taxamatch_rb 0.7.4 → 0.7.5

data/Gemfile

@@ -0,0 +1,20 @@
+source "http://rubygems.org"
+require 'yaml'
+YAML::ENGINE.yamler= 'syck'
+
+gem "biodiversity","~> 0.5.13"
+gem "biodiversity19","~> 0.5.13"
+
+
+group :development do
+  gem "rake-compiler"
+  gem "rspec", "~> 2.3.0"
+  gem "cucumber", ">= 0"
+  gem "bundler", "~> 1.0.0"
+  gem "jeweler", "~> 1.6.0"
+  gem "rcov", ">= 0"
+  gem "ruby-debug19"
+  gem "ruby-prof"
+  gem "shoulda"
+  gem "mocha"
+end

data/Makefile

@@ -0,0 +1,157 @@
+
+SHELL = /bin/sh
+
+#### Start of system configuration section. ####
+
+srcdir = /Users/dimus/.rvm/gems/ruby-1.9.2-p0/bin
+topdir = /Users/dimus/.rvm/rubies/ruby-1.9.2-p0/include/ruby-1.9.1
+hdrdir = /Users/dimus/.rvm/rubies/ruby-1.9.2-p0/include/ruby-1.9.1
+arch_hdrdir = /Users/dimus/.rvm/rubies/ruby-1.9.2-p0/include/ruby-1.9.1/$(arch)
+VPATH = $(srcdir):$(arch_hdrdir)/ruby:$(hdrdir)/ruby
+prefix = $(DESTDIR)/Users/dimus/.rvm/rubies/ruby-1.9.2-p0
+rubylibprefix = $(libdir)/$(RUBY_BASE_NAME)
+exec_prefix = $(prefix)
+vendorhdrdir = $(rubyhdrdir)/vendor_ruby
+sitehdrdir = $(rubyhdrdir)/site_ruby
+rubyhdrdir = $(includedir)/$(RUBY_BASE_NAME)-$(ruby_version)
+vendordir = $(rubylibprefix)/vendor_ruby
+sitedir = $(rubylibprefix)/site_ruby
+ridir = $(datarootdir)/$(RI_BASE_NAME)
+mandir = $(datarootdir)/man
+localedir = $(datarootdir)/locale
+libdir = $(exec_prefix)/lib
+psdir = $(docdir)
+pdfdir = $(docdir)
+dvidir = $(docdir)
+htmldir = $(docdir)
+infodir = $(datarootdir)/info
+docdir = $(datarootdir)/doc/$(PACKAGE)
+oldincludedir = $(DESTDIR)/usr/include
+includedir = $(prefix)/include
+localstatedir = $(prefix)/var
+sharedstatedir = $(prefix)/com
+sysconfdir = $(prefix)/etc
+datadir = $(datarootdir)
+datarootdir = $(prefix)/share
+libexecdir = $(exec_prefix)/libexec
+sbindir = $(exec_prefix)/sbin
+bindir = $(exec_prefix)/bin
+rubylibdir = $(rubylibprefix)/$(ruby_version)
+archdir = $(rubylibdir)/$(arch)
+sitelibdir = $(sitedir)/$(ruby_version)
+sitearchdir = $(sitelibdir)/$(sitearch)
+vendorlibdir = $(vendordir)/$(ruby_version)
+vendorarchdir = $(vendorlibdir)/$(sitearch)
+
+CC = gcc
+CXX = g++
+LIBRUBY = $(LIBRUBY_SO)
+LIBRUBY_A = lib$(RUBY_SO_NAME)-static.a
+LIBRUBYARG_SHARED = -l$(RUBY_SO_NAME)
+LIBRUBYARG_STATIC = -l$(RUBY_SO_NAME)-static
+OUTFLAG = -o 
+COUTFLAG = -o 
+
+RUBY_EXTCONF_H = 
+cflags   =  $(optflags) $(debugflags) $(warnflags)
+optflags = -O3
+debugflags = -ggdb
+warnflags = -Wextra -Wno-unused-parameter -Wno-parentheses -Wpointer-arith -Wwrite-strings -Wno-missing-field-initializers -Wshorten-64-to-32 -Wno-long-long
+CFLAGS   = -fno-common $(cflags)  -fno-common -pipe 
+INCFLAGS = -I. -I$(arch_hdrdir) -I$(hdrdir)/ruby/backward -I$(hdrdir) -I$(srcdir)
+DEFS     = 
+CPPFLAGS =  -D_XOPEN_SOURCE -D_DARWIN_C_SOURCE $(DEFS) $(cppflags)
+CXXFLAGS = $(CFLAGS) $(cxxflags)
+ldflags  = -L. 
+dldflags = -Wl,-undefined,dynamic_lookup -Wl,-multiply_defined,suppress -Wl,-flat_namespace
+ARCH_FLAG = 
+DLDFLAGS = $(ldflags) $(dldflags)
+LDSHARED = $(CC) -dynamic -bundle
+LDSHAREDXX = $(CXX) -dynamic -bundle
+AR = ar
+EXEEXT = 
+
+RUBY_BASE_NAME = ruby
+RUBY_INSTALL_NAME = ruby
+RUBY_SO_NAME = ruby.1.9.1
+arch = x86_64-darwin10.3.1
+sitearch = $(arch)
+ruby_version = 1.9.1
+ruby = /Users/dimus/.rvm/rubies/ruby-1.9.2-p0/bin/ruby
+RUBY = $(ruby)
+RM = rm -f
+RM_RF = $(RUBY) -run -e rm -- -rf
+RMDIRS = $(RUBY) -run -e rmdir -- -p
+MAKEDIRS = mkdir -p
+INSTALL = /usr/bin/install -c
+INSTALL_PROG = $(INSTALL) -m 0755
+INSTALL_DATA = $(INSTALL) -m 644
+COPY = cp
+
+#### End of system configuration section. ####
+
+preload = 
+
+libpath = . $(libdir)
+LIBPATH =  -L. -L$(libdir)
+DEFFILE = 
+
+CLEANFILES = mkmf.log
+DISTCLEANFILES = 
+DISTCLEANDIRS = 
+
+extout = 
+extout_prefix = 
+target_prefix = 
+LOCAL_LIBS = 
+LIBS = $(LIBRUBYARG_SHARED)  -lpthread -ldl -lobjc 
+SRCS = 
+OBJS = 
+TARGET = 
+DLLIB = 
+EXTSTATIC = 
+STATIC_LIB = 
+
+BINDIR        = $(bindir)
+RUBYCOMMONDIR = $(sitedir)$(target_prefix)
+RUBYLIBDIR    = $(sitelibdir)$(target_prefix)
+RUBYARCHDIR   = $(sitearchdir)$(target_prefix)
+HDRDIR        = $(rubyhdrdir)/ruby$(target_prefix)
+ARCHHDRDIR    = $(rubyhdrdir)/$(arch)/ruby$(target_prefix)
+
+TARGET_SO     = $(DLLIB)
+CLEANLIBS     = $(TARGET).bundle 
+CLEANOBJS     = *.o  *.bak
+
+all:    Makefile
+static: $(STATIC_LIB)
+.PHONY: all install static install-so install-rb
+.PHONY: clean clean-so clean-rb
+
+clean-rb-default::
+clean-rb::
+clean-so::
+clean: clean-so clean-rb-default clean-rb
+		@-$(RM) $(CLEANLIBS) $(CLEANOBJS) $(CLEANFILES)
+
+distclean-rb-default::
+distclean-rb::
+distclean-so::
+distclean: clean distclean-so distclean-rb-default distclean-rb
+		@-$(RM) Makefile $(RUBY_EXTCONF_H) conftest.* mkmf.log
+		@-$(RM) core ruby$(EXEEXT) *~ $(DISTCLEANFILES)
+		@-$(RMDIRS) $(DISTCLEANDIRS)
+
+realclean: distclean
+install: install-so install-rb
+
+install-so: Makefile
+install-rb: pre-install-rb install-rb-default
+install-rb-default: pre-install-rb-default
+pre-install-rb: Makefile
+pre-install-rb-default: Makefile
+
+site-install: site-install-so site-install-rb
+site-install-so: install-so
+site-install-rb: install-rb
+

data/Rakefile

@@ -0,0 +1,52 @@
+require 'rubygems'
+
+require 'bundler'
+begin
+  Bundler.setup(:default, :development)
+rescue Bundler::BundlerError => e
+  $stderr.puts e.message
+  $stderr.puts "Run `bundle install` to install missing gems"
+  exit e.status_code
+end
+
+require 'rake'
+require 'rake/extensiontask'
+
+begin
+  require 'jeweler'
+  Jeweler::Tasks.new do |gem|
+    gem.name = "taxamatch_rb"
+    gem.summary = 'Implementation of Tony Rees Taxamatch algorithms'
+    gem.description = 'This gem implements algorithm for fuzzy matching scientific names developed by Tony Rees'
+    gem.email = "dmozzherin@eol.org"
+    gem.homepage = "http://github.com/GlobalNamesArchitecture/taxamatch_rb"
+    gem.authors = ["Dmitry Mozzherin"]
+    gem.files = FileList["[A-Z]*", "*.gemspec", "{bin,generators,lib,spec}/**/*"]
+    gem.files -= FileList['lib/**/*.bundle', 'lib/**/*.dll', 'lib/**/*.so']
+    gem.files += FileList['ext/**/*.c']
+    gem.extensions = FileList['ext/**/extconf.rb']
+    # gem is a Gem::Specification... see http://www.rubygems.org/read/chapter/20 for additional settings
+  end
+
+rescue LoadError
+  puts "Jeweler (or a dependency) not available. Install it with: sudo gem install jeweler"
+end
+
+require 'rspec/core'
+require 'rspec/core/rake_task'
+RSpec::Core::RakeTask.new(:spec) do |spec|
+  spec.pattern = FileList['spec/**/*_spec.rb']
+end
+
+RSpec::Core::RakeTask.new(:rcov) do |spec|
+  spec.pattern = 'spec/**/*_spec.rb'
+  spec.rcov = true
+end
+
+Rake::ExtensionTask.new("damerau_levenshtein") do |extension|
+    extension.lib_dir = "lib"
+end
+
+Rake::Task[:spec].prerequisites << :compile
+
+task :default => :spec

data/VERSION

@@ -0,0 +1 @@
+0.7.5

data/ext/damerau_levenshtein/damerau_levenshtein.c

@@ -0,0 +1,112 @@
+#include "ruby.h"
+
+VALUE DamerauLevenshtein = Qnil;
+
+void Init_damerau_levenshtein();
+
+VALUE method_distance_utf(VALUE self, VALUE _s, VALUE _t, VALUE _block_size, VALUE _max_distance);
+
+void Init_damerau_levenshtein() {
+	DamerauLevenshtein = rb_define_module("DamerauLevenshtein");
+	rb_define_method(DamerauLevenshtein, "distance_utf", method_distance_utf, 4);
+}
+
+VALUE method_distance_utf(VALUE self, VALUE _s, VALUE _t, VALUE _block_size, VALUE _max_distance){
+  int i, i1, j, j1, k, half_tl, cost, *d, distance, del, ins, subs, transp, block;
+  int sl, tl, half_sl;
+  int stop_execution = 0;
+  int min = 0;
+  int current_distance = 0;
+
+  int block_size = NUM2INT(_block_size);
+  int max_distance = NUM2INT(_max_distance);
+
+  VALUE *sv = RARRAY_PTR(_s);
+  VALUE *tv = RARRAY_PTR(_t);
+
+  sl = (int) RARRAY_LEN(_s);
+  tl = (int) RARRAY_LEN(_t);
+
+  if (sl == 0) return INT2NUM(tl);
+  if (tl == 0) return INT2NUM(sl);
+  //case of lengths 1 must present or it will break further in the code
+  if (sl == 1 && tl == 1 && sv[0] != tv[0]) return INT2NUM(1);
+
+  int s[sl];
+  int t[tl];
+
+  for (i=0; i < sl; i++) s[i] = NUM2INT(sv[i]);
+  for (i=0; i < tl; i++) t[i] = NUM2INT(tv[i]);
+
+  sl++;
+  tl++;
+
+  //one-dimentional representation of 2 dimentional array len(s)+1 * len(t)+1
+  d = malloc((sizeof(int))*(sl)*(tl));
+  //populate 'vertical' row starting from the 2nd position (first one is filled already)
+  for(i = 0; i < tl; i++){
+    d[i*sl] = i;
+  }
+
+  //fill up array with scores
+  for(i = 1; i<sl; i++){
+    d[i] = i;
+    if (stop_execution == 1) break;
+    current_distance = 10000;
+    for(j = 1; j<tl; j++){
+
+      cost = 1;
+      if(s[i-1] == t[j-1]) cost = 0;
+
+      half_sl = (sl - 1)/2;
+      half_tl = (tl - 1)/2;
+
+      block = block_size < half_sl ? block_size : half_sl;
+      block = block < half_tl ? block : half_tl;
+
+      while (block >= 1){
+        int swap1 = 1;
+        int swap2 = 1;
+        i1 = i - (block * 2);
+        j1 = j - (block * 2);
+        for (k = i1; k < i1 + block; k++) {
+          if (s[k] != t[k + block]){
+            swap1 = 0;
+            break;
+          }
+        }
+        for (k = j1; k < j1 + block; k++) {
+          if (t[k] != s[k + block]){
+            swap2 = 0;
+            break;
+          }
+        }
+
+        del = d[j*sl + i - 1] + 1;
+        ins = d[(j-1)*sl + i] + 1;
+        min = del;
+        if (ins < min) min = ins;
+        //if (i == 2 && j==2) return INT2NUM(swap2+5);
+        if (i >= block && j >= block && swap1 == 1 && swap2 == 1){
+          transp = d[(j - block * 2) * sl + i - block * 2] + cost + block -1;
+          if (transp < min) min = transp;
+          block = 0;
+        } else if (block == 1) {
+          subs = d[(j-1)*sl + i - 1] + cost;
+          if (subs < min) min = subs;
+        }
+        block--;
+      }
+      d[j*sl+i]=min;
+      if (current_distance > d[j*sl+i]) current_distance = d[j*sl+i];
+    }
+    if (current_distance > max_distance) {
+      stop_execution = 1;
+    }
+  }
+  distance=d[sl * tl - 1];
+  if (stop_execution == 1) distance = current_distance;
+
+  free(d);
+  return INT2NUM(distance);
+}

data/lib/taxamatch_rb/damerau_levenshtein_mod.rb

@@ -1,6 +1,6 @@
 # encoding: UTF-8
 
-require File.join(File.dirname(__FILE__), 'damerau_levenshtein')
+require 'damerau_levenshtein'
 
 module Taxamatch
 

data/taxamatch_rb.gemspec

@@ -0,0 +1,96 @@
+# Generated by jeweler
+# DO NOT EDIT THIS FILE DIRECTLY
+# Instead, edit Jeweler::Tasks in Rakefile, and run 'rake gemspec'
+# -*- encoding: utf-8 -*-
+
+Gem::Specification.new do |s|
+  s.name = %q{taxamatch_rb}
+  s.version = "0.7.5"
+
+  s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
+  s.authors = ["Dmitry Mozzherin"]
+  s.date = %q{2011-06-23}
+  s.description = %q{This gem implements algorithm for fuzzy matching scientific names developed by Tony Rees}
+  s.email = %q{dmozzherin@eol.org}
+  s.extensions = ["ext/damerau_levenshtein/extconf.rb"]
+  s.extra_rdoc_files = [
+    "LICENSE",
+    "README.rdoc"
+  ]
+  s.files = [
+    "Gemfile",
+    "Gemfile.lock",
+    "LICENSE",
+    "Makefile",
+    "README.rdoc",
+    "Rakefile",
+    "VERSION",
+    "ext/damerau_levenshtein/damerau_levenshtein.c",
+    "lib/taxamatch_rb.rb",
+    "lib/taxamatch_rb/atomizer.rb",
+    "lib/taxamatch_rb/authmatch.rb",
+    "lib/taxamatch_rb/damerau_levenshtein_mod.rb",
+    "lib/taxamatch_rb/normalizer.rb",
+    "lib/taxamatch_rb/phonetizer.rb",
+    "spec/damerau_levenshtein_mod_test.txt",
+    "spec/spec.opts",
+    "spec/spec_helper.rb",
+    "spec/taxamatch_rb_spec.rb",
+    "spec/taxamatch_test.txt",
+    "taxamatch_rb.gemspec"
+  ]
+  s.homepage = %q{http://github.com/GlobalNamesArchitecture/taxamatch_rb}
+  s.require_paths = ["lib"]
+  s.rubygems_version = %q{1.3.7}
+  s.summary = %q{Implementation of Tony Rees Taxamatch algorithms}
+
+  if s.respond_to? :specification_version then
+    current_version = Gem::Specification::CURRENT_SPECIFICATION_VERSION
+    s.specification_version = 3
+
+    if Gem::Version.new(Gem::VERSION) >= Gem::Version.new('1.2.0') then
+      s.add_runtime_dependency(%q<biodiversity>, ["~> 0.5.13"])
+      s.add_runtime_dependency(%q<biodiversity19>, ["~> 0.5.13"])
+      s.add_runtime_dependency(%q<rake-compiler>, [">= 0"])
+      s.add_development_dependency(%q<rspec>, ["~> 2.3.0"])
+      s.add_development_dependency(%q<cucumber>, [">= 0"])
+      s.add_development_dependency(%q<bundler>, ["~> 1.0.0"])
+      s.add_development_dependency(%q<jeweler>, ["~> 1.6.0"])
+      s.add_development_dependency(%q<rcov>, [">= 0"])
+      s.add_development_dependency(%q<ruby-debug19>, [">= 0"])
+      s.add_development_dependency(%q<ruby-prof>, [">= 0"])
+      s.add_development_dependency(%q<shoulda>, [">= 0"])
+      s.add_development_dependency(%q<mocha>, [">= 0"])
+      s.add_runtime_dependency(%q<biodiversity>, [">= 0.5.13"])
+    else
+      s.add_dependency(%q<biodiversity>, ["~> 0.5.13"])
+      s.add_dependency(%q<biodiversity19>, ["~> 0.5.13"])
+      s.add_dependency(%q<rake-compiler>, [">= 0"])
+      s.add_dependency(%q<rspec>, ["~> 2.3.0"])
+      s.add_dependency(%q<cucumber>, [">= 0"])
+      s.add_dependency(%q<bundler>, ["~> 1.0.0"])
+      s.add_dependency(%q<jeweler>, ["~> 1.6.0"])
+      s.add_dependency(%q<rcov>, [">= 0"])
+      s.add_dependency(%q<ruby-debug19>, [">= 0"])
+      s.add_dependency(%q<ruby-prof>, [">= 0"])
+      s.add_dependency(%q<shoulda>, [">= 0"])
+      s.add_dependency(%q<mocha>, [">= 0"])
+      s.add_dependency(%q<biodiversity>, [">= 0.5.13"])
+    end
+  else
+    s.add_dependency(%q<biodiversity>, ["~> 0.5.13"])
+    s.add_dependency(%q<biodiversity19>, ["~> 0.5.13"])
+    s.add_dependency(%q<rake-compiler>, [">= 0"])
+    s.add_dependency(%q<rspec>, ["~> 2.3.0"])
+    s.add_dependency(%q<cucumber>, [">= 0"])
+    s.add_dependency(%q<bundler>, ["~> 1.0.0"])
+    s.add_dependency(%q<jeweler>, ["~> 1.6.0"])
+    s.add_dependency(%q<rcov>, [">= 0"])
+    s.add_dependency(%q<ruby-debug19>, [">= 0"])
+    s.add_dependency(%q<ruby-prof>, [">= 0"])
+    s.add_dependency(%q<shoulda>, [">= 0"])
+    s.add_dependency(%q<mocha>, [">= 0"])
+    s.add_dependency(%q<biodiversity>, [">= 0.5.13"])
+  end
+end
+

metadata

@@ -5,8 +5,8 @@ version: !ruby/object:Gem::Version
   segments: 
   - 0
   - 7
-  - 4
-  version: 0.7.4
+  - 5
+  version: 0.7.5
 platform: ruby
 authors: 
 - Dmitry Mozzherin
@@ -57,7 +57,7 @@ dependencies:
         segments: 
         - 0
         version: "0"
-  type: :runtime
+  type: :development
   prerelease: false
   version_requirements: *id003
 - !ruby/object:Gem::Dependency 
@@ -183,34 +183,6 @@ dependencies:
   type: :development
   prerelease: false
   version_requirements: *id012
-- !ruby/object:Gem::Dependency 
-  name: biodiversity
-  requirement: &id013 !ruby/object:Gem::Requirement 
-    none: false
-    requirements: 
-    - - ">="
-      - !ruby/object:Gem::Version 
-        segments: 
-        - 0
-        - 5
-        - 13
-        version: 0.5.13
-  type: :runtime
-  prerelease: false
-  version_requirements: *id013
-- !ruby/object:Gem::Dependency 
-  name: rake-compiler
-  requirement: &id014 !ruby/object:Gem::Requirement 
-    none: false
-    requirements: 
-    - - ">="
-      - !ruby/object:Gem::Version 
-        segments: 
-        - 0
-        version: "0"
-  type: :runtime
-  prerelease: false
-  version_requirements: *id014
 description: This gem implements algorithm for fuzzy matching scientific names developed by Tony Rees
 email: dmozzherin@eol.org
 executables: []
@@ -221,12 +193,17 @@ extra_rdoc_files:
 - LICENSE
 - README.rdoc
 files: 
+- Gemfile
 - Gemfile.lock
+- LICENSE
+- Makefile
 - README.rdoc
+- Rakefile
+- VERSION
+- ext/damerau_levenshtein/damerau_levenshtein.c
 - lib/taxamatch_rb.rb
 - lib/taxamatch_rb/atomizer.rb
 - lib/taxamatch_rb/authmatch.rb
-- lib/taxamatch_rb/damerau_levenshtein.bundle
 - lib/taxamatch_rb/damerau_levenshtein_mod.rb
 - lib/taxamatch_rb/normalizer.rb
 - lib/taxamatch_rb/phonetizer.rb
@@ -235,7 +212,7 @@ files:
 - spec/spec_helper.rb
 - spec/taxamatch_rb_spec.rb
 - spec/taxamatch_test.txt
-- LICENSE
+- taxamatch_rb.gemspec
 - ext/damerau_levenshtein/extconf.rb
 has_rdoc: true
 homepage: http://github.com/GlobalNamesArchitecture/taxamatch_rb
@@ -251,7 +228,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements: 
   - - ">="
     - !ruby/object:Gem::Version 
-      hash: -2865757795593253659
+      hash: -571731221339913870
       segments: 
       - 0
       version: "0"