Class: Omnizip::Parallel::ParallelCompressor

Inherits:
Object
  • Object
show all
Defined in:
lib/omnizip/parallel/parallel_compressor.rb

Overview

Parallel compression coordinator using Fractor

Manages parallel compression of files in a directory. Distributes compression work across multiple workers and writes results to archive in a thread-safe manner.

Examples:

Compress directory in parallel

compressor = Omnizip::Parallel::ParallelCompressor.new(threads: 4)
compressor.compress('files/', 'backup.zip')

With options

options = Omnizip::Models::ParallelOptions.new
options.threads = 8
compressor = Omnizip::Parallel::ParallelCompressor.new(options)
compressor.compress('files/', 'backup.zip', compression: :lzma2)

Defined Under Namespace

Classes: CompressionWork, CompressionWorker

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(options = nil, threads: nil) ⇒ ParallelCompressor

Initialize parallel compressor

Parameters:

  • options (Omnizip::Models::ParallelOptions, Hash) (defaults to: nil) —

    parallel options

  • threads (Integer) (defaults to: nil) —

    number of threads (overrides options)



129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
 
# File 'lib/omnizip/parallel/parallel_compressor.rb', line 129

def initialize(options = nil, threads: nil)
  @options = case options
             when Omnizip::Models::ParallelOptions
               options.dup
             when Hash
               Omnizip::Models::ParallelOptions.new.apply(options)
             else
               Omnizip::Models::ParallelOptions.new
             end

  @options.threads = threads if threads
  @options.validate!

  @stats = {
    files_processed: 0,
    bytes_processed: 0,
    bytes_compressed: 0,
    start_time: nil,
    end_time: nil,
  }
end

Instance Attribute Details

#options ⇒ Omnizip::Models::ParallelOptions (readonly)

Returns parallel options.

Returns:



120
121
122
 
# File 'lib/omnizip/parallel/parallel_compressor.rb', line 120

def options
  @options
end

#stats ⇒ Hash (readonly)

Returns compression statistics.

Returns:

  • (Hash) —

    compression statistics



123
124
125
 
# File 'lib/omnizip/parallel/parallel_compressor.rb', line 123

def stats
  @stats
end

Instance Method Details

#compress(dir, output, **options) ⇒ String

Compress directory to archive in parallel

Parameters:

  • dir (String) —

    directory path

  • output (String) —

    output archive path

  • options (Hash) —

    compression options

Options Hash (**options):

  • :compression (Symbol) —

    compression method

  • :level (Integer) —

    compression level

  • :recursive (Boolean) —

    include subdirectories

  • :progress (Proc) —

    progress callback

Returns:

  • (String) —

    path to created archive



161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
 
# File 'lib/omnizip/parallel/parallel_compressor.rb', line 161

def compress(dir, output, **options)
  unless ::File.exist?(dir)
    raise Errno::ENOENT,
          "Directory not found: #{dir}"
  end
  unless ::File.directory?(dir)
    raise ArgumentError,
          "Not a directory: #{dir}"
  end

  compression = options[:compression] || :deflate
  level = options[:level] || 6
  recursive = options.fetch(:recursive, true)
  options[:progress]

  @stats[:start_time] = Time.now

  # Scan directory for files
  files = scan_directory(dir, recursive: recursive)

  # Create job queue
  job_queue = JobQueue.new(max_size: @options.queue_size)

  # Schedule jobs
  JobScheduler.new(strategy: @options.strategy)
  files.each do |file_path|
    archive_path = file_path.sub("#{dir}/", "")
    file_size = ::File.size(file_path)

    job_queue.push_with_size(
      file: file_path,
      size: file_size,
      data: {
        archive_path: archive_path,
        compression: compression,
        level: level,
      },
    )
  end

  # Create work items from jobs
  work_items = []
  until job_queue.empty?
    job = job_queue.pop(timeout: 0.1)
    break unless job

    work_items << CompressionWork.new(
      file_path: job.file,
      archive_path: job.data[:archive_path],
      compression: job.data[:compression],
      level: job.data[:level],
    )
  end

  # Create worker pool
  pool = WorkerPool.new(
    worker_class: CompressionWorker,
    num_workers: @options.threads,
    continuous: false,
  )

  pool.start
  pool.submit_batch(work_items)
  pool.run

  # Collect results
  results = pool.successful_results
  errors = pool.failed_results

  # Handle errors
  unless errors.empty?
    error_msgs = errors.map do |e|
      "#{e.work&.file_path}: #{e.error}"
    end.join("\n")
    raise Omnizip::CompressionError, "Compression errors:\n#{error_msgs}"
  end

  # Write archive sequentially (thread-safe)
  write_archive(output, results, compression: compression)

  pool.shutdown

  @stats[:end_time] = Time.now
  @stats[:files_processed] = results.size

  output
end

#statistics ⇒ Hash

Get compression statistics

Returns:

  • (Hash) —

    statistics



252
253
254
255
256
257
258
259
260
261
262
263
264
 
# File 'lib/omnizip/parallel/parallel_compressor.rb', line 252

def statistics
  duration = if @stats[:start_time] && @stats[:end_time]
               @stats[:end_time] - @stats[:start_time]
             else
               0
             end

  @stats.merge(
    duration: duration,
    compression_ratio: calculate_compression_ratio,
    throughput_mbps: calculate_throughput(duration),
  )
end