Class: Droonga::Plugins::Search::DistributedSearchPlanner::QueryTransformer

Inherits:
Object
  • Object
show all
Defined in:
lib/droonga/plugins/search/distributed_search_planner.rb

Constant Summary collapse

ASCENDING_OPERATOR = 
"<"
DESCENDING_OPERATOR = 
">"

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(query) ⇒ QueryTransformer

Returns a new instance of QueryTransformer.



102
103
104
105
106
107
108
109
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 102

def initialize(query)
  @query = query
  @output = @query["output"]
  @reducers = {}
  @mappers = {}
  @output_records = true
  transform!
end

Instance Attribute Details

#mappers ⇒ Object (readonly)

Returns the value of attribute mappers.



100
101
102
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 100

def mappers
  @mappers
end

#reducers ⇒ Object (readonly)

Returns the value of attribute reducers.



100
101
102
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 100

def reducers
  @reducers
end

Instance Method Details

#array_style_attributes ⇒ Object 



306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 306

def array_style_attributes
  attributes = @output["attributes"] || []
  if attributes.is_a?(Hash)
    attributes.keys.collect do |key|
      attribute = attributes[key]
      case attribute
      when String
        {
          "label"  => key,
          "source" => attribute,
        }
      when Hash
        attribute["label"] = key
    attribute
      end
    end
  else
    attributes
  end
end

#build_count_mapper_and_reducer! ⇒ Object 



237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 237

def build_count_mapper_and_reducer!
  return unless @output["elements"].include?("count")

  @reducers["count"] = {
    "type" => "sum",
  }
  if unifiable?
    @query["sortBy"]["limit"] = -1 if @query["sortBy"].is_a?(Hash)
    @output["limit"] = -1
    mapper = {
      "target" => "records",
    }
    unless @output["elements"].include?("records")
      @records_limit = -1
      @output["elements"] << "records"
      @output["attributes"] ||= ["_key"]
      @output_records = false
    end
    @mappers["count"] = mapper
  end
end

#build_records_mapper_and_reducer! ⇒ Object 



259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 259

def build_records_mapper_and_reducer!
  # Skip reducing phase for a result with no record output.
  return if !@output["elements"].include?("records") || @records_limit.zero?

  # Append sort key attributes to the list of output attributes
  # temporarily, for the reducing phase. After all extra columns
  # are removed on the gathering phase.
  final_attributes = output_attribute_names
  update_output_attributes!

  @reducers["records"] = build_records_reducer

  mapper = {}
  if @output_records
    mapper["format"]     = @records_format unless @records_format == "simple"
    mapper["attributes"] = final_attributes unless final_attributes.empty?
    mapper["offset"]     = @records_offset unless @records_offset.zero?
    mapper["limit"]      = @records_limit unless @records_limit.zero?
  else
    mapper["no_output"] = true
  end
  @mappers["records"] = mapper
end

#build_records_reducer ⇒ Object 



362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 362

def build_records_reducer
  attributes = source_column_names
  key_column_index = attributes.index("_key")

  operators = @sort_keys.collect do |sort_key|
    operator = ASCENDING_OPERATOR
    if sort_key[0] == "-"
      operator = DESCENDING_OPERATOR
      sort_key = sort_key[1..-1]
    end
    {
      "operator" => operator,
      "column"   => attributes.index(sort_key),
    }
  end

  reducer = {
    "type"      => "sort",
    "operators" => operators,
  }
  if unifiable? and !key_column_index.nil?
    reducer["key_column"] = key_column_index
  end

  # On the reducing phase, we apply only "limit". We cannot apply
  # "offset" on this phase because the collector merges a pair of
  # results step by step even if there are three or more results.
  # Instead, we apply "offset" on the gathering phase.
  reducer["limit"] = @output["limit"]

  reducer
end

#calculate_offset_and_limit! ⇒ Object 



128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 128

def calculate_offset_and_limit!
  @original_sort_offset = sort_offset
  @original_output_offset = output_offset
  @original_sort_limit = sort_limit
  @original_output_limit = output_limit

  calculate_sort_offset!
  calculate_output_offset!

  # We have to calculate limit based on offset.
  # <A, B = limited integer (0...MAXINT)>
  # | sort limit | output limit | => | worker's sort limit      | worker's output limit   | final limit |
  # =============================    ====================================================================
  # | UNLIMITED  | UNLIMITED    | => | UNLIMITED                | UNLIMITED               | UNLIMITED   |
  # | UNLIMITED  | B            | => | final_offset + B         | final_offset + B        | B           |
  # | A          | UNLIMITED    | => | final_offset + A         | final_offset + A        | A           |
  # | A          | B            | => | final_offset + max(A, B) | final_offset + min(A, B)| min(A, B)   |

  # XXX final_limit and final_offset calculated in many times

  @records_offset = final_offset
  @records_limit = final_limit

  updated_sort_limit = nil
  updated_output_limit = nil
  if final_limit == UNLIMITED
    updated_output_limit = UNLIMITED
  else
    if rich_sort?
      updated_sort_limit = final_offset + [sort_limit, output_limit].max
    end
    updated_output_limit = final_offset + final_limit
  end

  if updated_sort_limit and updated_sort_limit != @query["sortBy"]["limit"]
    @query["sortBy"]["limit"] = updated_sort_limit
  end
  if updated_output_limit and @output["limit"] and updated_output_limit != @output["limit"]
    @output["limit"] = updated_output_limit
  end
end

#calculate_output_offset! ⇒ Object 



202
203
204
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 202

def calculate_output_offset!
  @output["offset"] = 0 if have_records? and @output["offset"]
end

#calculate_sort_offset! ⇒ Object 



170
171
172
173
174
175
176
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 170

def calculate_sort_offset!
  # Offset for workers must be zero, because we have to apply "limit" and
  # "offset" on the last gathering phase instead of each reducing phase.
  if rich_sort?
    @query["sortBy"]["offset"] = 0
  end
end

#final_limit ⇒ Object 



210
211
212
213
214
215
216
217
218
219
220
221
222
223
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 210

def final_limit
  if @original_sort_limit == UNLIMITED and
      @original_output_limit == UNLIMITED
    UNLIMITED
  else
    if @original_sort_limit == UNLIMITED
      @original_output_limit
    elsif @original_output_limit == UNLIMITED
      @original_sort_limit
    else
      [@original_sort_limit, @original_output_limit].min
    end
  end
end

#final_offset ⇒ Object 



206
207
208
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 206

def final_offset
  @original_sort_offset + @original_output_offset
end

#have_records? ⇒ Boolean

Returns:

  • (Boolean) 


225
226
227
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 225

def have_records?
  @output["elements"].include?("records")
end

#output_attribute_names ⇒ Object 



283
284
285
286
287
288
289
290
291
292
293
294
295
296
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 283

def output_attribute_names
  attributes = @output["attributes"] || []
  if attributes.is_a?(Hash)
    attributes.keys
  else
    attributes.collect do |attribute|
      if attribute.is_a?(Hash)
        attribute["label"] || attribute["source"]
      else
        attribute
      end
    end
  end
end

#output_limit ⇒ Object 



198
199
200
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 198

def output_limit
  @output["limit"] || 0
end

#output_offset ⇒ Object 



186
187
188
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 186

def output_offset
  @output["offset"] || 0
end

#rich_sort? ⇒ Boolean

Returns:

  • (Boolean) 


229
230
231
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 229

def rich_sort?
  @query["sortBy"].is_a?(Hash)
end

#sort_attribute_names ⇒ Object 



347
348
349
350
351
352
353
354
355
356
357
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 347

def sort_attribute_names
  sort_attributes = @sort_keys.collect do |key|
    key = key[1..-1] if key[0] == "-"
    key
  end
  attributes = source_column_names
  sort_attributes.reject! do |attribute|
    attributes.include?(attribute)
  end
  sort_attributes
end

#sort_limit ⇒ Object 



190
191
192
193
194
195
196
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 190

def sort_limit
  if rich_sort?
    @query["sortBy"]["limit"] || UNLIMITED
  else
    UNLIMITED
  end
end

#sort_offset ⇒ Object 



178
179
180
181
182
183
184
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 178

def sort_offset
  if rich_sort?
    @query["sortBy"]["offset"] || 0
  else
    0
  end
end

#source_column_names ⇒ Object 



327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 327

def source_column_names
  attributes = @output["attributes"] || []
  if attributes.is_a?(Hash)
    attributes_hash = attributes
    attributes = []
    attributes_hash.each do |key, attribute|
      attributes << attribute["source"] || key
    end
    attributes
  else
    attributes.collect do |attribute|
      if attribute.is_a?(Hash)
        attribute["source"] || attribute["label"]
      else
        attribute
      end
    end
  end
end

#transform! ⇒ Object 



111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 111

def transform!
  # The collector module supports only "simple" format search results.
  # So we have to override the format and restore it on the gathering
  # phase.
  @records_format = @output["format"] || "simple"
  if @output["format"] and @output["format"] != "simple"
    @output["format"] = "simple"
  end

  @sort_keys = @query["sortBy"] || []
  @sort_keys = @sort_keys["keys"] || [] if @sort_keys.is_a?(Hash)

  calculate_offset_and_limit!
  build_count_mapper_and_reducer!
  build_records_mapper_and_reducer!
end

#unifiable? ⇒ Boolean

Returns:

  • (Boolean) 


233
234
235
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 233

def unifiable?
  @output["unifiable"]
end

#update_output_attributes! ⇒ Object 



298
299
300
301
302
303
304
 
# File 'lib/droonga/plugins/search/distributed_search_planner.rb', line 298

def update_output_attributes!
  @output["attributes"] = array_style_attributes
  @output["attributes"] += sort_attribute_names
  if unifiable? and !source_column_names.include?("_key")
    @output["attributes"] << "_key"
  end
end