Its not yet officially available, but can be downloaded directly from the codebase:
google-web-toolkit.googlecode.com/svn/trunk/ prebuilt folder in here  is the gwt-dev-plugin.xpi, and it works like a charm :)

Previously i wrote when the plugin was updated to work with Firefox 7 and 8.

More info in the group discussion: “GWT Developer Plugin for Firefox 9”

What timing, as Mozilla is likely to release Firefox 10 this coming Tuesday. For the sake of web developers that use GWT, i hope the plugin updates can come sooner in the future.

UPDATE 2012-02-02: This thread in the google group follows plugin development for Firefox 10.

I ended up creating this little handy script based on webkit2png (download source)

#!/usr/bin/env sh
HOST="http://localhost:8080/myproject/viewer"
/usr/bin/python webkit2png -F -o "01dashboard" $HOST/dashboard
/usr/bin/python webkit2png -F -o "02showdetail" $HOST/detail
/usr/bin/python webkit2png -F -o "03listpage" $HOST/list

$ ls -1
01dashboard.png
02showdetail.png
03listpage.png

<!-- settings.xml -->
<mirrors>
  <mirror>
    <id>insideFirewallRepo</id>
    <mirrorOf>*</mirrorOf>
    <name>Mirror of the world outside.</name>
    <url>http://nexus_url/nexus/content/groups/public/</url>
  </mirror>
</mirrors>

The reason for this being evil was our project pom files which had the following references

<properties>
  <nexus.url>http://nexus_url/nexus</nexus.url>
  <subversion.url>http://svn_url/svn</subversion.url>
</properties>

  <scm>
    <connection>scm:svn:${subversion.url}/common-gwt-gui/trunk</connection>
</scm>

<repositories>
    <repository>
        <id>nexus-repo-releases</id>
        <name>Nexus Repo</name>
        <url>${nexus.url}/content/groups/public/</url>
        <releases> <enabled>true</enabled> </releases>
        <snapshots> <enabled>false</enabled> </snapshots>
    </repository>
    <repository>
        <id>nexus-repo-snapshots</id>
        <name>Nexus Repo</name>
        <url>${nexus.url}/content/groups/public-snapshots/</url>
        <releases> <enabled>false</enabled> </releases>
        <snapshots> <enabled>true</enabled> </snapshots>
    </repository>
 </repositories>

[repeat the <repositories> section for <pluginRepositories>]

<distributionManagement>
    <repository>
        <id>company.releases</id>
        <url>${nexus.url}/content/repositories/company/</url>
    </repository>
    <snapshotRepository>
        <id>company.snapshots</id>
        <url>${nexus.url}/content/repositories/company-snapshots/</url>
    </snapshotRepository>
</distributionManagement>

Finally we have the maven release-plugin, but this irrelevant for now:

<build>
    <plugins>
        <plugin>
            <artifactId>maven-release-plugin</artifactId>
            <configuration>
                <tagBase>${subversion.url}/project_name/releases</tagBase>
            </configuration>
        </plugin>
    </plugins>
 </build>

Error message: Using the wrong repository for snapshots

Now, the generic mirrorOf * in settings.xml actually overrides the specific repository information we set in the pom.xml. Because of that setting, the pom.xml <repository> is not used. And the result is an error message like this. (error message enhanced in bold)

Downloading: http://nexus_url/nexus/content/groups/public/com/company/dependency_project/1.18-SNAPSHOT/maven-metadata
.xml
Downloading: http://nexus_url/nexus/content/groups/public/com/company/dependency_project/1.18-SNAPSHOT/maven-metadata
.xml
Downloading: http://nexus_url/nexus/content/groups/public/com/company/dependency_project/1.18-SNAPSHOT/dependency_project
-1.18-SNAPSHOT.pom
[WARNING] The POM for com.company:dependency_project:jar:1.18-SNAPSHOT is missing, no dependency information available
Downloading: http://nexus_url/nexus/content/groups/public/com/company/dependency_project/1.18-SNAPSHOT/dependency_project
-1.18-SNAPSHOT.jar
[INFO] ------------------------------------------------------------------------
[INFO] Reactor Summary:
[INFO]
[INFO] project_name ...................................... FAILURE [1.108s]
[INFO] ------------------------------------------------------------------------
[INFO] BUILD FAILURE
[INFO] ------------------------------------------------------------------------
[INFO] Total time: 1.364s
[INFO] Finished at: Fri Jan 20 12:28:04 CET 2012
[INFO] Final Memory: 5M/15M
[INFO] ------------------------------------------------------------------------
[ERROR] Failed to execute goal on project project_name: Could not resolve dependencies for project com.company:project_name:jar:2.1.1
3-SNAPSHOT: Could not find artifact com.company:dependency_project:jar:1.18-SNAPSHOT in insideFirewallRepo (http://nexus_url
/nexus/content/groups/public/) -> [Help 1]
[ERROR]
[ERROR] To see the full stack trace of the errors, re-run Maven with the -e switch.
[ERROR] Re-run Maven using the -X switch to enable full debug logging.
[ERROR]
[ERROR] For more information about the errors and possible solutions, please read the following articles:
[ERROR] [Help 1] http://cwiki.apache.org/confluence/display/MAVEN/DependencyResolutionException
[ERROR]
[ERROR] After correcting the problems, you can resume the build with the command
[ERROR]   mvn <goals> -rf :project_name

Notice that the errormessage actually gave the ID of our <mirror> from settings.xml:
<!-- settings.xml -->
 <mirrors>
   <mirror>
     <id>insideFirewallRepo</id>

<repositories>
<repository><id>nexus-repo-releases</id><repository>
<repository><id>nexus-repo-snapshots</id><repository>

<!-- settings.xml -->
<mirrors>
  <mirror>
    <id>insideFirewallRepo</id>
    <mirrorOf>!nexus-repo-releases,!nexus-repo-snapshots,*</mirrorOf>
    <name>Mirror of the world outside.</name>
    <url>http://nexus_url/nexus/content/groups/public/</url>
  </mirror>
</mirrors>

Most users could probably just remove the <mirrors> block from settings.xml, but in this particular project, that is not possible since firewall blocks for access to maven central. So, in order for us to use external dependencies, we must use the <mirror> in settings.xml

Then, you could argue, to use <proxy> instead of <mirrors> in settings.xml. This may be a decent solution as well, but in my case i wanted to avoid that, since the company proxy required user/password. Our jenkins build machines would then have my username/password, and it would lock my account whenever my password expires.

Within those constraints, the <mirror> block with excluded internal repositories seems like a decent fit.

For the issues described in this article, I have been using Maven 3.0.3 and Sonatype Nexus 1.9.2

So i ended up with some short search/replace, which are really handy in IntelliJ.

Replace “assertTrue” and “assertFalse” with “assertThat”

Search template

assertTrue($msg$, $actual$);

Replacement template:

org.junit.Assert.assertThat.assertThat($msg$, $actual$, org.hamcrest.core.Is.is(true));

Remove “assertEquals” with delta values

Search template

assertEquals($msg$, $expected$, $actual$, $delta$);

Replacement template:

org.junit.Assert.assertThat($msg$, $actual$, org.hamcrest.core.Is.is($expected$));

Optional Replace “assertEquals” with “assertThat”

Search template

assertEquals($msg$, $expected$, $actual$);

Replacement template:

org.junit.Assert.assertThat.assertThat($msg$, $actual$, org.hamcrest.core.Is.is($expected$));

Starting February 1st, i will join NineConsult, which is a Danish consultant company. I am really looking forward to this change, and it will be interesting to be part of a 25 people organisation (as opposed to the 110.000+ employees in Capgemini).

Having good and skilled colleagues is one of the single most important criteria for happiness at work. And this makes it extra interesting in NineConsult, since they have a very high standard on the programmers they employ.

I think NineConsult is one of the best places to be for programmers at the moment. Paul Graham puts it this way when describing programmers in “great hackers“:

[...] what good hackers like is other good hackers. Great hackers tend to clump together– sometimes spectacularly so, as at Xerox Parc. So you won’t attract good hackers in linear proportion to how good an environment you create for them. The tendency to clump means it’s more like the square of the environment. So it’s winner take all. At any given time, there are only about ten or twenty places where hackers most want to work, and if you aren’t one of them, you won’t just have fewer great hackers, you’ll have zero.

I know a lot of my new colleagues from previous jobs, and all of them (no exceptions) are very skilled.

For this single reason alone, i am really looking forward to my new job :)

Since Firefox 7 was released, i have defered upgrading for one simple reason: At work I am dependent on GWT Developer plugin.

Today I found the ready-to-release plugin which actually works with Firefox 7:

http://google-web-toolkit.googlecode.com/svn/trunk/plugins/xpcom/prebuilt/ in here  is the gwt-dev-plugin.xpi, and it works like a charm :)

(In firefox, it asks you to download. And then you can drag the downloaded .xpi back into the browser to install. Why Firefox will not instantly recognize the .xpi file is kind of strange)

The tests were of two formats: 10 questions in 15 minutes, and 30 questions in 40 minutes (multiple choice questions).

For the tests I took, these were some links with background material which I found most helpful:

• IBM Developerworks: Understanding RUP roles
• Wikipedia Capability Maturity Model Integration (CMMI)
• Wikipedia RUP (Rational Unified Process)
• Wikipedia Software configuration management
• Wikipedia Enterprise architecture
• Unified Process for EDUcation: A Software Architecture Document example
• and finally a blog post about test specification techniques

As you may notice, the topics mostly covered were RUP, IAF (a capgemini architecture framework), A little CMMI, and hardly any agile methodologies at all. No SCRUM, Lean, Kanban or similar references. Although i think i saw one question on XP (eXtreme Programming).

Now the tests are passed, the application is submitted. I will have a verdict probably within a month.

These are typically from situations where I found a useful way, which I wanted to jot down. Also I have answered and asked some questions on StackOverflow. The good thing about it, is that I am always able to find it again this way.

Justaddwater.dk posts:

1. How to make Git ignore files that already exist in your project
2. Spoken Git Commit Log — Another Annoyance at the Office
3. How to add Git Pull Shortcut to Different Github Branches
4. Using Git for SVN Repositories Workflow
5. Using Local File-based Git — Server Laziness

StackOverflow posts:

1. git-svn: how do I create a new svn branch via git?
2. How to switch svn branches using git-svn?
3. Git how to create remote branch
4. How to undo “git commit –amend” done instead of “git commit”
5. How to grep git commits for a certain word
6. Git svn clone: How to defer fetch of revision history
7. Which Git commit stats are easy to pull
8. Git how to save a preset git log –format
9. View remote Git revision on Heroku
10. The shortest possible output from git log containing author and date
11. Git how to find out which single commits that done on one branch and not another

Notice how Chrome actually also shows occurrences of “ae” when searching for “æ”.

The same thing goes when searching for “Ø”, “Ö” or similar accents: They will show occurences of “o”.

Usability wise, I think this is a rather good principle for end-user solutions, just like Google Chrome. But, but. As an advanced user, i would like to know if i can turn it off? Not as a general setting. Because the usecase would be that you search, and find too much. Based on this, if you search for something with accented characters, a new checkbox should appear. My suggestion (mockup):

This way, you will start with a “wide” search and end up being able to filter in more correctly.

IN this spirit, i created a quick @WTF annotation for Java.

Besides being completely useless (its removed at compile time), its only purpose is to give developers a new way to express themselves in code.

The cartoon states “The only valid measurement of code quality: WTFs/minute”. With the <code>@WTF</code> annotations, I would like to take it a little further. By annotating the code, you can actually encourage colleagues to help you fix it.

The best fix for WTF-code is of course to refactor, so that it becomes readable.

Interesting article:

http://www.gridshore.nl/2008/03/29/how-wtfs-improve-code-quality-awareness/

The original cartoon strip:  http://www.osnews.com/story/19266/WTFs_m