dhash-vips 0.2.3.0 → 0.2.4.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: ce70f825a89117f682dad1e5ae6400ec3cd430d08d8bb34b36719dff4905894e
-  data.tar.gz: 17ff221fad7650632cefc04f7421efc79b4f0fe0f2f884d6f1019fc82d360c94
+  metadata.gz: 6b9bb7ce3d33457aa4a9562c223fc86404304e6a15d92db631462e7f3d3ba1cb
+  data.tar.gz: '09c3df795276527873062983d954ebbdeb07125eccca88551a2fd56f95e13be2'
 SHA512:
-  metadata.gz: '019fa1a1b89a5e734332eb1993ea4ccd9dd18b6add3677f065c41cbf0b624986220a094ee5e055cf4f965b56292abe9275ee96fdcbdae53481ea557b0317ad46'
-  data.tar.gz: 4fa0705056bc27ef272b9a4ca93657bada9646486e0a9532b21d4f53f8d23847188e0b156d66c105054d9a0f2745be4be89b3c9b0244e8e52bc593abd8757a96
+  metadata.gz: 48e976f177d8fe6dcd010bc0501ec862d2c2bfdad45bfd17e1f6c2f03abd7481e4930683018c7ea879b9c7b339e8676460bea135f943314d25d5dc1b70f76e04
+  data.tar.gz: 23a9ea2f316a226872940e7f0e1dfb3547a01db4ab0c24d981ba4edc68934a02a54be4007aeb7015977c72f7fbf90ead4d092113b45a77c4076feb07ebb421c0

data/bin/idhash

@@ -0,0 +1,57 @@
+#!/usr/bin/env ruby
+Signal.trap(:INT){ abort "\n(interrupted by SIGINT)" }
+
+unless 2 == ARGV.size
+  puts "this command is to compare two images"
+  puts "usage: #{__FILE__} <path to image1> <path to image2>"
+  exit
+end
+
+require_relative "../lib/dhash-vips"
+ha, hb = ARGV.map{ |filename| DHashVips::IDHash.fingerprint filename }
+puts "distance: #{d1 = DHashVips::IDHash.distance ha, hb}"
+size = 2 ** 3
+shift = 2 * size * size
+ai = ha >> shift
+ad = ha - (ai << shift)
+bi = hb >> shift
+bd = hb - (bi << shift)
+
+_127 = shift - 1
+_63 = size * size - 1
+# width = 800
+# height = 800
+
+d2 = 0
+a, b = [[ad, ai, ARGV[0]], [bd, bi, ARGV[1]]].map do |xd, xi, path|
+  puts File.basename path
+  hor = Array.new(size){Array.new(size){" "}}
+  ver = Array.new(size){Array.new(size){" "}}
+  _127.downto(0).each_with_index do |i, ii|
+    if i > _63
+      y, x = (_127 - i).divmod size
+    else
+      x, y = (_63 - i).divmod size
+    end
+    if xi[i] > 0
+      target, c = if i > _63
+        [ver, %w{ v ^ }[xd[i]]]
+      else
+        [hor, %w{ > < }[xd[i]]]
+      end
+      target[y][x] = c
+    end
+    if ai[i] + bi[i] > 0 && ad[i] != bd[i]
+      d2 += 1
+      target = if i > _63
+        ver
+      else
+        hor
+      end
+      target[y][x] = "\e[7m#{target[y][x]}\e[27m"
+    end
+  end
+  hor.zip(ver).each{ |_| puts _.join " " }
+end
+abort "something went wrong" unless d1 * 2 == d2
+puts "OK"

data/dhash-vips.gemspec

@@ -1,17 +1,19 @@
 Gem::Specification.new do |spec|
   spec.name          = "dhash-vips"
-  spec.version       = "0.2.3.0"
+  spec.version       = "0.2.4.0"
   spec.summary       = "dHash and IDHash perceptual image hashing/fingerprinting"
-  # spec.homepage      = "https://github.com/nakilon/dhash-vips"
 
-  spec.author        = "Victor Maslov"
+  spec.author        = "Victor Maslov aka Nakilon"
   spec.email         = "nakilon@gmail.com"
   spec.license       = "MIT"
+  spec.metadata      = {"source_code_uri" => "https://github.com/nakilon/dhash-vips"}
 
   spec.add_dependency "ruby-vips", "~> 2.0", "!= 2.1.0", "!= 2.1.1"
 
   spec.require_path  = "lib"
-  spec.test_files    = %w{ test.rb }
   spec.extensions    = %w{ extconf.rb }
-  spec.files         = %w{ extconf.rb Gemfile LICENSE.txt common.rb dhash-vips.gemspec idhash.c lib/dhash-vips-post-install-test.rb lib/dhash-vips.rb }
+  spec.files         = %w{ LICENSE.txt dhash-vips.gemspec lib/dhash-vips.rb idhash.c lib/dhash-vips-post-install-test.rb } + spec.extensions +
+                       %w{ bin/idhash }
+  spec.executables   = %w{     idhash }
+  spec.bindir        = "bin"
 end

data/extconf.rb

@@ -3,32 +3,26 @@ require "mkmf"
 File.write "Makefile", dummy_makefile(?.).join
 
 unless Gem::Version.new(RUBY_VERSION) < Gem::Version.new("2.3.8")
-  if ruby_source_dir = if File.directory? "/ruby"
-    "-I/ruby"  # for Github Actions: docker (currently disabled) and benchmark
-  elsif ENV["RBENV_ROOT"] && ENV["RBENV_VERSION"] && File.exist?(t = "#{ENV["RBENV_ROOT"]}/sources/#{ENV["RBENV_VERSION"]}/ruby-#{ENV["RBENV_VERSION"]}/bignum.c")   # https://github.com/rbenv/rbenv/issues/1199
-    "-I#{File.dirname t}"
-  end
-    append_cppflags ruby_source_dir
-    append_cppflags "-DRUBY_EXPORT" unless Gem::Version.new(RUBY_VERSION) < Gem::Version.new("2.4")
-    create_makefile "idhash"
-    # Why this hack?
-    # 1. Because I want to use Ruby and ./idhash.bundle for tests, not C.
-    # 2. Because I don't want to bother users with two gems instead of one.
-    File.write "Makefile", <<~HEREDOC + File.read("Makefile")
-      .PHONY: test
-      test: all
-      \t$(RUBY) -r./lib/dhash-vips.rb ./lib/dhash-vips-post-install-test.rb
-    HEREDOC
-  end
+  append_cppflags "-DRUBY_EXPORT" unless Gem::Version.new(RUBY_VERSION) < Gem::Version.new("2.4")
+  create_makefile "idhash"
+  # Why this hack?
+  # 1. Because I want to use Ruby and ./idhash.bundle for tests, not C.
+  # 2. Because I don't want to bother users with two gems instead of one.
+  File.write "Makefile", <<~HEREDOC + File.read("Makefile")
+    .PHONY: test
+    test: all
+    \t$(RUBY) -r./lib/dhash-vips.rb ./lib/dhash-vips-post-install-test.rb
+  HEREDOC
 end
 
 __END__
 
 # this unlike using rake is building to current directory
-#   that is vital to be able to require the native extention for benchmarking, etc.
+#   that is vital to be able to require the native extension for benchmarking, etc.
 $ ruby extconf.rb && make clean && make
 
-# to test: $ rake clean && rake install
+# to test the installation:
+$ rake clean && rake install
 
 $ ruby -e "require 'dhash-vips'; p DHashVips::IDHash.method(:distance3).source_location"  # using -r makes bundler mad
 # [".../dhash-vips.rb", 32] # if LoadError

data/idhash.c

@@ -1,29 +1,35 @@
-#include <bignum.c>
+#include <ruby.h>
 
-static VALUE idhash_distance(VALUE self, VALUE a, VALUE b){
-    BDIGIT* tempd;
-    long i, an = BIGNUM_LEN(a), bn = BIGNUM_LEN(b), templ, acc = 0;
-    BDIGIT* as = BDIGITS(a);
-    BDIGIT* bs = BDIGITS(b);
-    while (0 < an && as[an-1] == 0) an--; // for (i = an; --i;) printf("%u\n", as[i]);
-    while (0 < bn && bs[bn-1] == 0) bn--; // for (i = bn; --i;) printf("%u\n", bs[i]);
-    // printf("%lu %lu\n", an, bn);
-    if (an < bn) {
-      tempd = as; as = bs; bs = tempd;
-      templ = an; an = bn; bn = templ;
-    }
-    for (i = an; i-- > 4;) {
-      // printf("%ld : (%u | %u) & (%u ^ %u)\n", i, as[i], (i >= bn ? 0 : bs[i]), as[i-4], bs[i-4]);
-      acc += __builtin_popcountl((as[i] | (i >= bn ? 0 : bs[i])) & (as[i-4] ^ bs[i-4]));
-      // printf("%ld : %ld\n", i, acc);
-    }
-    RB_GC_GUARD(a);
-    RB_GC_GUARD(b);
-    return INT2FIX(acc);
+static VALUE idhash_distance(VALUE self, VALUE a, VALUE b) {
+    const size_t max_words = 256 / sizeof(uint64_t);
+
+    const size_t word_numbits = sizeof(uint64_t) * CHAR_BIT;
+    size_t n;
+    n = rb_absint_numwords(a, word_numbits, NULL);
+    if (n > max_words || n == (size_t)-1)
+        rb_raise(rb_eRangeError, "fingerprint #1 exceeds 256 bits");
+    n = rb_absint_numwords(b, word_numbits, NULL);
+    if (n > max_words || n == (size_t)-1)
+        rb_raise(rb_eRangeError, "fingerprint #2 exceeds 256 bits");
+
+    uint64_t as[max_words], bs[max_words];
+    rb_integer_pack(
+        a, as, max_words, sizeof(uint64_t), 0,
+        INTEGER_PACK_LSWORD_FIRST | INTEGER_PACK_NATIVE_BYTE_ORDER | INTEGER_PACK_2COMP
+    );
+    rb_integer_pack(
+        b, bs, max_words, sizeof(uint64_t), 0,
+        INTEGER_PACK_LSWORD_FIRST | INTEGER_PACK_NATIVE_BYTE_ORDER | INTEGER_PACK_2COMP
+    );
+
+    return INT2FIX(
+        __builtin_popcountll((as[3] | bs[3]) & (as[1] ^ bs[1])) +
+        __builtin_popcountll((as[2] | bs[2]) & (as[0] ^ bs[0]))
+    );
 }
 
 void Init_idhash() {
-  VALUE m = rb_define_module("DHashVips");
-  VALUE mm = rb_define_module_under(m, "IDHash");
-  rb_define_module_function(mm, "distance3_c", idhash_distance, 2);
+    VALUE m = rb_define_module("DHashVips");
+    VALUE mm = rb_define_module_under(m, "IDHash");
+    rb_define_module_function(mm, "distance3_c", idhash_distance, 2);
 }

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: dhash-vips
 version: !ruby/object:Gem::Version
-  version: 0.2.3.0
+  version: 0.2.4.0
 platform: ruby
 authors:
-- Victor Maslov
-autorequire: 
+- Victor Maslov aka Nakilon
+autorequire:
 bindir: bin
 cert_chain: []
-date: 2022-12-23 00:00:00.000000000 Z
+date: 2025-04-12 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: ruby-vips
@@ -36,27 +36,27 @@ dependencies:
     - - "!="
       - !ruby/object:Gem::Version
         version: 2.1.1
-description: 
+description:
 email: nakilon@gmail.com
-executables: []
+executables:
+- idhash
 extensions:
 - extconf.rb
 extra_rdoc_files: []
 files:
-- Gemfile
 - LICENSE.txt
-- common.rb
+- bin/idhash
 - dhash-vips.gemspec
 - extconf.rb
 - idhash.c
 - lib/dhash-vips-post-install-test.rb
 - lib/dhash-vips.rb
-- test.rb
-homepage: 
+homepage:
 licenses:
 - MIT
-metadata: {}
-post_install_message: 
+metadata:
+  source_code_uri: https://github.com/nakilon/dhash-vips
+post_install_message:
 rdoc_options: []
 require_paths:
 - lib
@@ -71,9 +71,8 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.3.25
-signing_key: 
+rubygems_version: 3.1.6
+signing_key:
 specification_version: 4
 summary: dHash and IDHash perceptual image hashing/fingerprinting
-test_files:
-- test.rb
+test_files: []

data/Gemfile

@@ -1,12 +0,0 @@
-source "https://rubygems.org"
-
-gem "rake"
-gem "rmagick"
-gem "dhash", github: "nakilon/dhash"
-gem "phamilie"
-gem "get_process_mem"
-gem "mll"
-gem "minitest"
-gem "byebug", "<11.1.0"   # for Ruby 2.3
-
-gemspec

data/common.rb

@@ -1,11 +0,0 @@
-def download_if_needed path
-  require "open-uri"
-  require "digest"
-  FileUtils.mkdir_p File.dirname path
-  URI("http://gems.nakilon.pro.storage.yandexcloud.net/dhash-vips/#{File.basename path}".tap do |url|
-    puts "downloading #{path} from #{url}"
-  end).open do |link|
-    File.open(path, "wb"){ |file| IO.copy_stream link, file }
-  end unless File.exist?(path) && Digest::MD5.file(path) == File.basename(path, File.extname(path))
-  path
-end

data/test.rb

@@ -1,126 +0,0 @@
-require "minitest/autorun"
-
-require_relative "lib/dhash-vips"
-
-# TODO tests about `fingerprint(4)`
-
-[
-  [DHashVips::DHash, :hamming, :calculate, 2, 23, 16, 50, 7, 0x7919395919191919],
-
-    # v0.2.3.0
-    # [[0, 18, 26, 28, 28, 24, 35, 31, 42, 42, 32, 34, 33, 29, 35, 40],
-    #  [18, 0, 30, 24, 38, 34, 33, 33, 44, 42, 38, 42, 41, 37, 37, 50],
-    #  [26, 30, 0, 16, 34, 38, 29, 35, 42, 40, 36, 38, 31, 31, 31, 36],
-    #  [28, 24, 16, 0, 36, 36, 31, 35, 40, 38, 34, 34, 41, 37, 27, 34],
-    #  [28, 38, 34, 36, 0, 14, 35, 33, 40, 42, 32, 26, 27, 33, 35, 28],
-    #  [24, 34, 38, 36, 14, 0, 43, 39, 38, 40, 24, 28, 25, 31, 29, 28],
-    #  [35, 33, 29, 31, 35, 43, 0, 8, 27, 25, 35, 33, 32, 32, 28, 31],
-    #  [31, 33, 35, 35, 33, 39, 8, 0, 27, 27, 35, 33, 34, 34, 28, 33],
-    #  [42, 44, 42, 40, 40, 38, 27, 27, 0, 2, 34, 32, 31, 33, 31, 28],
-    #  [42, 42, 40, 38, 42, 40, 25, 27, 2, 0, 34, 34, 33, 35, 31, 28],
-    #  [32, 38, 36, 34, 32, 24, 35, 35, 34, 34, 0, 10, 23, 31, 25, 18],
-    #  [34, 42, 38, 34, 26, 28, 33, 33, 32, 34, 10, 0, 23, 29, 25, 16],
-    #  [33, 41, 31, 41, 27, 25, 32, 34, 31, 33, 23, 23, 0, 20, 24, 19],
-    #  [29, 37, 31, 37, 33, 31, 32, 34, 33, 35, 31, 29, 20, 0, 22, 27],
-    #  [35, 37, 31, 27, 35, 29, 28, 28, 31, 31, 25, 25, 24, 22, 0, 23],
-    #  [40, 50, 36, 34, 28, 28, 31, 33, 28, 28, 18, 16, 19, 27, 23, 0]]
-
-  [DHashVips::IDHash, :distance, :fingerprint, 8, 22, 25, 70, 0, 0x1d5cdc0d1d0c1d9f5720a6fff2fe02013f00df9f005e1dc0ff670000ffff0080],
-
-    # v0.2.3.0
-    # [[0, 19, 34, 38, 57, 47, 50, 49, 45, 42, 55, 45, 60, 49, 51, 53],
-    #  [19, 0, 34, 33, 55, 47, 52, 49, 46, 49, 59, 46, 62, 54, 50, 57],
-    #  [34, 34, 0, 8, 48, 55, 46, 42, 55, 57, 45, 35, 51, 45, 46, 47],
-    #  [38, 33, 8, 0, 52, 61, 43, 38, 49, 50, 50, 37, 50, 39, 43, 51],
-    #  [57, 55, 48, 52, 0, 22, 46, 42, 70, 63, 51, 50, 35, 41, 46, 46],
-    #  [47, 47, 55, 61, 22, 0, 54, 54, 57, 53, 39, 44, 43, 41, 45, 38],
-    #  [50, 52, 46, 43, 46, 54, 0, 8, 35, 38, 51, 45, 50, 42, 43, 43],
-    #  [49, 49, 42, 38, 42, 54, 8, 0, 34, 36, 54, 49, 49, 42, 43, 44],
-    #  [45, 46, 55, 49, 70, 57, 35, 34, 0, 10, 53, 55, 56, 50, 49, 49],
-    #  [42, 49, 57, 50, 63, 53, 38, 36, 10, 0, 50, 52, 55, 48, 46, 45],
-    #  [55, 59, 45, 50, 51, 39, 51, 54, 53, 50, 0, 10, 32, 35, 40, 25],
-    #  [45, 46, 35, 37, 50, 44, 45, 49, 55, 52, 10, 0, 30, 27, 37, 26],
-    #  [60, 62, 51, 50, 35, 43, 50, 49, 56, 55, 32, 30, 0, 22, 26, 28],
-    #  [49, 54, 45, 39, 41, 41, 42, 42, 50, 48, 35, 27, 22, 0, 37, 35],
-    #  [51, 50, 46, 43, 46, 45, 43, 43, 49, 46, 40, 37, 26, 37, 0, 22],
-    #  [53, 57, 47, 51, 46, 38, 43, 44, 49, 45, 25, 26, 28, 35, 22, 0]]
-
-].each do |lib, dm, calc, min_similar, max_similar, min_not_similar, max_not_similar, bw_exceptional, hash|
-
-  describe lib do
-    require "fileutils"
-    require "digest"
-    require "mll"
-
-    require_relative "common"
-
-    # these are false positive by idhash
-    # 1b1d4bde376084011d027bba1c047a4b.jpg
-    # 6d97739b4a08f965dc9239dd24382e96.jpg
-    [
-      [:similar, %w{
-        1d468d064d2e26b5b5de9a0241ef2d4b.jpg 92d90b8977f813af803c78107e7f698e.jpg
-        309666c7b45ecbf8f13e85a0bd6b0a4c.jpg 3f9f3db06db20d1d9f8188cd753f6ef4.jpg
-        679634ff89a31279a39f03e278bc9a01.jpg df0a3b93e9412536ee8a11255f974141.jpg
-        54192a3f65bd03163b04849e1577a40b.jpg 6d32f57459e5b79b5deca2a361eb8c6e.jpg
-        4b62e0eef58bfbc8d0d2fbf2b9d05483.jpg b8eb0ca91855b657f12fb3d627d45c53.jpg
-        21cd9a6986d98976b6b4655e1de7baf4.jpg 9b158c0d4953d47171a22ed84917f812.jpg
-        9c2c240ec02356472fb532f404d28dde.jpg fc762fa286489d8afc80adc8cdcb125e.jpg
-        7a833d873f8d49f12882e86af1cc6b79.jpg ac033cf01a3941dd1baa876082938bc9.jpg
-      }, min_similar, max_similar, min_not_similar, max_not_similar], # slightly silimar images
-      [:bw_exceptional, %w{
-        71662d4d4029a3b41d47d5baf681ab9a.jpg ad8a37f872956666c3077a3e9e737984.jpg
-      }, bw_exceptional, bw_exceptional], # these are the same photo but of different size and colorspace
-    ].each do |test_name, _images, min, max, min_not, max_not|
-
-      images = _images.map{ |_| download_if_needed "test_images/#{_}" }
-
-      hashes = images.map &lib.method(calc)
-      table = MLL::table[lib.method(dm), [hashes], [hashes]]
-
-      # STDERR.puts ""
-      # require "pp"
-      # PP.pp table, STDERR
-      # STDERR.puts ""
-
-      it test_name do
-        similar = []
-        not_similar = []
-        hashes.size.times.to_a.repeated_combination(2) do |i, j|
-          case
-          when i == j
-            assert_predicate table[i][j], :zero?
-          when (j - i).abs == 1 && (i + j - 1) % 4 == 0
-            # STDERR.puts [table[i][j], min, max].inspect
-            similar.push table[i][j]
-          else
-            # STDERR.puts [table[i][j], min_not_similar, max_not_similar].inspect
-            not_similar.push table[i][j]
-          end
-        end
-        assert_equal [min, max], similar.minmax
-        assert_equal [min_not, max_not], not_similar.minmax if min_not
-      end
-
-    end
-
-    it "accepts Vips::Image" do
-      # https://github.com/libvips/ruby-vips/issues/349
-      lib.public_send calc, Vips::Image.new_from_buffer("GIF89a\x01\x00\x01\x00\x80\x01\x00\xFF\xFF\xFF\x00\x00\x00,\x00\x00\x00\x00\x01\x00\x01\x00\x00\x02\x02D\x01\x00;", "")
-    end
-
-    it "correct calculation" do
-      download_if_needed "alpha_only.png"
-      t = lib.public_send calc, "alpha_only.png"
-      assert_equal hash, t, ->{ "0x#{hash.to_s 16} != 0x#{t.to_s 16}" }
-    end
-
-  end
-
-end
-
-describe DHashVips::IDHash do
-  it "does not call distance3_ruby" do
-    assert_equal :distance3, DHashVips::IDHash.method(:distance3).original_name
-  end
-end