colincasey-activewarehouse-etl 0.9.4

Sign up to get free protection for your applications and to get access to all the features.
Files changed (86) hide show
  1. data/CHANGELOG +198 -0
  2. data/LICENSE +7 -0
  3. data/README +85 -0
  4. data/Rakefile +75 -0
  5. data/TODO +28 -0
  6. data/VERSION.yml +4 -0
  7. data/bin/etl +28 -0
  8. data/bin/etl.cmd +8 -0
  9. data/lib/etl.rb +81 -0
  10. data/lib/etl/batch.rb +2 -0
  11. data/lib/etl/batch/batch.rb +111 -0
  12. data/lib/etl/batch/directives.rb +55 -0
  13. data/lib/etl/builder.rb +2 -0
  14. data/lib/etl/builder/date_dimension_builder.rb +96 -0
  15. data/lib/etl/builder/time_dimension_builder.rb +31 -0
  16. data/lib/etl/commands/etl.rb +89 -0
  17. data/lib/etl/control.rb +3 -0
  18. data/lib/etl/control/control.rb +414 -0
  19. data/lib/etl/control/destination.rb +420 -0
  20. data/lib/etl/control/destination/csv_destination.rb +84 -0
  21. data/lib/etl/control/destination/database_destination.rb +95 -0
  22. data/lib/etl/control/destination/file_destination.rb +124 -0
  23. data/lib/etl/control/destination/yaml_destination.rb +74 -0
  24. data/lib/etl/control/source.rb +109 -0
  25. data/lib/etl/control/source/database_source.rb +220 -0
  26. data/lib/etl/control/source/enumerable_source.rb +11 -0
  27. data/lib/etl/control/source/file_source.rb +90 -0
  28. data/lib/etl/control/source/model_source.rb +39 -0
  29. data/lib/etl/core_ext.rb +1 -0
  30. data/lib/etl/core_ext/time.rb +5 -0
  31. data/lib/etl/core_ext/time/calculations.rb +42 -0
  32. data/lib/etl/engine.rb +574 -0
  33. data/lib/etl/execution.rb +20 -0
  34. data/lib/etl/execution/base.rb +9 -0
  35. data/lib/etl/execution/batch.rb +8 -0
  36. data/lib/etl/execution/job.rb +8 -0
  37. data/lib/etl/execution/migration.rb +85 -0
  38. data/lib/etl/generator.rb +2 -0
  39. data/lib/etl/generator/generator.rb +20 -0
  40. data/lib/etl/generator/surrogate_key_generator.rb +39 -0
  41. data/lib/etl/http_tools.rb +139 -0
  42. data/lib/etl/parser.rb +11 -0
  43. data/lib/etl/parser/apache_combined_log_parser.rb +49 -0
  44. data/lib/etl/parser/delimited_parser.rb +74 -0
  45. data/lib/etl/parser/fixed_width_parser.rb +65 -0
  46. data/lib/etl/parser/parser.rb +41 -0
  47. data/lib/etl/parser/sax_parser.rb +218 -0
  48. data/lib/etl/parser/spreadsheet_parser.rb +114 -0
  49. data/lib/etl/parser/xml_parser.rb +65 -0
  50. data/lib/etl/processor.rb +11 -0
  51. data/lib/etl/processor/block_processor.rb +14 -0
  52. data/lib/etl/processor/bulk_import_processor.rb +81 -0
  53. data/lib/etl/processor/check_exist_processor.rb +80 -0
  54. data/lib/etl/processor/check_unique_processor.rb +35 -0
  55. data/lib/etl/processor/copy_field_processor.rb +26 -0
  56. data/lib/etl/processor/encode_processor.rb +55 -0
  57. data/lib/etl/processor/hierarchy_exploder_processor.rb +55 -0
  58. data/lib/etl/processor/print_row_processor.rb +12 -0
  59. data/lib/etl/processor/processor.rb +25 -0
  60. data/lib/etl/processor/rename_processor.rb +24 -0
  61. data/lib/etl/processor/require_non_blank_processor.rb +26 -0
  62. data/lib/etl/processor/row_processor.rb +17 -0
  63. data/lib/etl/processor/sequence_processor.rb +23 -0
  64. data/lib/etl/processor/surrogate_key_processor.rb +53 -0
  65. data/lib/etl/processor/truncate_processor.rb +35 -0
  66. data/lib/etl/row.rb +20 -0
  67. data/lib/etl/screen.rb +14 -0
  68. data/lib/etl/screen/row_count_screen.rb +20 -0
  69. data/lib/etl/transform.rb +2 -0
  70. data/lib/etl/transform/block_transform.rb +13 -0
  71. data/lib/etl/transform/date_to_string_transform.rb +20 -0
  72. data/lib/etl/transform/decode_transform.rb +51 -0
  73. data/lib/etl/transform/default_transform.rb +20 -0
  74. data/lib/etl/transform/foreign_key_lookup_transform.rb +151 -0
  75. data/lib/etl/transform/hierarchy_lookup_transform.rb +49 -0
  76. data/lib/etl/transform/ordinalize_transform.rb +12 -0
  77. data/lib/etl/transform/sha1_transform.rb +13 -0
  78. data/lib/etl/transform/string_to_date_transform.rb +16 -0
  79. data/lib/etl/transform/string_to_datetime_transform.rb +14 -0
  80. data/lib/etl/transform/string_to_time_transform.rb +11 -0
  81. data/lib/etl/transform/transform.rb +61 -0
  82. data/lib/etl/transform/trim_transform.rb +26 -0
  83. data/lib/etl/transform/type_transform.rb +35 -0
  84. data/lib/etl/util.rb +59 -0
  85. data/lib/etl/version.rb +10 -0
  86. metadata +224 -0
data/lib/etl/batch.rb ADDED
@@ -0,0 +1,2 @@
1
+ require 'etl/batch/batch'
2
+ require 'etl/batch/directives'
@@ -0,0 +1,111 @@
1
+ module ETL #:nodoc:
2
+ module Batch
3
+ class Context
4
+ attr_reader :batch
5
+
6
+ class << self
7
+ # Create a context that is used when evaluating the batch file
8
+ def create(batch)
9
+ Context.new(batch).get_binding
10
+ end
11
+ end
12
+
13
+ def initialize(batch)
14
+ @batch = batch
15
+ end
16
+
17
+ def file
18
+ batch.file
19
+ end
20
+
21
+ def get_binding
22
+ binding
23
+ end
24
+
25
+ def run(file)
26
+ batch.run(File.dirname(self.file) + "/" + file)
27
+ end
28
+
29
+ def use_temp_tables(value=true)
30
+ batch.use_temp_tables(value)
31
+ end
32
+
33
+ end
34
+ class Batch
35
+ attr_accessor :file
36
+ attr_accessor :engine
37
+
38
+ class << self
39
+ # Resolve the given object to an ETL::Control::Control instance. Acceptable arguments
40
+ # are:
41
+ # * The path to a control file as a String
42
+ # * A File object referencing the control file
43
+ # * The ETL::Control::Control object (which will just be returned)
44
+ #
45
+ # Raises a ControlError if any other type is given
46
+ def resolve(batch, engine)
47
+ batch = do_resolve(batch)
48
+ batch.engine = engine
49
+ batch
50
+ end
51
+
52
+ protected
53
+ def parse(batch_file)
54
+ batch_file = batch_file.path if batch_file.instance_of?(File)
55
+ batch = ETL::Batch::Batch.new(batch_file)
56
+ eval(IO.readlines(batch_file).join("\n"), Context.create(batch), batch_file)
57
+ batch
58
+ end
59
+
60
+ def do_resolve(batch)
61
+ case batch
62
+ when String
63
+ ETL::Batch::Batch.parse(File.new(batch))
64
+ when File
65
+ ETL::Batch::Batch.parse(batch)
66
+ when ETL::Batch::Batch
67
+ batch
68
+ else
69
+ raise RuntimeError, "Batch must be a String, File or Batch object"
70
+ end
71
+ end
72
+ end
73
+
74
+ def initialize(file)
75
+ @file = file
76
+ end
77
+
78
+ def run(file)
79
+ directives << Run.new(self, file)
80
+ end
81
+
82
+ def use_temp_tables(value = true)
83
+ directives << UseTempTables.new(self)
84
+ end
85
+
86
+ def execute
87
+ engine.say "Executing batch"
88
+ before_execute
89
+ directives.each do |directive|
90
+ directive.execute
91
+ end
92
+ engine.say "Finishing batch"
93
+ after_execute
94
+ engine.say "Batch complete"
95
+ end
96
+
97
+ def directives
98
+ @directives ||= []
99
+ end
100
+
101
+ def before_execute
102
+
103
+ end
104
+
105
+ def after_execute
106
+ ETL::Engine.finish # TODO: should be moved to the directive?
107
+ ETL::Engine.use_temp_tables = false # reset the temp tables
108
+ end
109
+ end
110
+ end
111
+ end
@@ -0,0 +1,55 @@
1
+ module ETL #:nodoc:
2
+ module Batch #:nodoc:
3
+ # Abstract base class for directives
4
+ class Directive
5
+ # Method to access the batch object
6
+ attr_reader :batch
7
+
8
+ # Initialize the directive with the given batch object
9
+ def initialize(batch)
10
+ @batch = batch
11
+ end
12
+
13
+ # Execute the directive
14
+ def execute
15
+ do_execute
16
+ end
17
+
18
+ protected
19
+ # Implemented by subclasses
20
+ def do_execute
21
+ raise RuntimeError, "Directive must implement do_execute method"
22
+ end
23
+ end
24
+
25
+ # Directive indicating that the specified ETL control file should be
26
+ # run
27
+ class Run < Directive
28
+ # The file to execute
29
+ attr_reader :file
30
+
31
+ # Initialize the directive with the given batch object and file
32
+ def initialize(batch, file)
33
+ super(batch)
34
+ @file = file
35
+ end
36
+
37
+ protected
38
+ # Execute the process
39
+ def do_execute
40
+ batch.engine.process(file)
41
+ end
42
+ end
43
+
44
+ # Directive indicating temp tables should be used.
45
+ class UseTempTables < Directive
46
+ def initialize(batch)
47
+ super(batch)
48
+ end
49
+ protected
50
+ def do_execute
51
+ ETL::Engine.use_temp_tables = true
52
+ end
53
+ end
54
+ end
55
+ end
@@ -0,0 +1,2 @@
1
+ require 'etl/builder/date_dimension_builder'
2
+ require 'etl/builder/time_dimension_builder'
@@ -0,0 +1,96 @@
1
+ module ETL #:nodoc:
2
+ module Builder #:nodoc:
3
+ # A builder which will build a data structure which can be used to populate a date dimension using
4
+ # commonly used date dimension columns.
5
+ class DateDimensionBuilder
6
+ # Specify the start date for the first record
7
+ attr_accessor :start_date
8
+
9
+ # Specify the end date for the last record
10
+ attr_accessor :end_date
11
+
12
+ # Define any holiday indicators
13
+ attr_accessor :holiday_indicators
14
+
15
+ # Add offset month for fiscal year
16
+ attr_accessor :fiscal_year_offset_month
17
+
18
+ # Define the weekday indicators. The default array begins on Sunday and goes to Saturday.
19
+ cattr_accessor :weekday_indicators
20
+ @@weekday_indicators = ['Weekend','Weekday','Weekday','Weekday','Weekday','Weekday','Weekend']
21
+
22
+ # Initialize the builder.
23
+ #
24
+ # * <tt>start_date</tt>: The start date. Defaults to 5 years ago from today.
25
+ # * <tt>end_date</tt>: The end date. Defaults to now.
26
+ def initialize(start_date=Time.now.years_ago(5), end_date=Time.now, fiscal_year_offset_month=10)
27
+ @start_date = start_date.to_date
28
+ @end_date = end_date.to_date
29
+ @fiscal_year_offset_month = fiscal_year_offset_month.to_i
30
+ @holiday_indicators = []
31
+ end
32
+
33
+ # Returns an array of hashes representing records in the dimension.
34
+ def build(options={})
35
+ (start_date..end_date).map { |date| record_from_date(date) }
36
+ end
37
+
38
+ private
39
+
40
+ # Returns a hash representing a record in the dimension. The values for each record are
41
+ # accessed by name.
42
+ def record_from_date(date)
43
+ time = date.to_time # need methods only available in Time
44
+ record = {}
45
+ record[:date] = time.strftime("%m/%d/%Y")
46
+ record[:full_date_description] = time.strftime("%B %d,%Y")
47
+ record[:day_of_week] = time.strftime("%A")
48
+ record[:day_in_week] = record[:day_of_week] # alias
49
+ #record[:day_number_in_epoch] = time.to_i / 24
50
+ #record[:week_number_in_epoch] = time.to_i / (24 * 7)
51
+ #record[:month_number_in_epoch] = time.to_i / (24 * 7 * 30)
52
+ record[:day_number_in_calendar_month] = time.day
53
+ record[:day_number_in_calendar_year] = time.yday
54
+ record[:day_number_in_fiscal_month] = time.day # should this be different from CY?
55
+ record[:day_number_in_fiscal_year] = time.fiscal_year_yday(fiscal_year_offset_month)
56
+ #record[:last_day_in_week_indicator] =
57
+ #record[:last_day_in_month_indicator] =
58
+ #record[:calendar_week_ending_date] =
59
+ record[:calendar_week] = "Week #{time.week}"
60
+ record[:calendar_week_number] = time.week
61
+ record[:calendar_week_number_in_year] = time.week # DEPRECATED
62
+ record[:calendar_month_name] = time.strftime("%B")
63
+ record[:calendar_month_number_in_year] = time.month # DEPRECATED
64
+ record[:calendar_month_number] = time.month
65
+ record[:calendar_year_month] = time.strftime("%Y-%m")
66
+ record[:calendar_quarter] = "Q#{time.quarter}"
67
+ record[:calendar_quarter_number] = time.quarter
68
+ record[:calendar_quarter_number_in_year] = time.quarter # DEPRECATED
69
+ record[:calendar_year_quarter] = "#{time.strftime('%Y')}-#{record[:calendar_quarter]}"
70
+ #record[:calendar_half_year] =
71
+ record[:calendar_year] = "#{time.year}"
72
+ record[:fiscal_week] = "FY Week #{time.fiscal_year_week(fiscal_year_offset_month)}"
73
+ record[:fiscal_week_number_in_year] = time.fiscal_year_week(fiscal_year_offset_month) # DEPRECATED
74
+ record[:fiscal_week_number] = time.fiscal_year_week(fiscal_year_offset_month)
75
+ record[:fiscal_month] = time.fiscal_year_month(fiscal_year_offset_month)
76
+ record[:fiscal_month_number] = time.fiscal_year_month(fiscal_year_offset_month)
77
+ record[:fiscal_month_number_in_year] = time.fiscal_year_month(fiscal_year_offset_month) # DEPRECATED
78
+ record[:fiscal_year_month] = "FY#{time.fiscal_year(fiscal_year_offset_month)}-" + time.fiscal_year_month(fiscal_year_offset_month).to_s.rjust(2, '0')
79
+ record[:fiscal_quarter] = "FY Q#{time.fiscal_year_quarter(fiscal_year_offset_month)}"
80
+ record[:fiscal_year_quarter] = "FY#{time.fiscal_year(fiscal_year_offset_month)}-Q#{time.fiscal_year_quarter(fiscal_year_offset_month)}"
81
+ record[:fiscal_quarter_number] = time.fiscal_year_quarter(fiscal_year_offset_month) # DEPRECATED
82
+ record[:fiscal_year_quarter_number] = time.fiscal_year_quarter(fiscal_year_offset_month)
83
+ #record[:fiscal_half_year] =
84
+ record[:fiscal_year] = "FY#{time.fiscal_year(fiscal_year_offset_month)}"
85
+ record[:fiscal_year_number] = time.fiscal_year(fiscal_year_offset_month)
86
+ record[:holiday_indicator] = holiday_indicators.include?(date) ? 'Holiday' : 'Nonholiday'
87
+ record[:weekday_indicator] = weekday_indicators[time.wday]
88
+ record[:selling_season] = 'None'
89
+ record[:major_event] = 'None'
90
+ record[:sql_date_stamp] = date
91
+
92
+ record
93
+ end
94
+ end
95
+ end
96
+ end
@@ -0,0 +1,31 @@
1
+ module ETL #:nodoc:
2
+ module Builder #:nodoc:
3
+ # Builder that creates a simple time dimension.
4
+ class TimeDimensionBuilder
5
+ def initialize
6
+ # Returns an array of hashes representing records in the dimension. The values for each record are
7
+ # accessed by name.
8
+ def build(options={})
9
+ records = []
10
+ 0.upto(23) do |t_hour|
11
+ 0.upto(59) do |t_minute|
12
+ 0.upto(59) do |t_second|
13
+ t_hour_string = t_hour.to_s.rjust(2, '0')
14
+ t_minute_string = t_minute.to_s.rjust(2, '0')
15
+ t_second_string = t_second.to_s.rjust(2, '0')
16
+ record = {}
17
+ record[:hour] = t_hour
18
+ record[:minute] = t_minute
19
+ record[:second] = t_second
20
+ record[:minute_description] = "#{t_hour_string}:#{t_minute_string}"
21
+ record[:full_description] = "#{t_hour_string}:#{t_minute_string}:#{t_second_string}"
22
+ records << record
23
+ end
24
+ end
25
+ end
26
+ records
27
+ end
28
+ end
29
+ end
30
+ end
31
+ end
@@ -0,0 +1,89 @@
1
+ #--
2
+ # Copyright (c) 2006 Anthony Eden
3
+ #
4
+ # Permission is hereby granted, free of charge, to any person obtaining
5
+ # a copy of this software and associated documentation files (the
6
+ # "Software"), to deal in the Software without restriction, including
7
+ # without limitation the rights to use, copy, modify, merge, publish,
8
+ # distribute, sublicense, and/or sell copies of the Software, and to
9
+ # permit persons to whom the Software is furnished to do so, subject to
10
+ # the following conditions:
11
+ #
12
+ # The above copyright notice and this permission notice shall be
13
+ # included in all copies or substantial portions of the Software.
14
+ #
15
+ # THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
16
+ # EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
17
+ # MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
18
+ # NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
19
+ # LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
20
+ # OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
21
+ # WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
22
+ #++
23
+
24
+ require 'benchmark'
25
+ require 'getoptlong'
26
+
27
+ # Print a usage statement
28
+ def usage #:nodoc:
29
+ puts "Usage: etl file [file file ...]" # TODO: add the command line options
30
+ end
31
+
32
+ def execute
33
+ opts = GetoptLong.new(
34
+ [ '--version', '-v', GetoptLong::NO_ARGUMENT],
35
+ [ '--help', '-h', GetoptLong::NO_ARGUMENT ],
36
+ [ '--config', '-c', GetoptLong::REQUIRED_ARGUMENT ],
37
+ [ '--limit', '-l', GetoptLong::REQUIRED_ARGUMENT ],
38
+ [ '--offset', '-o', GetoptLong::REQUIRED_ARGUMENT],
39
+ [ '--newlog', '-n', GetoptLong::NO_ARGUMENT ],
40
+ [ '--skip-bulk-import', '-s', GetoptLong::NO_ARGUMENT ],
41
+ [ '--read-locally', GetoptLong::NO_ARGUMENT],
42
+ [ '--rails-root', GetoptLong::REQUIRED_ARGUMENT]
43
+ )
44
+
45
+ options = {}
46
+ opts.each do |opt, arg|
47
+ case opt
48
+ when '--version'
49
+ puts "ActiveWarehouse ETL version #{ETL.version}"
50
+ return
51
+ when '--help'
52
+ usage
53
+ return
54
+ when '--config'
55
+ options[:config] = arg
56
+ when '--limit'
57
+ options[:limit] = arg.to_i
58
+ when '--offset'
59
+ options[:offset] = arg.to_i
60
+ when '--newlog'
61
+ options[:newlog] = true
62
+ when '--skip-bulk-import'
63
+ puts "skip bulk import enabled"
64
+ options[:skip_bulk_import] = true
65
+ when '--read-locally'
66
+ puts "read locally enabled"
67
+ options[:read_locally] = true
68
+ when '--rails-root'
69
+ options[:rails_root] = arg
70
+ puts "rails root set to #{options[:rails_root]}"
71
+ end
72
+ end
73
+
74
+ if ARGV.length < 1
75
+ usage
76
+ else
77
+ puts "Starting ETL process"
78
+
79
+ ETL::Engine.init(options)
80
+ ARGV.each do |f|
81
+ ETL::Engine.realtime_activity = true
82
+ ETL::Engine.process(f)
83
+ end
84
+
85
+ puts "ETL process complete\n\n"
86
+ end
87
+ end
88
+
89
+ execute
@@ -0,0 +1,3 @@
1
+ require 'etl/control/control'
2
+ require 'etl/control/source'
3
+ require 'etl/control/destination'
@@ -0,0 +1,414 @@
1
+ module ETL #:nodoc:
2
+ module Control #:nodoc:
3
+ # The Context is passed to eval.
4
+ class Context
5
+ require 'test/unit/assertions'
6
+ include Test::Unit::Assertions
7
+ attr_reader :control
8
+
9
+ class << self
10
+ # Create a Context instance
11
+ def create(control)
12
+ Context.new(control).get_binding
13
+ end
14
+ end
15
+
16
+ # Initialize the context
17
+ def initialize(control)
18
+ @control = control
19
+ end
20
+
21
+ # Get the control file
22
+ def file
23
+ control.file
24
+ end
25
+
26
+ # Set the allowed error threshold
27
+ def set_error_threshold(error_threshold)
28
+ control.error_threshold = error_threshold
29
+ end
30
+
31
+ # Define a list of control files that this file depends on. Those control
32
+ # files will be executed prior to this control file. The list may
33
+ # contain symbols that will be converted to file names by calling
34
+ # to_s + '.ctl', or they may be strings in which case they will be used
35
+ # as is
36
+ def depends_on(*args)
37
+ (dependencies << args).flatten!
38
+ end
39
+
40
+ # Get the defined dependencies
41
+ def dependencies
42
+ control.dependencies
43
+ end
44
+
45
+ def context(context_helper)
46
+ case context_helper
47
+ when String, Symbol
48
+ extend("#{context_helper.to_s.camelize}Context".constantize)
49
+ when Module
50
+ extend(context_helper)
51
+ end
52
+ end
53
+
54
+ # Define a source.
55
+ def source(name, configuration={}, definition={})
56
+ if configuration[:type]
57
+ case configuration[:type]
58
+ when Class
59
+ source_class = configuration[:type]
60
+ sources << source_class.new(self, configuration, definition)
61
+ when String, Symbol
62
+ source_class = ETL::Control::Source.class_for_name(configuration[:type])
63
+ sources << source_class.new(self, configuration, definition)
64
+ else
65
+ if configuration[:type].is_a?(ETL::Control::Source)
66
+ sources << configuration[:type]
67
+ else
68
+ raise ControlError, "Type must be a Class, String, Symbol or object extending ETL::Control::Source"
69
+ end
70
+ end
71
+ else
72
+ source_types.each do |source_type|
73
+ if configuration[source_type]
74
+ source_class = ETL::Control::Source.class_for_name(source_type)
75
+ sources << source_class.new(self, configuration, definition)
76
+ break
77
+ end
78
+ raise ControlError, "A source was specified but no matching type was found"
79
+ end
80
+ end
81
+ end
82
+
83
+ # Get the defined source
84
+ def sources
85
+ control.sources
86
+ end
87
+
88
+ # Define a destination
89
+ def destination(name, configuration={}, mapping={})
90
+ if configuration[:type]
91
+ case configuration[:type]
92
+ when Class
93
+ dest_class = configuration[:type]
94
+ destinations << dest_class.new(self, configuration, mapping)
95
+ when String, Symbol
96
+ dest_class = ETL::Control::Destination.class_for_name(configuration[:type])
97
+ destinations << dest_class.new(self, configuration, mapping)
98
+ else
99
+ if configuration[:type].is_a?(ETL::Control::Destination)
100
+ destinations << configuration[:type]
101
+ else
102
+ raise ControlError, "Type must be a Class, String, Symbol or object extending ETL::Control::Destination"
103
+ end
104
+ end
105
+ else
106
+ destination_types.each do |dest_type|
107
+ if configuration[dest_type]
108
+ dest_class = ETL::Control::Destination.class_for_name(dest_type)
109
+ destinations << dest_class.new(self, configuration, mapping)
110
+ break
111
+ end
112
+ raise ControlError, "A destination was specified but no matching destination type was found"
113
+ end
114
+ end
115
+ end
116
+
117
+ # Get the defined destinations
118
+ def destinations
119
+ control.destinations
120
+ end
121
+
122
+ # Define a transform
123
+ def transform(name, transformer=nil, configuration={}, &block)
124
+ if transformer
125
+ case transformer
126
+ when String, Symbol
127
+ class_name = "#{transformer.to_s.camelize}Transform"
128
+ begin
129
+ transform_class = ETL::Transform.const_get(class_name)
130
+ transforms << transform_class.new(self, name, configuration)
131
+ rescue NameError => e
132
+ raise ControlError, "Unable to find transformer #{class_name}: #{e}"
133
+ end
134
+ when Class
135
+ transforms << transformer.new(self, transformer.name, configuration)
136
+ else
137
+ #transformer.class.inspect
138
+ if transformer.is_a?(ETL::Transform::Transform)
139
+ Engine.logger.debug "Adding transformer #{transformer.inspect} for field #{name}"
140
+ t = transformer.dup
141
+ t.name = name
142
+ transforms << t
143
+ else
144
+ raise ControlError, "Transformer must be a String, Symbol, Class or Transform instance"
145
+ end
146
+ end
147
+ elsif block_given?
148
+ transforms << ETL::Transform::BlockTransform.new(self, name, :block => block)
149
+ else
150
+ raise ControlError, "Either a transformer or a block must be specified"
151
+ end
152
+ end
153
+
154
+ # Get the defined transforms
155
+ def transforms
156
+ control.transforms
157
+ end
158
+
159
+ # Define a before post-process screen block. The type argument must be
160
+ # one of :fatal, :error or :warn
161
+ def screen(type, &block)
162
+ screens[type] << block
163
+ end
164
+
165
+ # Get the before post-process screen blocks
166
+ def screens
167
+ control.screens
168
+ end
169
+
170
+ # Define an after post-proces screen block. The type argument must be
171
+ # one of :fatal, :error or :warn
172
+ def after_post_process_screen(type, &block)
173
+ after_post_process_screens[type] << block
174
+ end
175
+
176
+ # Get the after post-process screen blocks
177
+ def after_post_process_screens
178
+ control.after_post_process_screens
179
+ end
180
+
181
+ # Rename the source field to the destination field
182
+ def rename(source, destination)
183
+ after_read :rename, :source => source, :dest => destination
184
+ end
185
+
186
+ # Copy the source field to the destination field
187
+ def copy(source, destination)
188
+ after_read :copy_field, :source => source, :dest => destination
189
+ end
190
+
191
+ protected
192
+ # This method is used to define a processor and insert into the specified processor
193
+ # collection.
194
+ def define_processor(name, processor_collection, configuration, proc)
195
+ case name
196
+ when String, Symbol, nil
197
+ name ||= 'block'
198
+ class_name = "#{name.to_s.camelize}Processor"
199
+ begin
200
+ processor_class = ETL::Processor.const_get(class_name)
201
+ if name == 'block'
202
+ raise ControlError, "A block must be passed for block processor" if proc.nil?
203
+ configuration[:block] = proc
204
+ end
205
+ processor_collection << processor_class.new(self, configuration)
206
+ rescue NameError => e
207
+ raise ControlError, "Unable to find processor #{class_name}: #{e}"
208
+ end
209
+ when Class
210
+ processor_collection << name.new(self, configuration)
211
+ else
212
+ raise ControlError, "The process declaration requires a String, Symbol, Class, or a Block to be passed"
213
+ end
214
+ end
215
+
216
+ public
217
+ # Define an "after read" processor. This must be a row-level processor.
218
+ def after_read(name='block', configuration={}, &block)
219
+ define_processor(name, after_read_processors, configuration, block)
220
+ end
221
+
222
+ # Get the defined "after read" processors
223
+ def after_read_processors
224
+ control.after_read_processors
225
+ end
226
+
227
+ # Define a "before write" processor. This must be a row-level processor.
228
+ def before_write(name='block', configuration={}, &block)
229
+ define_processor(name, before_write_processors, configuration, block)
230
+ end
231
+
232
+ # Get the defined "before write" processors
233
+ def before_write_processors
234
+ control.before_write_processors
235
+ end
236
+
237
+ # Define a pre-processor
238
+ def pre_process(name='block', configuration={}, &block)
239
+ define_processor(name, pre_processors, configuration, block)
240
+ end
241
+
242
+ # Get the defined pre-processors
243
+ def pre_processors
244
+ control.pre_processors
245
+ end
246
+
247
+ # Define a post-processor
248
+ def post_process(name='block', configuration={}, &block)
249
+ define_processor(name, post_processors, configuration, block)
250
+ end
251
+
252
+ # Get the defined post-processors
253
+ def post_processors
254
+ control.post_processors
255
+ end
256
+
257
+ # Get the binding object
258
+ def get_binding
259
+ binding
260
+ end
261
+
262
+ protected
263
+ # Get an array of supported source types
264
+ def source_types
265
+ control.source_types
266
+ end
267
+
268
+ # Get an array of supported destination types
269
+ def destination_types
270
+ control.destination_types
271
+ end
272
+
273
+ end
274
+
275
+ # Object representation of a control file
276
+ class Control
277
+ # The File object
278
+ attr_reader :file
279
+
280
+ # The error threshold
281
+ attr_accessor :error_threshold
282
+
283
+ class << self
284
+ # Parse a control file and return a Control instance
285
+ def parse(control_file)
286
+ control_file = control_file.path if control_file.instance_of?(File)
287
+ control = ETL::Control::Control.new(control_file)
288
+ # TODO: better handling of parser errors. Return the line in the control file where the error occurs.
289
+ eval(IO.readlines(control_file).join("\n"), Context.create(control), control_file)
290
+ control.validate
291
+ control
292
+ end
293
+
294
+ def parse_text(text)
295
+ control = ETL::Control::Control.new(nil)
296
+ eval(text, Context.create(control), 'inline')
297
+ control.validate
298
+ control
299
+ end
300
+
301
+ # Resolve the given object to an ETL::Control::Control instance. Acceptable arguments
302
+ # are:
303
+ # * The path to a control file as a String
304
+ # * A File object referencing the control file
305
+ # * The ETL::Control::Control object (which will just be returned)
306
+ #
307
+ # Raises a ControlError if any other type is given
308
+ def resolve(control)
309
+ case control
310
+ when String
311
+ ETL::Control::Control.parse(File.new(control))
312
+ when File
313
+ ETL::Control::Control.parse(control)
314
+ when ETL::Control::Control
315
+ control
316
+ else
317
+ raise ControlError, "Control must be a String, File or Control object"
318
+ end
319
+ end
320
+ end
321
+
322
+ # Initialize the instance with the given File object
323
+ def initialize(file)
324
+ @file = file
325
+ end
326
+
327
+ # Get a list of dependencies
328
+ def dependencies
329
+ @dependencies ||= []
330
+ end
331
+
332
+ # Get the defined source
333
+ def sources
334
+ @sources ||= []
335
+ end
336
+
337
+ # Get the defined destinations
338
+ def destinations
339
+ @destinations ||= []
340
+ end
341
+
342
+ # Get the transforms with the specified name
343
+ # def transform(name)
344
+ # transforms[name] ||= []
345
+ # end
346
+
347
+ def after_read_processors
348
+ @after_read_processors ||= []
349
+ end
350
+
351
+ # Get all of the "before write" processors
352
+ def before_write_processors
353
+ @before_write_processors ||= []
354
+ end
355
+
356
+ # Get an Array of preprocessors
357
+ def pre_processors
358
+ @pre_processors ||= []
359
+ end
360
+
361
+ # Get an Array of post processors
362
+ def post_processors
363
+ @post_processors ||= []
364
+ end
365
+
366
+ # Get an Array of all transforms for this control
367
+ def transforms
368
+ @transforms ||= []
369
+ end
370
+
371
+ # A hash of the screens executed before post-process
372
+ def screens
373
+ @screens ||= {
374
+ :fatal => [],
375
+ :error => [],
376
+ :warn => []
377
+ }
378
+ end
379
+
380
+ # A hash of the screens executed after post-process
381
+ def after_post_process_screens
382
+ @after_post_process_screens ||= {
383
+ :fatal => [],
384
+ :error => [],
385
+ :warn => []
386
+ }
387
+ end
388
+
389
+ # Get the error threshold. Defaults to 100.
390
+ def error_threshold
391
+ @error_threshold ||= 100
392
+ end
393
+
394
+ # Validate the control file
395
+ def validate
396
+ #unless sources.length > 0
397
+ # raise ControlError, "Configuration must include one of the following for the source: #{source_types.join(',')}"
398
+ #end
399
+ #unless destinations.length > 0
400
+ # raise ControlError, "Configuration must include one of the following for the destination: #{destination_types.join(',')}"
401
+ #end
402
+ end
403
+
404
+ def source_types
405
+ [:file, :database]
406
+ end
407
+
408
+ def destination_types
409
+ [:file, :database]
410
+ end
411
+
412
+ end
413
+ end
414
+ end