James Crisp

Coding, calisthenics, book reviews, mind hacks and the occasional personal bit.

Month: December 2007

Practical JRuby on Rails (Web 2.0 Projects) by Ola Bini

The fine folk at Apress sent me a copy of Ola Bini‘s new book to review. The full title is “Practical JRuby on Rails Web 2.0 Projects – Bringing Ruby On Rails to the Java Platform”. Overall, it was a good read, and extremely valuable to anyone who is developing in JRuby. JRuby information and documentation is scarce and most of the time, a Google trawl does not give you good results on a JRuby related query. Ola’s is the first, and currently the only JRuby book available, and in my experience, the most valuable resource available to give you an all-round picture of JRuby capabilities and usage.

Despite comments on the cover, I would suggest that this book is not ideal for people new to Ruby / Rails. Ola jumps in the deep end quite quickly, and being a talented Ruby programmer, makes use of lots of shorthand, procs, code blocks etc which would likely be hard to follow for someone new to Ruby. Although there is a section at the back called “Ruby for Java programmers”, I think this would not be sufficient for somebody new to Ruby to understand all the code examples.

To get the most value out of the book, it would be good to have at least a basic understanding of Ruby and Rails (eg, having read Agile Web Development with Rails or messed around with Ruby/Rails a bit) and a basic understanding of Java syntax, deployment and Java EE.

What’s Covered?
The book is project based, so as to give context and useful examples of JRuby functionality. There are 4 projects:

  • The Store (Shoplet) – a standard Rails app running under JRuby using Active Record JDBC.
  • Content management system – general Java integration and using Java libraries for content rendering.
  • Administration System – using EJBs, JMX and discussion of JRuby deployment options.
  • Library System – JRuby as the “glue that never sets”. Using Java Web service frameworks and JMS from JRuby.

The Good

  • Teaches you how to do all those tricky bits which are half-Java and half-Ruby and can’t be easily found online, such as converting between Ruby and Java types, including JAR files, implementing Java interfaces, etc
  • Clever and concise Ruby code – I picked up some Ruby tricks reading Ola’s code.
  • Complex code snippets are generally well explained in text.
  • Useful tips on when to use Java libraries and when to use Ruby ones.
  • Generally good and interesting example projects which justified the use of JRuby and the techniques shown in the book.
  • Helpful discussion of JDBC and database connectivity options for JRuby.
  • Nice overview of the many JRuby deployment options.
  • Helpful “sidebars” about Java Enterprise Edition technologies.
  • Covers the strong areas of JRuby well – web applications and system integration.
  • Appendices provide useful reference information.
  • Nice section at the end on how you can get involved in JRuby.

The Less Good

  • Example views often contain table layouts, inline styles and other layout information that would be better done in separate CSS files.
  • Variable names in code could be more descriptive. This would make example code easier to follow.
  • Occasional odd spelling like “sur_name” and use of deprecated Rails features, such as “start_form_tag” (to be fair though, Rails API does change very quickly).
  • The title suggests that the book is about Web 2.0. There is a little token AJAX, and I suppose a content management system is a bit Web 2.0, but overall, buy the book if you want to know about JRuby, not Web 2.0.
  • Although REST is only mentioned briefly in a little sidebar, and not a focus of the book, I found the description of REST and CRUD a bit misleading, especially when considering PUT vs POST.
  • The discussion of JRuby deployment provides a good overview, but more in depth discussion of major options (eg, GoldSpike), and production configurations would be great.

As the best and only JRuby reference, I’d highly recommend you buy a copy if you are working in, or planning to work in JRuby. The book will help you to write JRuby applications which make good use of Ruby, Rails, Java libraries and Java Enterprise Edition features.

Treo 650 incompatible with Optus Postpaid SIMs

Over the last 6 months, my old Treo 600 touch screen was dying. Cleaning and re-seating helped for a while, but eventually, the touch screen stopped working. As the Treo 680 barely improves on the Treo 650, and costs a lot more, I bought a new-ish Treo 650 on eBay. All went well until I tried to send SMSes using my Optus SIM. No luck – I got an error 4027 every time I tried to send a SMS, although voice and GPRS worked fine.

First attempt at a solution was to set the Message Centre Number in the phone. Strangely, the Message Centre number (configured though the SMS app) was always empty, and when I tried to set it to the correct number for Optus, the value wouldn’t stick past a phone reset.

Thanks to Google, I found the outline of a solution on a message board.

In summary, the Treo 650 (unlike the Treo 600) is not compatible with Optus’s standard post-paid/plan SIMs used in Australia. However, it is compatible with the Optus pre-paid SIMs. If you have a Treo 650, and a post-paid/plan SIM, you can buy yourself a new pre-paid SIM for $4 from your local Optus store, then ring the Optus call centre to have them do a “SIM replacement” to move your account and number to the new SIM. This solves the problem within minutes of the “SIM replacement”.

Automating MySQL Database Backups

Great, so you’ve got your web application up and running on some server somewhere, and thousands of users are hard at work entering their valuable data into your application, and thence to a MySQL database. Now people depend on your system to keep their data safe. What are you going to do about it?

The approach I’ve taken is to run a backup script on a cron job, which dumps data from the database a few times a daily, compresses it, and emails it off to my GMail account. All those GMail gigabytes have got to be good for something, right :-)?

I started writing a simple backup shell script using mysqldump, gzip and sendmail, but ran into problems on RailsPlayground as sendmail is not supported there. RailsPlayground suggested this perl script for mysql backup (mirrored on my server). The script does pretty much what I was looking for, and also supports skipping tables (useful for ignoring things like session tables, which, for my app, are not really worth backing up).

A little more investigation revealed that the script wouldn’t work out of the box, as RailsPlayground requires SMTP auth. The backup script is not hard to enhance to support SMTP auth. It can be done as follows:

In the configuration section of the file (around line 203), add 2 new variables:

$smtp_user           = "my username"
$smtp_password       = "my password"

Around line 1200, add in the AuthUser and AuthPass:

MIME::Lite->send("$send_method", "$mailprog_or_smtp_host", Timeout=>60,
            AuthUser=>$smtp_user, AuthPass=>$smtp_password);

As an aside, this script can also do backups via FTP.

Restore of mysql dump files is pretty easy. You just use the standard client:

mysql -uMyUser -p MyDatabaseName < SqlDumpFile.sql

Some Murakami Quotes

Recently, I’ve been catching up on my reading. Haruki Murakami is one of my all time favourite authors. His novels are always interesting, and the writing style is generally gorgeous, although it does vary a little between different books. I’m not sure if this is Haruki Murakami changing his style in the original Japanese, or simply results from different translators.

Something I’ve been noticing lately is that in all of the Murakami novels I can recall, the protagonist always has a lot of time. This is so different to most other novels which try to rush from one exciting event to the next. Perhaps this is part of the reason why I really like Murakami’s writing so much. Anyway, here’s some memorable quotes from Murakami novels that I’ve read recently:

From Kafka on the Shore:
“Perhaps most people in the world aren’t trying to be free, Kafka. They just think they are. It’s all an illusion. If they really were set free, most people would be in a real pickle. You’d better remember that. People actually prefer not being free.”

“Pointless thinking is worse than no thinking at all”.

From Hard-Boiled Wonderland and the End of the World:
“Huge organisations and me don’t get along. They’re too inflexible, waste too much time, have too many stupid people.”

“It’s frightening,” she said. “Most of my salary disappears into my stomach.”

From The Wind Up Bird Chronicles: (language warning on this one)
Show quote, although it contains 4-letter words.

Opera Browser – Lean and quick

I’ve been using Firefox since its inception, and its glorious introduction of tabbed browsing. However, every new update, Firefox seems to get ever slower, and use more memory and more processor.

Finally, last week, after Firefox hung yet again using 100% of processor and over 300mb of memory, I decided to give Opera a go. So far, I’ve been very satisfied. It was a small download, easy install and uses about 1/3 of the memory and significantly less processor than Firefox does, especially when you have many tabs open and leave the same instance of the browser running for several days. I’ve had one crash when setting up Adobe Acrobat preferences, but in general browsing, it is yet to break. Pages load faster, and rendering of HTML and CSS seems fine. The only problem I’ve had is with Lotus Notes Webmail saying Opera is “not supported” and not wanting to run.

So far, I’ve installed Opera on older Windows and Mac boxes, and been very happy with the improved browsing experience. I still keep my Firefox handy though, mainly for its great web development plug-ins (Firebug and friends).

Powered by WordPress & Theme by Anders Norén