wukong 0.1.1
Sign up to get free protection for your applications and to get access to all the features.
- data/LICENSE.textile +107 -0
- data/README.textile +166 -0
- data/bin/cutc +30 -0
- data/bin/cuttab +5 -0
- data/bin/greptrue +8 -0
- data/bin/hdp-cat +3 -0
- data/bin/hdp-catd +3 -0
- data/bin/hdp-du +81 -0
- data/bin/hdp-get +3 -0
- data/bin/hdp-kill +3 -0
- data/bin/hdp-ls +10 -0
- data/bin/hdp-mkdir +3 -0
- data/bin/hdp-mv +3 -0
- data/bin/hdp-parts_to_keys.rb +77 -0
- data/bin/hdp-ps +3 -0
- data/bin/hdp-put +3 -0
- data/bin/hdp-rm +11 -0
- data/bin/hdp-sort +29 -0
- data/bin/hdp-stream +29 -0
- data/bin/hdp-stream-flat +18 -0
- data/bin/hdp-sync +17 -0
- data/bin/hdp-wc +67 -0
- data/bin/md5sort +20 -0
- data/bin/tabchar +5 -0
- data/bin/uniqc +3 -0
- data/bin/wu-hist +3 -0
- data/bin/wu-lign +177 -0
- data/bin/wu-sum +30 -0
- data/doc/INSTALL.textile +41 -0
- data/doc/LICENSE.textile +107 -0
- data/doc/README-tutorial.textile +163 -0
- data/doc/README-wulign.textile +59 -0
- data/doc/README-wutils.textile +128 -0
- data/doc/TODO.textile +61 -0
- data/doc/UsingWukong-part1-setup.textile +2 -0
- data/doc/UsingWukong-part2-scraping.textile +2 -0
- data/doc/UsingWukong-part3-parsing.textile +132 -0
- data/doc/code/api_response_example.txt +20 -0
- data/doc/code/parser_skeleton.rb +38 -0
- data/doc/hadoop-nfs.textile +51 -0
- data/doc/hadoop-setup.textile +29 -0
- data/doc/index.textile +124 -0
- data/doc/intro_to_map_reduce/MapReduceDiagram.graffle +0 -0
- data/doc/links.textile +42 -0
- data/doc/overview.textile +91 -0
- data/doc/pig/PigLatinExpressionsList.txt +122 -0
- data/doc/pig/PigLatinReferenceManual.html +19134 -0
- data/doc/pig/PigLatinReferenceManual.txt +1640 -0
- data/doc/tips.textile +116 -0
- data/doc/usage.textile +102 -0
- data/doc/utils.textile +48 -0
- data/examples/README.txt +17 -0
- data/examples/and_pig/sample_queries.rb +128 -0
- data/examples/apache_log_parser.rb +53 -0
- data/examples/count_keys.rb +56 -0
- data/examples/count_keys_at_mapper.rb +57 -0
- data/examples/graph/adjacency_list.rb +74 -0
- data/examples/graph/breadth_first_search.rb +79 -0
- data/examples/graph/gen_2paths.rb +68 -0
- data/examples/graph/gen_multi_edge.rb +103 -0
- data/examples/graph/gen_symmetric_links.rb +53 -0
- data/examples/package-local.rb +100 -0
- data/examples/package.rb +96 -0
- data/examples/pagerank/README.textile +6 -0
- data/examples/pagerank/gen_initial_pagerank_graph.pig +57 -0
- data/examples/pagerank/pagerank.rb +88 -0
- data/examples/pagerank/pagerank_initialize.rb +46 -0
- data/examples/pagerank/run_pagerank.sh +19 -0
- data/examples/rank_and_bin.rb +173 -0
- data/examples/run_all.sh +47 -0
- data/examples/sample_records.rb +44 -0
- data/examples/size.rb +60 -0
- data/examples/word_count.rb +95 -0
- data/lib/wukong.rb +11 -0
- data/lib/wukong/and_pig.rb +62 -0
- data/lib/wukong/and_pig/README.textile +12 -0
- data/lib/wukong/and_pig/as.rb +37 -0
- data/lib/wukong/and_pig/data_types.rb +30 -0
- data/lib/wukong/and_pig/functions.rb +50 -0
- data/lib/wukong/and_pig/generate.rb +85 -0
- data/lib/wukong/and_pig/generate/variable_inflections.rb +82 -0
- data/lib/wukong/and_pig/junk.rb +51 -0
- data/lib/wukong/and_pig/operators.rb +8 -0
- data/lib/wukong/and_pig/operators/compound.rb +29 -0
- data/lib/wukong/and_pig/operators/evaluators.rb +7 -0
- data/lib/wukong/and_pig/operators/execution.rb +15 -0
- data/lib/wukong/and_pig/operators/file_methods.rb +29 -0
- data/lib/wukong/and_pig/operators/foreach.rb +98 -0
- data/lib/wukong/and_pig/operators/groupies.rb +212 -0
- data/lib/wukong/and_pig/operators/load_store.rb +65 -0
- data/lib/wukong/and_pig/operators/meta.rb +42 -0
- data/lib/wukong/and_pig/operators/relational.rb +129 -0
- data/lib/wukong/and_pig/pig_struct.rb +48 -0
- data/lib/wukong/and_pig/pig_var.rb +95 -0
- data/lib/wukong/and_pig/symbol.rb +29 -0
- data/lib/wukong/and_pig/utils.rb +0 -0
- data/lib/wukong/bad_record.rb +18 -0
- data/lib/wukong/boot.rb +47 -0
- data/lib/wukong/datatypes.rb +24 -0
- data/lib/wukong/datatypes/enum.rb +123 -0
- data/lib/wukong/dfs.rb +80 -0
- data/lib/wukong/encoding.rb +111 -0
- data/lib/wukong/extensions.rb +15 -0
- data/lib/wukong/extensions/array.rb +18 -0
- data/lib/wukong/extensions/blank.rb +93 -0
- data/lib/wukong/extensions/class.rb +189 -0
- data/lib/wukong/extensions/date_time.rb +24 -0
- data/lib/wukong/extensions/emittable.rb +82 -0
- data/lib/wukong/extensions/hash.rb +120 -0
- data/lib/wukong/extensions/hash_like.rb +119 -0
- data/lib/wukong/extensions/hashlike_class.rb +47 -0
- data/lib/wukong/extensions/module.rb +2 -0
- data/lib/wukong/extensions/pathname.rb +27 -0
- data/lib/wukong/extensions/string.rb +65 -0
- data/lib/wukong/extensions/struct.rb +17 -0
- data/lib/wukong/extensions/symbol.rb +11 -0
- data/lib/wukong/logger.rb +53 -0
- data/lib/wukong/models/graph.rb +27 -0
- data/lib/wukong/rdf.rb +104 -0
- data/lib/wukong/schema.rb +37 -0
- data/lib/wukong/script.rb +265 -0
- data/lib/wukong/script/hadoop_command.rb +111 -0
- data/lib/wukong/script/local_command.rb +14 -0
- data/lib/wukong/streamer.rb +13 -0
- data/lib/wukong/streamer/accumulating_reducer.rb +89 -0
- data/lib/wukong/streamer/base.rb +76 -0
- data/lib/wukong/streamer/count_keys.rb +30 -0
- data/lib/wukong/streamer/count_lines.rb +26 -0
- data/lib/wukong/streamer/filter.rb +20 -0
- data/lib/wukong/streamer/line_streamer.rb +12 -0
- data/lib/wukong/streamer/list_reducer.rb +20 -0
- data/lib/wukong/streamer/preprocess_with_pipe_streamer.rb +22 -0
- data/lib/wukong/streamer/rank_and_bin_reducer.rb +145 -0
- data/lib/wukong/streamer/set_reducer.rb +14 -0
- data/lib/wukong/streamer/struct_streamer.rb +48 -0
- data/lib/wukong/streamer/summing_reducer.rb +29 -0
- data/lib/wukong/streamer/uniq_by_last_reducer.rb +44 -0
- data/lib/wukong/typed_struct.rb +12 -0
- data/lib/wukong/wukong_class.rb +21 -0
- data/spec/bin/hdp-wc_spec.rb +4 -0
- data/spec/spec_helper.rb +0 -0
- data/wukong.gemspec +179 -0
- metadata +214 -0
@@ -0,0 +1,189 @@
|
|
1
|
+
#
|
2
|
+
# This is taken in whole from the extlib gem. Thanks y'all.
|
3
|
+
#
|
4
|
+
|
5
|
+
|
6
|
+
# Copyright (c) 2004-2008 David Heinemeier Hansson
|
7
|
+
#
|
8
|
+
# Permission is hereby granted, free of charge, to any person obtaining
|
9
|
+
# a copy of this software and associated documentation files (the
|
10
|
+
# "Software"), to deal in the Software without restriction, including
|
11
|
+
# without limitation the rights to use, copy, modify, merge, publish,
|
12
|
+
# distribute, sublicense, and/or sell copies of the Software, and to
|
13
|
+
# permit persons to whom the Software is furnished to do so, subject to
|
14
|
+
# the following conditions:
|
15
|
+
#
|
16
|
+
# The above copyright notice and this permission notice shall be
|
17
|
+
# included in all copies or substantial portions of the Software.
|
18
|
+
#
|
19
|
+
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
|
20
|
+
# EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
|
21
|
+
# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
|
22
|
+
# NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
|
23
|
+
# LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
|
24
|
+
# OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
|
25
|
+
# WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
26
|
+
|
27
|
+
# Allows attributes to be shared within an inheritance hierarchy, but where
|
28
|
+
# each descendant gets a copy of their parents' attributes, instead of just a
|
29
|
+
# pointer to the same. This means that the child can add elements to, for
|
30
|
+
# example, an array without those additions being shared with either their
|
31
|
+
# parent, siblings, or children, which is unlike the regular class-level
|
32
|
+
# attributes that are shared across the entire hierarchy.
|
33
|
+
class Class
|
34
|
+
# Defines class-level and instance-level attribute reader.
|
35
|
+
#
|
36
|
+
# @param *syms<Array> Array of attributes to define reader for.
|
37
|
+
# @return <Array[#to_s]> List of attributes that were made into cattr_readers
|
38
|
+
#
|
39
|
+
# @api public
|
40
|
+
#
|
41
|
+
# @todo Is this inconsistent in that it does not allow you to prevent
|
42
|
+
# an instance_reader via :instance_reader => false
|
43
|
+
def cattr_reader(*syms)
|
44
|
+
syms.flatten.each do |sym|
|
45
|
+
next if sym.is_a?(Hash)
|
46
|
+
class_eval(<<-RUBY, __FILE__, __LINE__ + 1)
|
47
|
+
unless defined? @@#{sym}
|
48
|
+
@@#{sym} = nil
|
49
|
+
end
|
50
|
+
|
51
|
+
def self.#{sym}
|
52
|
+
@@#{sym}
|
53
|
+
end
|
54
|
+
|
55
|
+
def #{sym}
|
56
|
+
@@#{sym}
|
57
|
+
end
|
58
|
+
RUBY
|
59
|
+
end
|
60
|
+
end
|
61
|
+
|
62
|
+
# Defines class-level (and optionally instance-level) attribute writer.
|
63
|
+
#
|
64
|
+
# @param <Array[*#to_s, Hash{:instance_writer => Boolean}]> Array of attributes to define writer for.
|
65
|
+
# @option syms :instance_writer<Boolean> if true, instance-level attribute writer is defined.
|
66
|
+
# @return <Array[#to_s]> List of attributes that were made into cattr_writers
|
67
|
+
#
|
68
|
+
# @api public
|
69
|
+
def cattr_writer(*syms)
|
70
|
+
options = syms.last.is_a?(Hash) ? syms.pop : {}
|
71
|
+
syms.flatten.each do |sym|
|
72
|
+
class_eval(<<-RUBY, __FILE__, __LINE__ + 1)
|
73
|
+
unless defined? @@#{sym}
|
74
|
+
@@#{sym} = nil
|
75
|
+
end
|
76
|
+
|
77
|
+
def self.#{sym}=(obj)
|
78
|
+
@@#{sym} = obj
|
79
|
+
end
|
80
|
+
RUBY
|
81
|
+
|
82
|
+
unless options[:instance_writer] == false
|
83
|
+
class_eval(<<-RUBY, __FILE__, __LINE__ + 1)
|
84
|
+
def #{sym}=(obj)
|
85
|
+
@@#{sym} = obj
|
86
|
+
end
|
87
|
+
RUBY
|
88
|
+
end
|
89
|
+
end
|
90
|
+
end
|
91
|
+
|
92
|
+
# Defines class-level (and optionally instance-level) attribute accessor.
|
93
|
+
#
|
94
|
+
# @param *syms<Array[*#to_s, Hash{:instance_writer => Boolean}]> Array of attributes to define accessor for.
|
95
|
+
# @option syms :instance_writer<Boolean> if true, instance-level attribute writer is defined.
|
96
|
+
# @return <Array[#to_s]> List of attributes that were made into accessors
|
97
|
+
#
|
98
|
+
# @api public
|
99
|
+
def cattr_accessor(*syms)
|
100
|
+
cattr_reader(*syms)
|
101
|
+
cattr_writer(*syms)
|
102
|
+
end
|
103
|
+
|
104
|
+
# Defines class-level inheritable attribute reader. Attributes are available to subclasses,
|
105
|
+
# each subclass has a copy of parent's attribute.
|
106
|
+
#
|
107
|
+
# @param *syms<Array[#to_s]> Array of attributes to define inheritable reader for.
|
108
|
+
# @return <Array[#to_s]> Array of attributes converted into inheritable_readers.
|
109
|
+
#
|
110
|
+
# @api public
|
111
|
+
#
|
112
|
+
# @todo Do we want to block instance_reader via :instance_reader => false
|
113
|
+
# @todo It would be preferable that we do something with a Hash passed in
|
114
|
+
# (error out or do the same as other methods above) instead of silently
|
115
|
+
# moving on). In particular, this makes the return value of this function
|
116
|
+
# less useful.
|
117
|
+
def class_inheritable_reader(*ivars)
|
118
|
+
instance_reader = ivars.pop[:reader] if ivars.last.is_a?(Hash)
|
119
|
+
|
120
|
+
ivars.each do |ivar|
|
121
|
+
self.class_eval <<-RUBY, __FILE__, __LINE__ + 1
|
122
|
+
def self.#{ivar}
|
123
|
+
return @#{ivar} if defined?(@#{ivar})
|
124
|
+
return nil if self.object_id == #{self.object_id}
|
125
|
+
ivar = superclass.#{ivar}
|
126
|
+
return nil if ivar.nil?
|
127
|
+
@#{ivar} = ivar.try_dup
|
128
|
+
end
|
129
|
+
RUBY
|
130
|
+
|
131
|
+
unless instance_reader == false
|
132
|
+
self.class_eval <<-RUBY, __FILE__, __LINE__ + 1
|
133
|
+
def #{ivar}
|
134
|
+
self.class.#{ivar}
|
135
|
+
end
|
136
|
+
RUBY
|
137
|
+
end
|
138
|
+
end
|
139
|
+
end
|
140
|
+
|
141
|
+
# Defines class-level inheritable attribute writer. Attributes are available to subclasses,
|
142
|
+
# each subclass has a copy of parent's attribute.
|
143
|
+
#
|
144
|
+
# @param *syms<Array[*#to_s, Hash{:instance_writer => Boolean}]> Array of attributes to
|
145
|
+
# define inheritable writer for.
|
146
|
+
# @option syms :instance_writer<Boolean> if true, instance-level inheritable attribute writer is defined.
|
147
|
+
# @return <Array[#to_s]> An Array of the attributes that were made into inheritable writers.
|
148
|
+
#
|
149
|
+
# @api public
|
150
|
+
#
|
151
|
+
# @todo We need a style for class_eval <<-HEREDOC. I'd like to make it
|
152
|
+
# class_eval(<<-RUBY, __FILE__, __LINE__), but we should codify it somewhere.
|
153
|
+
def class_inheritable_writer(*ivars)
|
154
|
+
instance_writer = ivars.pop[:instance_writer] if ivars.last.is_a?(Hash)
|
155
|
+
ivars.each do |ivar|
|
156
|
+
self.class_eval <<-RUBY, __FILE__, __LINE__ + 1
|
157
|
+
def self.#{ivar}=(obj)
|
158
|
+
@#{ivar} = obj
|
159
|
+
end
|
160
|
+
RUBY
|
161
|
+
unless instance_writer == false
|
162
|
+
self.class_eval <<-RUBY, __FILE__, __LINE__ + 1
|
163
|
+
def #{ivar}=(obj) self.class.#{ivar} = obj end
|
164
|
+
RUBY
|
165
|
+
end
|
166
|
+
end
|
167
|
+
end
|
168
|
+
|
169
|
+
# Defines class-level inheritable attribute accessor. Attributes are available to subclasses,
|
170
|
+
# each subclass has a copy of parent's attribute.
|
171
|
+
#
|
172
|
+
# @param *syms<Array[*#to_s, Hash{:instance_writer => Boolean}]> Array of attributes to
|
173
|
+
# define inheritable accessor for.
|
174
|
+
# @option syms :instance_writer<Boolean> if true, instance-level inheritable attribute writer is defined.
|
175
|
+
# @return <Array[#to_s]> An Array of attributes turned into inheritable accessors.
|
176
|
+
#
|
177
|
+
# @api public
|
178
|
+
def class_inheritable_accessor(*syms)
|
179
|
+
class_inheritable_reader(*syms)
|
180
|
+
class_inheritable_writer(*syms)
|
181
|
+
end
|
182
|
+
end
|
183
|
+
|
184
|
+
|
185
|
+
class Fixnum
|
186
|
+
def try_dup
|
187
|
+
self
|
188
|
+
end
|
189
|
+
end
|
@@ -0,0 +1,24 @@
|
|
1
|
+
require 'time'
|
2
|
+
DateTime.class_eval do
|
3
|
+
#
|
4
|
+
# Parses the time but never fails.
|
5
|
+
#
|
6
|
+
# A flattened time -- a 12-digit YYYYmmddHHMMMSS -- is treated as a UTC
|
7
|
+
# datetime.
|
8
|
+
#
|
9
|
+
def self.parse_safely dt
|
10
|
+
begin
|
11
|
+
if dt.to_s =~ /\A\d{12}Z?\z/
|
12
|
+
parse(dt+'Z', true)
|
13
|
+
else
|
14
|
+
parse(dt, true)
|
15
|
+
end
|
16
|
+
rescue
|
17
|
+
nil
|
18
|
+
end
|
19
|
+
end
|
20
|
+
|
21
|
+
def self.parse_and_flatten str
|
22
|
+
parse_safely(str).to_flat
|
23
|
+
end
|
24
|
+
end
|
@@ -0,0 +1,82 @@
|
|
1
|
+
|
2
|
+
Object.class_eval do
|
3
|
+
def to_flat() [to_s] end
|
4
|
+
end
|
5
|
+
|
6
|
+
module Enumerable
|
7
|
+
alias_method :to_flat, :to_a
|
8
|
+
end
|
9
|
+
|
10
|
+
Struct.class_eval do
|
11
|
+
#
|
12
|
+
# The last portion of the class in underscored form
|
13
|
+
# note memoization
|
14
|
+
#
|
15
|
+
def self.resource_name
|
16
|
+
@resource_name ||= self.to_s.gsub(%r{.*::}, '').underscore.to_sym
|
17
|
+
end
|
18
|
+
#
|
19
|
+
# Flatten for packing as resource name followed by all fields
|
20
|
+
#
|
21
|
+
def to_flat include_key=true
|
22
|
+
if include_key.is_a? Proc
|
23
|
+
sort_key = include_key.call(self)
|
24
|
+
elsif (! include_key.blank?) && respond_to?(:key)
|
25
|
+
sort_key = [self.class.resource_name, key].flatten.join("-")
|
26
|
+
else
|
27
|
+
sort_key = self.class.resource_name
|
28
|
+
end
|
29
|
+
[sort_key, *to_a] # .map(&:to_flat).flatten
|
30
|
+
end
|
31
|
+
end
|
32
|
+
|
33
|
+
module HashLike
|
34
|
+
#
|
35
|
+
# Flatten for packing as resource name followed by all fields
|
36
|
+
#
|
37
|
+
def to_flat include_key=true
|
38
|
+
if include_key.is_a? Proc
|
39
|
+
sort_key = include_key.call(self)
|
40
|
+
elsif include_key && respond_to?(:key)
|
41
|
+
sort_key = [self.class.resource_name, key].flatten.join("-")
|
42
|
+
else
|
43
|
+
sort_key = self.class.resource_name
|
44
|
+
end
|
45
|
+
[sort_key, *to_a] # .map(&:to_flat).flatten
|
46
|
+
end
|
47
|
+
end
|
48
|
+
|
49
|
+
Hash.class_eval do
|
50
|
+
def to_flat
|
51
|
+
map do |k, v|
|
52
|
+
[k.to_flat, v.to_flat].join(":")
|
53
|
+
end
|
54
|
+
end
|
55
|
+
end
|
56
|
+
|
57
|
+
class Time
|
58
|
+
# strftime() format to flatten a date
|
59
|
+
FLAT_FORMAT = "%Y%m%d%H%M%S"
|
60
|
+
# Flatten
|
61
|
+
def to_flat
|
62
|
+
strftime(FLAT_FORMAT)
|
63
|
+
end
|
64
|
+
end
|
65
|
+
|
66
|
+
class Date
|
67
|
+
# strftime() format to flatten a date
|
68
|
+
FLAT_FORMAT = "%Y%m%d"
|
69
|
+
# Flatten
|
70
|
+
def to_flat
|
71
|
+
strftime(FLAT_FORMAT)
|
72
|
+
end
|
73
|
+
end
|
74
|
+
|
75
|
+
class DateTime < Date
|
76
|
+
# strftime() format to flatten a date
|
77
|
+
FLAT_FORMAT = "%Y%m%d%H%M%S"
|
78
|
+
# Flatten
|
79
|
+
def to_flat
|
80
|
+
strftime(FLAT_FORMAT)
|
81
|
+
end
|
82
|
+
end
|
@@ -0,0 +1,120 @@
|
|
1
|
+
#
|
2
|
+
# h2. extensions/hash.rb -- hash extensions
|
3
|
+
#
|
4
|
+
|
5
|
+
require 'set'
|
6
|
+
class Hash
|
7
|
+
|
8
|
+
# Slice a hash to include only the given keys. This is useful for
|
9
|
+
# limiting an options hash to valid keys before passing to a method:
|
10
|
+
#
|
11
|
+
# def search(criteria = {})
|
12
|
+
# assert_valid_keys(:mass, :velocity, :time)
|
13
|
+
# end
|
14
|
+
#
|
15
|
+
# search(options.slice(:mass, :velocity, :time))
|
16
|
+
# Returns a new hash with only the given keys.
|
17
|
+
def slice(*keys)
|
18
|
+
allowed = Set.new(respond_to?(:convert_key) ? keys.map { |key| convert_key(key) } : keys)
|
19
|
+
reject{|key,| !allowed.include?(key) }
|
20
|
+
end
|
21
|
+
#
|
22
|
+
# Replace the hash with only the given keys.
|
23
|
+
#
|
24
|
+
def slice!(*keys)
|
25
|
+
replace(slice(*keys))
|
26
|
+
end
|
27
|
+
#
|
28
|
+
# #values_of is an alias for #values_at, but can be called on a Hash, a
|
29
|
+
# Struct, or an instance of a class that includes HashLike
|
30
|
+
#
|
31
|
+
alias_method :values_of, :values_at
|
32
|
+
|
33
|
+
#
|
34
|
+
# Create a hash from an array of keys and corresponding values.
|
35
|
+
#
|
36
|
+
def self.zip(keys, values, default=nil, &block)
|
37
|
+
hash = block_given? ? Hash.new(&block) : Hash.new(default)
|
38
|
+
keys.zip(values){|key,val| hash[key]=val }
|
39
|
+
hash
|
40
|
+
end
|
41
|
+
|
42
|
+
# lambda for recursive merges
|
43
|
+
Hash::DEEP_MERGER = proc do |key,v1,v2|
|
44
|
+
(v1.respond_to?(:merge) && v2.respond_to?(:merge)) ? v1.merge(v2.compact, &Hash::DEEP_MERGER) : (v2.nil? ? v1 : v2)
|
45
|
+
end
|
46
|
+
|
47
|
+
#
|
48
|
+
# Merge hashes recursively.
|
49
|
+
# Nothing special happens to array values
|
50
|
+
#
|
51
|
+
# x = { :subhash => { 1 => :val_from_x, 222 => :only_in_x, 333 => :only_in_x }, :scalar => :scalar_from_x}
|
52
|
+
# y = { :subhash => { 1 => :val_from_y, 999 => :only_in_y }, :scalar => :scalar_from_y }
|
53
|
+
# x.deep_merge y
|
54
|
+
# => {:subhash=>{1=>:val_from_y, 222=>:only_in_x, 333=>:only_in_x, 999=>:only_in_y}, :scalar=>:scalar_from_y}
|
55
|
+
# y.deep_merge x
|
56
|
+
# => {:subhash=>{1=>:val_from_x, 222=>:only_in_x, 333=>:only_in_x, 999=>:only_in_y}, :scalar=>:scalar_from_x}
|
57
|
+
#
|
58
|
+
# Nil values always lose.
|
59
|
+
#
|
60
|
+
# x = {:subhash=>{:nil_in_x=>nil, 1=>:val1,}, :nil_in_x=>nil}
|
61
|
+
# y = {:subhash=>{:nil_in_x=>5}, :nil_in_x=>5}
|
62
|
+
# y.deep_merge x
|
63
|
+
# => {:subhash=>{1=>:val1, :nil_in_x=>5}, :nil_in_x=>5}
|
64
|
+
# x.deep_merge y
|
65
|
+
# => {:subhash=>{1=>:val1, :nil_in_x=>5}, :nil_in_x=>5}
|
66
|
+
#
|
67
|
+
def deep_merge hsh2
|
68
|
+
merge hsh2, &Hash::DEEP_MERGER
|
69
|
+
end
|
70
|
+
|
71
|
+
def deep_merge! hsh2
|
72
|
+
merge! hsh2, &Hash::DEEP_MERGER
|
73
|
+
end
|
74
|
+
|
75
|
+
|
76
|
+
#
|
77
|
+
# Treat hash as tree of hashes:
|
78
|
+
#
|
79
|
+
# x = { 1 => :val, :subhash => { 1 => :val1 } }
|
80
|
+
# x.deep_set(:subhash, 3, 4)
|
81
|
+
# # => { 1 => :val, :subhash => { 1 => :val1, 3 => 4 } }
|
82
|
+
# x.deep_set(:subhash, 1, :newval)
|
83
|
+
# # => { 1 => :val, :subhash => { 1 => :newval, 3 => 4 } }
|
84
|
+
#
|
85
|
+
#
|
86
|
+
def deep_set *args
|
87
|
+
hsh = self
|
88
|
+
head_keys = args[0..-3]
|
89
|
+
last_key = args[-2]
|
90
|
+
val = args[-1]
|
91
|
+
# grab last subtree (building out if necessary)
|
92
|
+
head_keys.each{|key| hsh = (hsh[key] ||= {}) }
|
93
|
+
# set leaf value
|
94
|
+
hsh[last_key] = val
|
95
|
+
end
|
96
|
+
|
97
|
+
# Stolen from ActiveSupport::CoreExtensions::Hash::ReverseMerge.
|
98
|
+
def reverse_merge(other_hash)
|
99
|
+
other_hash.merge(self)
|
100
|
+
end
|
101
|
+
|
102
|
+
# Stolen from ActiveSupport::CoreExtensions::Hash::ReverseMerge.
|
103
|
+
def reverse_merge!(other_hash)
|
104
|
+
replace(reverse_merge(other_hash))
|
105
|
+
end
|
106
|
+
|
107
|
+
#
|
108
|
+
# remove all key-value pairs where the value is nil
|
109
|
+
#
|
110
|
+
def compact
|
111
|
+
reject{|key,val| val.nil? }
|
112
|
+
end
|
113
|
+
#
|
114
|
+
# Replace the hash with its compacted self
|
115
|
+
#
|
116
|
+
def compact!
|
117
|
+
replace(compact)
|
118
|
+
end
|
119
|
+
|
120
|
+
end
|
@@ -0,0 +1,119 @@
|
|
1
|
+
module Wukong
|
2
|
+
#
|
3
|
+
# A hashlike has to
|
4
|
+
#
|
5
|
+
# *
|
6
|
+
# * The arguments to your initializer should be the same as the keys, in order
|
7
|
+
# If not, you must override #from_hash
|
8
|
+
#
|
9
|
+
#
|
10
|
+
module HashLike
|
11
|
+
|
12
|
+
# List of possible keys --
|
13
|
+
# delegates to the class
|
14
|
+
def keys
|
15
|
+
self.class.keys
|
16
|
+
end
|
17
|
+
|
18
|
+
#
|
19
|
+
# Return a Hash containing only values for the given keys.
|
20
|
+
#
|
21
|
+
# Since this is intended to mirror Hash#slice it will harmlessly ignore keys
|
22
|
+
# not present in the struct. They will be unset (hsh.include? is not true)
|
23
|
+
# as opposed to nil.
|
24
|
+
#
|
25
|
+
def slice *keys
|
26
|
+
keys.inject({}) do |hsh, key|
|
27
|
+
hsh[key] = send(key) if respond_to?(key)
|
28
|
+
hsh
|
29
|
+
end
|
30
|
+
end
|
31
|
+
|
32
|
+
#
|
33
|
+
# values_at like a hash
|
34
|
+
#
|
35
|
+
# Since this is intended to mirror Hash#values_at it will harmlessly ignore
|
36
|
+
# keys not present in the struct
|
37
|
+
#
|
38
|
+
def values_of *keys
|
39
|
+
keys.map{|key| self.send(key) if respond_to?(key) }
|
40
|
+
end
|
41
|
+
|
42
|
+
#
|
43
|
+
# Convert to a hash
|
44
|
+
#
|
45
|
+
def to_hash
|
46
|
+
slice(*self.class.members)
|
47
|
+
end
|
48
|
+
|
49
|
+
#
|
50
|
+
# Analagous to Hash#each_pair
|
51
|
+
#
|
52
|
+
def pairs
|
53
|
+
self.class.members.map{|attr| [attr, self[attr]] }
|
54
|
+
end
|
55
|
+
|
56
|
+
#
|
57
|
+
# Analagous to Hash#each_pair
|
58
|
+
#
|
59
|
+
def each_pair *args, &block
|
60
|
+
pairs.each(*args, &block)
|
61
|
+
end
|
62
|
+
|
63
|
+
#
|
64
|
+
# Analagous to Hash#merge
|
65
|
+
#
|
66
|
+
def merge *args
|
67
|
+
self.dup.merge! *args
|
68
|
+
end
|
69
|
+
def merge! hsh, &block
|
70
|
+
raise "can't handle block arg yet" if block
|
71
|
+
hsh.each_pair{|key, val| self.send("#{key}=", val) if self.respond_to?("#{key}=") }
|
72
|
+
self
|
73
|
+
end
|
74
|
+
alias_method :update, :merge!
|
75
|
+
|
76
|
+
#
|
77
|
+
# Merge hashes recursively.
|
78
|
+
# Nothing special happens to array values
|
79
|
+
#
|
80
|
+
# x = { :subhash => { 1 => :val_from_x, 222 => :only_in_x, 333 => :only_in_x }, :scalar => :scalar_from_x}
|
81
|
+
# y = { :subhash => { 1 => :val_from_y, 999 => :only_in_y }, :scalar => :scalar_from_y }
|
82
|
+
# x.deep_merge y
|
83
|
+
# => {:subhash=>{1=>:val_from_y, 222=>:only_in_x, 333=>:only_in_x, 999=>:only_in_y}, :scalar=>:scalar_from_y}
|
84
|
+
# y.deep_merge x
|
85
|
+
# => {:subhash=>{1=>:val_from_x, 222=>:only_in_x, 333=>:only_in_x, 999=>:only_in_y}, :scalar=>:scalar_from_x}
|
86
|
+
#
|
87
|
+
def deep_merge hsh2
|
88
|
+
merge hsh2, &Hash::DEEP_MERGER
|
89
|
+
end
|
90
|
+
|
91
|
+
module ClassMethods
|
92
|
+
#
|
93
|
+
# Instantiate an instance of the struct from a hash
|
94
|
+
#
|
95
|
+
# Specify has_symbol_keys if the supplied hash's keys are symbolic;
|
96
|
+
# otherwise they must be uniformly strings
|
97
|
+
#
|
98
|
+
def from_hash(hsh, has_symbol_keys=false)
|
99
|
+
keys = self.keys
|
100
|
+
keys = keys.map(&:to_sym) if has_symbol_keys
|
101
|
+
self.new *hsh.values_of(*keys)
|
102
|
+
end
|
103
|
+
#
|
104
|
+
# The last portion of the class in underscored form
|
105
|
+
# note memoization
|
106
|
+
#
|
107
|
+
def self.resource_name
|
108
|
+
@resource_name ||= self.to_s.gsub(%r{.*::}, '').underscore.to_sym
|
109
|
+
end
|
110
|
+
end
|
111
|
+
|
112
|
+
def self.included base
|
113
|
+
base.class_eval do
|
114
|
+
extend ClassMethods
|
115
|
+
end
|
116
|
+
end
|
117
|
+
end
|
118
|
+
|
119
|
+
end
|