footle

It took me a bit of digging to figure this out (it kept failing with readline errors, and then iconv went missing), so I thought I’d share:

rvm package install readline
rvm package install iconv
rvm install 1.9.2 -C --enable-shared,--with-iconv-dir=$HOME/.rvm/usr,\
--with-readline-dir=$HOME/.rvm/usr,--build=x86_64-apple-darwin10

While poking around the database in my rails console, I often found myself jumping to the mysql console just so I could get an easier-to-digest view of the data. I finally had enough of this silliness and wrote up a quick function to dump of set of ActiveRecord objects in the mysql report format.

>> report(records, :id, :amount, :created_at)
+------+-----------+--------------------------------+
| id   | amount    | created_at                     |
+------+-----------+--------------------------------+
| 8301 | $12.40    | Sat Feb 28 09:20:47 -0800 2009 |
| 6060 | $39.62    | Sun Feb 15 14:45:38 -0800 2009 |
| 6061 | $167.52   | Sun Feb 15 14:45:38 -0800 2009 |
| 6067 | $12.00    | Sun Feb 15 14:45:40 -0800 2009 |
| 6059 | $1,000.00 | Sun Feb 15 14:45:38 -0800 2009 |
+------+-----------+--------------------------------+
5 rows in set

Grab it from GitHub and stick it in your .irbrc.

A colleague was asking today how he could see if there’s data waiting on STDIN in ruby (on Linux). On OS X, this is pretty straightforward:

if $stdin.stat.size > 0
  puts "got something: #{STDIN.read}"
else
  puts "nada"
end

That doesn’t work in Linux, though. After some digging, I found this post, which lead to:

require 'fcntl'

flags = STDIN.fcntl(Fcntl::F_GETFL, 0)
flags |= Fcntl::O_NONBLOCK
STDIN.fcntl(Fcntl::F_SETFL, flags)

begin
  puts "got something: #{STDIN.read}"
rescue Errno::EAGAIN
  puts "nada"
end

Which works on both platforms, but is (a) ugly (catching an exception), and (b) requires you to actually try to read from STDIN.

In playing around with that, though, I noticed that STDIN.fcntl(Fcntl::F_GETFL, 0) returned 0 if there was something on STDIN, and something non-zero (2 in OSX and 32770 in Linux) if STDIN was empty. So now the code is simple again:

require 'fcntl'

if STDIN.fcntl(Fcntl::F_GETFL, 0) == 0
  puts "got something: #{STDIN.read}"
else
  puts "nada"
end

I’m not sure how reliable that is on other platforms (it won’t work on Windows—fcntl is a Unix-y thing). If I’m understanding fcntl.h correctly, a return value of 2/32770 (0×8002) seems to indicate that the io stream in question is open for writing. I’m not sure if that is intended or just a side-effect. Anyone know?

I hadn’t posted this before because it seemed kind of trivial and silly, but I used it again the other day and decided maybe I should share the joy after all.

Far too often I find myself jumping through hoops building ActiveRecord find conditions arrays dynamically depending on what arguments were passed in to the method. So I cooked up a simple little class to clean that up a bit:

# class to construct conditions for AR queries
# cc = ConditionsConstructor.new('foobar > ?', 7)
# cc.add('blah is not null')
# cc.add('baz in (?)', [1,2,3])
# cc.conditions
# # => ["foobar > ? and blah is not null and baz in (?)", 7, [1, 2, 3]]
class ConditionsConstructor
  def initialize(str = nil, *args)
    @conditions_strs = str ? [str] : []
    @conditions_args = args
  end

  def add(str, *args)
    @conditions_strs << str
    @conditions_args += args
  end

  def conditions
    [@conditions_strs.join(' and ')] + @conditions_args
  end
end

I ran across a bug in Date::Format today, and after spending a few hours hacking away at a fix (the date/format.rb code is *uuuuugly* and *sloooow*…someone should really rewrite that. Better yet, rewrite it in C), I thought I’d submit a patch. So I grabbed the ruby_1_8 branch and lo and behold, my issue had already been fixed!

So the question now was how to monkeypatch the entire Date::Format module? Simply require-ing it as a plugin doesn’t work, since Date::Format is already loaded at that point. The trick then is to use load instead of require.

First, I tried this:

load 'date/format.rb'

(note: load needs the actual filename; it doesn’t have the magic that require does) but that gave me an error:

in `load': wrong number of arguments (1 for 0) (ArgumentError)

Turns out Rails::Plugin::Loader defines its own load, so:

Kernel::load 'date/format.rb'

et voila!

My colleague Coda Hale just released a sweet bcrypt-ruby gem that does password hashing right. It also provides for future-proofing by enabling you to assign a computational cost for generating a password hash and letting you version your password hashes. You have no more excuses.