Class: FiberStream::Flow

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

Class Method Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(&attach) ⇒ Flow

Returns a new instance of Flow.



379
380
381
 
# File 'lib/fiber_stream/flow.rb', line 379

def initialize(&attach)
  @attach = attach
end

Class Method Details

.async ⇒ Object

Creates a scheduler-backed asynchronous boundary.

The boundary starts its producer on the first downstream demand and requires an installed ‘Fiber.scheduler` at that point. Upstream stages run in a non-blocking producer fiber, downstream stages remain in the caller’s current fiber, and each downstream pull resumes at most one upstream pull. Closing the boundary closes upstream and requests producer cancellation. FiberStream does not depend on Async at runtime.



223
224
225
 
# File 'lib/fiber_stream/flow.rb', line 223

def self.async
  new { |upstream| Pull.async(upstream) }
end

.buffer(count) ⇒ Object

Creates a bounded asynchronous buffer.

The buffer starts its producer on the first downstream demand and requires an installed ‘Fiber.scheduler` at that point. It preserves element order, stores at most `count` messages, and closes upstream while requesting producer cancellation when closed. `count` must be a positive Integer. FiberStream does not depend on Async at runtime.

Raises:

  • (TypeError) 


234
235
236
237
238
239
 
# File 'lib/fiber_stream/flow.rb', line 234

def self.buffer(count)
  raise TypeError, "count must be an Integer" unless count.is_a?(Integer)
  raise ArgumentError, "count must be positive" unless count.positive?

  new { |upstream| Pull.buffer(upstream, count) }
end

.build(&attach) ⇒ Object

:nodoc:



292
293
294
 
# File 'lib/fiber_stream/flow.rb', line 292

def self.build(&attach) # :nodoc:
  new(&attach)
end

.compact ⇒ Object

Creates a nil-dropping flow.

The flow drops ‘nil` elements and passes every non-`nil` element through unchanged, including `false`.



32
33
34
 
# File 'lib/fiber_stream/flow.rb', line 32

def self.compact
  new { |upstream| Pull.compact(upstream) }
end

.drop(count) ⇒ Object

Creates a fixed-prefix dropping flow.

The flow discards the first ‘count` upstream elements, then passes later elements through unchanged. `drop(0)` behaves as pass-through. Negative counts raise `ArgumentError`; non-Integer counts raise `TypeError`.

Raises:

  • (TypeError) 


158
159
160
161
162
163
 
# File 'lib/fiber_stream/flow.rb', line 158

def self.drop(count)
  raise TypeError, "count must be an Integer" unless count.is_a?(Integer)
  raise ArgumentError, "count must be non-negative" if count.negative?

  new { |upstream| Pull.drop(upstream, count) }
end

.drop_while(&block) ⇒ Object

Creates a predicate-based prefix-dropping flow.

The flow drops leading elements while the block result is truthy. The first false or nil result, and all later elements, pass through unchanged. After that boundary the block is not called again. Exceptions raised by the block fail the stream and are re-raised from ‘Source#run_with`.

Raises:

  • (ArgumentError) 


209
210
211
212
213
 
# File 'lib/fiber_stream/flow.rb', line 209

def self.drop_while(&block)
  raise ArgumentError, "missing block" unless block

  new { |upstream| Pull.drop_while(upstream, block) }
end

.filter_map(&block) ⇒ Object

Creates a transform-and-filter flow.

The block is called once for each upstream element observed by this stage. Truthy block results are emitted downstream as transformed values; false and nil results are dropped. Exceptions raised by the block fail the stream and are re-raised from ‘Source#run_with`.

Raises:

  • (ArgumentError) 


22
23
24
25
26
 
# File 'lib/fiber_stream/flow.rb', line 22

def self.filter_map(&block)
  raise ArgumentError, "missing block" unless block

  new { |upstream| Pull.filter_map(upstream, block) }
end

.grouped(count) ⇒ Object

Creates a fixed-size grouping flow.

The flow emits arrays containing up to ‘count` adjacent upstream elements. Full groups contain exactly `count` elements; normal upstream completion emits one final partial group when one exists. `count` must be a positive Integer.

Raises:

  • (TypeError) 


171
172
173
174
175
176
 
# File 'lib/fiber_stream/flow.rb', line 171

def self.grouped(count)
  raise TypeError, "count must be an Integer" unless count.is_a?(Integer)
  raise ArgumentError, "count must be positive" unless count.positive?

  new { |upstream| Pull.grouped(upstream, count) }
end

.lines(chomp: true, max_length: nil) ⇒ Object

Creates a line-splitting flow.

The flow accepts String chunks and emits lines split on “n”. By default it chomps the trailing newline and one preceding “r”. ‘max_length` is an optional per-line bytesize limit. With `max_length: nil`, one unterminated line can buffer without bound. Set a positive `max_length` for untrusted, network-facing, or otherwise unbounded streams.

Raises:

  • (TypeError) 


261
262
263
264
265
266
267
268
269
 
# File 'lib/fiber_stream/flow.rb', line 261

def self.lines(chomp: true, max_length: nil)
  raise TypeError, "chomp must be true or false" unless [true, false].include?(chomp)
  unless max_length.nil? || max_length.is_a?(Integer)
    raise TypeError, "max_length must be nil or an Integer"
  end
  raise ArgumentError, "max_length must be positive" if max_length&.<= 0

  new { |upstream| Pull.lines(upstream, chomp, max_length) }
end

.map(&block) ⇒ Object

Creates a mapping flow.

The block is called once for each element pulled through this flow. Exceptions raised by the block fail the stream and are re-raised from ‘Source#run_with`.

Raises:

  • (ArgumentError) 


10
11
12
13
14
 
# File 'lib/fiber_stream/flow.rb', line 10

def self.map(&block)
  raise ArgumentError, "missing block" unless block

  new { |upstream| Pull.map(upstream, block) }
end

.map_concat(&block) ⇒ Object

Creates a one-to-many mapping flow.

The block is called once for each upstream element whose expansion is needed. It must return an object that responds to ‘#each`; yielded values are emitted in order before the next upstream element is pulled. Exceptions raised by the block or by the returned object’s ‘#each` fail the stream and are re-raised from `Source#run_with`.

Raises:

  • (ArgumentError) 


43
44
45
46
47
 
# File 'lib/fiber_stream/flow.rb', line 43

def self.map_concat(&block)
  raise ArgumentError, "missing block" unless block

  new { |upstream| Pull.map_concat(upstream, block) }
end

.parallel_map(concurrency:, &block) ⇒ Object

Creates an ordered scheduler-backed parallel mapping flow.

The stage starts internal scheduled fibers on first downstream demand and requires an installed ‘Fiber.scheduler` in a non-blocking fiber at that point. At most `concurrency` mapping blocks run at the same time, and at most `concurrency` upstream elements are pulled but not yet emitted downstream. Results are emitted in input order. Closing the boundary closes upstream and requests internal worker cancellation. FiberStream does not depend on Async at runtime.

Raises:

  • (ArgumentError) 


70
71
72
73
74
75
76
 
# File 'lib/fiber_stream/flow.rb', line 70

def self.parallel_map(concurrency:, &block)
  raise ArgumentError, "missing block" unless block
  raise TypeError, "concurrency must be an Integer" unless concurrency.is_a?(Integer)
  raise ArgumentError, "concurrency must be positive" unless concurrency.positive?

  new { |upstream| Pull.parallel_map(upstream, concurrency, block) }
end

.parallel_unordered_map(concurrency:, &block) ⇒ Object

Creates an unordered scheduler-backed parallel mapping flow.

The stage starts internal scheduled fibers on first downstream demand and requires an installed ‘Fiber.scheduler` in a non-blocking fiber at that point. At most `concurrency` mapping blocks run at the same time, and at most `concurrency` upstream elements are pulled but not yet emitted downstream. Results are emitted in completion order and input order is not preserved. Closing the boundary closes upstream and requests internal worker cancellation. FiberStream does not depend on Async at runtime.

Raises:

  • (ArgumentError) 


87
88
89
90
91
92
93
 
# File 'lib/fiber_stream/flow.rb', line 87

def self.parallel_unordered_map(concurrency:, &block)
  raise ArgumentError, "missing block" unless block
  raise TypeError, "concurrency must be an Integer" unless concurrency.is_a?(Integer)
  raise ArgumentError, "concurrency must be positive" unless concurrency.positive?

  new { |upstream| Pull.parallel_unordered_map(upstream, concurrency, block) }
end

.ractor_map(workers:, input_transfer: :copy, output_transfer: :copy, &block) ⇒ Object

Creates an ordered Ractor-backed mapping flow.

The mapper runs inside worker ractors and must be shareable, typically created with ‘Ractor.shareable_proc`. Results are emitted in input order, and at most `workers` upstream elements are pulled but not yet emitted. `input_transfer` and `output_transfer` must be `:copy` or `:move` and are passed to Ractor message sends for element and result transfer.

Raises:

  • (ArgumentError) 


102
103
104
105
106
107
108
109
110
111
112
 
# File 'lib/fiber_stream/flow.rb', line 102

def self.ractor_map(workers:, input_transfer: :copy, output_transfer: :copy, &block)
  raise ArgumentError, "missing block" unless block
  raise TypeError, "workers must be an Integer" unless workers.is_a?(Integer)
  raise ArgumentError, "workers must be positive" unless workers.positive?

  Internal::RactorTransferPolicy.validate!(:input_transfer, input_transfer)
  Internal::RactorTransferPolicy.validate!(:output_transfer, output_transfer)
  raise TypeError, "block must be shareable" unless Ractor.shareable?(block)

  new { |upstream| Pull.ractor_map(upstream, workers, input_transfer, output_transfer, block) }
end

.reject(&block) ⇒ Object

Creates a complement filtering flow.

The block is called for upstream elements until it returns ‘false` or `nil`, or upstream completes. Truthy predicate results drop the original element; false and nil results pass the element through unchanged. Exceptions raised by the block fail the stream and are re-raised from `Source#run_with`.

Raises:

  • (ArgumentError) 


133
134
135
136
137
 
# File 'lib/fiber_stream/flow.rb', line 133

def self.reject(&block)
  raise ArgumentError, "missing block" unless block

  new { |upstream| Pull.reject(upstream, block) }
end

.scan(initial, &block) ⇒ Object

Creates a running-accumulator flow.

The block is called as ‘block.call(accumulator, element)` for each upstream element, matching `Sink.fold`. The block result becomes the new accumulator and is emitted downstream. The initial accumulator is not emitted before the first upstream element.

Raises:

  • (ArgumentError) 


184
185
186
187
188
 
# File 'lib/fiber_stream/flow.rb', line 184

def self.scan(initial, &block)
  raise ArgumentError, "missing block" unless block

  new { |upstream| Pull.scan(upstream, initial, block) }
end

.select(&block) ⇒ Object

Creates a filtering flow.

The block is called for upstream elements until it returns a truthy value or upstream completes. Matching elements pass through unchanged. Exceptions raised by the block fail the stream and are re-raised from ‘Source#run_with`.

Raises:

  • (ArgumentError) 


120
121
122
123
124
 
# File 'lib/fiber_stream/flow.rb', line 120

def self.select(&block)
  raise ArgumentError, "missing block" unless block

  new { |upstream| Pull.select(upstream, block) }
end

.split(separator, keep_separator: false, max_length: nil) ⇒ Object

Creates a delimiter-splitting flow.

The flow accepts String chunks and emits frames split on the non-empty String ‘separator`. Separator matching is byte-oriented. By default emitted frames exclude the separator; `keep_separator: true` preserves it on separator-terminated frames. `max_length` is an optional per-frame body bytesize limit. With `max_length: nil`, one unterminated frame can buffer without bound. Set a positive `max_length` for untrusted, network-facing, or otherwise unbounded streams.

Raises:

  • (TypeError) 


280
281
282
283
284
285
286
287
288
289
290
 
# File 'lib/fiber_stream/flow.rb', line 280

def self.split(separator, keep_separator: false, max_length: nil)
  raise TypeError, "separator must be String" unless separator.is_a?(String)
  raise ArgumentError, "separator must not be empty" if separator.empty?
  raise TypeError, "keep_separator must be true or false" unless [true, false].include?(keep_separator)
  unless max_length.nil? || max_length.is_a?(Integer)
    raise TypeError, "max_length must be nil or an Integer"
  end
  raise ArgumentError, "max_length must be positive" if max_length&.<= 0

  new { |upstream| Pull.split(upstream, separator, keep_separator, max_length) }
end

.take(count) ⇒ Object

Creates a limiting flow.

The flow emits at most ‘count` elements. `take(0)` completes without pulling upstream and closes upstream on the first downstream demand. After the limit is reached, upstream is closed during the pull that forwards the final element. Negative counts raise `ArgumentError`; non-Integer counts raise `TypeError`.

Raises:

  • (TypeError) 


146
147
148
149
150
151
 
# File 'lib/fiber_stream/flow.rb', line 146

def self.take(count)
  raise TypeError, "count must be an Integer" unless count.is_a?(Integer)
  raise ArgumentError, "count must be non-negative" if count.negative?

  new { |upstream| Pull.take(upstream, count) }
end

.take_while(&block) ⇒ Object

Creates a predicate-based limiting flow.

The flow emits leading elements while the block result is truthy. The first false or nil result completes the stream without emitting that element and closes upstream during the same downstream pull. Exceptions raised by the block fail the stream and are re-raised from ‘Source#run_with`.

Raises:

  • (ArgumentError) 


197
198
199
200
201
 
# File 'lib/fiber_stream/flow.rb', line 197

def self.take_while(&block)
  raise ArgumentError, "missing block" unless block

  new { |upstream| Pull.take_while(upstream, block) }
end

.tap(&block) ⇒ Object

Creates a pass-through observing flow.

The block is called once for each element before that element is emitted downstream. The block return value is ignored and the original element is passed through unchanged. Exceptions raised by the block fail the stream and are re-raised from ‘Source#run_with`.

Raises:

  • (ArgumentError) 


55
56
57
58
59
 
# File 'lib/fiber_stream/flow.rb', line 55

def self.tap(&block)
  raise ArgumentError, "missing block" unless block

  new { |upstream| Pull.tap(upstream, block) }
end

.throttle(**options) ⇒ Object

Creates a scheduler-aware throttling flow.

The ‘rate:` form creates a fresh `RateLimiter` for each materialization. The `limiter:` form uses the supplied limiter object, which must respond to `acquire(permits:)` and return only after permits are acquired. When FiberStream-owned waiting is required, the current fiber must be non-blocking with an installed `Fiber.scheduler`.



248
249
250
251
252
 
# File 'lib/fiber_stream/flow.rb', line 248

def self.throttle(**options)
  limiter = build_throttle_limiter(options)

  new { |upstream| Pull.throttle(upstream, limiter.call) }
end

Instance Method Details

#attach_to(upstream) ⇒ Object

:nodoc:



385
386
387
 
# File 'lib/fiber_stream/flow.rb', line 385

def attach_to(upstream) # :nodoc:
  @attach.call(upstream)
end

#to(sink) ⇒ Object

Returns a sink that runs this flow before ‘sink`.

The composed sink accepts this flow’s input elements and returns the wrapped sink’s materialized value. It closes the attached flow chain after normal completion, failure, or early sink completion.

Raises:

  • (TypeError) 


356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
 
# File 'lib/fiber_stream/flow.rb', line 356

def to(sink)
  raise TypeError, "expected FiberStream::Sink" unless sink.is_a?(Sink)

  Sink.build do |stream|
    attached_stream = nil
    primary_error = nil

    begin
      attached_stream = attach_to(stream)
      sink.run_stream(attached_stream)
    rescue StandardError => error
      primary_error = error
      raise
    ensure
      begin
        attached_stream&.close
      rescue StandardError => close_error
        raise close_error unless primary_error
      end
    end
  end
end

#via(flow) ⇒ Object

Returns a reusable flow that applies this flow and then ‘flow`.

Construction is lazy. No upstream stream is attached and no elements are pulled until the composed flow is materialized by a source or sink.

Raises:

  • (TypeError) 


332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
 
# File 'lib/fiber_stream/flow.rb', line 332

def via(flow)
  raise TypeError, "expected FiberStream::Flow" unless flow.is_a?(Flow)

  self.class.build do |upstream|
    attached_stream = attach_to(upstream)

    begin
      flow.attach_to(attached_stream)
    rescue StandardError
      begin
        attached_stream.close
      rescue StandardError
        nil
      end
      raise
    end
  end
end