threach
Deprecated and archived. Never really worked reliably. Archived in case I ever want to revisit this space.
threach
monkeypatches the Enumerable module with a new method #threach
that provides a threaded version of #each
(or whatever enumerator you throw at it). It's a very simple producer-consumer model.
Installation
threach
is on rubygems.org, so you should just be able to do
gem install threach
# or jruby -S gem install threach
Use
require 'rubygems'
require 'threach'
# Process with 2 threads. It assumes you want 'each'
# as your iterator.
(1..10).threach(2) {|i| puts i.to_s}
# If you want to watch it work...
(1..50).threach(2) do |i|
puts "Thread #{Thread.current[:tnum]}: #{i}"
end
# You can also specify the iterator as the second argument
File.open('mybigfile') do |f|
f.threach(3, :each_line) do |line|
processLine(line)
end
end
# threach does not care what the arity of your block is
# as long as it matches the iterator specifed
('A'..'Z').threach(3, :each_with_index) do |letter, index|
puts "#{index}: #{letter}"
end
# Same thing with a hash, where the default #each actually returns two values
h = {'a' => 1, 'b'=>2, 'c'=>3}
h.threach(2) do |letter, i|
puts "#{i}: #{letter}"
end
Things you need to know
- The number you provide to
threach
is the number of consumer threads. It's assumed that the time to iterate once on the producer is much less than the work done by a consumer, so you need multiple consumers to keep up. -
threach
doesn't magically make your code thread-safe. That's still up to you. - Using
break
under JRuby works as expected but writes a log line to STDERR. This is something internal to JRuby and I don't know how to stop it. - Throwing exceptions as
raise "oops'
under JRuby is so slow that if you have more than one consumer, the time between theraise
and the time you exit thethreach
loop is long enough that a lot of work will still get done. You need to use use the three-argument formraise WhateverError, value, nil
. The lastnil
tells JRuby to not bother making a full stack trace and reduces the penalty, but you shouldn't useraise
for flow control; usecatch
(or, if you can, just regular oldbreak
).
Why and when to use it?
Well, if you're using stock (MRI) ruby -- you probably shouldn't bother with threach
unless you're doing IO-intensive stuff. It'll just slow things down. But if you're using a ruby implementation that has real threads, like JRuby, this will give you relatively painless multi-threading.
You can always do something like:
if defined? JRUBY_VERSION
numthreads = 3
else
numthreads = 0
end
my_enumerable.threach(numthreads) {|i| ...}
...since threach(0)
is exactly the same as each
Note the "relatively" in front of "painless" up there. The block you pass still has to be thread-safe, and there are many data structures you'll encounter that are not thread-safe. Scalars, arrays, and hashes are, though, under JRuby, and that'll get you pretty far.
Change Notes
- 0.3 Successfully deal with
break
without deadlocks by using another SizedQueue as, basically, a thread-safe counter of how many threads have finished. - 0.2 Undo attempts to deal with non-local exit
- 0.1 first release
Note on Patches/Pull Requests
- Fork the project.
- Make your feature addition or bug fix.
- Add tests for it. This is important so I don't break it in a future version unintentionally.
- Commit, do not mess with rakefile, version, or history. (if you want to have your own version, that is fine but bump version in a commit by itself I can ignore when I pull)
- Send me a pull request. Bonus points for topic branches.
Copyright
Copyright (c) 2010-2011 Bill Dueber. See LICENSE for details.