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.



276
277
278
 
# File 'lib/fiber_stream/flow.rb', line 276

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.



165
166
167
 
# File 'lib/fiber_stream/flow.rb', line 165

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) 


176
177
178
179
180
181
 
# File 'lib/fiber_stream/flow.rb', line 176

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:



221
222
223
 
# File 'lib/fiber_stream/flow.rb', line 221

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


100
101
102
103
104
105
 
# File 'lib/fiber_stream/flow.rb', line 100

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) 


151
152
153
154
155
 
# File 'lib/fiber_stream/flow.rb', line 151

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

  new { |upstream| Pull.drop_while(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) 


113
114
115
116
117
118
 
# File 'lib/fiber_stream/flow.rb', line 113

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) 


190
191
192
193
194
195
196
197
198
 
# File 'lib/fiber_stream/flow.rb', line 190

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

.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) 


25
26
27
28
29
30
31
 
# File 'lib/fiber_stream/flow.rb', line 25

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) 


42
43
44
45
46
47
48
 
# File 'lib/fiber_stream/flow.rb', line 42

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) 


57
58
59
60
61
62
63
64
65
66
67
 
# File 'lib/fiber_stream/flow.rb', line 57

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

.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) 


126
127
128
129
130
 
# File 'lib/fiber_stream/flow.rb', line 126

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) 


75
76
77
78
79
 
# File 'lib/fiber_stream/flow.rb', line 75

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) 


209
210
211
212
213
214
215
216
217
218
219
 
# File 'lib/fiber_stream/flow.rb', line 209

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) 


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

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) 


139
140
141
142
143
 
# File 'lib/fiber_stream/flow.rb', line 139

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

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

Instance Method Details

#attach_to(upstream) ⇒ Object

:nodoc:



282
283
284
 
# File 'lib/fiber_stream/flow.rb', line 282

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) 


253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
 
# File 'lib/fiber_stream/flow.rb', line 253

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) 


229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
 
# File 'lib/fiber_stream/flow.rb', line 229

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