The Joy of Prefetch

If you heard me speak at YAPC or you’ve had any kind of conversation with me over the last few weeks then it’s likely you’ve heard me mention the secret project that I’ve been writing for my wife’s school.

To give you a bit of background, there’s one afternoon a week where the students at the school don’t follow the normal academic timetable. On that afternoon, the teachers all offer classes on wider topics. This year’s topics include Acting, Money Management and Quilt-Making. It’s a wide-ranging selection. Each student chooses one class per term.

This year I offered to write a web app that allowed the students to make their selections. This seemed better than the spreadsheet-based mechanisms that have been used in the past. Each student registers with their school-based email address and then on a given date, they can log in and make their selections.

I wrote the app in Dancer2 (my web framework of choice) and the site started allowing students to make their selections last Thursday morning. In the run-up to the go-live time, Google Analytics showed me that about 180 students were on the site waiting to make their selections. At 7am the selections part of the site went live.

And immediately stopped working. Much to my embarrassment.

It turned out that a disk failed on the server moments after the site went live. It’s the kind of thing that you can’t predict.But it leads to lots of frustrated teenagers and doesn’t give a very good impression.

To give me time to rebuild and stress-test the site we’ve decided to relaunch at 8pm this evening. I’ve spent the weekend rebuilding the app on a new (and more powerful) server.

I’m pretty sure that the timing of the failure was coincidental. I don’t think that my app caused the disk failure. But a failure of this magnitude makes you paranoid, so I spent a lot of yesterday tuning the code.

The area I looked at most closely was the number of database queries that the app was making. There are two main actions that might be slow – the page that builds the list of courses that a student can choose from and the page which saves a student’s selections.

I started with the first of these. I set DBIC_TRACE to 1 and fired up a development copy of the app. I was shocked to see the app run about 120 queries – many of which were identical.

Of course I should have tested this before. And, yes, it’s an idiotic way to build an application. But I’m afraid that using an ORM like DBIx::Class can make it all too easy to write code like this. Fortunately, it makes it easy to fix it too. The secret is “prefetch”.

“Prefetch” is an option you can pass to the the “search” method on a resultset. Here’s an example of the difference that can make.

There are seven year groups in a British secondary school. Most schools call them Year 7 to Year 13 (the earlier years are in primary school). Each year group will have a number of forms. So there’s a one to many relationship between years and forms. In database terms, the form table holds a foreign key to the year table. In DBIC terms, the Year result class has a “has_many” relationship with the Form result class and the Form result class has a “belongs_to” relation with the Year result class.

A naive way to list the years and their associated forms would look like this:

Run code like that with DBIC_TRACE turned on and you’ll see the proliferation of database queries. There’s one query that selects all of the years and then for each year, you get another query to get all of its associated forms.

Of course, if you were writing raw SQL, you wouldn’t do that. You’d write one query that joins the year and form tables and pulls all of the data back at once. And the “prefetch” option gives you a way to do that in DBIC as well.

All we have done here is to interpose a call to “search” which adds the “prefetch” option. If you run this code with DBIC_TRACE turned on, then you’ll see that there’s only one database query and it’ll be very similar to the raw SQL that you would have written – it brings back the data from both of the tables at the same time.

But that’s not all of the cleverness of the “prefetch” option. You might be wondering what the difference is between “prefetch” and the rather similar-sounding “join” option. Well, with “join” the columns from the joined table would be added to your main table’s result set. This would, for example, create some kind of mutant Year resultset object that you could ask for Form data using calls like “get_column(‘’)”. [Update: I was trying to simplify this explanation and I ended up over-simplifying to the point of complete inaccuracy – joined columns only get added to your result set if you use the “columns” or “select/as” attributes. And the argument to “get_column()” needs to be the column name that you have defined using those options.] And that’s useful sometimes, but often I find it easier to use “prefetch” as that uses the data from the form table to build Form result objects which look exactly as they would if you pulled them directly from the database.

So that’s the kind of change that I made in my code. By prefetching a lot of associated tables I was able to drastically cut down the number of queries made to build that course selection page. Originally, it was about 120 queries. I got it down to three. Of course, each of those queries is a lot larger and is doing far more work. But there’s a lot less time spent compiling SQL and pulling data from the database.

The other page I looked at – the one that saves a student’s selections – wasn’t quite so impressive. Originally it was about twenty queries and I got it down to six.

Reducing the number of database queries is a really useful way to make your applications more efficient and DBIC’s “prefetch” option is a great tool for enabling that. I recommend that you take a close look at it.

After crowing about my success on Twitter I got a reply from a colleague pointing me at Test::DBIC::ExpectedQueries which looks like a great tool for monitoring the number of queries in your app.

Driving a Business with Perl

I’ve been a freelance programmer for over twenty years. One really important part of the job is getting paid for the work I do. Back in 1995 when I started out there wasn’t all of the accounting software available that you get now and (if I recall correctly) the little that was available was all pretty expensive stuff.

At some point I thought to myself “I don’t need to buy one of these expensive systems, I’ll write something myself”. So I sat down and sketched out a database schema and wrote a few Perl programs to insert data about the work I had done and generate invoices from that data.

I don’t remember much about the early versions. I do remember coming to the conclusion that the easiest way to generate PDFs of the invoices was using LaTex and then wasting a lot of time trying to bend LaTeX to my will. I got something that looked vaguely ok eventually, but it was always incredibly painful if I ever needed to edit it in any way. These days, I use wkhtmltopdf and my life is far easier. I understand HTML and CSS in a way that I will never understand LaTeX.

Why am I telling you this, twenty years after I started using this code? Well, during this last week, I finally decided it was time to put the code on Github. There were two reasons for this. Firstly, I thought that it might be useful for other people. And secondly, I’m ashamed to admit that this is the first time that the code has ever been put under any kind of version control (and, yes, this is an embarrassing case of “do as I say, not as I do“). I have no excuses. The software I used to drive my business was in a few files on a single hard drive. Files that I was hacking away at with gay abandon when I thought they needed changing. I am a terrible role model.

Other than all the obvious reasons, I’m sad that it wasn’t in version control as it would have been interesting to trace the evolution of the software over the last twenty years. For example, the database access started as raw DBI, spent a brief time using Class::DBI and at some point all got moved to DBIx::Class. It’s likely that I wasn’t using the Template Toolkit when I started – but I can’t remember what I was using in its place.

Anyway, the code is there now. I don’t give any guarantees for its quality, but it does the job for me. Let me know if you find any of it interesting or useful (or, even, laughable).

p.s. An interesting side effect of putting it under (public) version control – since I uploaded it to Github I have been constantly tweaking it. The potential embarrassment of having my code available for anyone to see means that I’ve made more improvements to it in the last week that I have in the previous five years. I’m even considering replacing all the command line programs with a Dancer app.

p.p.s. I actually use FreeAgent for all my accounting these days. It’s wonderful and I highly recommend it. But I still use my own system to generate invoices.

Subroutines and Ampersands

I’ve had this discussion several times recently, so I thought it was worth writing a blog post so that I have somewhere to point people the next time it comes up.

Using ampersands on subroutine calls (&my_sub or &my_sub(...)) is never necessary and can have potentially surprising side-effects. It should, therefore, never be used and should particularly be avoided in examples aimed at beginners.

Using an ampersand when calling a subroutine has three effects.

  1. It disambiguates the code so the the Perl compiler knows for sure that it has come across a subroutine call.
  2. It turns off prototype checking.
  3. If you use the &my_sub form (i.e. without parentheses) then the current value of @_ is passed on to the called subroutine.

Let’s look at these three effects in a little more detail.

Disambiguating the code is obviously a good idea. But adding the ampersand is not the only way to do it. Adding a pair of parentheses to the end of the call (my_sub()) has exactly the same effect. And, as a bonus, it looks the same as subroutine calls do in pretty much every other programming language ever invented. I can’t think of a single reason why anyone would pick &my_sub over my_sub().

I hope we’re agreed that prototypes are unnecessary in most Perl code (perhaps that needs to be another blog post at some point). Of course there are a few good reasons to use them, but most of us won’t be using them most of the time. If you’re using them, then turning off prototype checking seems to be a bad idea. And if you’re not using them, then it doesn’t matter whether they’re checked or not. There’s no good argument here for  using ampersands.

Then we come to the invisible passing of @_ to the called subroutine. I have no idea why anyone ever thought this was a good idea. The perlsub documentation calls it “an efficiency mechanism” but admits that is it one “that new users may wish to avoid”. If you want @_ to be available to the called subroutine then just pass it in explicitly. Your maintenance programmer (and remember, that could be you in six months time) will be grateful and won’t waste hours trying to work out what is going on.

So, no, there is no good reason to use ampersands when calling subroutines. Please don’t use them.

There is, of course, one case where ampersands are still useful when dealing with subroutines – when you are taking a reference to an existing, named subroutine. But that’s the only case that I can think of.

What do you think? Have I missed something?

It’s unfortunate that a lot of the older documentation on CPAN (and, indeed, some popular beginners’ books) still perpetuate this outdated style. It would be great if we could remove it from all example code.

Dev Assistant

A couple of days ago, I updated to my laptop to Fedora 21. One of the new features was an application called DevAssistant which claimed that:

It does not matter if you only recently discovered the world of software development, or if you have been coding for two decades, there’s always something DevAssistant can do to make your life easier.

I thought it was worth investigating – particularly when I saw that it had support for Perl.

Starting the GUI and pressing the Perl button gives me two options: “Basic Class” and “Dancer”. I chose the “Basic Class” option. That gave me an dialogue box where I could give my new project a name. I chose “MyClass” (it’s only an example!) This created a directory called MyClass in my home directory and put two files in that directory. Here are the contents of those two files.

It’s great, of course, that the project wants to support Perl. I think that we should do everything we can to help them. But it’s clear to me that they don’t have anyone on the team who knows anything about modern Perl practices.

So who wants to volunteer to help them?

Update: So it turns out that the dev team are really responsive to pull requests 🙂

“I Do Not Want To Use Any Modules”

Almost every day on the Perl groups on LinkedIn (or Facebook, or StackOverflow, or somewhere like that) I see a question that includes the restriction “I do not want to use any modules”.

There was one on LinkedIn yesterday. He wanted to create a MIME message to pass to sendmail, but he didn’t want to install any modules. Because “getting a module installed will have to go though a long long process of approvals”.

And I understand that. I really do. We’ve all seen places where getting new software installed is a problem. But I see that problem as a bug in the development process. A bug that needs to be fixed before anything can get done in a reasonable manner. Here’s what I’ve just written in reply:

Of course it can be achieved without modules. Just create an email in the correct format and pass it to sendmail.

Ah, but what’s the right format? Well, that is (of course) the tricky bit. I have no idea what the correct format is. Oh, I could Google a bit and come up with some ideas. I might even find the RFC that defines the MIME format. And then I’d be able to knock up some code that created something that looked like it would work. But would I be sure that it works? In every case? With all the weird corner-cases that people might throw at it?

This is where CPAN modules come in handy. You’re using someone else’s knowledge. Someone who is (hopefully) an expert in the field. And because modules are used by lots of people, bugs get found and fixed.

A lot of modern Perl programming is about choosing the right set of CPAN modules and plumbing them together. That’s what makes Perl so powerful. That’s what makes Perl programmers so efficient. We’re standing on the shoulders of giants and re-using other people’s code.

If you’re not going to use CPAN then you might as well use shell-scripting or awk.

If you’re in a situation where getting CPAN modules installed is hard, then fixing that problem should be your first priority. Because that’s a big impediment to your Perl programming. And investing time in fixing that will be massively beneficial to you in a very short amount of time.

The obvious solution is to install your own module tree (alongside your own Perl) as part of your application. But that might be overkill in some situations, so you could also consider using the system Perl and asking your sysadmin to install packages from your distribution’s repositories. Of course, that might need a change in process. But it’s a change that is well worth making; a change that will improve your (programming) life immensely.

Update: Some very interesting discussion about this over on Reddit.