graphmatcher 0.3.4 → 0.3.5

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 16aafb8ddede700354cfac5795c95c8f16ecf3c4076b082d4e409b7b211d4310
-  data.tar.gz: 742e07a51018dae3dc9de0b046a82e9f35b39825784c9f126964685f77366e3e
+  metadata.gz: 1a07b061aee6d3c909fa21e7dcc3ed815c556b358f0e233338614f97dc26d90f
+  data.tar.gz: ac1149e8f375be540ba8d77d77276f9d309df1dc44c7e7057eef9def8ff38be8
 SHA512:
-  metadata.gz: 16f27d89dae6474a68ec4ed01c800f0047f9f5f46967771172a7a3150174359bddd84d4b5d1467410abfe2f8a8e35bf6458aaf1a1a951d4e5b8524d03259a94f
-  data.tar.gz: bd84006820c720fecfd948113423a90e8aa980ac281e9342a7ac5b38874c4f5e84412d9c5a32ff8a27d29260c8cb01f03b1317f3041a58d994efc6341f1da359
+  metadata.gz: be30bea5f4045cefb90288e632e43e4ea9d2472879c2e4d339fef667ddf842f863601064052e969bb1f7e6afa8f3d17daa5457ed52208e311faf76575ada5ca4
+  data.tar.gz: 6ce6b7bd8a89fbc44c06079c014353220ed8f5cc3e938040bb7da9eb9b2226d967c11eadef2e1fe50c95989fbfd303d637d90cb057ea948adc3c785ba48ad142

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: graphmatcher
 version: !ruby/object:Gem::Version
-  version: 0.3.4
+  version: 0.3.5
 platform: ruby
 authors:
 - Emre Unlu
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2020-12-19 00:00:00.000000000 Z
+date: 2020-12-20 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -42,44 +42,30 @@ dependencies:
   name: rspec
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
-        version: 3.5.0
+        version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
-        version: 3.5.0
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: ruby-prof
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
-        version: 0.16.2
+        version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: 0.16.2
-- !ruby/object:Gem::Dependency
-  name: rspec-prof
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: 0.0.7
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
-        version: 0.0.7
+        version: '0'
 description: An effective subgraph matching gem based on DualIso algorithm of M. Saltz
   et al.
 email:
@@ -87,8 +73,7 @@ email:
 executables: []
 extensions: []
 extra_rdoc_files: []
-files:
-- lib/graphmatcher.rb
+files: []
 homepage: https://github.com/forvelin/graphmatcher
 licenses:
 - MIT

data/lib/graphmatcher.rb

@@ -1,291 +0,0 @@
-require 'logger'
-# require 'ruby-prof'
-
-# @query_graph & @data_graph: Arrays which are represented as an
-# array such as g=[[[1,2],[3],[3],[]],['x','y','y','z']]. g[0] is
-# adjacency array which points children of given vertex of array index,
-# -e.g. vertex 0 has 1 and 2 as children- and g[1] is labels for vertices.
-# -e.g. vertex 0 has label 'x'.
-#
-# @limit: Upper limit for number of matches. Procedure terminates after
-# reaching to an upper limit.
-# @max_allowed_time: Maximum allowed time for procedure to run.
-# @self_loops: Boolean value for representing if query is cyclic or not.
-class Graphmatcher
-  @@logger = Logger.new(STDOUT)
-  @@logger.level = Logger::FATAL
-
-  def initialize(args)
-    @query_graph = args[:query_graph].to_a
-    @data_graph = args[:data_graph].to_a
-    @limit = (args[:limit] || 1).to_i
-    @max_allowed_time = (args[:max_allowed_time] || 4.000).to_f
-    @cost_matrix = args[:cost_matrix] || nil
-    @self_loops = args[:self_loops] || false
-    validate!
-  end
-
-  # Function for generating feasible matches for query
-  # graph based on labels of vertices of data graph.
-  def label_match
-    data_labels = @data_graph[1]
-    query_labels = @query_graph[1]
-
-    feasible = query_labels.map.with_index do |ql, _index|
-      data_labels.each_index.select { |i| data_labels[i] == ql }
-    end
-
-    if @cost_matrix
-
-      feasible = assess_cost(feasible, @cost_matrix)
-
-      feasible = feasible.select { |f| f[1] }.map do |feasible_set|
-        feasible_set.sort_by { |f| f[1] }
-      end
-
-      feasible = feasible.map do |f_set|
-        f_set.map do |f|
-          f[0]
-        end
-      end
-    end
-
-    @@logger.info('Label matches(phi) are: ' + feasible.to_s)
-    feasible
-  end
-
-  # Public interface for Graphmatcher class.
-  #
-  # @matches: Array of matching indices of query graph in data graph.
-  def find_matches
-    @matches = []
-    @t0 = Time.now.to_f
-    phi = label_match
-
-    dual_iso(dual_simulation(phi), 0)
-    # @@logger.info("FINISHED matches=#{@matches}")
-    if @cost_matrix
-      # if cost matrix is available, get costs of found matches.
-
-      @matches = assess_cost(@matches, @cost_matrix)
-
-      # sort matches by sum of costs of matched resources.
-      #               MATCHES
-      # [ [[1,100],[2,10]],[[3,500],[4,800]] ]
-      #              MATCH COSTS
-      #         110              1300
-
-      # The behaviour here is important !
-      # Sum of costs vs. max of costs, depends which one is relevant.
-
-      @matches.reject! { |match_set| match_set.map { |e| e[1] }.include?(nil) }
-
-      @matches = @matches.sort_by do |match_set|
-        match_set.reduce(0) { |sum, e| sum + e[1] }
-      end
-
-    end
-    @matches
-  end
-
-  def get_resource_property(_match, property)
-    truncated_data_graph = @data_graph.truncate
-    @matches.map { |match_set| match_set.map { |match| truncated_data_graph[match[0]][2][property] } }
-  end
-
-  def get_resource_cost(costs, resource_position, query_index)
-    # costs = { resource_id => { query_index => cost } }
-    # e.g.
-    # costs = { 40 => { 0 => 5, 1 => 10 } }
-    if costs[resource_position][query_index]
-      cost = (costs[resource_position][query_index])
-      cost
-    end
-  end
-
-  def assess_cost(matches, costs)
-    # resource_graph =
-    #   [
-    #       [[],[],[],[],[],[],[]], #adj.
-    #       ['SPO2','NRF52','SPO2','NRF52','SPO2','NRF52','SPO2'], #types
-    #       ['img_x','img_y','img_z','img_t','img_z','img_q','img_z'], #images
-    #       ['12','52','25','61','74','95','11']            #resource_id
-    #   ]
-
-    # request_graph =
-    #   [
-    #       [[],[]],
-    #       ['NRF52','NRF52'],
-    #       ['img_y','img_z'],
-    #       ['NODE_A','HUB_A']
-    #   ]
-
-    # costs = {
-    #   52 => {0 => 0, 1 => 50} , #y
-    #   61 => {0 => 30, 1 => 70} , #t
-    #   95 => {0 => 40, 1 => 55} , #q
-    # }
-
-    # matches = [
-    #   [[1],[2]],[[1],[4]],[[1],[6]],
-    #   [[3],[2]],[[3],[4]],[[3],[6]],
-    #   [[5],[2]],[[5],[4]],[[5],[6]]
-    # ]
-
-    matches.map do |match_set| # [ [1],[2] ]
-      match_set.flatten.map.with_index do |match, query_index| # 1
-        [match, get_resource_cost(costs, match, query_index).to_f]
-      end
-    end
-  end
-
-  # INFO: Function that uses parameter phi -which is generated by label_match-
-  # to determine which matches of data have expected relations in query graph.
-  # phi = ...
-  def dual_simulation(phi)
-    # One directional adjacency array for data graph and query graphs.
-    data_children = @data_graph[0]
-    query_children = @query_graph[0]
-    # @@logger.info("Data children: #{data_children.to_s}")
-    # @@logger.info("Query children: #{query_children.to_s}")
-    changed = true
-    while changed
-      changed = false
-      return nil if (Time.now.to_f - @t0) > @max_allowed_time
-
-      # children = query_edges
-      # q_index = query_vertex_index
-      query_children.each_with_index do |children, q_index|
-        # query_child = query_edge_target
-        children.each do |query_child|
-          # Create a temporary phi object.
-          temp_phi = []
-          # Loop over candidates of each vertex in data graph.
-          to_delete = []
-
-          phi[q_index].map do |child| # loop 3
-            # @@logger.debug("u=#{q_index}, u_c=#{query_child}, child=#{child}")
-
-            # Find intersection of children of 'child' in data graph and
-            # candidates of 'query child' in data graph.
-            phi_intersection = data_children[child] & phi[query_child]
-            # @@logger.debug("datachildren[child]=#{data_children[child]}")
-            # @@logger.debug("phi[query_child]=#{phi[query_child]}")
-            # @@logger.debug("Intersection=#{phi_intersection}")
-            if phi_intersection.nil? || phi_intersection.empty?
-              to_delete.push(child)
-              return phi if phi[q_index].empty?
-
-              changed = true
-            end
-            temp_phi |= phi_intersection
-          end
-
-          unless to_delete.empty?
-            to_delete.each do |td|
-              phi[q_index].delete(td)
-            end
-          end
-          return phi if temp_phi.flatten.empty?
-
-          changed = true if temp_phi.size < phi[query_child].size
-          if @self_loops && query_child == q_index
-            phi[query_child] &= temp_phi
-          else
-            # @@logger.debug("phi=#{phi} and phi[#{query_child}]=#{temp_phi}")
-            phi[query_child] = temp_phi
-          end
-        end
-      end
-    end
-    @@logger.info("Returning phi=#{phi}")
-    phi
-  end
-
-  # INFO: Function call to collect matches from phi object.
-  # phi = ...
-  # depth = ...
-  # matches = ...
-  def dual_iso(phi, depth)
-    if depth == @query_graph[0].length
-      unless phi.nil? || phi.empty?
-        @matches <<
-          if phi.include?([]) # Unable to match this vertex in graph.
-            [nil]
-          else
-            phi
-          end
-      end
-    elsif !(phi.nil? || phi.empty?)
-      phi[depth].sort_by { |value| @cost_matrix ? (@cost_matrix[value][depth] || Float::INFINITY) : next }.each do |value|
-        next if contains(phi, depth, value)
-
-        # keys are indices 0...n, values are possible values for that index
-        phicopy = phi.map(&:clone)
-        # @@logger.info("phicopy=#{phicopy},depth=#{depth},value=#{value}")
-        phicopy[depth] = [value]
-        if @matches.length >= @limit
-          @@logger.info("FINISHED matches=#{@matches}")
-          return @matches
-        end
-        dual_iso(dual_simulation(phicopy), depth + 1)
-      end
-    end
-  end
-
-  # INFO: Checks if vertex J is contained in any of previous matches.
-  # TODO: Change with find method.
-  def contains(phi, depth, vertex_j)
-    false if depth <= 0
-    (0..depth - 1).each do |i|
-      # @@logger.info("phi[#{i}]=#{phi[i]},depth=#{depth},vertex_j=#{vertex_j}")
-      return true if phi[i].include?(vertex_j)
-    end
-    false
-  end
-
-  # EXPERIMENTAL
-  # INFO: Produce a GraphViz-compliant directed graph syntax.
-  # INFO: Needs dot/graphviz tools installed as a dependency.
-  # TODO: Unable to handle multiple results, color each result different.
-  # Indices are IDs, labels are labels adjencency array is outgoing edges.
-  def dot_graph(data, subgraph = nil, prefix = '')
-    output = ['digraph {']
-    data.transpose.each_with_index do |node, id|
-      output <<
-        ["#{id} [label=\"#{node[1]}##{id}\"]",
-         "#{id}->{#{node[0].join(' ')}}"].join("\n")
-    end
-    if subgraph
-      subgraph.each_with_index do |node, _id|
-        output << "#{node} [fontcolor=\"Red\"]"
-      end
-    end
-    output << '}'
-    tstamp = Time.new.to_i.to_s
-    File.write("#{prefix}#{tstamp}.dot", output.join("\n"))
-    dot_produce = ['dot', '-Tpng', "#{prefix}#{tstamp}.dot",
-                   '-o', "#{prefix}#{tstamp}.png"].join(' ')
-    `#{dot_produce}`
-  end
-
-  def validate!
-    unless @query_graph.is_a?(Array) && @data_graph.is_a?(Array)
-      raise ArgumentError,
-            'Type mismatch for graphs in initialization !'
-    end
-    unless @limit.is_a?(Numeric) && @max_allowed_time.is_a?(Numeric)
-      raise ArgumentError,
-            'Type mismatch for limit or timeout value in initialization !'
-    end
-    unless @query_graph.length >= 2 && @data_graph.length >= 2
-      raise ArgumentError,
-            'Input graphs must have at least two dimensions !'
-    end
-    unless @query_graph.map(&:length).uniq.size == 1 &&
-           @data_graph.map(&:length).uniq.size == 1
-      raise ArgumentError,
-            'Input graphs\' adjencency and label arrays must be sized equal !'
-    end
-  end
-end