Class: Embulk::Filter::RubyProc

Inherits:
FilterPlugin
  • Object
show all
Defined in:
lib/embulk/filter/ruby_proc.rb

Defined Under Namespace

Classes: Evaluator

Class Method Summary collapse

Instance Method Summary collapse

Class Method Details

.procs ⇒ Object 



75
76
77
 
# File 'lib/embulk/filter/ruby_proc.rb', line 75

def self.procs
  @procs
end

.row_procs ⇒ Object 



79
80
81
 
# File 'lib/embulk/filter/ruby_proc.rb', line 79

def self.row_procs
  @row_procs
end

.transaction(config, in_schema) {|task, out_columns| ... } ⇒ Object

Yields:

  • (task, out_columns) 


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
 
# File 'lib/embulk/filter/ruby_proc.rb', line 31

def self.transaction(config, in_schema, &control)
  task = {
    "columns" => config.param("columns", :array, default: []),
    "rows" => config.param("rows", :array, default: []),
    "requires" => config.param("requires", :array, default: []),
    "variables" => config.param("variables", :hash, default: {}),
  }

  out_columns = in_schema.map do |col|
    target = task["columns"].find { |filter_col| filter_col["name"] == col.name }
    if target
      type = target["type"] ? target["type"].to_sym : col.type
      Embulk::Column.new(index: col.index, name: col.name, type: type || col.type, format: target["format"] || col.format)
    else
      col
    end
  end

  task["requires"].each do |lib|
    require lib
  end

  evaluator_binding = Evaluator.new(task["variables"]).get_binding

  # In order to avoid multithread probrem, initialize procs here
  @procs = Hash[task["columns"].map {|col|
    if col["proc"]
      [col["name"], eval(col["proc"], evaluator_binding)]
    else
      [col["name"], eval(File.read(col["proc_file"]), evaluator_binding, File.expand_path(col["proc_file"]))]
    end
  }]
  @row_procs = task["rows"].map {|rowdef|
    if rowdef["proc"]
      eval(rowdef["proc"], evaluator_binding)
    else
      eval(File.read(rowdef["proc_file"]), evaluator_binding, File.expand_path(rowdef["proc_file"]))
    end
  }.compact
  raise "Need columns or rows parameter" if @row_procs.empty? && @procs.empty?

  yield(task, out_columns)
end

Instance Method Details

#add(page) ⇒ Object 



92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
 
# File 'lib/embulk/filter/ruby_proc.rb', line 92

def add(page)
  page.each do |record|
    if row_procs.empty?
      record_hashes = [hashrize(record)]
    else
      record_hashes = row_procs.each_with_object([]) do |pr, arr|
        result = pr.call(hashrize(record))
        case result
        when Array
          result.each do |r|
            arr << r
          end
        when Hash
          arr << result
        else
          raise "row proc return value must be a Array or Hash"
        end
      end
    end

    record_hashes.each do |record_hash|
      procs.each do |col, pr|
        next unless record_hash.has_key?(col)
        next if record_hash[col].nil? && skip_nils[col]

        if pr.arity == 1
          record_hash[col] = pr.call(record_hash[col])
        else
          record_hash[col] = pr.call(record_hash[col], record_hash)
        end
      end
      page_builder.add(record_hash.values)
    end
  end
end

#close ⇒ Object 



89
90
 
# File 'lib/embulk/filter/ruby_proc.rb', line 89

def close
end

#finish ⇒ Object 



128
129
130
 
# File 'lib/embulk/filter/ruby_proc.rb', line 128

def finish
  page_builder.finish
end

#init ⇒ Object 



83
84
85
86
87
 
# File 'lib/embulk/filter/ruby_proc.rb', line 83

def init
  @skip_nils = Hash[task["columns"].map {|col|
    [col["name"], col["skip_nil"].nil? ? true : !!col["skip_nil"]]
  }]
end