Echo in UK

2008-12-09T04:22:00.000-08:00

Hi All

I've made it!!!!

Against all odds, lack of proper training and my injured knee, I managed to finish the 8 miles (12.8 km) race, in 1:18 hour, in 648th place, from a total of 2.500 participants. Not bad at all : )

It wasn't easy though: a circuit loaded with huge ponds, 1/2 meter deep mud, hills, downhills and steep eroded parts, it was for sure the hardest physical activity I ever did so far in my life. At the end of the race your body just ask you to stop and you try to not listen to it and just carry on until the end. I think it will take one week for me to be able to walk normally again : )

But it was definitely worth it and I will surely do it again next year, with a healthy knee and plenty of training!!!

The rain have stopped in my state Santa Catarina, leaving a total of 122 deaths and more than 80 people missing, but there are still thousands of people homeless and in need of basic help.

My initiative even attracted the attention of the Brazilian media. See here a report that was broadcasted last week. it seems I'm sort of famous in Brazil now : )

http://uk.youtube.com/watch?v=QUThtCOEhUI

Grim Challenge 8 2008 Race

2008-12-07T08:26:00.000-08:00

Just a short update, I went quite well in the race, I finished it!!! It was muh much garder than I could had ever imagined!!!

I was with my knee injured and I thougth even that wouldn't be able to compete.

Tomorrow I will put more info herewith the results, timing and position.

They will be published here:

http://www.grimchallenge.co.uk/grim8/results.html

A few videos:

http://uk.youtube.com/watch?v=tPBtmC_p0I8

A few pictures of the event:

Getting ready!!!!

Where is wally?

Almost there....:

HELP FOR THE VICTIMS OF THE FLOODINGS IN BRAZIL. My 8 miles race challenge!!

2008-11-29T04:12:00.000-08:00

Hi All

I am going to be running 8 miles in the mud on Saturday 6th December
to raise money for the victims of the flooding in the south of Brazil
(in my home state Santa Catarina). The flood has claimed more than 109
lives and left more than 80000 people without homes, food or water. The situation is overwhelming, people really need help now!!!

For more details of the challenge I have set myself see:
http://www.grimchallenge.co.uk/
For more information of the Brazilian flooding:
http://news.bbc.co.uk/1/hi/world/americas/7753744.stm
http://english.aljazeera.net/news/americas/2008/11/20081129101739685948.html

Update:

If you woul like to donate directly, please use the bank accounts in brasil provided in the civil defense page:

http://www.desastre.sc.gov.br/index.php?option=com_content&view=article&id=5:contas-bancarias&catid=13:doacoes&Itemid=2

I decided to do this only 2 days ago. I was going to do a donation myself, then I thought: "I think I could get more people to help", and then a friend of mine told me she couldn't run in the race as she got injured, my chance then, I set-up this separated account in my bank, sent an email around the work asking for sponsorship, and now I make this request online, to anyone who would like to help. I consider myself a relatively fit person :) But I think 8 miles (13 kms) on the mud won't be an easy job, so I started training harder at the gym since yesterday. I did 4 kms yesterday, not too bad for the first day. Now I will try to increase 2 kms every day, so being very optmistic by tuesday/wednesday I would be reaching the necessary 13 kms!!! :) Then put a huge pond ont he top of it, and you've got a real challenge. I'm totally sure it will be worth it to help my countryman in need!!!

Map with cities and roads affectted:

Thanks
Emerson Cargnin

TOMCAT LOGS: Rotating catalina.out on a solaris box

2008-11-06T09:31:00.003-08:00

I tried several different solutions for getting a rotated log on tomcat. I tried configuring a new tomcat.log and putting log4j and commonslog jars inside common/lib, but that just created a new log. When I thought I would have to live with the endless catalina.out, I tried logadm, which with a few configurations actually worked just as I wanted!!!

So here it goes.

1) In a directory which the user you are running tomcat can write create a file called logadm.conf with the following content (assuming tomcat is in the /opt/tomcat/ folder):

/opt/tomcat/logs/catalina.out -C 15 -c -t '/opt/tomcat/logs/$basename.%Y-%m-%d'

2) Create a cron job for the same user (assuming you saved the configuration in /opt/tomcat/logrotate):

You can add the cron entry to the user that is running tomcat in the following way:

export EDITOR=vi
crontab -e

add the line:

0 0 * * * logadm -f /opt/tomcat/logrotate/logadm.conf

This will make it run every midnight of everyday.

Explanation on the parameters of logadm

This command, that exists since the solaris 9, will rotate the catalina.out, will rotate the specified log file. The -C tells how many old copies of the files will be kept. The -c is the secret here. It will truncate the log file instead of renaming. This way tomcat keeps the handle to the file and can keep logging. -t uses a template to create the rotated files. Without that it will just append a .1, .2, etc.

You should run it manually before adding to your crontab.

That should be all!!!!

Update to new version Chrome 0.2.149.30 - Release notes?

2008-09-18T18:00:00.000-07:00

I just updated my version of Chrome to 0.2.149.30. Surprisingly it doesn't come with a "new release detector", so you will have to go to tools button and them choose "about google chrome" and then it will detect it's outdated and you can press in the link to upgrade.

I had posted before with my first impressions of my first week using google chrome:

http://echofloripa.blogspot.com/2008/09/google-chrome-7-days-after-updated.html

I could notice that the problem with the link was solved. Also the mouse wheel problem now seems to be solved as well. But still the lack of add-ons is really annoying.

What I see as a problem is the lack of a release notes with the changes made to this new version. Is it or not an open-source software? It should come with a proper release notes of the major changes made.

[edit] Just found out that there is a blog with the release notes. Why google doesn't make our lives easier putting a link in the download page or in the official google blog, I don't know :)

http://googlechromereleases.blogspot.com/

[edit] Aparently they've made some improvement for the "on-demand installation".

[edit] Just found another problem, if you are inside a text box and try to select with the left button and use the mouse wheel (to go down or up and continuing to select) it doesn't work.

regards

Emerson

Adding XML to your Posts in the Blogger

2008-09-07T17:58:00.000-07:00

Have you ever tried to post some XML to your posts and got only some values of the tags, loosing all of the XML?

So that an XML like:


<dependency>
 <groupId>javanettasks</groupId>
 <artifactId>httpunit</artifactId>
 <version>1.7</version>
 <scope>compile</scope>
</dependency>

becomes:

javanettasks
httpunit
1.7
compile

Why is that????

The problem is that the XML is been interpreted as part of the HTML.
To avoid this you can use the following site:

http://www.elliotswan.com/postable/

It will replace the '<' and '>' symbols (among others) to XML entities, that will be shown correctly in the html but won't make your browser think that they are part of the HTML. You paste the xml there, and as a result you will get something like:

& lt;dependency& gt;
& lt;groupId& gt;javanettasks& lt;/groupId& gt;
& lt;artifactId& gt;httpunit& lt;/artifactId& gt;
& lt;version& gt;1.7& lt;/version& gt;
& lt;scope& gt;compile& lt;/scope& gt;
& lt;/dependency& gt;

That then you can paste safely in your post. To get the best formatting, add a </pre></pre> tags before and after the text.

Google Chrome - 7 days after - Updated review

2008-09-07T17:03:00.001-07:00

So, after 6 days I've been using Chrome, I believe I got some updates on my views.

First of all, I'm using Chrome's last release, that I updated via a very intelligently auto upgradeable. The current version is 0.2.149.29.

Pros

Speed: I can't deny that Chrome, even lacking the maturity of other browsers, is pretty fast. The start up time to get an empty browser is impressive.
Omnibox: I don't care that much if google is storing all my searches, the omnibox is very good, especially after a bit of use. I was used to using this style of browsing with firefox already, just typing the first letters of the website, but Chrome takes it one level above searching the possible websites even if you never accessed that before.
Task Manager: The task manager shows exactly which page or plugin are putting your system down. I believe it's really useful.

Memory Consumption: The way it control the memory access with a process per tab and other optmizations make Chrome the best browser in terms of memory consumption. In firefox I have to restart once in a while my whole browser when it's getting above 500-700 megabytes.

Cons

I still have some critics, which I believe will soon be dealt with.

Task manager:

The task manager could allow for sorting using its fields, the same way the task manager does.
It could also show the totals directly in the main task manager (you can see that in the "Stats for nerds").

And Chrome has something very interesting, if you click "Stats for nerds" link in the task manager it shows the total memory used by all I tried to put it to test then. I put the same tabs in fresh just new chrome, firefox and internet explorer. Here are the results:

It seems like Chrome is the one that occupies more memory for the same exact tabs opened.

Find Box: The find box seems interesting, counting which occurrence it is at the moment (eg: 1 of 20). It doesn't seem to work properly when what you are looking for is inside a text box.

Integration with google's services: It could have a better integration with google's services. I would like to have all the tools in the Google's Toolbar that I use in firefox, for example the super useful auto-fill feature.

Error in loading some pages: This page gets loaded without its html tabs when some of the sections are chosen (try "International" tab for example).

How it should be (in firefox):

How it is shown in Chrome:

So the only way to get the main sections is returning to the main page.

Links not working: I just tested again and now it seems to work. But this site had the main news links not working (in red and blue).

Plugins: Some plugins are not detected correctly and some just don't work. The following image is while trying to use the Photos Uploader of facebook.

Whell-button scroll:

The scroll goes down but don't go up!!

Conclusion is that it is fast, good, but still in its beginning and lacks a lot of things to make me replace firefox.

Loading a page with javascript programatically - Httpclient x HtttUnit

2008-09-07T15:38:00.000-07:00

Hi Guys

You might know that I appreciate very much surfing. As a sport for practicing myself, although after moving to England I haven't done as much as I could/should, and for viweing. I quite like following WCT, the main and closed championship with the best 45 surfers int he world, and its qualifier, the WQS. In my spare time, I was getting together a simple app to load the current rankings, then based on the current situation of incomplete competitions that might be running, generate an updatet ranking.

I thought in using httpclient to load the current rankings page, but the page uses javascript with document.writeln(' to write the table, and httpclient doesn't run javascript.

HttpUnit

Then I remembered that in the past I had done some website client automation with httpunit, then decided to use it.

Httpunit Maven dependency POM

First problem is to add it as a maven dependency. The last release, 1.7 from march/2008), is not available in the ibiblio main maven2 repository. I found it at the java.net maven 2 repository. To configure maven2 to use it add a settings.xml to your .m2 folder in your home directory with the following content:


<settings>
<profiles>
<profile>
 <id>myprofile</id>
 <repositories>
  <repository>
  <id>java.net-maven2-repository</id>
  <url>http://download.java.net/maven/2/</url>
</repository>
 </repositories>
</profile>
</profiles>
<activeProfiles>
<activeProfile>myprofile</activeProfile>
</activeProfiles>
</settings>

The following xml need to be added to your POM in order to import the right libraries into my project:


<dependency>
  <groupId>javanettasks</groupId>
  <artifactId>httpunit</artifactId>
  <version>1.7</version>
  <scope>compile</scope>
</dependency>

Missing something?

Ok, I re-run the mvn eclipse:eclipse to re-generate my eclipse project files, then I got a java snippet with a simple request of a webpage and printing it. That should show my ranking, as HttpUnit in this page states that it supports "document.writeln".
First problem: It gives a java.lang.NoClassDefFoundError: org/mozilla/javascript/Scriptable. After a bit more researching, I found the library that implements it, the mozilla Rhino, and added it to my POM.xml:


<dependency>
<groupId>rhino</groupId>
<artifactId>js</artifactId>
<version>1.6R5</version>
<scope>compile</scope>
</dependency>

Error during javascript processing:

Regenerating again the pom and regreshing the project in eclipse, just to find out that I would get still a further problem:

RHINO USAGE WARNING: Missed Context.javaToJS() conversion:
Rhino runtime detected object [Ljava.lang.Object;@110003 of class [Ljava.lang.Object; where it expected String, Number, Boolean or Scriptable instance. Please check your code for missing Context.javaToJS() call.
java.net.ConnectException: Connection refused: connect

I noticed in the 1.7 release notes for HttpUnit that the Rhino library was upgraded to the version 1.6R5. I tried changing to that version, but it doesn't seem to help much more.

Depending on the page I try to load I get a different message, for example the fotolog.net site:

failed: org.mozilla.javascript.EcmaError: TypeError: Cannot find function attachEvent in object [object Window]. (httpunit#10(eval)#1)

Any suggestion to make this work?

My resolved classpath (created by maven) is:
httpunit-1.7.jar
js-1.6R5.jar
jtidy-4aug2000r7-dev.jar
junit-3.8.1.jar
nekohtml-0.9.5.jar
servlet-api-2.4.jar
xercesImpl-2.4.0.jar

I'm using the following piece of code below to test:


import com.meterware.httpunit.GetMethodWebRequest;
import com.meterware.httpunit.WebConversation;
import com.meterware.httpunit.WebRequest;
import com.meterware.httpunit.WebResponse;

public class Main {
public static void main(String[] args) throws Exception {
String URL="http://www.aspworldtour.com/2008/ratings.asp?rView=w&rpage=menwqs&rRat=menwqs1&rNav=Men";
WebConversation wc = new WebConversation();
WebRequest     req = new GetMethodWebRequest( URL );
WebResponse   resp = wc.getResponse( req );
String text = resp.getText();
System.out.println(text);
}
}

UPDATE:

I tried then to use HtmlUnit (like mrs.gredler suggested) onthe same site, and that's the error I get instead:

Exception in thread "main" ======= EXCEPTION START ======== Exception class=[org.mozilla.javascript.WrappedException] com.gargoylesoftware.htmlunit.ScriptException: Wrapped com.gargoylesoftware.htmlunit.ScriptException: Wrapped com.gargoylesoftware.htmlunit.ScriptException: Connection refused: connect (http://ad.outsidehub.com/st?ad_type=ad&ad_size=728x90&section=337979#1) (http://oascentral.surfline.com/RealMedia/ads/adstream_mjx.ads/www.aspworldtour.com/1552025455@Top?#8)

Google Chrome - first impressions and the crash

2008-09-02T17:16:00.000-07:00

Hi Guys

You've all have heard about the google Chrome.
First impression was that it doens't really have anything that special. Maybe only a slightly impression it's a bit faster, but still lacks the normal google bar, that I use quite a bit on firefox, and maybe easier ways to get googles apps integrated. Google's notebook, for example, I don't think there is any way to install it yet.

Addons

I quite use a couple of addons for firefox. Dictionary tip, for example, that I can use inside a browser, would be something Google could add easily to the Chrome. Add-block is another one that hsould be integrated somehow. I don't know if the strategy of google will be to not allow developers to create their own plugins to not risk security and reliability. Most of memory leaks with firefox are due to bad-behaving plugins.

Crash

Suddenly, when I was messing with the task manager (which is quite interesting by the way), the whole Chrome crashed.

Google's advertisement: "Every tab you're using is run independently in the browser, so if one app crashes it won't take anything else down."

Not quite....

[edited]

Today I tried to use chrome all night at home. I faced then, the tab crash. Just note the app that made the tab crash. Exactly, gmail.

Summary

Positives points: task manager (try the "stats for nerds view"), the thumbnails view, the element inspector for developers, the history, among others.
Negative points: no integration with google toolbar and add-ons (no yet at least), seems like lacking something... And also, of course the fact that it crashed in the first run.

See ya next time and good browsing
emerson

Resin vs Tomcat Benchmark

2008-06-23T09:40:00.000-07:00

A company that I know have been using resin for years so far. Now that resin changed its license so that it will be charged per CPU core, they are looking in migrating to tomcat.

I've been told ( by someone in this company) that resin was chosen because a benchmark done years ago showed that it was much much faster.

Ok, then, I've been doing a bit of tests and benchmark to find out if that statement was still true.

Another statement found in the caucho's blog was that:

"Resin Open Source is about 45% faster than Tomcat, Resin Professional is about 80% faster than Tomcat."

Quite a bold statement, that hasn't proved to be true.

Let's go to the benchmark, if you allow me.

Tools

I used Jmeter to simulate concurrent users, and got a couple of URL's from a live log to simulate production conditions. I then used excel to plot the charts and a in-house tool to show CPU usage. The application uses heavily network access to a separate backend.

I used resin 3.0.14 and tomcat 5.5.26.

Tests

I run three types of tests:

Long run (or overnight) test: running over night . I run this only with tomcat to prove that it would be reliable over a greater period of time

Progressive load: I made separated test runs, increasing the number of users, from 5-170
Ramp-up comparison: In the same test I used the jmeter ramp-up mode to simulate an increment of users.

Results:

Long run.

No. of samples: 205.289
Average response time: 9106 ms
Minimum response time: 3.150 ms
Maximum response time: 22 seconds
Percentage of error: 0%
Final throughput 6.2 per second

This test only shows that tomcat is indeed reliable.

Progressive Load

The chart below shows the final chart with the different tests executed.

The chart shows that the response time is almost the same, however tomcat is slightly superior in all of the tests.

The main thing I noticed in this test is the CPU consumption. Tomcat is visibly lighter in this regard. The graph below shows a run using tomcat, following a run with resin:

This chart was done when I was testing 100 concurrent threads sending requests to one server at a time.

Ramp-up comparison

On this test I used the rampup optin in Jmeter to simulate an increment of users during the same test. I then got the final report ("View Results in Table" listener in Jmeter) and plotted a graph.

Tomcat Results:

Samples: 1067
Average: 10427
Min: 3216
Max: 27350
Throughput: 4.4/sec
Kb/Sec: 66.34KB/sec

Resin Results:

Samples: 1090
Average: 10450
Min: 3434
Max: 32360
Throughput: 4.5/sec
Kb/Sec: 66.6/sec

Not much difference here (a little advantage for tomcat though), although I would say that the resin is a little bit more erratic on its response time.

Again, the CPU monitoring shows that the resin consumed a lot more CPU than tomcat (the graph shows tomcat and ):

Conclusion

From the different tests I've made, we can say that tomcat is slighly superior in terms of performance and visibly superior in terms of memory consumption. I want to clarify here that I used an specific version of resin, not the latest one available (3.1.6 been the latest). I also found several references stating that tomcat 6 uses has several improvements regarding performance, but we have decided to stick to 5.5.

Besides the performance, during the port of the application to tomcat, I also found out some other issues with resin:

- Resin xml validation is less strict: this means that a web.xml that work in resin might fail in tomcat. Resin also allows to use a non-compliant web.xml, based on a resin's schema.
- Resin allow mounting external path-mappings. This "feature" is used, for example to locate JSP's outside the war. Tomcat, in its version 6 had received a patch that allowed to do the same, but the patch was vetoed as this is against the specification.

- Shutdown and Start up: Contrary to my expectation, resin had a fast startup, which I'm counting until it comes back with the request response. It took 20 seconds both. From my experience I would say resin start was quite slow. Shutting resin down also is a bit problematic.

Security interface in Anthill - Adding permissions to N resources at a time

2008-06-10T08:29:00.000-07:00

If I was to point one of the most annoying weaknesses in Anthill I would say that it is it's web interface when setting security.

There is no way to apply the security for several resources at once, or to apply several permission at once.

I had to add execution permission to our server operations team, but I didn't want to spend 1 hour opening all of the 20 projects and 60 workflows and add one by one.

So I decided to use remote scripting to achieved what I wanted. Remote scripting enables you to use Beanshell language with the AnthillPro API and add the permissions programatically.

The script that follows add a read permission to all projects and execution permission to all workflows that have deploy in its name. You can change it as you want to tailor it to your needs.

To execute it, you need to :
1) save it in a text file ("script.bsh", for example)
2) Get the remote scripting API at your local AHP server: http://subversion:18080/tools/remoting/anthill3-remoting.zip
3) Expand it and run your script using:
ah3client.cmd [file_name]

import com.urbancode.anthill3.main.client.AnthillClient;
import com.urbancode.anthill3.persistence.UnitOfWork;
import com.urbancode.anthill3.domain.project.*;
import com.urbancode.anthill3.domain.workflow.*;
import com.urbancode.anthill3.domain.security.*;

String serverHost = "server";
int serverPort = 4567;
String userName = "admin";
String password = "adminpasswd";
// obtain connection to the Anthill server
AnthillClient anthill = AnthillClient.connect(serverHost, serverPort,
userName, password);
// create a Unit of Work
UnitOfWork uow = anthill.createUnitOfWork();

String roleName = "Tech Lead"; // change to the name of the role you want to grant permission to
Role role = RoleFactory.getInstance().restoreForName(roleName);

//adding role permission to all deployment workflows
Project[] projects = ProjectFactory.getInstance().restoreAll();
for (int i = 0; i < projects.length; i++) {
Project project=projects[i];
//giving permission for the project
Resource resource =ResourceFactory.getInstance().restoreForPersistent(project);
Permission perm = new Permission(true, resource, "read",role);
perm.store();

Workflow[] workflowArray = project.getWorkflowArray();
for (int j = 0; j < workflowArray.length; j++) {
Workflow workflow = workflowArray[j];
if (workflow.getName().indexOf("deploy")>0) {
System.out.print("Adding server support permission to: "+project.getName()+" - "+workflow.getName()+"--"+workflow.getDescription());
System.out.println();
Resource resource =ResourceFactory.getInstance().restoreForPersistent(workflow);
Permission perm = new Permission(true, resource, Workflow.SEC_PERM_EXEC,role);
perm.store();
}
}

}
uow.commitAndClose();

Some more info here in the docs:
http://www.anthillpro.com/anthill3-help-3.5/html/ch46.html

Snowboard Chamrousse - March 2008

2008-04-21T17:10:00.001-07:00

A video made from a couple of pictures and movies. Some fun wipe outs but I quite managed to do some nice jumps, and even I got the way in carving switched : ) I snowed so much the last days, and got so foggy, you couldn't see actually anything (see in the movie).
This was from 17th to 22nd of march in Chamrousse, southeast of France, in a friend's girlfriend flat.

Note after the beginning of the movie after a do a jump (a few of the succeeded ones :)) a guy comes behing and come rolling down the hill :) And that thing horrible at the end is myself trying to do a rodeo... Almost broke my neck : )

background music: joe satriani-Crushing day and Red Hot Chilly Peppers-Dani California

Android in Kaiser - T-Mobile MDA VARIO III

2008-04-16T18:39:00.000-07:00

Hi guys, I tested a modified version of android image for Kaiser, t-mobile MDA VARIO III.

Download the files here:
http://it029000.massey.ac.nz/vogue/

Don't try to download directly in the phone, I think some files it
downloads as text. Or maybe it just didn't work first time.

Thread at xda-developers:
Original thread:
http://forum.xda-developers.com/showthread.php?p=2083117
Another branched thread:
http://forum.xda-developers.com/showthread.php?t=383695

Also is good to have some spare memory, I removed some apps i had store on the main memory.

I had it working without proper keyboard and touch screens. A combination of keys enable some navigation ont he menus: s+ok(keyboard ok) and z+ok. See the main post to see more details.

Aparently using
the remote console, it's possible to have it full working. Haven't tried this yet though.

regards
Emerson

Creating Nightly Builds with Anthill Pro

2008-04-08T11:05:00.000-07:00

Hi There

From today on I will start to post a series of posts on AnthillPro.
Just implemented a nightly build for one of our workflows. Here is the main points:

Stamp strategy

You will want to copy or continuous integration workflow first. After doing so, you will want to change the stamp strategy. I used the following beanshell expression (put it all in one line):

NIGHTLY-build-${bsh:java.util.GregorianCalendar calendar = new java.util.GregorianCalendar(); calendar.add(java.util.Calendar.DATE,1); Date date = calendar.getTime(); new java.text.SimpleDateFormat("MM/dd/yyyy-HH:mm").format(date)}

This will create a stamp of the following pattern: "NIGHTLY-build-04/09/2008-19:38", you can change to suit your needs.

Cron schedule

Then you will need a cron based schedule. Go to System-Schedules and click at "create schedule" button. Then choose "Cron Expression Schedule". Then give it a name, "Nightly builds" might be a good one.
I'm not that expert in cron expressions, but that's the one I used: "0 0 3 ? * MON-FRI"
First 0 stands for 0 seconds, then 0 minutes, and 3 is the hour (3am), then not specifying the day of the month (with '?') and for last from monday to friday, as we don't usually work at weekends.

Trigger

The last thing is the workflow trigger, where we will associate with the schedule we create before. Choose for the type "Scheduled trigger", give it a name, select the "Nightly builds" schedule you create before and set any workflow property you might have. For example I have two property to execute or not the tests and to choose the branch.

Now just wait until 3am to see if works : )

In the next articles I will show:

How to create a workflow to build from the tag
How to run QTP (quick test professional) from anthill
How to best reuse jobs and workflow definitions
How to deploy resources for various environments using ant and anthill without duplicating files
and more... : )

WindDirStat - A great Disk space manager

2008-03-30T08:41:00.000-07:00

Since a couple of months I've been using this great tool to quickly do a clean up on my disks at work, and today while I was trying to burn a DVD and needed to freee up some disk space, I remembered of this tool and also installed at home.

It's name is WinDirStat, and opensource tool hosted in sourceforge. It also shows the files as TreeMaps. Treemaps were invented by Ben Shneiderman, and they provide a compact visualization of directory tree structures. See the history of TreeMaps here.

Below is a screenshot of windirstat in action:

It's a great tool, very easy to use and will definitelly help you to get rid of those hidden old big files : )

-Emerson

Powerball and Tendinitis

2008-01-28T22:29:00.000-08:00

Hi, quite a time since the last post, busy with holidays : )

In december when I was going around amazon I saw an ad of a "Powerball".

At first i thought it was some kind of portable batteries, eheheh, power+ball : )
After some research I found out that it was a rotor that works at high speeds and generate a strong force and that was recommended to tendinitis and carpal tunnel syndrome. I used to have tendinitis and playing guitar and typing too much never went very well for me. It seemed quite fun, then I decided to give it a try and I got three of these, also as a present to my girlfriend.

Well, it turned up a really good for forearm muscles, and I really feel that it helped a lot with my tendinitis. I can play guitar now and then, and also working without feeling some annoying pains I would get normally.

And they even have a life-time guarantee for changing parts or broken ball. One of the ones I bought does a bit of strange noise and doesn't go too high anymore, they might be sending the changing parts next days.

And it's really fun and addictive. It has a sort of speedometer that shows the number of rotations for minute. My record so far is 12.400. I know not too high, but I haven't spend so much time also. You usually do sessions, that you keep it rotating with the weak hand (my left) and then transfer to the right to get the highest speed. It's funny that when I start I get around 11.000, after 10-15 minutes I get close to my record, 12.000.

It's also funny when you see someone else doing it (don't' think of anything else). In this movie, the powerball world champion at more than 16.000 rotations per minute speed.

I have it for a bit more than a month but I can tell that it's really worthwhile. If you have tendinitis or any other hand or arm injure, start slow, but you definitely will feel improvements.

To buy:

I bought with amazon. A "professional" version, which is the one with the counter (no fun if you can't see how fast you go) in amazon is 13 pounds. In the original website is 23!! Don't ask me how amazon manage to sell so cheap...

You can buy cheap here:

A couple of links:
http://www.powerballs.com/benefits_rehab.php
http://www.powerballs.com/forum/

Oracle, XMLType, Spring and a couple of wasted hours

2008-01-18T11:09:00.000-08:00

I had written an application that reads a set of XML files and writes them to an Oracle database.
I was using Spring configuration injection to get a DataSource and Connection and Spring JDBCTemplate to read and write to the database.

The field in the database I should write to was an Oracle XMLType, which is a kind of clob.

At the beggining to write using the JDBCTemplate I would use the following:

LobHandler lobHandler = new DefaultLobHandler();
Object clob = new SqlLobValue(xmlString, lobHandler);
insert.update(new Object[] {new Long(key),...

That did the job until today, when we needed XML files bigger than 4k bytes to be written.
I started getting the following error:

java.sql.SQLException: ORA-01461: can bind a LONG value only for insert into a LONG column

It happens that the LONG for oracle is the CLOB, but it just treats as CLOB when the sizes is bigger than 4k (or somewhere around).

First I try to use OracleLobHandler instead of the Default one, then it would give:

org.springframework.dao.InvalidDataAccessApiUsageException: OracleLobCreator needs to work on [oracle.jdbc.OracleConnection], not on [org.apache.commons.dbcp.PoolableConnection]: specify a corresponding NativeJdbcExtractor; nested exception is java.lang.ClassCastException: org.apache.commons.dbcp.PoolableConnection
java.lang.ClassCastException: org.apache.commons.dbcp.PoolableConnection

If I use directly an OracleDataSource on my spring configuration or if I would apply this NativeJdbcExtractor I would then got:

ORA-00932: inconsistent datatypes: expected NUMBER got CLOB

Ok, now I'm totally stuck I thought. Then I completely gave up using JdbcTemplate and went straight to JDBC. Well, even then things didn't go as smooth as I thought they would.

First thing, after some research I had to use the following getClob method, which I got in this Oracle OTN page here:

private CLOB getCLOB(String xmlData, Connection conn) throws Exception{
CLOB tempClob = null;
tempClob = CLOB.createTemporary(conn, true, CLOB.DURATION_SESSION);
tempClob.open(CLOB.MODE_READWRITE);
Writer tempClobWriter = tempClob.getCharacterOutputStream();
tempClobWriter.write(xmlData);
tempClobWriter.flush();
tempClobWriter.close();
tempClob.close();

return tempClob;
}

Then when setting it inside your preparedstatement:

CLOB clob = getCLOB(myStringObj, connection);
stmt.setClob(4, clob);

Also, in your SQL you should have something like:

insert into table (XML) values(XMLType(?)

Ok, after all this I still got the following stacktrace:

java.lang.ClassCastException: org.apache.commons.dbcp.PoolingDataSource$PoolGuardConnectionWrapper
at oracle.sql.CLOB.createTemporary(CLOB.java:754)

Then I tried to use instead of org.apache.commons.dbcp.BasicDataSource to use directly OracleDataSource, which worked, but the properties to set are different from those I used for DBCP, url is URL and username is user and I would like to keep it generic. That's when I tried to use Spring's org.springframework.jdbc.datasource.DriverManagerDataSource for my datasource configuration, which then finally got it working.

After this I tried to get back to have it working through JdbcTemplate and the new DataSource, but with no luck.

Maven2 Dashboard Plugin

2007-12-05T02:26:00.000-08:00

What a trouble to have the maven 2 dashboard plugin working...

I will later tell you my saga on upgrading my project to maven 2, but now I just would like to write down the procedure to have a maven 2 multiproject built with the dashboard plugin.
In the plugin page (this one) it miss some information.

1- Add the repository to your parent pom.xml

As the plugin is a snapshot version, you need to add a pluginRepository section to your parent pom (this can be done at the end of the pom, before the ) :
<pluginRepositories>
<pluginRepository>
<id>Codehaus Snapshots</id>
<url>http://snapshots.repository.codehaus.org/</url>
<releases>
<enabled>true</enabled>
</releases>
<snapshots>
<enabled>true</enabled>
</snapshots>
</pluginRepository>
</pluginRepositories>

2- Insert the report section into the POM

This need to be inserted as the last report into your parent pom.xml

<plugin>
<groupId>org.codehaus.mojo</groupId>
<artifactId>dashboard-maven-plugin</artifactId>
<version>1.0-SNAPSHOT</version>
</plugin>

what was missing here in the official page was the version, as it doesn't auto-resolve the version for snapshot releases.

3 - Run the site and the dashboard

This is a two step process. First you need to run the site:

mvn site:site

and this is the catch, when running the dashboard plugin (which aggregates results and regenerates the dashboard html) you need to specify the version. This is not on the usage page as well:

mvn org.codehaus.mojo:dashboard-maven-plugin:1.0-SNAPSHOT:dashboard

If you want to have the aggregate page as in the maven 1 dashboard plugin, you should use:

<plugin>
<groupId>org.codehaus.mojo</groupId>
<artifactId>dashboard-maven-plugin</artifactId>
<version>1.0-SNAPSHOT</version>
<configuration>
<m1LikeRendering>true</m1LikeRendering>
</configuration>
</plugin>

After this you might have the snapshot report on your parent page of your site (reports section), the report that follows is the sub-project one. I used the m1LikeRendering property set to true.

One problem though that I couldn't fix: the links from the the aggregate dashboard report doesn't work for all submodules, just got one to work, aparently, if your sub-module name has a '.' it gets clipped when creating the link.

Emerson

multiple rows for firefox tabs

2007-11-13T07:34:00.000-08:00

A while ago when I had to reinstall everything on my computer I couldn't find again which was the plugin that enables the tabs to be shown in multiple rows. When searching in google all I found was a tweak

.tabbrowser-tabs > hbox {
display: block !important;
}

.tabbrowser-tabs tab {
-moz-appearance: none !important;
display: inline !important;
width: 16% !important;
min-width: 16% !important;
max-width: 16% !important;
}

Actually it's not that elegant. I prefer to use some addon for it. So i found again what I had been using before. It's the Tab Mix Plus, that among several features, enables to show tabs in as many rows as you want:

The configuration for this is a bit hidden. Tools -> Options -> tabs -> Tab Mix Plus Options -> Display -> Tab Bar -> When tabs don't fit width -> Multi- row, and then specify the number of rows you need.

Centralising enviroment variables in resource files

2007-11-12T03:06:00.001-08:00

It's always a pain to manage different resource files, which most of times, share non-environmental properties with environmental dependent properties. Keeping one copy of each file per environment is not always a good option, for the following reasons:

To create a new environment, it's needed to go through all files looking for environment dependent ones.
Changes to environment dependent files need to be in sync with all different environment
Duplication of the same properties in different resource files

One solution of this problems is to parametrise the environment dependent values into a central and single file, which is then processed using Ant (or other tool if you prefer).

In this article, Pete Eakle, shows how to use Ant to replace tokens in css files and resource files.

I think it was missing what Ant replacement can do best, to separate the environmental dependencies in one central file per environment (and possibly also per application).

The following figure shows how the approach would work:

The env files (eg: dev.properties, qa.properties) would contain only environmental name=values entries. Those names would be tokens in the central resource files.

For example:

dev.properties:
...
#HOST FOR THE DB
DB_HOST=xdbserver
...

and inside my central resource file:

myspring.properties :

db_host=${DB_HOST}

In this way, i can change the properties on the fly during build, and before deploying. Using the replacement token ${xxx} is not mandatory and can be changed in the ant script.

The example should be used with the following structure:


root
|-resources-|-myspring.properties
|           |-x.properties
|-build.xml
|-dev.properties
|-live.properties
|-target-
|-resources

The build file can be something like the following:

<?xml version="1.0"?>
<project name="resources_merging" basedir="." default="filter_resources">
<target name="filter_resources" depends="describe" if="env_type">
<echo message="Replacing tokens for resource files using ${env_type}.properties">
<copy todir="./target/resources/" overwrite="true">
<filterset begintoken="$${" endtoken="}">
<filtersfile file="./${env_type}.properties" />
</filterset>
<fileset dir="./resources" includes="**/*.*" />
</copy>
</target>
<target name="describe" unless="env_type">
<echo message="Deployment type required, eg.: ant -Denv_type=local filter_resources" />
</target>

</project>

To use this scrip, you need to provide an environment parameter, like:

ant -Denv_type=dev filter_resources

This will replace the values provided in dev.properties in all the resource files, and copy all of them to the target/resources folder (that needs to be created beforehand).

It uses the filterset option of the copy ant task, which will search for all the parameters in all the files (in this case in a recursive way).

If you have any doubt about this approach or would like to discuss it, please let me know!!

Emerson Cargnin

Google Presentation Reloaded (export as PDF)

2007-11-02T05:44:00.000-07:00

Hi all

In september I wrote this post about google's presentation webapp.

Today i tried again to see which improvements we would have after 1.5 months. And guess what? I didn't notice any major improvement. Still unable to export as PDF, still without a image library or vector graphics, and the list continues.

But actually i wrote this post to help you to export your google presentation as PDF.
There is this presentation sharing website called www.myplick.com, which now accepts google presentation to import. And it supplies an export as PDF.

So to do it you go to the next page:
http://www.myplick.com/upload.do?t=g

And you can add the link to your presentation and even an mp3 with sound (good missing feature for google presentation to copy...)

And after submitting the presentation, which doesn require you to register, you can export it as a pdf (button on the botton right).

Here is my presentation there:
http://www.myplick.com/view/eACtc3maRQ2/Test

Changing windows title for dos command line and putty windows

2007-11-02T04:26:00.000-07:00

Have you ever got confused with lots of different putty/command line windows for several different servers/different users and wanted to distinguish them?

Now all your problems came to an end!!!!

Change Putty window's title bar:

PROMPT_COMMAND='echo -ne "\033]0;london box - tomcat \007"'

Change command's window title bar:

cmd /k title that's my dev trunk build window

Emerson

Exporting youtube videos to dvd format

2007-10-28T15:33:00.000-07:00

Hey guys, a while I haven't been here. Since two weeks ago I started going to a capoeira class here in UK, maybe trying to get back to my roots : ) I found some good lessons on youtube and other similar sites, but as my room is not that big to train capoeira inside : ) I had thought in recording them into dvd format. It came out quite tricky to find the right set of tools to do the job.

So here goes my how-to:

1- Downloading youtube videos in flv format:

You need first to get the video from any video website. For this we will use a firefox plugin called Video DownloadHelper. If you are not using firefox, a good chance for you to get rid of evil ms iexplorer. Install the plugin in the following page:
https://addons.mozilla.org/en-US/firefox/addon/3006

After installing it and going to your favorite video page, you will see an animated icon at the right side of the address bar:

Don't get scared with the image above, that's me playing harmonica with a sort of Andy Warhol's painting filter to the camera :)

2- Converting from flv to mpg4

This one was the hardest to find. There is quite a lot of apps that do the job, but most that i found you could use for free with a logo watermark on it, so not that useful, or you would have to pay for it. well, I ended up finding SUPER, which is a Universal Player Encoder & Renderer. It does the job well, even working in batch mode with loads of files, as it was my case. They have a good slogan which i confirm as true : "SUPER © does for FREE what other encoders CAN'T do for money.". So if you want to donate and make them happy, click here :) and if you want to make me happy, please check the google links on the top or botton :P Finding the link to download that would work was the trickest part. You can try the following ones:
http://www.erightsoft.info/GetFile.php?SUPERsetup.exe
http://www.erightsoft.net/S6Kg1.html
If it doesn't work, try in another browser, i had to use evil ms internet explorer, can you believe it? : ) well, nothing can be so easy, isn't? Just to give some challenge to the task :)

They say the following related to the problem of downloading:

To download from this site without looping back to the same page you need to:
Enable your JavaScript, clear your MSIE cache, do not block your http_referrer (with ZoneAlarm or similar), do not use a proxy.
Queries concerning this issue will not be answered. We do not know any other answer to give.

After installing, when running, to do a quick transformation use the following settings:

Actually the only thing I've changed was the first line: output container -> MPG4, output video codec -> DIVX and output audio codec-> AAC. The rest was default already. One thing that is not so intuitive is how to set the output folder. You need to right click and choose the "Output File Save Management". To choose the files you want to convert, you can right click and choose "Add Multimedia File(s)". After everything ready, hit "Encode (Active files)" button and wait the files to be processed.

3- Creating a dvd with the files

Up to this moment you could already burn a CD with the files and play at most of new dvds. But if you want to burn a DVD you will need an app that can convert conventional MPG files into DVD format. Most os applications that burn DVD are paid, like nero. But as I wouldn't let you without an alternative, I discovered this really cool opensource project that came out quite good and easy to use. It's called dvdflick, an opensource project from sourceforge. Use the following link to download it:
http://www.dvdflick.net/download.php
It was released last month (september), but it looks really good and stable, and with the same features of the other apps.

Main window- Adding media files

Configuring the project:

You need to choose the second option only (burn the project to the disc), if you choose both you will get an error message and if you don't choose any nothing will be saved in the DVD.

Generating the dvd.

This following windows will appear when you press the "Create DVD" button in the main window:

Conclusion

As you could see, it's not rocket science to get a bunch of videos from youtube playing in a DVD. It's just about finding the right tools for the right job : ) And all free (and mainly: without limitations) tools so that might facilitate things for everyone.

Also, most of the previous softwares shown here use ffmpeg command line tool. This is the big guy when talking about sound and video conversion.

Hope this might be useful to you guys, feel free to leave suggestions or critics.

Emerson

Oracle and Dates

2007-10-15T08:35:00.000-07:00

Something i usually suffer is on how to format date on Oracle. Here goes some tips on fomatting, inserting and comparing oracle dates:

The following brings the current date/time:

SELECT sysdate FROM dual;

SYSDATE
———
14-SEP-05

This simple select statement returns the date in the standard format (typically DD-MON-YY.)

SELECT systimestamp FROM DUAL;

SYSTIMESTAMP
—————————————————————————
14-SEP-05 04.06.31.264201 PM -04:00

Here we see that systimestamp reports more detail than sysdate, including the offset from GMT.

Displaying Dates

The to_char function will allow you to describe how you want dates displayed and will convert them to a character string in that format. For Oracle , the default is DD-MON-YY.

SELECT to_char(sysdate, 'MM/DD/YYYY') FROM dual;

TO_CHAR(SYSDATE,'mm/DD/YYYY')
———-
09/14/2005

the to_char function requires two parameters: a date to display, and the format you want it to be in. There are several formatting options, but here are some common ones:

SELECT to_char(sysdate, 'MM/DD/YY') FROM dual;

09/14/05

SELECT to_char(sysdate, ‘MM/DD/YYYY HH:MI:SS’) FROM dual;

09/14/2005 04:09:03

SELECT to_char(sysdate, ‘DAY, MONTH DD, HH12:MI AM’) FROM dual;

WEDNESDAY, SEPTEMBER 14, 04:09 PM

SELECT to_char(sysdate, ‘YYYY BC’) FROM dual;

2005 AD

SELECT to_char(systimestamp, ‘HH24:MI:SS.FF3′) FROM dual;

16:09:24.606

Inserting Dates

The function to_date works similarly to the to_char function above. You must specify a date, typically enclosed by single quotes, then describe the format with the date components as above. To demonstrate this let's create a table we can insert some dates into.

CREATE TABLE dates ( entry NUMBER, entry_date DATE, CONSTRAINT pk_dates PRIMARY KEY (entry) );

Now a few inserts:

INSERT INTO dates (entry, entry_date) VALUES (1, sysdate);

Inserts the current date and time to the second.

INSERT INTO dates (entry, entry_date) VALUES (2, to_date(’09/27/05′, ‘MM/DD/YY’));

INSERT INTO dates (entry, entry_date)
VALUES (3, to_date(’10/02/2005 10:05:33 PM’, ‘MM/DD/YYYY HH:MI:SS AM’));

INSERT INTO dates (entry, entry_date)
VALUES (4, to_date(’17:01:24′, ‘HH24:MI:SS’));

INSERT INTO dates (entry, entry_date)
VALUES (5, to_date(’Monday, September 12, 2:30 PM’, ‘DAY, MONTH DD, HH:MI AM’));

COMMIT;

Now let’s take a look at the data in the dates table:

SELECT entry, to_char(entry_date, 'MM/DD/YYYY HH:MI:SS AM') FROM dates;

1 09/14/2005 09:08:32 PM
2 09/27/2005 12:00:00 AM
3 10/02/2005 10:05:33 PM
4 09/01/2005 05:01:24 PM
5 09/12/2005 02:30:00 PM

We can see that the current date and time was entered in entry 1 down to the second.

Entry 2 contains the date we entered, but since we did not specify the time it has defaulted to midnight.

Entry 3 shows a complete timestamp exactly as we specified.

In entry 4 we see the time as we specified, but since we didn’t specify a date it has defaulted to the first of this month. I have a feeling this varies from database to database. Probably best not to rely on this.

Entry 5 shows the date and time, however since we did not specify seconds they display as :00.

Comparing Dates

Dates can be compared much like other values. To demonstrate this we’ll do some quick selects on the table we just created.

select entry, to_char(entry_date, 'MM/DD/YYYY HH:MI:SS AM') from dates where entry_date > to_date(’09/20/2005′, ‘MM/DD/YYYY’);

2 09/27/2005 12:00:00 AM
3 10/02/2005 10:05:33 PM

select entry, to_char(entry_date, ‘MM/DD/YYYY HH:MI:SS AM’)
from dates
where entry_date <>

1 09/14/2005 09:08:32 PM
4 09/01/2005 05:01:24 PM
5 09/12/2005 02:30:00 PM

select entry, to_char(entry_date, ‘MM/DD/YYYY HH:MI:SS AM’)
from dates
order by entry_date;

4 09/01/2005 05:01:24 PM
5 09/12/2005 02:30:00 PM
1 09/14/2005 09:08:32 PM
2 09/27/2005 12:00:00 AM
3 10/02/2005 10:05:33 PM

These are the main things you can do with dates in Oracle, if you think of anything else, please let me know.

Emerson

Tropa de Elite, the new (and the best, according to some) brazilian movie

2007-10-08T14:09:00.000-07:00

Hi guys, yesterday I've finished watching the movie "Tropa de Elite", a new brazilian movie just released a couple of days ago. This movie created such a comotion even before been released, been so by far the most copied (brazilian) movie in brazil. Because of this the movie was re-scheduled to the 5th of October instead of 2th, the original schedule.
The movie is a story based on real facts about the BOPE, in english "Especial Operations Police Battalion", a special police of Rio de Janeiro city. It shows the cruel reality of the poverty in the favelas of Rio de Janeiro, where the criminality and drug dealing goes along (and well) with the police, that by getting such a small salary almost always end up getting corrupted. The BOPE isn't like that (at least in the movie), and it's feared by the other kinds of police and the criminals in the favelas, giving a hope that not all police would be corrupt. But this special operations group also show a different view that seems to be more and more accepted by the society, not that this movie created, but just made it brought it up: the acceptance of the rule of "an eye for an eye and a tooth for a tooth" (Hammurabi's law). As the brazilian people see their corrupt politicians (pleonasm?) not been punished, the drug dealers commanding the favelas and the police charging to do their work, people start to accept this radical way of solving things.

Following is the first 7 minutes of the movie, with subtitles in English. Some weird translations here and there : )

Before anything I would like to highlight that the reality shown in this movie is NOT what you will see everywhere in Brazil. Even in Rio de Janeiro, apart if you really look for trouble, you won't see things that happen in this movie, as police and criminals going around shooting randomly.

The danger of movies that explore the extremes of Brazil's poverty and criminality like "City of God" or now "Tropa de Elite" is that people will create this stereotype that Brazil is just that everywhere. I lost count of how many times people told me in Europe: "Are you brazilian? Ahhh, I've seen 'City of God', is it really like that? ". At least it doesn't go the the extremes as the movie "Tourists", which shows americans and british tourists been kidnapped and having their organs taken alive... Really something I've NEVER heard about....

People don't know that for example, in the south of Brazil we have places that you would just feel like being in Germany, from the architecture or even from the language in some places. Blumenau, for example, a city founded by Germanic immigrants, 140 kms from my city Florianopolis, has an Oktoberfest every year, that is one of the biggest parties in Brazil.

People outside Brazil use to have a stereotype of brazilian been people that just know how to play football and dance samba. I hope I had done my job so far of showing to people in Europe that Brazil and brazilian people has much more to offer than only this.

To finish, the official trailer of the movie, unfortunately only in portuguese: