSION} VERSION_GREATER
+                                            3.13)
+    include(CTest)
+    enable_testing()
+    add_test(NAME stringzilla_test COMMAND stringzilla_test)
   endif()
 endif()
-
diff --git a/README.md b/README.md
@@ -35,8 +35,8 @@ Coming soon.
 
 ## Quick Start: Python 🐍
 
-1️. Install via pip: `pip install stringzilla`  
-1. Import the classes you need: `from stringzilla import Str, Strs, File`  
+1. Install via pip: `pip install stringzilla`  
+2. Import the classes you need: `from stringzilla import Str, Strs, File`  
 
 ### Basic Usage
 
@@ -115,13 +115,13 @@ There is an ABI-stable C 99 interface, in case you have a database, an operating
 #include "stringzilla.h"
 
 // Initialize your haystack and needle
-sz_haystack_t haystack = {your_text, your_text_length};
-sz_needle_t needle = {your_subtext, your_subtext_length, your_quadgram_offset};
+sz_string_view_t haystack = {your_text, your_text_length};
+sz_string_view_t needle = {your_subtext, your_subtext_length};
 
 // Perform string-level operations
 size_t character_count = sz_count_char(haystack, 'a');
 size_t character_position = sz_find_unigram(haystack, 'a');
-size_t substring_position = sz_find_substr(haystack, needle);
+size_t substring_position = sz_find_substring(haystack, needle);
 
 // Perform collection level operations
 sz_sequence_t array = {your_order, your_count, your_get_start, your_get_length, your_handle};

diff --git a/scripts/bench.ipynb b/scripts/bench.ipynb
@@ -88,7 +88,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "150 ms ± 2.01 ms per loop (mean ± std. dev. of 100 runs, 1 loop each)\n"
+      "152 ms ± 3.24 ms per loop (mean ± std. dev. of 100 runs, 1 loop each)\n"
      ]
     }
    ],
@@ -106,7 +106,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "37.8 ms ± 286 µs per loop (mean ± std. dev. of 100 runs, 1 loop each)\n"
+      "38.1 ms ± 312 µs per loop (mean ± std. dev. of 100 runs, 1 loop each)\n"
      ]
     }
    ],
@@ -124,7 +124,8 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "173 ns ± 23.7 ns per loop (mean ± std. dev. of 1000 runs, 1 loop each)\n"
+      "The slowest run took 7.28 times longer than the fastest. This could mean that an intermediate result is being cached.\n",
+      "186 ns ± 41.1 ns per loop (mean ± std. dev. of 1000 runs, 1 loop each)\n"
      ]
     }
    ],
@@ -142,8 +143,8 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "The slowest run took 82.51 times longer than the fastest. This could mean that an intermediate result is being cached.\n",
-      "94.3 ns ± 108 ns per loop (mean ± std. dev. of 1000 runs, 1 loop each)\n"
+      "The slowest run took 120.95 times longer than the fastest. This could mean that an intermediate result is being cached.\n",
+      "99.6 ns ± 155 ns per loop (mean ± std. dev. of 1000 runs, 1 loop each)\n"
      ]
     }
    ],
@@ -152,6 +153,13 @@
     "sz_str.find(pattern)"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
   {
    "cell_type": "code",
    "execution_count": null,
@@ -176,7 +184,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.13"
+   "version": "3.10.11"
   },
   "orig_nbformat": 4
  },

diff --git a/scripts/test.c b/scripts/test.c
@@ -18,8 +18,8 @@ void populate_random_string(char *buffer, int length, int variability) {
     buffer[length] = '\0';
 }
 
-// Test function for sz_find_substr
-void test_sz_find_substr() {
+// Test function for sz_find_substring
+void test_sz_find_substring() {
     char buffer[MAX_LENGTH + 1];
     char pattern[6]; // Maximum length of 5 + 1 for '\0'
 
@@ -39,19 +39,19 @@ void test_sz_find_substr() {
             needle.length = pattern_length;
 
             // Comparing the result of your function with the standard library function.
-            sz_string_ptr_t result_libc = strstr(buffer, pattern);
-            sz_string_ptr_t result_stringzilla =
-                sz_find_substr(haystack.start, haystack.length, needle.start, needle.length);
+            sz_string_start_t result_libc = strstr(buffer, pattern);
+            sz_string_start_t result_stringzilla =
+                sz_find_substring(haystack.start, haystack.length, needle.start, needle.length);
 
-            assert(((result_libc == NULL) ^ (result_stringzilla == NULL)) && "Test failed for sz_find_substr");
+            assert(((result_libc == NULL) ^ (result_stringzilla == NULL)) && "Test failed for sz_find_substring");
         }
     }
 }
 
 int main() {
     srand((unsigned int)time(NULL));
 
-    test_sz_find_substr();
+    test_sz_find_substring();
     // Add calls to other test functions as you implement them
 
     printf("All tests passed!\n");