package tools
{
	import flash.display.Sprite;

	import mx.controls.Alert;
	import mx.core.IFlexModuleFactory;
	import mx.core.mx_internal;

	/**
	 * Alert-Box which makes using HTML text possible
	 */
	public class HTMLAlert extends Alert
	{
		/**
		 * Given text is interpreted as HTML text,
		 * Example: "bold text" is rendered as bold text.
		 */
		public static function show(text:String = "", title:String = "",
						flags:uint = 0x4,
						parent:Sprite = null,
						closeHandler:Function = null,
						iconClass:Class = null,
						defaultButtonFlag:uint = 0x4,
						moduleFactory:IFlexModuleFactory = null):Alert
		{
			var alert:Alert = Alert.show(text, title, flags, parent,
				closeHandler, iconClass, defaultButtonFlag, moduleFactory);
			alert.mx_internal::alertForm.mx_internal::textField.htmlText = text;
			return alert;
		}
	}
}

Just use it by calling it like the regular Alert-class:

HTMLAlert.show("A bold Alert text", "HTMLAlert", Alert.OK);

Which then looks like this:

Since i don’t want to sit in front of my browser hitting F5 again and again (especially if it takes some minutes), i found it nice to watch the HTTP Server with a little script, which can run in background and send a little OS notification when the Site or Webservice is up again.

I’ll share it and hope anyone also has a use for this. For the notifications, you need the OS-independent notification library pynotify. If you are running Debian/Ubuntu, get it with apt-get install python-notify

#!/usr/bin/python
# Partly "borrowed" from:
# http://code.activestate.com/recipes/267197-urllib2-for-actions-depending-on-http-response-cod/
import urllib2
import pynotify
import time
import sys
from urllib2 import Request, urlopen, URLError, HTTPError

url = sys.argv[1]

print "Waiting for Site " + url + " to come up"

while True:
  try:
    urllib2.urlopen(url)
  except urllib2.HTTPError, e:
    if e.code == 401:
      print 'not authorized'
    elif e.code == 404:
      print 'not found'
    elif e.code == 503:
      print 'service unavailable'
    else:
      print 'unknown error: '
    time.sleep(3)
  else:
    print 'success'
    if pynotify.init("HTTP Monitor"):
      title_string = "HTTP Monitor"
      author_string = "Site " + url + " is available"
      n = pynotify.Notification(title_string, author_string, "emblem-shared")
      n.set_timeout(0)
      n.show()
      print author_string
      break

How the script is called:

python ~/tools/watchsite.py http://192.168.178.152:8080/UserService/UserServiceWSService?wsdl

Which looks like this:

Waiting for Site http://192.168.178.152:8080/UserService/UserServiceWSService?wsdl to come up....

And gives this nice notification (on Gnome) when the Server/Site/Service is up:

So today i ended my misery with converting text-files and tabs by hand and finally wrote a Bash script. It works both on Linux and on Cygwin.

It solves both problems at once: It converts Windows CR/LF line breaks (only when there is one!), AND replaces those superflous tabs with spaces. And all that with one single line.

find . -type f -regextype posix-awk -regex "(.*.java|.*.scala)" -exec sed -i -e 's/^M$//' -e 's/\t/  /g' {} ';'

What does it?

find .
recursively finding files in the current directory

-type f
only use files (d for directory)

-regextype posix-awk -regex “(.*.java|.*.scala)”
Find all files with extension Java OR Scala (you can add more)

-exec sed -i -e ’s/^M$//’
Remove CR/LF line break. -i means that the changes are written to the source immediately, -e is to concatenate multiple replacements

-e ’s/\t/ /g’
replace all Tabs with 2 spaces

{} ‘;’
use files found by find as parameters for sed, end

After downloading the Glassfish Tools from the download site and importing a simple Weld-JEE-project (i was using the Maven archetype “weld-jsf-jee-archetype”), an error showed up, pointing to the beans.xml of my project and saying

Referenced file contains errors (jar:file:/D:/Program Files (x86)/GlassFish-Tools-Bundle-For-Eclipse-1.2/dropins/feature-1.0.50/eclipse/plugins/com.sun.enterprise.jst.server.sunappsrv_1.0.50.jar!/ee6_schemas/beans_1_0.xsd).

So i opened the file com.sun.enterprise.jst.server.sunappsrv_1.0.50.jar in the mentioned directory with an archiver, where i found the beans_1_0.xsd in the folder ee6_schemas, which i opened with a text editor.

The targetNamespace of the schema file was pointing to the Seamframework, which sounded strange to me. So i googled to find another beans_1_0.xsd schema file and found one in the Weld trunk, with the targetNamespace=”http://java.sun.com/xml/ns/javaee” – which sounded much better to me.

So i replaced the schema file in the archive with the one from the Weld repo (you have to shutdown the GlassFish Tools Bundle first) and restarted the bundle, made a clean of my project – and voilà, the error vanished!

[EDIT]
I switched back to Eclipse and downloaded the Glassfish adapter seperately. Seems to work faster and more stable to me.

Also, there are some really annoying issues when deploying a Maven JEE6-app on Glassfish within Eclipse (both in bundled and non-bundle version with GF3 adapter). When “enable dependency management” in M2Eclipse is enabled, i can’t deploy the project. When it’s disabled, deploying works – but programming is hard because of the missing dependencies (although it works somehow).
Will write an article about that issue soon…

Steps:

1. Add the following dependencies to the pom.xml:

<!-- SL4J API -->
 <dependency>
 <groupId>org.slf4j</groupId>
 <artifactId>slf4j-api</artifactId>
 <version>1.6.0</version>
 </dependency>

 <!-- SLF4J JDK14 Binding  -->
 <dependency>
 <groupId>org.slf4j</groupId>
 <artifactId>slf4j-jdk14</artifactId>
 <version>1.6.0</version>
 </dependency>

 <!-- Injectable Weld-Logger -->
 <dependency>
 <groupId>org.jboss.weld</groupId>
 <artifactId>weld-logger</artifactId>
 <version>1.0.0-CR2</version>
 </dependency>

2. Inject the Logger and use it as usual:

import javax.inject.Inject;
import org.slf4j.Logger;

public class Example {
  @Inject
  private Logger logger;

  public void exampleFunc() {
    logger.info("Hello Logger!");
  }
}

The problem with Ant-contrib is that this library doesn’t come with the regular Ant-distribution. So it has to be downloaded separately.
But inside the Maven-antrun-plugin, Ant can’t access the ant-contrib.jar without explicitely referencing it with a hard-coded path (for example classpath=”C:/Dev/Ant/ant-contrib-1.0b3.jar”) – although it’s in the system classpath.

As it’s not the best practice to refer to local filesystem dependencies and rather downloading and using them within the Maven lifecycle, i’m referencing the ant-contrib.jar that Maven downloads when resolving the dependencies.

In my example i’m checking if the environment variable GLASSFISH has been set, failing the complete build if it’s not.
So my pom.xml looks like this:

<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
 xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/maven-v4_0_0.xsd">
 <modelVersion>4.0.0</modelVersion>
 <groupId>mackaz</groupId>
 <artifactId>antcontrib</artifactId>
 <version>1.0-SNAPSHOT</version>

 <dependencies>
  <!-- Ant-Contrib for Special Ant Tasks (IF) -->
  <dependency>
   <groupId>ant-contrib</groupId>
   <artifactId>ant-contrib</artifactId>
   <version>1.0b3</version>
   <exclusions>
    <exclusion>
     <groupId>ant</groupId>
     <artifactId>ant</artifactId>
    </exclusion>
   </exclusions>
  </dependency>
  <dependency>
   <groupId>ant</groupId>
   <artifactId>ant-nodeps</artifactId>
   <version>1.6.5</version>
  </dependency>
 </dependencies>

 <build>
  <plugins>
  <!-- Ant Task with Contrib -->
   <plugin>
    <artifactId>maven-antrun-plugin</artifactId>
    <executions>
     <execution>
     <phase>validate</phase>
     <configuration>
     <tasks>
       <!-- point to the dependency in the local Maven repository -->
       <taskdef resource="net/sf/antcontrib/antcontrib.properties" classpath="${settings.localRepository}/ant-contrib/ant-contrib/1.0b3/ant-contrib-1.0b3.jar" />
        <if>
         <not><isset property="env.GLASSFISH" /></not>
         <then>
          <fail>Glassfish Environment Property not set!</fail>
         </then>
         <else>
          <echo>Glassfish Environment Property is correct.</echo>
         </else>
        </if>
       </tasks>
      </configuration>
      <goals>
       <goal>run</goal>
      </goals>
     </execution>
    </executions>
   </plugin>
  </plugins>
 </build>
</project>

It’s also not perfect, but better then the examples i saw. The only downside is that when upgrading or downgrading Ant-contrib, the referenced path has to be modified.

C:\>83conv C:\Program Files (x86)\Java\jdk1.6.0_20
C:\PROGRA~2\Java\JDK16~1.0_2

C:\>

The source:

@echo off
rem # Shorten long Windows pathname to 8.3 pathname
rem # e.g. C:\Program Files (x86)\Java\jdk1.6.0_20\lib --> C:\PROGRA~2\Java\JDK16~1.0_2\lib
if %1!==! goto WRONGPARAM
if NOT exist "%*" goto :NODIR
set filename="%*"
FOR /F "delims=" %%I in ('echo %filename%') do @echo %%~sI
goto :eof

:WRONGPARAM
echo Usage:
echo    83conv [PATH]
echo.

:NODIR
echo 83conv: Not a directory 1>&2

No, i wasn’t the pilot. Not yet

Just solved one another annoying problem:
Everytime i had to clean my maven-eclipse-project with “mvn eclipse:clean” and re-created the project, i had to manually re-add the “src/test/scala”-folder to the eclipse project.
Now i found a way to tell maven additional source folders to integrate into the eclipse project: the build-helper-plugin.

So to add my src/test/scala-folder permanently i added the following plugin-configuration to my pom.xml (i also had to add the source) :

<!-- ================== MAVEN BUILDER PLUGIN =================== -->
<plugins>
(...)
<plugin>
  <groupId>org.codehaus.mojo</groupId>
  <artifactId>build-helper-maven-plugin</artifactId>
  <version>1.1</version>
  <executions>
    <execution>
      <id>add-source</id>
      <phase>generate-sources</phase>
      <goals>
        <goal>add-source</goal>
      </goals>
      <configuration>
        <sources>
          <source>src/main/scala</source>
        </sources>
      </configuration>
    </execution>
    <execution>
      <id>add-test-source</id>
      <phase>generate-sources</phase>
      <goals>
        <goal>add-test-source</goal>
      </goals>
      <configuration>
        <sources>
          <source>src/test/scala</source>
        </sources>
      </configuration>
    </execution>
  </executions>
</plugin>
(...)
</plugins>

Wonderful plugin. You can add infinite source folders to your project.
I will surely need it again, when i’m adding the next programming language to my project

[ERROR] error: error while loading Function1, class file needed by Function1 is missing.
[INFO] reference value Unit of package scala refers to nonexisting symbol.
[ERROR] one error found
[INFO] ————————————————————————
[ERROR] BUILD ERROR
[INFO] ————————————————————————
[INFO] wrap: org.apache.commons.exec.ExecuteException: Process exited with an error: 1(Exit value: 1)

Nothing worked anymore – no compiling, cleaning, testing or anything else concerning scala and maven.
My solution: Delete all scala-folders/files in your Maven Repository (%USER%/.M2), so that maven re-downloads them.
After that, anything worked again for me.