Class: ContentData::ContentData

Inherits:
Object
  • Object
show all
Defined in:
lib/content_data/content_data.rb

Overview

Content Data(CD) object holds files information as contents and instances Files info retrieved from hardware: checksum, size, time modification, server, device and path Those attributes are divided into content and instance attributes:

unique checksum, size are content attributes
time modification, server, device and path are instance attributes

The relationship between content and instances is 1:many meaning that a content can have instances in many servers. content also has time attribute, which has the value of the time of the first instance. This can be changed by using unify_time method which sets all time attributes for a content and it’s instances to the min time off all. Different files(instances) with same content(checksum), are grouped together under that content. Interface methods include:

iterate over contents and instances info,
unify time, add/remove instance, queries, merge, remove directory and more.

Content info data structure:

@contents_info = { Checksum -> [size, *instances*, content_modification_time] }
  *instances* = {[server,path] -> [instance_modification_time,index_time] }

Notes:

1. content_modification_time is the instance_modification_time of the first
   instances which was added to @contents_info

Constant Summary collapse

CHUNK_SIZE = 
5000

Instance Method Summary collapse

Constructor Details

#initialize(other = nil) ⇒ ContentData

Returns a new instance of ContentData.



34
35
36
37
38
39
40
41
42
 
# File 'lib/content_data/content_data.rb', line 34

def initialize(other = nil)
  if other.nil?
    @contents_info = {}  # Checksum --> [size, paths-->time(instance), time(content)]
    @instances_info = {}  # location --> checksum to optimize instances query
  else
    @contents_info = other.clone_contents_info
    @instances_info = other.clone_instances_info  # location --> checksum to optimize instances query
  end
end

Instance Method Details

#==(other) ⇒ Object 



230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
 
# File 'lib/content_data/content_data.rb', line 230

def ==(other)
  return false if other.nil?
  return false if @contents_info.length != other.contents_size
  other.each_instance { |checksum, size, content_mod_time, instance_mod_time, server, path|
    return false if instance_exists(path, server) != other.instance_exists(path, server)
    local_content_info = @contents_info[checksum]
    return false if local_content_info.nil?
    return false if local_content_info[0] != size
    return false if local_content_info[2] != content_mod_time
    #check instances
    local_instances =  local_content_info[1]
    return false if other.checksum_instances_size(checksum) != local_instances.length
    location = [server, path]
    local_instance_mod_time, _ = local_instances[location]
    return false if local_instance_mod_time.nil?
    return false if local_instance_mod_time != instance_mod_time
  }
  true
end

#add_instance(checksum, size, server, path, modification_time, index_time = Time.now.to_i) ⇒ Object 



156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
 
# File 'lib/content_data/content_data.rb', line 156

def add_instance(checksum, size, server, path, modification_time, index_time=Time.now.to_i)
  location = [server, path]

  # file was changed but remove_instance was not called
  if (@instances_info.include?(location) && @instances_info[location] != checksum)
    Log.warning("#{server}:#{path} file already exists with different checksum")
    remove_instance(server, path)
  end

  content_info = @contents_info[checksum]
  if content_info.nil?
    @contents_info[checksum] = [size,
                                {location => [modification_time,index_time]},
                                modification_time]
  else
    if size != content_info[0]
      Log.warning('File size different from content size while same checksum')
      Log.warning("instance location:server:'#{location[0]}'  path:'#{location[1]}'")
      Log.warning("instance mod time:'#{modification_time}'")
    end
    #override file if needed
    content_info[0] = size
    instances = content_info[1]
    instances[location] = [modification_time, index_time]
  end
  @instances_info[location] = checksum
end

#checksum_instances_size(checksum) ⇒ Object 



142
143
144
145
146
 
# File 'lib/content_data/content_data.rb', line 142

def checksum_instances_size(checksum)
  content_info = @contents_info[checksum]
  return 0 if content_info.nil?
  content_info[1].length
end

#clone_contents_info ⇒ Object 



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
 
# File 'lib/content_data/content_data.rb', line 60

def clone_contents_info
  clone_contents_info = {}
  contents_info_enum = @contents_info.each_key
  loop {
    checksum = contents_info_enum.next rescue break
    instances = @contents_info[checksum]
    size = instances[0]
    content_time = instances[2]
    instances_db = instances[1]
    instances_db_cloned = {}
    instances_db_enum = instances_db.each_key
    loop {
      location =  instances_db_enum.next rescue break
      inst_mod_times = instances_db[location]
      # we use deep clone for location since map key is using shallow clone.
      # we dont want references between new content data
      # and orig object. This will help the GC dispose the orig object if not used any more.
      instances_db_cloned[[location[0].clone,location[1].clone]] = inst_mod_times.clone
    }
    clone_contents_info[checksum] = [size,
                          instances_db_cloned,
                          content_time]
  }
  clone_contents_info
end

#clone_instances_info ⇒ Object 



50
51
52
53
54
55
56
57
58
 
# File 'lib/content_data/content_data.rb', line 50

def clone_instances_info
  clone_instances_info = {}
  instances_info_enum = @instances_info.each_key
  loop {
    location = instances_info_enum.next rescue break
    clone_instances_info[[location[0].clone, location[1].clone]] = @instances_info[location].clone
  }
  clone_instances_info
end

#content_each_instance(checksum, &block) ⇒ Object

iterator of instances over specific content block is provided with: checksum, size, content modification time,

instance modification time, server and file path



121
122
123
124
125
126
127
128
129
130
131
132
 
# File 'lib/content_data/content_data.rb', line 121

def content_each_instance(checksum, &block)
  content_info = @contents_info[checksum]
  instances_db_enum = content_info[1].each_key
  loop {
    location = instances_db_enum.next rescue break
    # provide the block with: checksum, size, content modification time,instance modification time,
    #   server and path.
    inst_mod_time,_ = content_info[1][location]
    block.call(checksum,content_info[0], content_info[2], inst_mod_time,
               location[0], location[1])
  }
end

#content_exists(checksum) ⇒ Object 



188
189
190
 
# File 'lib/content_data/content_data.rb', line 188

def content_exists(checksum)
  @contents_info.has_key?(checksum)
end

#contents_size ⇒ Object 



134
135
136
 
# File 'lib/content_data/content_data.rb', line 134

def contents_size()
  @contents_info.length
end

#each_content(&block) ⇒ Object

iterator over @contents_info data structure (not including instances) block is provided with: checksum, size and content modification time



88
89
90
91
92
93
94
95
96
 
# File 'lib/content_data/content_data.rb', line 88

def each_content(&block)
  contents_enum = @contents_info.each_key
  loop {
    checksum = contents_enum.next rescue break
    content_val = @contents_info[checksum]
    # provide checksum, size and content modification time to the block
    block.call(checksum,content_val[0], content_val[2])
  }
end

#each_instance(&block) ⇒ Object

iterator over @contents_info data structure (including instances) block is provided with: checksum, size, content modification time,

instance modification time, server and file path



101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
 
# File 'lib/content_data/content_data.rb', line 101

def each_instance(&block)
  contents_enum = @contents_info.each_key
  loop {
    checksum = contents_enum.next rescue break
    content_info = @contents_info[checksum]
    content_info_enum = content_info[1].each_key
    loop {
      location = content_info_enum.next rescue break
      # provide the block with: checksum, size, content modification time,instance modification time,
      #   server and path.
      inst_mod_time, inst_index_time = content_info[1][location]
      block.call(checksum,content_info[0], content_info[2], inst_mod_time,
                 location[0], location[1], inst_index_time)
    }
  }
end

#empty? ⇒ Boolean

Returns:

  • (Boolean) 


184
185
186
 
# File 'lib/content_data/content_data.rb', line 184

def empty?
  @contents_info.empty?
end

#from_file(filename) ⇒ Object

TODO validation that file indeed contains ContentData missing Loading db from file using chunks for better memory performance



336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
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
 
# File 'lib/content_data/content_data.rb', line 336

def from_file(filename)
  # read first line (number of contents)
  # calculate line number (number of instances)
  # read number of instances.
  # loop over instances lines (using chunks) and add instances

  File.open(filename, 'r') { |file|
    # Get number of contents (at first line)
    number_of_contents = file.gets  # this gets the next line or return nil at EOF
    unless (number_of_contents and number_of_contents.match(/^[\d]+$/))  # check that line is of Number format
      return reset_load_from_file(filename, file, "number of contents should be a number. We got:#{number_of_contents}")
    end
    number_of_contents = number_of_contents.to_i
    # advance file lines over all contents. We need only the instances data to build the content data object
    # use chunks and GC
    contents_chunks = number_of_contents / CHUNK_SIZE
    contents_chunks += 1 if (contents_chunks * CHUNK_SIZE < number_of_contents)
    chunk_index = 0
    while chunk_index < contents_chunks
      chunk_size = CHUNK_SIZE
      if chunk_index + 1 == contents_chunks
        # update last chunk size
        chunk_size = number_of_contents - (chunk_index * CHUNK_SIZE)
      end
      return unless read_contents_chunk(filename, file, chunk_size)
      GC.start
      chunk_index += 1
    end

    # get number of instances
    number_of_instances = file.gets
    unless (number_of_instances and number_of_instances.match(/^[\d]+$/))  # check that line is of Number format
      return reset_load_from_file(filename, file, "number of instances should be a Number. We got:#{number_of_instances}")
    end
    number_of_instances = number_of_instances.to_i
    # read in instances chunks and GC
    instances_chunks = number_of_instances / CHUNK_SIZE
    instances_chunks += 1 if (instances_chunks * CHUNK_SIZE < number_of_instances)
    chunk_index = 0
    while chunk_index < instances_chunks
      chunk_size = CHUNK_SIZE
      if chunk_index + 1 == instances_chunks
        # update last chunk size
        chunk_size = number_of_instances - (chunk_index * CHUNK_SIZE)
      end
      return unless read_instances_chunk(filename, file, chunk_size)
      GC.start
      chunk_index += 1
    end
  }
end

#get_instance_mod_time(checksum, location) ⇒ Object 



148
149
150
151
152
153
154
 
# File 'lib/content_data/content_data.rb', line 148

def get_instance_mod_time(checksum, location)
  content_info = @contents_info[checksum]
  return nil if content_info.nil?
  instances = content_info[1]
  instance_time,_ = instances[location]
  instance_time
end

#get_query(variable, params) ⇒ Object

TODO simplify conditions This mehod is experimental and shouldn't be used nil is used to define +/- infinity for to/from method arguments from/to values are exlusive in condition’a calculations Need to take care about ‘==’ operation that is used for object’s comparison. In need of case user should define it’s own ‘==’ implemementation.



598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
 
# File 'lib/content_data/content_data.rb', line 598

def get_query(variable, params)
  raise RuntimeError.new 'This method is experimental and shouldn\'t be used'

  exact = params['exact'].nil? ? Array.new : params['exact']
  from = params['from']
  to = params ['to']
  is_inside = params['is_inside']

  unless ContentInstance.new.instance_variable_defined?("@#{attribute}")
    raise ArgumentError "#{variable} isn't a ContentInstance variable"
  end

  if (exact.nil? && from.nil? && to.nil?)
    raise ArgumentError 'At least one of the argiments {exact, from, to} must be defined'
  end

  if (!(from.nil? || to.nil?) && from.kind_of?(to.class))
    raise ArgumentError 'to and from arguments should be comparable one with another'
  end

  # FIXME add support for from/to for Strings
  if ((!from.nil? && !from.kind_of?(Numeric.new.class))\
        || (!to.nil? && to.kind_of?(Numeric.new.class)))
    raise ArgumentError 'from and to options supported only for numeric values'
  end

  if (!exact.empty? && (!from.nil? || !to.nil?))
    raise ArgumentError 'exact and from/to options are mutually exclusive'
  end

  result_index = ContentData.new
  instances.each_value do |instance|
    is_match = false
    var_value = instance.instance_variable_get("@#{variable}")

    if exact.include? var_value
      is_match = true
    elsif (from.nil? || var_value > from) && (to.nil? || var_value < to)
      is_match = true
    end

    if (is_match && is_inside) || (!is_match && !is_inside)
      checksum = instance.checksum
      result_index.add_content(contents[checksum]) unless result_index.content_exists(checksum)
      result_index.add_instance instance
    end
  end
  result_index
end

#instance_exists(path, server) ⇒ Object 



192
193
194
 
# File 'lib/content_data/content_data.rb', line 192

def instance_exists(path, server)
  @instances_info.has_key?([server, path])
end

#instances_size ⇒ Object 



138
139
140
 
# File 'lib/content_data/content_data.rb', line 138

def instances_size()
  @instances_info.length
end

#read_contents_chunk(filename, file, chunk_size) ⇒ Object 



388
389
390
391
392
393
394
395
396
 
# File 'lib/content_data/content_data.rb', line 388

def read_contents_chunk(filename, file, chunk_size)
  chunk_index = 0
  while chunk_index < chunk_size
    return reset_load_from_file(filename, file, "Expecting content line but " +
        "reached end of file after line #{$.}") unless file.gets
    chunk_index += 1
  end
  true
end

#read_instances_chunk(filename, file, chunk_size) ⇒ Object 



398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
 
# File 'lib/content_data/content_data.rb', line 398

def  read_instances_chunk(filename, file, chunk_size)
  chunk_index = 0
  while chunk_index < chunk_size
    instance_line = file.gets
    return reset_load_from_file(filename, file, "Expected to read Instance line but reached EOF") unless instance_line
    parameters = instance_line.split(',')
    # bugfix: if file name consist a comma then parsing based on comma separating fails
    if (parameters.size > 6)
      (4..parameters.size-3).each do |i|
        parameters[3] = [parameters[3], parameters[i]].join(",")
      end
      (4..parameters.size-3).each do |i|
        parameters.delete_at(4)
      end
    end

    add_instance(parameters[0],        #checksum
                 parameters[1].to_i,   # size
                 parameters[2],        # server
                 parameters[3],        # path
                 parameters[4].to_i,   # mod time
                 parameters[5].to_i)   # index time
    chunk_index += 1
  end
  true
end

#remove_content(checksum) ⇒ Object 



250
251
252
253
254
255
256
257
258
 
# File 'lib/content_data/content_data.rb', line 250

def remove_content(checksum)
  content_info = @contents_info[checksum]
  if content_info
    content_info[1].each_key { |location|
      @instances_info.delete(location)
    }
    @contents_info.delete(checksum)
  end
end

#remove_directory(dir_to_remove, server) ⇒ Object

removes all instances records which are located under input param: dir_to_remove. found records are removed from both @instances_info and @instances_info. input params: server & dir_to_remove - are used to check each instance unique key (called location) removes also contents, if a contents becomes empty after removing instances



214
215
216
217
218
219
220
221
222
223
224
225
226
227
 
# File 'lib/content_data/content_data.rb', line 214

def remove_directory(dir_to_remove, server)
  contents_enum = @contents_info.each_key
  loop {
    checksum = contents_enum.next rescue break
    instances =  @contents_info[checksum][1]
    instances.each_key { |location|
      if location[0] == server and location[1].scan(dir_to_remove).size > 0
        instances.delete(location)
        @instances_info.delete(location)
      end
    }
    @contents_info.delete(checksum) if instances.empty?
  }
end

#remove_instance(server, path) ⇒ Object

removes an instance record both in @instances_info and @instances_info. input params: server & path - are the instance unique key (called location) removes also the content, if content becomes empty after removing the instance



199
200
201
202
203
204
205
206
207
208
 
# File 'lib/content_data/content_data.rb', line 199

def remove_instance(server, path)
  location = [server, path]
  checksum = @instances_info[location]
  content_info = @contents_info[checksum]
  return nil if content_info.nil?
  instances = content_info[1]
  instances.delete(location)
  @contents_info.delete(checksum) if instances.empty?
  @instances_info.delete(location)
end

#reset_load_from_file(file_name, file_io, err_msg) ⇒ Object 



425
426
427
428
429
430
431
 
# File 'lib/content_data/content_data.rb', line 425

def reset_load_from_file(file_name, file_io, err_msg)
  Log.error("unexpected error reading file:#{file_name}\nError message:#{err_msg}")
  @contents_info = {}  # Checksum --> [size, paths-->time(instance), time(content)]
  @instances_info = {}  # location --> checksum to optimize instances query
  file_io.close
  nil
end

#to_file(filename) ⇒ Object

Write content data to file. Write is using chunks (for both content chunks and instances chunks) Chunk is used to maximize GC affect. The temporary memory of each chunk is GCed. Without the chunks used in a dipper stack level, GC keeps the temporary objects as part of the stack context.



281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
 
# File 'lib/content_data/content_data.rb', line 281

def to_file(filename)
  content_data_dir = File.dirname(filename)
  FileUtils.makedirs(content_data_dir) unless File.directory?(content_data_dir)
  File.open(filename, 'w') { |file|
    file.write("#{@contents_info.length}\n")
    contents_enum = @contents_info.each_key
    content_chunks = @contents_info.length / CHUNK_SIZE + 1
    chunks_counter = 0
    while chunks_counter < content_chunks
      to_file_contents_chunk(file,contents_enum, CHUNK_SIZE)
      GC.start
      chunks_counter += 1
    end
    file.write("#{@instances_info.length}\n")
    contents_enum = @contents_info.each_key
    chunks_counter = 0
    while chunks_counter < content_chunks
      to_file_instances_chunk(file,contents_enum, CHUNK_SIZE)
      GC.start
      chunks_counter += 1
    end
  }
end

#to_file_contents_chunk(file, contents_enum, chunk_size) ⇒ Object 



305
306
307
308
309
310
311
312
313
 
# File 'lib/content_data/content_data.rb', line 305

def to_file_contents_chunk(file, contents_enum, chunk_size)
  chunk_counter = 0
  while chunk_counter < chunk_size
    checksum = contents_enum.next rescue return
    content_info = @contents_info[checksum]
    file.write("#{checksum},#{content_info[0]},#{content_info[2]}\n")
    chunk_counter += 1
  end
end

#to_file_instances_chunk(file, contents_enum, chunk_size) ⇒ Object 



315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
 
# File 'lib/content_data/content_data.rb', line 315

def to_file_instances_chunk(file, contents_enum, chunk_size)
  chunk_counter = 0
  while chunk_counter < chunk_size
    checksum = contents_enum.next rescue return
    content_info = @contents_info[checksum]
    instances_db_enum = content_info[1].each_key
    loop {
      location = instances_db_enum.next rescue break
      # provide the block with: checksum, size, content modification time,instance modification time,
      #   server and path.
      instance_modification_time,instance_index_time = content_info[1][location]
      file.write("#{checksum},#{content_info[0]},#{location[0]},#{location[1]}," +
                 "#{instance_modification_time},#{instance_index_time}\n")
    }
    chunk_counter += 1
    break if chunk_counter == chunk_size
  end
end

#to_s ⇒ Object 



260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
 
# File 'lib/content_data/content_data.rb', line 260

def to_s
  return_str = ""
  contents_str = ""
  instances_str = ""
  each_content { |checksum, size, content_mod_time|
    contents_str << "%s,%d,%d\n" % [checksum, size, content_mod_time]
  }
  each_instance { |checksum, size, content_mod_time, instance_mod_time, server, path|
    instances_str <<  "%s,%d,%s,%s,%d\n" % [checksum, size, server, path, instance_mod_time]
  }
  return_str << "%d\n" % [@contents_info.length]
  return_str << contents_str
  return_str << "%d\n" % [@instances_info.length]
  return_str << instances_str
  return_str
end

#unify_time ⇒ Object

for each content, all time fields (content and instances) are replaced with the min time found, while going through all time fields.



435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
 
# File 'lib/content_data/content_data.rb', line 435

def unify_time()
  contents_enum = @contents_info.each_key
  loop {
    checksum = contents_enum.next rescue break
    content_info = @contents_info[checksum]
    min_time_per_checksum = content_info[2]
    instances = content_info[1]
    instances_enum = instances.each_key
    loop {
      location = instances_enum.next rescue break
      instance_mod_time = instances[location][0]
      if instance_mod_time < min_time_per_checksum
        min_time_per_checksum = instance_mod_time
      end
    }
    # update all instances with min time
    instances_enum = instances.each_key
    loop {
      location = instances_enum.next rescue break
      instances[location][0] = min_time_per_checksum
    }
    # update content time with min time
    content_info[2] = min_time_per_checksum
  }
end

#unique_id ⇒ ID

Content Data unique identification

Returns:

  • (ID) —

    hash identification



46
47
48
 
# File 'lib/content_data/content_data.rb', line 46

def unique_id
  @instances_info.hash
end

#validate(params = nil) ⇒ Boolean

Validates index against file system that all instances hold a correct data regarding files that they represents.

There are two levels of validation, controlled by instance_check_level system parameter:

  • shallow - quick, tests instance for file existence and attributes.

  • deep - can take more time, in addition to shallow recalculates hash sum.

Parameters:

  • params (Hash) (defaults to: nil) —

    hash of parameters of validation, can be used to return additional data.

    Supported key/value combinations:

    • key is :failed value is ContentData used to return failed instances

Returns:

  • (Boolean) —

    true when index is correct, false otherwise

Raises:

  • (ArgumentError) —

    when instance_check_level is incorrect



473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
 
# File 'lib/content_data/content_data.rb', line 473

def validate(params = nil)
  # used to answer whether specific param was set
  param_exists = Proc.new do |param|
    !(params.nil? || params[param].nil?)
  end

  # used to process method parameters centrally
  process_params = Proc.new do |values|
    if param_exists.call(:failed)
      info = values[:details]
      unless info.nil?
        checksum = info[0]
        content_mtime = info[1]
        size = info[2]
        inst_mtime = info[3]
        server = info[4]
        file_path = info[5]
        params[:failed].add_instance(checksum, size, server, file_path, inst_mtime)
      end
    end
  end

  is_valid = true
  contents_enum = @contents_info.each_key
  loop {
    checksum = contents_enum.next rescue break
    instances = @contents_info[checksum]
    content_size = instances[0]
    content_mtime = instances[2]
    instances_enum = instances[1].each_key
    loop {
      unique_path = instances_enum.next rescue break
      instance_mtime = instances[1][unique_path][0]
      instance_info = [checksum, content_mtime, content_size, instance_mtime]
      instance_info.concat(unique_path)
      unless check_instance(instance_info)
        is_valid = false

        unless params.nil? || params.empty?
          process_params.call({:details => instance_info})
        end
      end
    }
  }
  is_valid
end