jmacaulay.net - Home

Active Shipping

2008-04-21T21:05:00Z

Active Shipping is now available under an open source license. It's on GitHub, here. Active Shipping is an extension for Active Merchant which I wrote to let Shopify users provide carrier-calculated shipping rates to their customers. You can easily include Active Shipping in your Rails project as a plugin (or in any other ruby project, for that matter) and start doing things like this:

include ActiveMerchant::Shipping
usps = USPS.new(:login => '123JAMES4567')

origin = Location.new(:country => 'US', :zip => '90210')
destination = Location.new(:country => 'CA', :postal_code => 'K2P 0K3')

grams = 100
centimetres = [20,10,10]
packages = [Package.new(grams, centimetres)]

response = usps.find_rates(origin, destination, packages)
rates = response.rates.sort_by(&:price)
cheapest = rates.map {|r| [ r.service_name,
                            r.price.to_f / 100, # rates returned in cents
                            r.currency]}.first
# => ["USPS First-Class Mail International", 1.62, "USD"]

It's under active development and more details for contributing to the project can be found in the README.

Rails migrations in local git branches: not so atrocious now

2008-02-14T04:29:00Z

One of the really annoying things about working on a Rails project with other developers is the inevitable conflicts with migration numbering. We recently switched to Git for version control on Shopify, and we all really love the ease of branching and merging to our hearts’ content. Here is a snippet of my workflow today:

Edit in my local “shipping” branch, adding some functionality to Shopify’s shipping system.
Commit a simple migration which adds a column that’s necessary for the aforementioned functionality.
Switch to my master branch and pull from our central repository to get the latest stuff.
Notice that Cody has added some crazy migration with an SQL query that I don’t really want to understand.

Now, at this point, my usual required course of action would be to switch back to my shipping branch, migrate down, rename my migration (thankfully there was just one this time), then git-rebase onto the updated master branch and migrate back up again. This happens fairly often, and it makes the world a darker place.

So I wrote this script. Save it as .git/hooks/pre-rebase in your git repository and make it executable. Now whenever you rebase one of your local development branches it will detect potential migration conflicts with the target of the rebase and do most of the dirty work for you: it migrates down, renames your branch’s migrations so they’ll be at the end of the line, generates a new commit for you with the new renaming, then lets the rebase proceed.

It prompts before it does the renaming or committing, so if you really, really want to have “23_foo.rb” and “23_bar.rb” after your rebase, then all the power to you. The prompts also give you a chance to ctrl-C out of the whole thing if you realize that the migrations will need more tweaking.

#!/usr/bin/env ruby
# Save to .git/hooks/pre-rebase in your rails project and chmod +x
# 
# Checks for migration numbering conflicts and sorts them out automagically.
# Intended for use in multi-dev environments where you have a local branch for
# your own development that you rebase every so often onto an updated (i.e.
# freshly pulled) master branch. Conflicts will still come up, but this covers
# 90%+ of cases.

root_dir = File.expand_path(File.join(File.dirname(__FILE__), "..", ".."))
orig_dir = `pwd`.chomp

`cd #{root_dir}`

# uses parameters passed by git-rebase
upstream = ARGV[0]
subject = ARGV[1] || "HEAD" 
migration_changes = `git diff #{subject} #{upstream} --summary | grep db/migrate`
migration_ary= migration_changes.split(/\n/)

# not really deletions in terms of the rebase; just named so by the diff output
# really these "deletions" are just the current branch's migrations that don't exist
# in the target of the rebase
deletions = migration_ary.grep(/^\s*delete/)
other_changes = migration_ary - deletions

# filenames of any migrations affected in one way or another by the rebase
migration_files = migration_ary.map {|line| line.split.last }

version_pattern = /^db\/migrate\/(\d+)/

branch_migration_files = deletions.map {|line| line.split.last }.sort
deletion_versions = branch_migration_files.map {|file| file.match(version_pattern)[1].to_i }.sort
other_changes_versions = other_changes.map {|line| line.split.last.match(version_pattern)[1].to_i }.sort

# detect duplicate versions by looking at intersection of arrays
conflicts = deletion_versions & other_changes_versions
conflict_files = migration_files.select {|f| conflicts.include? f.match(version_pattern)[1].to_i }

if conflicts.empty?
  puts "\nNo migration conflicts detected." 
else
  puts "\n*** Migration conflicts detected" 
  puts "\nLocal branch migrations not present in rebase target:" 
  puts(deletions.map {|line| line.gsub(/^.*(db\/migrate)/,      ' \1')}.join("\n") + "\n")
  puts "\nMigrations in rebase target to be added or modified:" 
  puts(other_changes.map {|line| line.gsub(/^.*(db\/migrate)/,  ' \1')}.join("\n") + "\n")

  print "\nFix by migrating down and renaming branch migrations? (Y/n) " 
  response = STDIN.readline.chomp.strip
  if response == '' or response =~ /^[Yy]/

    # migrate down
    puts (cmd = "rake db:migrate VERSION=#{(migration_files.map {|file| file.match(version_pattern)[1].to_i }).min - 1}")
    puts `#{cmd}`
    last_version = other_changes_versions.max

    # move each migration unique to the current branch to the end of the migration list
    branch_migration_files.each do |file|
      last_version += 1
      `git mv #{file} #{new_name}`
      puts "\nmoved #{file}" 
      puts "to    #{new_name}" 
    end

    print "Generate a new commit? (Y/n) " 
    response = STDIN.readline.chomp.strip
    if response == '' or response =~ /^[Yy]/
      puts(cmd = "git commit -a -m \"Renamed migrations\"")
      puts `#{cmd}`
    end

    puts "\n*** Be sure to run rake db:migrate after rebase has completed.\n" 
  else
    puts "Okay, not doing anything then." 
  end
end

`cd #{orig_dir}`

exit 0

Modified rss_reader Radiant extension

2007-11-09T03:37:00Z

Part of my work at jaded Pixel a few months ago was development and writing for a new “brochure” site at www.shopify.info. The current site, like the old one, is built with the fantastic Radiant CMS and uses a modified version of the rss_reader extension from BJ Clark, Loren Johnson, and Alessandro Preite Martinez.

We are using it to grab feeds that we publish in other places and re-display them to fit with the design of the site. For example our screenshots page is generated from our blog’s Shop of the Moment feed, and our list of supported methods for accepting payment is fed directly by Shopify itself.

Here is a modified version of the extension, available for your enjoyment and scrutiny:

rss_reader-0.2a-jadedpixelmod.tgz

Just unpack with tar xvzf rss_reader-0.2a-jadedpixelmod.tgz in your Radiant app’s root directory and it’ll put the necessary stuff in lib and vendor.

The modifications that I made add a few new features and, thanks to Tobi’s help, improved the robustness of the feed fetching code quite a bit. So for example, now you can order the feed however you want using a syntax similar to SQL’s ORDER BY:


<ul>

  <r:feed:items
      url="http://feeds.boingboing.net/boingboing/iBag" 
      order="creator ASC">

    <li><r:feed:link /></li>

  </r:feed:items>

</ul>

You can also do headers to mark off sections:


<ul>

  <r:feed:items
      url="http://feeds.boingboing.net/boingboing/iBag" 
      order="creator ASC">

    <r:feed:header for="creator">
      <h2><r:feed:creator /></h2>
    </r:feed:header>

    <li><r:feed:link /></li>

  </r:feed:items>

</ul>

This would show the code within the <r:feed:header /> tag if and only if the “creator” attribute of the item is different from the previous item in the list. Thus, it would only make reasonable output if the feed were ordered by creator with the above method. Another more obvious grouping for headers would be the item’s “date” attribute. Yes, the date of a feed item has hours and minutes and seconds, but I made it so that a new header only appears on new days of the month.

You can sort items and group headers by date, title, content, creator, or link (i.e. the URL of the item). There are a few other options for the feed tags that I haven’t mentioned here but which are documented within the code.

Facebook find: father-daughter moment

2007-06-02T00:01:00Z

A browser history for the command line

2007-05-25T15:16:00Z

I'm not a shell expert by any means. I'm certainly comfortable enough on the command line and can pipe I/O fairly effectively; however, I know very little about bash scripting, and irb and svn really cover the vast majority of my Terminal usage aside from the ubiquitous ls and cd.

Even humble cd has features which I had not initially fathomed: it was a wonderful moment when I recently saw Tobi use cd - to flip back and forth from the last directory he had visited. I had been vaguely yearning for a way to go back directories for a while, but never got around to figuring it out. This cd - thing wasn't really what I was looking for, though, because what I wanted was the equivalent of the back and forward buttons in a web browser and cd - doesn't have a memory of anything more than the last directory you were in, via the environment variable OLDPWD.

So I did some googling and found the necessary components to come up with something that is pretty darn satisfying.

How it works

I added three short functions in ~/.bash_profile:

function cd {
    pushd "$@" > /dev/null
}

Instead of its regular behaviour, cd is now mapped to a nifty command called pushd. This command takes a directory path as an argument and changes to that directory, just like cd does, but it also stores the new path in a stack stored in an environment variable.

It outputs the contents of that stack whenever you use pushd, but I don't want to see it every time I change directory which is why I've redirected standard output to oblivion. STDERR is left untouched so if I type a nonexistent directory then I will still find out about it. The "$@" provides pushd with all the arguments I pass to cd separated by spaces, so everything stays the way I typed it; if I were to use "$1" here, then it would choke on directory names with spaces in them.

Now that we are building onto our stack of visited directories every time we cd anywhere, we can use this as a browser history for the shell:

function bd {
    if [ -z $1 ]; then
        n=1
    else
        n=$1
    fi
    pushd +$n > /dev/null
}

function fd {
    if [ -z $1 ]; then
        n=0
    else
        n=$[$1-1]
    fi
    pushd -$n > /dev/null
}

bd is for "back a directory" and fd is for "forward a directory". We are still just using pushd to manipulate the stack, but now we are using a different kind of argument for it. pushd +n will rotate the stack backwards n steps and change to that directory. We give n a default value of 1 when no arguments are passed to bd, and do pushd +n.

For fd, we rotate the other way. The default for n is now 0 and we decrement the argument by 1 otherwise because pushd -0 refers to the bottom element of the stack. If we go forward before going backward, it doesn't care and will just rotate back to the first pushed directory (which is probably ~) and continue rotating from there, which is fine by me.

How it plays

So now I can go back and forward with bd and fd on their own, or do bd 8 to go back wherever I was eight directories ago and then maybe a little fd 3 to revise my search.

Each shell gets its own history which just accumulates and accumulates. If I feel like keeping a Terminal window open for a few years then I can do dirs -c to clear the stack. The way I have this set up, cd without any arguments will swap back and forth between the current and last directory as cd - would normally do, because that's how pushd behaves when you type it by itself. Normally cd by itself would be the equivalent of cd ~, but I never used cd like that in the first place so I'm not missing anything.

There's obviously room to modify this formula and use this wonderful directory stack in different ways. Here is a straightforward reference for the directory stack commands in bash. There's also a few tilde expansions that reference the directory stack which some people might find useful.

Wikipedia find: the refined nobility of the Harpy Eagle

2007-05-05T19:12:00Z

Oh and besides monkeys, they also tear apart sloths on a regular basis, for crying out loud. When I ponder what pure evil might be, I think I will now invariably picture one of these bastards.

My bitchin' new job

2007-05-03T02:31:00Z

My exams are all done and I am now working full-time for jaded Pixel, developers of the incomparable Shopify. I can't really think of a better place to work: I get to make money by programming with my favourite tools, working on a product which I love, on a team comprised of such a stellar group of co-workers I could almost weep.

Oh, and my commute is a five minute walk.

Yet another nice thing about this is that my wonderfully forward-thinking university is taking care of much of my salary, specifically for the work that I'm doing on the various open-source projects which jaded Pixel uses and maintains.

I first met the jaded Pixel crew after installing Typo, clicking through one of the default sidebar links to Tobi's blog, and then scrolling down and seeing this post. The post informed me that this Tobias fellow whose last name I wasn't sure how to pronounce did not live half-way around the world as I expected, but rather right here in Ottawa (and as I discovered shortly thereafter, within spitting distance of my apartment).

So I started attending meetups for OGRE, the local Ruby users' group mentioned in said blog post, and ended up getting a short gig writing help docs when Shopify officially launched almost a year ago. Now I'm back for the foreseeable future and doing a little bit of everything: back-end, front-end, user support, making coffee, and stealing things from Scott's desk whenever possible. It rawks.

Microsoft's innovative approach to user security

2007-04-21T21:12:00Z

An excerpt out of my Adium chat logs from a Thursday night chat over MSN. To set the stage: a number of jovially intoxicated barhoppers had just made their way loudly past my window…

James: hmm…screaming hooligans outside
Could not send; a connection error occurred.
James: hmm…screaming hooligans outside
Could not send; a connection error occurred.
James: you there?
Caitlin: yeah?
James: not letting me send a message a couple times
James: hmm…screaming hooligans outside
Could not send; a connection error occurred.
James: bah, it’s happening again…didn’t this happen before? maybe it was with someone else
James: some random string of characters will cause a connection error
James: screaming hooligans outside
James: hmm…
James: hmm… screaming hooligans outside
James: hmm…screaming hooligans outside
Could not send; a connection error occurred.
Caitlin: weird
James: only does it when I remove the space between hmm… and screaming
Caitlin: tell them to shut up
James: .screaming
Could not send; a connection error occurred.
James: .s
James: .scre
Could not send; a connection error occurred.
James: .sc
James: .scr
Could not send; a connection error occurred.
James: dot-s-c-r doesn’t work
James: scr
James: ..sc
James: strange
Caitlin: what’s wrong?
James: nothing…just this string of characters being rejected by msn for whatever reason
James: .scr
Could not send; a connection error occurred.
James: try sending me [period]scr
James: like . scr without the space
James: .sca
James: .scb
James: .scc
James: .scr
Could not send; a connection error occurred.
James: .scd
James: .sce
James: .sar
James: .bcr
James: weird
Caitlin: it’s gotta be your messenger
James: did you send one to me?
Caitlin: yup
James: didn’t get it
James: yeah I assume it’s my client (Editor’s note: I’m sorry I doubted you, Adium.)
Caitlin: why were you trying to send it to me
James: because I narrowed it down to that sequence of characters from my original message, which was “hmm…[no space]screaming hooligans outside”
James: which gave me the initial connection error messages
Caitlin: strange
James: damn straight

If you haven’t guessed it yet, here is the punchline, courtesy of Stu Tomlinson from the newly-renamed Pidgin project (née Gaim):

If you try to send a message containing any of the words “.pif”, “.scr”, “download.php” or “gallery.php” the message will be silently discarded. The official client from Microsoft will provide no indication to either the sender or recipient that the message didn’t go through.

(For those of you who might not know, .scr is the extension of Windows screen-saver files, which are basically just disguised .exe’s. This makes them a popular choice for those in the business of spreading worms and viruses and so forth.)

I tried a bunch of “.scr” messages using Microsoft’s official messenger app for OS X instead of Adium, and it did tell me they weren’t sent, but only after a long delay: I would send a bunch of the messages, and then twenty or thirty seconds later I would get the error messages for all of them. Regardless, MSN gives you no indication why your message wasn’t sent, leaving people to wonder at the fickle gods of the internets for not letting them talk about screaming hooligans (or what-have-you).

I don’t know how the MSN protocol works, but it seems pretty absurd that it would be designed in such a way that this kind of nightmarish hack would become necessary in the first place. What would be even scarier, of course, is if MSN actually was a reasonably well-designed protocol, and Microsoft freely chose to subject their users to this nonsense instead of implementing one of the multitude of obviously better solutions.

Blue eyes

2007-03-19T19:19:00Z

One of my favourite webcomics is xkcd (“a webcomic of romance, sarcasm, math, and language”). A while ago, the author posted the most satisfying logic puzzle I’ve ever mulled over. Take a minute to read it before coming back.

Intriguing, no? Depending on what kind of person you are, you might want to take a minute or a week trying to figure it out before either succeeding or wanting to just find the answer by any means necessary. I don’t have much patience for these things, so I didn’t wait long to read the solution. Thankfully it was incredibly satisfying nonetheless.

I absolutely love the follow-up questions Randall provides. They are like little doorways into those strange and vast mindrealms of logic, highlighting all the ways that one can easily misunderstand the problem. If you’re having fun with all of this, there’s a fantastic amount of meaty discussion to chew on in this xkcd forum thread. Once you get to the second page of the thread, you get into some really thought-provoking arguments suggesting that the solution may not be valid after all. I think I agree with the solution provided, but the process of peeling away the layers of superfluous information and getting to the crux of the issue is really fun.

Guerilla GrÃ¤fenbergology

2007-03-08T05:31:00Z

Last Saturday I saw a staging of the Vagina Monologues put on at Carleton, which a friend of mine was performing in. I thought it was great. They did a good job of expressing the main thesis of the piece. For those of you who don’t know, the main thesis is that vaginas are pretty frigging awesome and they should be treated as such.

But the performances were just the first wave of the vaginamania to come. This thoughtfully placed educational material was posted in the men’s washroom just outside Rooster’s:

SS Cardiacs

2007-02-13T05:27:00Z

I was recently reminded of this band called SS Cardiacs, which I know only because of a random CBC Radio 3 podcast that I listened to a year or so ago. This podcast contained a particularly ass-kicking track that I ended up listening to quite a bit over the period of a few weeks.

Thanks to Radio 3’s resident pre-cogs and telepaths, they were able to have a track playing from this very band as soon as I visited the Radio 3 website. I happened to have another browser tab open at the time with the band’s label’s website and initially assumed that the music was coming from that page instead. Seriously, I think CBC has telepaths. People should investigate this.

Anyways, thanks to Radio 3’s resident web developers and the fine job they seem to be doing, I am able to provide you with a pleasant little link for you to hear that same kickass track which introduced me to the band way back when:

http://radio3.cbc.ca/play/band/SS-CARDIACS/Tell-Me-Again/

Is it working?

2007-01-26T06:23:00Z

Mr. Overwijk Draws a Perfect Circle

2007-01-16T05:20:00Z

This guy taught me grade 12 math:

A friend from high school passed the link my way just a minute ago. The funny thing is that when you Google his name now, it’s all different copies and references to this suddenly popular video. Go Mr. Overwijk!

Christmas sketch

2006-12-25T23:24:00Z

Dragging and dropping with ease

2006-12-11T07:37:00Z

I have no idea how well-known this is, but I figured I’d better share this OS X tip because I have been pining for something like it for ages without knowing it existed.

Extended preamble: it’s been a little more than a year since switching to the other side, and during that time a particular nicety has gradually solidified in my mind as the one thing that I really miss about Windows (until tonight, that is): right-click-dragging. If you drag a file from point A to point B in Windows, but use the right button to drag instead of the left, then you get a great mini-menu at the cursor wherever you drop the file. This menu lets you specify the nature of the drag, turning it into a copy, a move, or the creation of a shortcut. I use this all the time when I’m on a Windows PC, because it is such an efficient and dare-I-say elegant method of getting shit done.

At least OS X has the same default left-click-drag behaviour: dragging between two mounted volumes produces a copy, and dragging within the same volume results in a move. This makes sense from an OS standpoint, because each of those operations is simpler (and therefore less error-prone) for the computer to perform in the respective context. It also makes sense from a user standpoint, because you usually don’t want too many copies of the same content on one drive, but between multiple drives you are probably intending to keep a copy on the original.

No right-click-drag menu on a Mac, though. Not too surprising, since on most Macs the right mouse button is still a key and a button. I actually don’t mind ctrl-clicking, and one reason is that I’m not forced to do it very often because context menus are not the central focus in OS X that they are in Windows.

So that brings me to the great news: a right-click-drag menu would be completely redundant in OS X, because Apple has included something much nicer:

Drag the file to the target location, while holding down any of these keys:
- Option to perform a copy
- Command to perform a move
- Option and command to create an alias
Release the mouse button

The mouse cursor will change appearance as you hold down the modifier keys to indicate a copy or alias operation. In fact you can change your mind mid-drag and switch up the keys whenever you want. The operation performed is only dependent on the keys being held when you release the mouse button.

The resulting file will be named identically to the first, unless a copy or alias is being put in the same folder as the original (in which case the relevant suffix is appended).

I came upon all this wonderful knowledge through a session of inspired experimentation: this TUAW post reminded me about Apple’s great little collection of random tips, which happens to contain this tip about command-click-dragging between volumes, which led me to try out the other aforementioned key combinations.

So is this as relieving to everyone else as it is to me?