for

This little shard implements a couple macros for additional for-loop-like looping constructs.

It implements a simple declarative style for-like iteration loop called iterate, which traverses all of the elements of a collection, passing each in turn to the provided block. It also implements a C-style for looping construct, which can be assigned to a variable for execution later (or for repeated execution), with initialization, test, and update code lines, and a body that runs as a closure with its own lexical scope.

The C-style loop structure is useful. You can use it to make your code more terse. So, you can type this:

require "prime"
require "for"

prime = uninitialized Int32
do_until({max = 2147483647}, ->(){prime.prime?}, {prime = rand(max)})

puts "random prime number: #{prime}"

Instead of typing this (which is basically what is generated from the above code):

require "prime"
require "for"

prime = uninitialized Int32
->() do
  max = 2147483647
  loop do
    prime = rand(max)
    break if prime.prime?
  end
end.call

puts "random prime number: #{prime}"

Because for loops are encapsulated inside of a Proc closure, you can also assign them to variables, pass them around, call them at will, and reuse them. This has a lot of potential uses. Consider the following for one of them:

require "for"
require "http/server"

jobs = [] of HTTP::Server::Context
handlers = [] of Proc(Nil)
queue = Channel(Tuple(HTTP::Server::Context?, Channel(Nil))).new(1000)

8.times do
  handlers << for(
    {counter = 1},
    ->{ tup = queue.receive? },
    {counter += 1},
    run: false) do
    puts "REQ #{counter} -- #{tup[0]}"
    tup[1].send(nil)
  end
end

server = HTTP::Server.new do |context|
  pong = Channel(Nil).new
  queue.send({context, pong})
  pong.receive # Worker has finished; return response
end

spawn(name: "work loop") do
  handlers.each { |handler| spawn(name: "worker") { handler.call } }
end

server.bind_tcp 8080
server.listen

That code simulates an HTTP server that accepts requests, each of which represents a job of some sort to be handled by a worker, and 8 workers, implemented as C-style for loops running in fibers.

Installation

1. Add the dependency to your shard.yml:

   dependencies:
     for:
       github: wyhaines/for.cr
   

2. Run shards install

Usage

require "for"

This defines four macros at the top level.

using

The first is using TARGET, which is a syntactic sugar macro. using foo results in inserting foo where it is executed.

This means that the following are equivalent:

for(count, ary) { puts count }

for(count, using ary) { puts count }

And the following is acceptable syntax:

ary = (1..10).to_a

pp using ary

iterate

The second is a for style loop, called iterate, that iterates over all of the members of a collection.

ary = [1, 2, 3]

iterate(num, using ary) { puts num }

# This is the same as:

ary.each { |num| puts num }

This works with any collection that has an #each method.

for

The third structure is a C-style for loop.

for(initialization, test, update) { # block contents to iterate on }

The initialization, *test, and update items are code that will be evaluated during the execution of the loop. The code should be provided as either a String or a Proc(Nil, Nil).

The initialization code gets executed a single time, before the loop starts, and can be used to setup any state that is required before the loop runs.

The test code will be evaluated at the start of each iteration. If it evaluates to false, then the loop body will execute. If it evaluates to true then the loop exits immediately.

The update code executes after the loop body, and can be used to change counters or other state that the test code may depend on to determine if the loop has finished execution.

This structure could be legitimately useful. The loop so created is a closure, and can be captured in a variable.

# Specify code with strings.
for(%(t = 0), %( t < 10), %(t += 1)) do
  # Do Stuff
end

# Specify code with procs, and leverage the fact that it is a closure.
t = uninitialized Int32
for(->{ t = 0 }, ->{ t < 10 }, ->{ t += 1 }) do
  # Do Stuff
end
puts t
# => 10

# The *for* loop can be assigned to a variable and called later.
ary = (1..10).to_a

t = uninitialized Int32
transform = for(->{ t = 0 }, ->{ t < 10 }, ->{ t += 1 }, run: false) { ary[t] = ary[t] * ary[t] }

# Do some stuff here...

transform.call

pp ary
# => [1, 4, 9, 16, 25, 36, 49, 64, 81, 100]

do_until

The do_until loop is like the for loop, except that it is an exit-controlled loop. This means that the condition to test for exiting the loop is checked at the end of the loop body instead of at the beginning of the loop body. This guarantees that the loop body will execute at least one time. Additionally, the loop will be exited when the condition is true.

Thus, it is functionally the opposite of a for loop.

t = uninitialized Int32
do_until(->(){ t = 1 }, ->(){ t > 0 }, ->{ t += 1}) {puts t}

# => 1

puts t

# => 2

Other Information

There is also an article on The Relicans that might have some additional context and details about how this shard works.

Contributing

1. Fork it (https://github.com/wyhaines/for.cr/fork)
2. Create your feature branch (git checkout -b my-new-feature)
3. Commit your changes (git commit -am 'Add some feature')
4. Push to the branch (git push origin my-new-feature)
5. Create a new Pull Request

Contributors

• Kirk Haines - creator and maintainer

git@github.com:h2o/picohttpparser.git