Class: Coupler::Models::Scenario::Runner

Inherits:
Object
  • Object
show all
Defined in:
lib/coupler/models/scenario/runner.rb

Constant Summary collapse

LIMIT = 
10000

Instance Method Summary collapse

Constructor Details

#initialize(parent, &progress) ⇒ Runner

Returns a new instance of Runner.



7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
 
# File 'lib/coupler/models/scenario/runner.rb', line 7

def initialize(parent, &progress)
  @parent = parent
  @progress = progress
  @matcher = parent.matcher
  @type = parent.linkage_type
  if @type == 'cross-linkage'
    @resources = [parent.resource_1, parent.resource_1]
  else
    @resources = parent.resources
  end
  @run_number = @parent.run_count + 1
  @mutex = Mutex.new
  @group_number = 0
  setup_pairs
  create_tables
end

Instance Method Details

#run! ⇒ Object 



24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
 
# File 'lib/coupler/models/scenario/runner.rb', line 24

def run!
  @parent.local_database do |scenario_db|
    @groups_dataset = scenario_db[@groups_table_name]
    @groups_buffer = ImportBuffer.new(@groups_column_names, @groups_dataset)
    @join_dataset = scenario_db[@join_table_name]
    @join_buffer = ImportBuffer.new([:record_id, :which, :group_id], @join_dataset)

    # Group records for each dataset.  This step is the same for both
    # self-linkage and dual-linkage.  However, this is the only step
    # for self-linkage.  Dual-linkage requires another pass after
    # this.
    #
    # Cross-matching on a single dataset is treated as a dual-linkage.
    #
    @pairs = @phase_one_pairs
    tw = ThreadsWait.new
    databases_to_close = []
    @resources.each_with_index do |resource, i|
      dataset = resource.final_dataset
      databases_to_close << dataset.db
      primary_key = resource.primary_key_sym
      which = @type == 'self-linkage' ? nil : i
      resource_thread = phase_one_thread(dataset, primary_key, which)
      tw.join_nowait(resource_thread)
    end
    tw.all_waits
    databases_to_close.each do |db|   # tidy up
      db.disconnect
      ::Sequel::DATABASES.delete(db)
    end

    if @type != 'self-linkage'
      # Phase 2!
      secondary_groups_ds = scenario_db[@secondary_groups_table_name].order(:group_1_id, :group_2_id)
      @join_buffer = ImportBuffer.new([:group_1_id, :group_2_id], secondary_groups_ds)
      @pairs = @phase_two_pairs
      phase_two(@groups_dataset, :id)
      @join_buffer.flush

      # Update groups and groups_records
      tw = ThreadsWait.new
      count = secondary_groups_ds.count
      offset = 0
      last_group_id = @group_number
      while offset < count
        dataset = secondary_groups_ds.limit(10, offset)
        offset += 10
        dataset.each do |row|
          thread = Thread.new(row[:group_1_id], row[:group_2_id]) do |group_1_id, group_2_id|
            new_group_id = get_next_group_id
            @join_dataset.filter(:group_id => [group_1_id, group_2_id]).update(:group_id => new_group_id)
            @groups_dataset.filter(:id => group_1_id).update(:id => new_group_id)
            @groups_dataset.filter(:id => group_2_id).delete
          end
          thread.abort_on_exception = true
          tw.join_nowait(thread)
        end
        tw.all_waits
      end

      # Clean up groups and records that don't match
      @groups_dataset.filter(:id <= last_group_id).delete
      @join_dataset.filter(:group_id <= last_group_id).delete

      # Don't need the secondary table anymore
      scenario_db.drop_table(@secondary_groups_table_name)
    end

    # Calculate some summary stats
    @join_dataset.group_and_count(:group_id, :which).each do |row|
      col = row[:which] ? :"resource_#{row[:which]+1}_count" : :"resource_1_count"
      @groups_dataset.filter(:id => row[:group_id]).update(col => row[:count])
    end
  end
end