Over the week-end, I implemented new groovy methods, as defined in GROOVY-208 Plus a few complementary methods.

• getText():
  • BufferedReader.getText()
  • File.getText()
  • File.getText(encoding)
  • Process.getText()

You can now easily read the full content of a file or of a buffered reader and get it as a String.

// retrieve the content of the file  
def content = new File("myFile.txt").text  
// you can specify the encoding of the file  
// note that since getText() has a parameter,   
// you cannot call it with something like text("UTF-8")   
content = new File("myFile.txt").getText("UTF-8") 

There is also a getText() method on Process which gathers the output of a process:

When you want to make some optimizations to your code, you often use the good old System.currentTimeMillis() method to time certain parts of your application. I wanted to do a similar thing when hacking some Groovy scripts recently, and the idea came to me that I could simply create a timing closure! So here it is, for your eyes only:

timer = { closure ->
    start = System.currentTimeMillis()
    closure()
    println System.currentTimeMillis() - start
}

Then you can use your brand new timing closure:

A year ago, Codehaus saw the light of day. And since, it gathered a lot of bright and talented persons working on innovative and very good quality projects. And moreover, those projects have a friendly licence scheme (I’m not a GPL/LGPL lover).

For a few months, I’ve been part of the Hausmates, thanks to my efforts in developing some code for Groovy, and I’m really proud of beeing part of it.

I wish a very happy birthday to Codehaus, and to all those who made it happen. Thank you Bob, and thank you James for Groovy ;-)

Perhaps you noticed recently that there’s a new interesting page on Groovy’s website ? Well, all pages are interesting of course! But there’s a new page describing the Groovy methods enhancing the core JDK classes.

In groovy, you have additional methods that you can call on standard Java classes. For instance, you can use the eachLine() method on java.io.File. With this method, you’ll be able to easily read a text file line after line, and do whatever with this line inside a closure without having to care about things like closing streams. Let’s illustrate this with an example :

Yesterday, I was granted commit rights on the Groovy source tree, at Codehaus. Of special interest for me :

• I developed a utility class which helps Groovy create new Readers for text files with the correct encoding already set (I’m really keen on charset/encodings issues),
• Adding some new Groovy methods to the core JDK classes,
• Working on an automatic documentation generation engine “a la” Javadoc, so that it may be possible to browse all the methods that have been added to the core JDK classes.

James asked me if I wanted a CVS access to deal with those matters of interest instead of tunneling through him, and I gladfully accepted. But, alas, here starts the nightmare…

That’s incredible, but Robert S. Sfeir is at it again in the EAP forum. He produced one more prayer that we should all be repeating again and again. (I hope he won’t mind me from quoting his holy words)

Our IDEA who art in heaven
Hollowed be thy name
Thy IDE come, the EAP done
On earth as it is in heaven,
Give us this day our daily build,
And forgive our bug reports,
As we forgive those who report against us,
But lead us not into frustration,
Now and forever…

Read more...

After my suggestion of Miranda as a code name for the upcoming EAP of IntelliJ IDEA, there was a funny quote from Robert S. Sfeir in the forums regarding this suggestion. It’s a new version of the Miranda Rights :

You have the right to refactor. Anything you refactor can be used against you in your code and calls. You have the right to have IDEA present now and during any future refactorings. If you cannot afford an IDEA license, one will be appointed to you, free of charge, during EAP, if you wish.

Read more...

IntelliJ IDEA 4.0 is not yet out of its Early Access Program phase that a discussion already started in the forums for the code name of the next EAP cycle.

There are already a few propositions for the next code name. Among them:

• StarGate (hmm, not that good)
• Avani (which means earth in some indian languages)
• Stella (which will not be used because it was already used for IDEA 2.0 for those who remember)
• Avalon (already used by Microsoft’s new UI framework, and the Apache Avalon framework)
• Rhea (suggested internally at JetBrains by Eugene Belyaev)
• Aardvark (maybe too harsh to pronounce)
• Atlantis (mythological sunken island)
• Athena (greek goddess)
• Lachesis, Clotho or Athropos (the three moirae)
• Ananke (necessity), Erebus
• Vedra (as in Vedra Valles)
• Miranda (that’s my suggestion, already used by Miranda-IM, and it reminds something to outlaws in the US)
• Olesya (CVS integration developer), Katja or Laika (first female dog in space)
• Derivations of Simple (Simplex, Simplicity, SimpleA…)
• Dione (another Greek goddess)
• Aphrodite, Athena, Adonis (to stick with “A” and the Greek mythology)
• Artemida (tiger hunter… JDK 1.5 hunter)
• Archimedes (another sinking idea…)
• Revolution (somewhat related to the Aurora cruiser)
• Banana (quite funny one)
• Donut (for hungry developers)

I like Miranda, of course, because that’s my suggestion, but I also like Rhea and Ananke, they sound good to my ears. What would you suggest ?

In the IntelliJ forums, I came across an off-topic (but funny) post by Robert Gibson who was wondering :

Somebody told me once that there are only two words in the English language which contain each vowel, once only, in alphabetical order. Anybody know what the other one is?

Indeed, there are more than two words corresponding to those constraints. I then wrote a little Java class which took all the words of a words file (with 100k words) and tested if they matched a regexp corresponding to those constraints.

Some time ago I came across an interesting article on Apache’s wiki entitled (Apache Wiki: The Elements Of Ant Style. It’s a very thourough overview of good practices for writing clean, readable, reusable and efficient Ant build files.

Today, through magpiebrain, I read on OnJava another very interesting and concise article giving 15 Ant best practices. It’s much shorter than the previous article on Apache’s wiki, but still, with those 15 practices you’ll manage to improve some messy build scripts. Believe me !