StrIdx 0.1.0 → 0.1.2

data/rubyext/ruby_interf.cpp

@@ -36,12 +36,12 @@ VALUE StringIndexAddSegments(VALUE self, VALUE str, VALUE fileId) {
 
   void *data;
   TypedData_Get_Struct(self, int, &str_idx_type, data);
-  ((StringIndex *)data)->addStrToIndex(s1, fid, '/');
+  ((StringIndex *)data)->addStrToIndex(s1, fid);
 
   return self;
 }
 
-VALUE StringIndexFind(VALUE self, VALUE str, VALUE minChars) {
+VALUE StringIndexFind(VALUE self, VALUE str) {
   VALUE ret;
   std::string s1 = StringValueCStr(str);
 
@@ -50,7 +50,7 @@ VALUE StringIndexFind(VALUE self, VALUE str, VALUE minChars) {
   StringIndex *idx = (StringIndex *)data;
 
   ret = rb_ary_new();
-  const std::vector<std::pair<float, int>> &results = idx->findSimilar(s1, NUM2INT(minChars));
+  const std::vector<std::pair<float, int>> &results = idx->findSimilar(s1, 2);
   int limit = 15;
   int i = 0;
   for (const auto &res : results) {
@@ -66,14 +66,35 @@ VALUE StringIndexFind(VALUE self, VALUE str, VALUE minChars) {
   return ret;
 }
 
+VALUE StringIndexSetDirSeparator(VALUE self, VALUE str) {
+  char c = '/';
+  if (TYPE(str) == T_STRING) {
+    std::string s = StringValueCStr(str);
+    if (s.size() >= 1) {
+      c = s[0];
+    }
+  } else {
+    c = '\0'; // No separator
+    // if (TYPE(obj) == T_NIL)
+  }
+
+  void *data;
+  TypedData_Get_Struct(self, int, &str_idx_type, data);
+  StringIndex *idx = (StringIndex *)data;
+  idx->setDirSeparator(c);
+
+  return self;
+}
+
 void Init_stridx(void) {
 
-  VALUE cFoo = rb_define_class("CppStringIndex", rb_cObject);
+  VALUE mStrIdx = rb_define_module("StrIdx");
+  VALUE classStringIndex = rb_define_class_under(mStrIdx, "StringIndex", rb_cObject);
 
-  rb_define_alloc_func(cFoo, str_idx_alloc);
-  rb_define_method(cFoo, "add", StringIndexAddSegments, 2);
-  rb_define_method(cFoo, "find", StringIndexFind, 2);
+  rb_define_alloc_func(classStringIndex, str_idx_alloc);
+  rb_define_method(classStringIndex, "add", StringIndexAddSegments, 2);
+  rb_define_method(classStringIndex, "find", StringIndexFind, 1);
+  rb_define_method(classStringIndex, "setDirSeparator", StringIndexSetDirSeparator, 1);
 }
 
 } // End extern "C"
-

data/stridx.hpp

@@ -214,7 +214,7 @@ public:
   }
 
   /**
-   * Add a string to the index to be search for afterwards
+   * Add a string to the index to be searched for afterwards
    *
    * @param filePath String to index (e.g. /home/user/Project/main.cpp).
    * @param fileId Unique identifier for filePath. Will be return as result from findSimilar.
@@ -272,7 +272,7 @@ public:
   }
 
   /**
-  * The search will find filepaths similar to the input string
+  The search will find filepaths similar to the input string
 
   To be considered a candidate path, the file component of the path (e.g. file.txt)
   is required to have at least a substring of two characters in common with the
@@ -286,8 +286,8 @@ public:
       is also included in the PathSegment
     - take the lenght of that substring as score
   sum up the scores for each character c and divide by (string length)^2
-  
-  For example, if query = "rngnomadriv" 
+
+  For example, if query = "rngnomadriv"
   and candidate is "./drivers/char/hw_random/nomadik-rng.c", then scores are calculated
   as follows:
     rngnomadriv
@@ -296,13 +296,17 @@ public:
     score1=(3+3+3+5+5+5+5+5+(4+4+4)*0.7)
 
     In final score, give a small penalty for larger candidate filenames:
-    Divide main part of score with (query string length)^2 
+    Divide main part of score with (query string length)^2
     and minor part by (query string length)*(candidate string length)
     score = score1/(11*11)*0.97 + score1/(11*38)*0.03 = 0.342944
 
   @param query String to search for inside the index
   */
 
+  std::vector<std::pair<float, int>> findSimilar(std::string query) {
+    return findSimilar(query, 2);
+  }
+
   std::vector<std::pair<float, int>> findSimilar(std::string query, int minChars) {
     CandMap fileCandMap;
     CandMap dirCandMap;

data/{rubyext/test.rb → test.rb}

@@ -1,34 +1,31 @@
 #!/usr/bin/env ruby
 
-$:.unshift File.dirname(__FILE__)
 require "stridx"
-idx = CppStringIndex.new
+idx = StrIdx::StringIndex.new
 
 t = Time.new
-fn = File.expand_path("../flist.txt")
+fn = File.expand_path("flist.txt")
 lines = IO.read(fn).lines.collect { |x| x.strip }
-i = 1
+i = 0
 for x in lines
   idx.add(x, i)
   i += 1
 end
 
 idx_time = Time.new 
-puts "\nIndexing time: #{idx_time - t}"
-query = "helbind.h"
-res = idx.find(query, 2)
+puts "\nIndexing time (#{lines.size} files}): #{(idx_time - t).round(4)} seconds"
+
+query = "rngnomadriv"
+res = idx.find(query)
 puts "query: #{query}"
 puts "\nResults:"
 puts "Filename, score"
 puts "==============="
-for x in res
-  fn = lines[x[0] - 1]
-  score = x[1]
+for id, score in res
+  fn = lines[id]
   puts "#{fn}, #{score.round(4)}"
-  # pp [lines[x[0] - 1], x[1]]
 end
 
-
 query_time = Time.new 
 
-puts "\nSearch time: #{query_time - idx_time}"
+puts "\nSearch time: #{(query_time - idx_time).round(4)} seconds"

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: StrIdx
 version: !ruby/object:Gem::Version
-  version: 0.1.0
+  version: 0.1.2
 platform: ruby
 authors:
 - Sami Sieranoja
-autorequire: 
+autorequire:
 bindir: exe
 cert_chain: []
-date: 2024-05-05 00:00:00.000000000 Z
+date: 2024-05-08 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -50,16 +50,19 @@ files:
 - Makefile
 - README.md
 - demo.cpp
+- flist.txt
 - rubyext/extconf.rb
 - rubyext/ruby_interf.cpp
-- rubyext/test.rb
 - stridx.hpp
+- test.rb
 - unordered_dense.h
 homepage: https://github.com/SamiSieranoja/stridx
 licenses:
 - LGPL-2.0+
-metadata: {}
-post_install_message: 
+metadata:
+  source_code_uri: https://github.com/SamiSieranoja/stridx
+  homepage_uri: https://github.com/SamiSieranoja/stridx
+post_install_message:
 rdoc_options: []
 require_paths:
 - lib
@@ -76,7 +79,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubygems_version: 3.3.26
-signing_key: 
+signing_key:
 specification_version: 4
 summary: StrIdx
 test_files: []