1#@!#!123s

: / proc / self / root / opt / alt / ruby34 / share / ri / system / Fiber /
Filename : cdesc-Fiber.ri
back
U:RDoc::NormalClass[iI"
Fiber:ET@I"Object;To:RDoc::Markup::Document:@parts[o;;[$o:RDoc::Markup::Paragraph;[
I"EFibers are primitives for implementing light weight cooperative ;TI"Mconcurrency in Ruby. Basically they are a means of creating code blocks ;TI"Lthat can be paused and resumed, much like threads. The main difference ;TI"Nis that they are never preempted and that the scheduling must be done by ;TI"#the programmer and not the VM.;To:RDoc::Markup::BlankLine�o;	;[	I"OAs opposed to other stackless light weight concurrency models, each fiber ;TI"Jcomes with a stack.  This enables the fiber to be paused from deeply ;TI"Dnested function calls within the fiber block.  See the ruby(1) ;TI"9manpage to configure the size of the fiber stack(s).;T@o;	;[
I"KWhen a fiber is created it will not run automatically. Rather it must ;TI"?be explicitly asked to run using the Fiber#resume method. ;TI"FThe code running inside the fiber can give up control by calling ;TI"EFiber.yield in which case it yields control back to caller (the ;TI"!caller of the Fiber#resume).;T@o;	;[I"JUpon yielding or termination the Fiber returns the value of the last ;TI"executed expression;T@o;	;[I"For instance:;T@o:RDoc::Markup::Verbatim;[
I"fiber = Fiber.new do
;TI"  Fiber.yield 1
;TI"	  2
;TI"	end
;TI"
;TI"puts fiber.resume
;TI"puts fiber.resume
;TI"puts fiber.resume
;T:@format0o;	;[I"<em>produces</em>;T@o;;[I"1
;TI"2
;TI"#FiberError: dead fiber called
;T;0o;	;[	I"HThe Fiber#resume method accepts an arbitrary number of parameters, ;TI"Dif it is the first call to #resume then they will be passed as ;TI"Eblock arguments. Otherwise they will be the return value of the ;TI"call to Fiber.yield;T@o;	;[I"
Example:;T@o;;[I""fiber = Fiber.new do |first|
;TI"&  second = Fiber.yield first + 2
;TI"	end
;TI"
;TI"puts fiber.resume 10
;TI"!puts fiber.resume 1_000_000
;TI"Kputs fiber.resume "The fiber will be dead before I can cause trouble"
;T;0o;	;[I"<em>produces</em>;T@o;;[I"12
;TI"
1000000
;TI"#FiberError: dead fiber called
;T;0S:RDoc::Markup::Heading:
leveli:	textI"Non-blocking Fibers;T@o;	;[
I"LThe concept of <em>non-blocking fiber</em> was introduced in Ruby 3.0. ;TI"OA non-blocking fiber, when reaching a operation that would normally block ;TI"Mthe fiber (like <code>sleep</code>, or wait for another process or I/O) ;TI"Lwill yield control to other fibers and allow the <em>scheduler</em> to ;TI"Mhandle blocking and waking up (resuming) this fiber when it can proceed.;T@o;	;[	I"TFor a Fiber to behave as non-blocking, it need to be created in Fiber.new with ;TI"J<tt>blocking: false</tt> (which is the default), and Fiber.scheduler ;TI"Nshould be set with Fiber.set_scheduler. If Fiber.scheduler is not set in ;TI"Qthe current thread, blocking and non-blocking fibers' behavior is identical.;T@o;	;[I"QRuby doesn't provide a scheduler class: it is expected to be implemented by ;TI"1the user and correspond to Fiber::Scheduler.;T@o;	;[I"SThere is also Fiber.schedule method, which is expected to immediately perform ;TI"Rthe given block in a non-blocking manner. Its actual implementation is up to ;TI"the scheduler.;T:
@fileI"cont.c;T:0@omit_headings_from_table_of_contents_below0;0;0[�[�[�[[I"
class;T[[:public[[I"[];TI"cont.c;T[I"[]=;T@y[I"
blocking;T@y[I"blocking?;T@y[I"current;T@y[I"current_scheduler;T@y[I"new;T@y[I"
schedule;T@y[I"scheduler;T@y[I"set_scheduler;T@y[I"
yield;T@y[:protected[�[:private[�[I"
instance;T[[;[[I"alive?;T@y[I"backtrace;T@y[I"backtrace_locations;T@y[I"blocking?;T@y[I"inspect;T@y[I"	kill;T@y[I"
raise;T@y[I"resume;T@y[I"storage;T@y[I"
storage=;T@y[I"	to_s;T@y[I"
transfer;T@y[;[�[;[�[�[U:RDoc::Context::Section[i�0o;;[�;0;0[@m@mcRDoc::TopLevel