On the binary playground_

My experience with unit testing in software development

2011-09-24T23:13:00.000+07:00

I am not a fan of software engineering which is taught in school. Waterfall model, spiral model, many blah blah phases ,diagrams and documents. I usually thought of those as useless works or not worth to do. I thought that software development should be simple. Get requirement -> Draft design -> Go coding! -> Debug -> Someone test it manually -> Debug -> Test again (until it's acceptable) -> OK, Finished

Documentation should exist only if it is essential and the source code is one of the documentation. This methodology works fine because I usually develop alone or in a small team and the software does not have many modules.

Unit testing is one of those which I thought it's not worth to do. I agree that's it's useful for regression test of the code but I thought I know what I've done with my code and know what are affected.

Around two years ago, I and my colleagues started a new project. One of them wanted to try a kind of Agile development model (I don't know the name. I've said I'm not a fan of these models) and we tried.

He encouraged me to write the unit test for each class and a few acceptance tests for overall project behaviors. I did it because I wanted to give it a try too.

Then I know I was wrong about testing. I usually see some bugs because the tests fail. In addition, I feels great when all the tests go green. The tests guarantee that the software still work as it should. However, I see some disadvantages or difficulties in writing test.

First, White box testing in the unit test for each class are painful to maintain. When I change the code or the behaviors, It always breaks. Most of the causes are silly for example, syntax errors, missing mock and over verification. I also have to write the new white box tests for the new code which I definitely know that it will pass.

Furthermore, it usually be the acceptance tests which show the bugs. Unit tests don't.

The problem is the unit test granularity. As a result of these, I recently try to write unit test in black box way, testing only method effects to others (including return result) and I think about how small unit test I will write. If I write tests for every classes, as one-to-one, it will be the same as white box testing above. Then I decide to write only a unit test for a module or a group of classes. The group is grouped by top-most functional. The dependencies of the group which directly involves the file system or network are cut off to their own tests and replaced by mocks. They will be like smaller acceptance tests.

Lastly, I don't feel constructive while I'm writing the tests. I love programming because I want to create something or have computer to do something but, for writing test, I don't get that feeling. I feel like a QA engineer who plans the tests and tests the software but instead of manually test, I write the code to test it. I have to mock various possible inputs, tests various output conditions and prepare many possible scenarios by writing lifeless code to do them.

Yes, you got it. I want someone else do this task for me. LOL

BTW, I'm learning and I am thinking of how to destroy these difficulties and will update what I found in this blog in the future.

VisualVM on ubuntu(lucid) failed to initialize the Profiler engine

2011-08-14T23:53:00.000+07:00

I installed VisualVM on my lucid(amd64), luanch it and it shows an error.

Failed to initialize the Profiler engine: Problem with a required JFluid installation directory or file /usr/lib/jvm/java-6-openjdk/lib/visualvm/profiler3/lib/deployed/jdk16/linux-amd64/libprofilerinterface.so Original message: File does not exist.

In sprite of this error, the visualvm still can run but profiler/sampler doesn't show up.
Someone have filed this bug here. To fix this problem, recompile the missing lib manually and move to the right place.
Here are the steps

Download visualvm source here (visualvm_1.2.2.orig.tar.gz at the right pane)
Extract it and cd to the build directory
tar -xzvf visualvm_1.2.2.orig.tar.gz
tar -xzvf netbeans-profiler-visualvm_release68.tar.gz
cd lib.profiler/native/build
You will see build script ( buildnative-linux64.sh ) then edit it.

Replace gcc32 with gcc
Change 2 BuildJDK lines to below. /usr/bin/jvm/default-java/ is my jdk path; yours may differ.
BuildForJDK "/usr/lib/jvm/default-java" "jdk16"

Make directory for release
mkdir -p ../../release/lib/deployed/jdk16/linux-amd64/
Run this build script. It will compile the source against jni headers in our jdk and output as libprofilerinterface.so
./buildnative-linux64.sh
Move libprofilerinterface.so to the right place in your jdk path
sudo cp ../../release/lib/deployed/jdk16/linux-amd64/libprofilerinterface.so /usr/lib/jvm/default-java/lib/visualvm/profiler3/lib/deployed/jdk16/linux-amd64/
Launch VisualVM again.

How to reset the main menu in Ubuntu

2010-09-05T14:07:00.000+07:00

I rearranged and deleted some menu items in the main menu. Finally they are in a mess. Some menu is missing and I can't remember the original position or command. I want to reset them to default like fresh install. I don't see any way to reset it easily. There's a revert button but It doesn't work right. So I searched and dug in some configurations.
Then I learn that there're default menu configuration somewhere. One of them is at /etc/xdg/menu which there're applications.menu and settings.menu for application menu and system setting menu respectively. And each user has its own customization of menu in ~/.config/menus which has applications.menu and settings.menu too. Those files contain some exclude configurations and they are merged with default configurations when it is used.

So deleting directory ~/.config/menus should reset the main menu to default state and that's all.

rm -rf ~/.config/menus

and logout and login again.

but if it's still not be reset like in my case, it need to delete *.desktop in your home directory too.

find ~/ -name "*.desktop" -delete

How to forward port by ssh

2010-07-28T12:42:00.000+07:00

There are 3 machine

My machine
Host A with public IP
Target Host B with can be reached from Host A

The goal is that I and my friends want to access target host B services in convenience. We can ssh to Host A with public IP.

Local Forwarding

Run this command on my machine

ssh -L 123:targetB:80 userA@hostA

We can access targetB:80 by myMachine:123 instead
The traffic will be forward from myMachine:123 through ssh tunnel to hostA and the forward to targetB finally.

Remote Forwarding

Run this command on hostA

ssh -R 123:targetB:80 me@myMachine

We can access targetB:80 by myMachine:123 instead
It's similar to local forwarding but the ssh tunnel is created by hostA which is the forwarder
The traffic will be forward from myMachine:123 through ssh tunnel to hostA and the forward to targetB finally.

Dynamic Forwarding

Run this command on my machine

ssh -D 123 userA@hostA

It's similar to local forwarding but it doesn't specify the target:port because the target address and the port will be determined automatically by application protocol such as SOCKS.
We can access targetB:80 by SOCKS proxy in browser or OS network settings.
Currently SOCKS4 and SOCKS5 protocols are supported.

Remarks

There might have firewall issues. It might have to change some port unless the packets will be blocked.
This method is for TCP traffic. For UDP, you may have to try other methods.

How to setup remote profiling session with JProfiler

2010-06-30T21:18:00.000+07:00

What does JProfiler do?
JProfiler is a tool to profile a JVM in runtime locally or remotely. Its profiling features are mainly in 2 parts which are CPU and memory.

In CPU part, it can shows all threads situation in JVM and its state(RUNNING,BLOCKING,WAITING,..) and also traces which part of instruction the thread currently runs. It can show object locking graph to identify a locking problem which is difficult to figure out by normal debugging.

In memory part, it collects and shows memory allocation stats, object counts and also traces the methods or classes which do most of memory allocation. Furthermore, it can shows what and when java garbage collector do with current memory consumption as a time-based graph.

How to setup?

Download JProfiler and its evaluation key

Install JProfiler in your computer.

Copy JProfiler agent library to a target which you want to do remote profiling. JProfiler agent library is in [JProfiler path]/bin/

Add an argument in JVM which you want to do profiling. (I use FreeBSD.)
-agentpath:/usr/home/tmp/jprofiler6/bin/freebsd-x86/libjprofilerti.so=port=8849

** 8849 is a port for JProfiler client to communicate with its agent. You can change this
if you have firewall, you may need to allow the port
ipfw add 6000 allow tcp from any to me dst-port 8849 keep-state

Open JProfiler client. -> Start center -> New session tab -> New session button -> select remote -> configure host and port of target jvm -> Start

Start target JVM. it will pause and wait JProfiler to attach it.

In JProfiler client, click Attach and OK

It should be work now.
Next time I may write about how to setup JProfiler with Eclipse to do local profiling. If you can’t wait, find it in JProfiler help.
** When the evaluation ends, the client still can open but it doesn’t work

Picasa's bugs cause my images to be deleted.

2010-02-05T00:28:00.000+07:00

I use Picasa client and Picasa Web Album for a time and it works enough for me. In addition to managing my photos in Picasa client, I also use Picasa Web Album as images hosting for my thai blog. It's really easy and safe in an album.

Yesterday, an unlucky day, I wanted to sync the blog album to my Picasa client and tried setting them to keep syncing in the future therefore I can upload an image via either the client or the web album. It would be great, isn't it?
I clicked the "download album to picasa" button in the web client. The album was downloaded to the client beautifully and it was shown as a downloaded web album. Then I clicked the "sync to web" button in the client. It asked something and uploaded all images in the album. Why did they need to be uploaded again?
The problem is here. In the web client, there were 2 album now in the same name, the original and the recently uploaded. At first I thought it was not a bug, you might need to have a fresh uploaded album at the first time to keep syncing in the future and these twin albums was different album and unique.
and this idea leads to the mistaken operation.
I right-clicked at the downloaded web album in the client and chose to delete the album with "delete the online copies" checked. Then both original and recently uploaded album were gradually deleted! and they were deleted successfully.
So I don't have any copies of this album left. It's a bug of Picasa and I reported to its forum yesterday.
but good luck of me that the album has only 25 images and google still cache some of them. :P

How to change 'open link in new tab' behavior in Firefox 3.6 back to the old style.

2010-01-24T08:32:00.000+07:00

I downloaded Firefox 3.6 yesterday and I see that the 'open link new tab' behavior is changed. In the previous version, an opened new tab will be inserted at the rightmost in the tab pane but now the new tab will be inserted next to the current related tab. It's just a style but I like the old one.
To change it back to an old style. Steps are below.

Enter about:config in the location bar.

Press "I'll be careful, I promise!" button. yes, you have to be careful about this.

Type tab in filter box as in image.
Find the preference name browser.tabs.insertRelatedAfterCurrent and double click it. The value of this preference will be toggled to false.

That's it.

URL Regular Expression Pattern

2009-11-16T10:57:00.000+07:00

Tonight I search for a regex pattern to parsing ftp or http url. I found some but I choose Ivan's pattern and then I modify it to suit my problem. this is the result.

^(ht|f)tps?:(//)(w+(:w+)?@)?([-w]+.?)*[-w]+(:[d]{1,5})?(/([-w~!$+|.,=]|%[a-fd]{2})+)*(/)?(?([-w~!$+|.,*:]|%[a-fd{2}])+=?([-w~!$+|.,*:=]|%[a-fd]{2})*(&([-w~!$+|.,*:]|%[a-fd{2}])+=?([-w~!$+|.,*:=]|%[a-fd]{2})*)*)?(#([-w~!$+|.,*:=]|%[a-fd]{2})*)?$

Some paths derive from Ivan's pattern and some I rewrite it. I note this for further use in the future. Hope it maybe useful to you too.
And I also found regexpal which is a great online regex tool.

GoogleSurfKey will be outdated soon.

2009-10-28T06:03:00.000+07:00

Today I just see that Google is doing keyboard shortcut feature for their search result page. Although now it's in google lab and just a experimental feature but I think it will be out of the lab soon. It's a good news for keyboard-centric user like me. :D So my GoogleSurfKey script will be outdated soon.

You can try this new Google's experimental feature out at the Google Experimental page. Now it can use only in main english google.

Google Surf Key 0.3.0 is released!!

2009-08-16T03:06:00.000+07:00

I've not updated Google Surf Key script for a long time. Now I update it and release the new version 0.3.0. Recently google search has updated new result page layout and the DOM layout has changed (more complex) so my script hasn't worked anymore.

Install it here

This is version 0.3.0 changelog.

version 0.3.0 released.(15 Aug 2009)
- Updated for new Google's search result page layout.
- Fixed the 10 results limitation.(Only j,k traversing not number keys)
- Add l command to toggle show the index label.
- Change the implementation to use JQuery 1.2.6.
- Improve the link focusing.

Hope you like it!

Convert all column data type from VARCHAR2 to NVARCHAR2 and vice versa in Oracle 11g

2009-06-24T00:39:00.000+07:00

I found a problem about the encoding and some data size issues. So I need to convert all column in all table in my database from VARCHAR2 to NVARCHAR2 with the same size. Definitely I can convert the data type column by column manually but it takes time and not a smart way. So this is the way.

I logged in as SYS and run this sql statement.

select 'ALTER TABLE '||OWNER||'.'||TABLE_NAME||' MODIFY('||COLUMN_NAME||' NVARCHAR2('||DATA_LENGTH||'));' AS GEN_SQL_STATEMENTfrom   SYS.DBA_tab_columnswhere  owner = 'ownername' and DATA_TYPE = 'VARCHAR2';

The result is the rows of ALTER statement filled with the table name, column name and new data type.
Then I copied all rows and concatenate them with some editor. and then run them as a sql script.

If you want to convert from NVARCHAR2 to VARCHAR2, just switch 'VARCHAR2' and NVARCHAR2' in above sql statement. You can modify above script as well to suit your propose.

Finally, you cannot convert the column which is not empty otherwise the error will occur.

IE6 crashes on JQuery append() function

2009-06-02T13:23:00.000+07:00

This is a problem only on IE6 as I know. (IE7 and Firefox3 are tested.) This is my JQuery code that causes IE6 crashed.

var table = $("#contact_list");table.empty();$("group",data).each(function(){var group_name = $(this).attr("name");table.append("<tr><td>"+group_name+"</td></tr>");$("user",$(this)).each(function(){var username = $(this).text();table.append("<tr><td>"+username+"</td></tr>");});});

When I saw IE6 crashed, I tried to figure out what was the problem and found it was an append function. So I tried to implement the code in many ways and I saw something.

It will crashes if the html string as an argument of the function is not a properly dom object or closed tag. For example

var box = $("#contact_list");box.append('<table>');//do something.....box.append('</table>');

So I solved this problem by define an variable for collecting all the html string which I want to append and append it by append() in the end. And the code works fine on IE6.

Switch input language by Caps Lock #2

2009-04-03T00:51:00.000+07:00

Due to comments in this post, I tried to find another way to use caps lock key for switching input language. The limit of that post is it works only Windows which has grave accent(`) option for switch language but many computer hasn't that option. It usually has Left Alt+Shift option.

So I have to remap caps lock key to left alt+shift. There is no way to do this in Windows registry. I found AutoHotkey can do this for me.

Steps are

Download and install AutoHotkey in your computer.

Write .ahk script like below
Capslock::Send {LAlt down}{Shift}{LAlt up}

Save this file and try double click to run it.

Test it by press caps lock and see.

Create shortcut to this .ahk file and move it to Startup folder in Start menu. [Do this to run this script at startup]

Hope you like it!!

Major Limitation of Gmail and Google Calendar in Offline Mode

2009-03-10T05:57:00.000+07:00

Recently Google launched an offline mode of Gmail in labs and just few days ago Google Calendar offline mode had been released in beta after the news of this feature over a year.

Both of them use the capabilities of Google Gears for running in offline. So I installed Google Chrome for try out these things.

It works nicely but I found a major limitation. A few days ago ,I was disconnected and I opened the Gmail. I expected that I could browse my email easily and see my drafts but it didn’t work.

Google Chrome showed an error. It said “105 (net::ERR_NAME_NOT_RESOLVED)” that is about Chrome cannot resolve an IP address of website (gmail.com). So it cannot work offline. The same error was shown when I tried the Calendar.

But Google Document or Remember The Milk which support Gears work correctly.

Gmail and Calendar work in offline mode correctly if they have already been opened before your computer is disconnected. They will switch to offline mode and continue running correctly.

I wish this limitation will be removed soon.

Booting Process Tuning in FreeBSD 7.1

2009-03-02T23:38:00.000+07:00

This is the last semester in my undergraduate student life, I attended the FreeBSD class and the last project is "Special Problem: Subsystem Tuning".
The project's objective is to tune or optimize a subsystem of FreeBSD and/or its environs (applications, utilities, etc.) to make it perform better.
I choose to tune the booting process of FreeBSD. Due to my laziness and lack of time, so I didn’t do tuning as much as I expected. but if I have time……if I have time…..
Here is my presentation.

Upgrade Thinkpad X200 Harddisk and Keep Service Partition Safe By Puppy Linux.

2009-02-22T22:07:00.000+07:00

I bought a Thinkpad X200 with 80 GB harddisk 2 months ago. I'm very impressed with this thinkpad. Now I'd like to upgrade my harddisk from 80 GB to 320 GB because I need more space.
The usual procedure , replace it with a new harddisk and format, is fine but that procedure will blow your Service partition away. In term of Thinkpad , this kind of partition is called "Predesktop Area".
Predesktop Area or you'll see a partition named "Service003" in Computer is needed for booting your Thinkpad in Rescue And Recovery mode by Press Blue ThinkVantage at the boot time.
This mode can format your harddisk and recover it to the factory-installed state.
So I'd like to keep Service partition. Many people in the forums recommend to use Acronis True Image to clone the whole disk. It's a good application. I tried it and succeeded the cloning process. It works except booting in Rescue And Recovery mode.
So I searched for another way. and found the PuppyLinux is a good way.
Disclaimer : Do it at your own risk!!!. It works for me but I can't gurantee it'll work for your X200 or other model too.
Objective : To migrate all the data in the old harddisk to the new bigger harddisk and all data must remains. Can boot to windows with no problem and boot to Rescue and Recovery mode by pressing the blue button at the boot time.
Tools :

New bigger harddisk.

The USB externel harddisk enclosure.

PuppyLinux CD

A Screwdriver

Environment : In my old harddisk I have layout like this.

Method :

[optional but recommend] Create the Product Recovery Media (if you've created it) by Start->All Programs -> ThinkVantage -> Create Product Recovery Media
The Product Recovery Media needs 1 blank CD for start CD and 2 blank DVD. It's for using to rescue system when you cannont boot to your harddisk or you want to recovery your whole disk to factory-installed state (I think it includes Service partition).
** Product Recovery Media can create only once. It you lose it, you may contact Lenevo Support to have a new one.(maybe free or not)

[optional] Recover the Lenevo Partition (after create Product Recovery Media) by double click on this parition in Computer and choose Recover space.

Reboot to Puppy Linux CD. (you may have to select boot order by press F12 at the boot time)

Wait for the system is ready.

Integrate the new harddisk with the external HD enclosure.

Plug the external to the Thinkpad. Puppy may see new the drive if it was formatted.

Run Pudd (Menu -> Utility -> Pudd copy drive/partition) to clone the internal harddisk to usb external harddisk. The main point is that cloning the all partition from disk to disk in "As is" mode that means the partition size will not expand propotionally. From 80 GB to 320 GB, you'll see the unallocated area 240GB approx in the rest of the external harddisk.

It takes about half an hour for clone the disk. The limitation of Pudd is that no progress showing. Please wait patiently and wish it's working.

When Pudd finished his work, run GParted (Menu -> System -> Gparted). You'll see the partition layout of the external harddisk as same as the old 80GB one.

Now using the GParted to resize/move/create/delete some partitions except Service partition as you like.

This is mine. (you'll see the 1 MB unallocated area at the beginning of disk. I think this results in why Acronis didn't works. )
Shut down the computer.

Switch the harddisk 80GB and 320GB. You may need a good screw driver. Do it by the Thinkpad Manual. It's very easy.

Now you can continue working with the new bigger space.
Wish it works for you too.

My favourite Dijkstra's epigrams

2009-01-26T09:50:00.000+07:00

Do not try to change the world. Give the world the opportunity to change itself.

The question of whether computers can think is like the question of whether submarines can swim.

Computer Science is no more about computers than astronomy is about telescopes.

A formula is worth a thousand pictures.

Progress is possible only if we train ourselves to think about programs without thinking of them as pieces of executable code.

Program testing can at best show the presence of errors but never their absence.

I pray daily that more of my fellow programmers may find the means of freeing themselves from the curse of compatibility.

If you carefully read its literature and analyze what its devotees actually do, you will discover that software engineering has accepted as its charter, “How to program if you cannot.”

Waiting is a very funny activity: you can’t wait twice as fast.

Mathematicians are like managers; they want improvement without change.

source : Memorial Resolution - Edsger Wyber Dijkstra

Facebook lost my data!

2008-11-30T08:11:00.000+07:00

I received an email 10 minute ago. The mail body is below

Unfortunately, the settings that control which email notifications get sent to you were lost. We're sorry for the inconvenience.
To reset your email notification settings, go to:
http://www.facebook.com/<wbr>editaccount.php?notifications
Thanks,
The Facebook Team

I was surprised very much with this. I thought that it is a phishing email but I see the sender and I logged in to my account and found this message.

Oh! it's real! I don't know what's happen at Facebook.(wait for news) but it made the image of facebook look bad. The very big website lost their member data. How about backup system?
There is no backup for the data?
After I googled about this issue, many bloggers have commented about this too. Really I love facebook more than other social networking websites. I wait for the news to know what happen to the Facebook now?

Emesene crashes during logging in

2008-11-04T09:47:00.000+07:00

I use emesene 1.0 on Debian. It usually works great and I prefer it to pidgin.

But there are a few bugs. Emesene crashed during logging in. It'll crash everytime after it occurs the first time. I don't know why. It hardly occurs and it crashed unexpectedly. I used to see this bug a few times in the past. My friend found it too and has already found the solution. Deleting all the files in ~/.config/emesene1.0/[your_account] is the solution but you have to sacrifice your config files and preferences.

I got it crashed again. But this time I found the better solution. I digged in the file and found that deleting the PersonalMessage.conf in ~/.config/emesene1.0/[your_account] is enough.

The file contains you personal message. After I deleted it, emesene can logging in and online again. The preferences still be but only personal message not.

I took experiment something. I set presonal message to the previous personal message that's in the PersonalMessage.conf before It's deleted. It contains Thai and English word in mix. It crashed immediatly. I run it and set it again. It crashed. This time I set the personal message to only English message like "test". It didn't crash. Now I can set the previous message and crashing haven't occured anymore.

Maybe this is the clue of this bug. I'll report it to the bug tracker later.

Time conflicts in Dual-boot, UTC or Local time in BIOS?

2008-10-22T07:20:00.000+07:00

I'm surprised that I've used dual-boot many times but I've just found this problem. There is a problem about time between Windows and others such as Debian.

The problem is
Debian (or unix-like) uses time in BIOS as UTC time
but
Windows uses time in BIOS as local time.

So the problem shows out. I'm in Thailand so my timezone is +7. If UTC time is 10:00, I boot to Debian then the time in BIOS should be 10:00 and Debian shows 17:00 as a time. Then I switch to windows , it shows the time 10:00 as my time. The time is in the past. I adjust to the right time 17:00 and windows sets it in BIOS too. If I switched to Debian again, it will show the time is 00:00 (17:00+7 hours). That's it.

The root of the problem is each OS treat time in BIOS in different ways.
So the question is "Which time should be in BIOS? UTC or local time?"

UTC - more standards and every systems can treat it in their way. It makes sense.
Local time - After I googled , why windows does it like that, there are reasons. First is backward compatibility. The former windows system such as MS-DOS or Windows 3.1 do this way so they have to do the same way to remain compatibilites and can dual-boot between Windows' family. Last is to avoid confusing on BIOS time itself. If you're in the BIOS setting and you've to set the BIOS time. Which time will you set? your local time or UTC. It makes sense that you will set it by local time. If BIOS asks you UTC time, it's a bit strange.

So I still don't know the answer of this question. But there are some workarounds to fix this problems
You need to do only one of these.

In Linux :

Edit /etc/default/rcS
by set UTC=Yes to UTC=No

In Windows:

Add the registry
HKEY_LOCAL_MACHINE System CurrentControlSet Control TimeZoneInformation RealTimeIsUniversal
and set its value to 1.
Uncheck the "Automatically synchronize with an Internet time server" in Date and Time Properties.
*I don't recommend this way because I found many people told about few following problems.

but I still don't know why I didn't found this problem when I was dual-booting Ubuntu and Windows. OR Didn't I notice this problem?

Gumstix-OE Connex on QEmu

2008-10-16T23:22:00.000+07:00

I'm newbies to Gumstix and OpenEmbedded platform so I try to emulate it for working easier and more safety with my experiments.

From this and this tutorial , those're for verdex , you can change a bit for you connex easily.

First you have to install qemu (version 0.9.1 or later) to you system. I'm using Debian so do this way.

sudo apt-get update
sudo apt-get install qemu

Second you need to find the 3 file for creating the qemu image. These consist of

U-boot
This is the boot loader for you gumstix. You can build your own from source or get the pre-built file from the gumstix webstie here.
Example is "u-boot-connex-400-r1604.bin".

Kernel Image
It's the kernel for your system. It's usually start with 'uImage'. Of course you can build your own via this tutorial or pick the right pre-built image from gumstix website here. About the difference between glibc and uclibc you can googling to find out yourself.
Example is "uImage-2.6.21-r1-gumstix-custom-connex.bin".

Filesystem Image
It's the files for your gumstix including many packages installed. You can built it your own from the same tutorial above or pick it here.
Example is "Angstrom-gumstix-basic-image-glibc-ipk-2007.9-test-20080512-gumstix-custom-connex.rootfs.jffs2".

Now you can create the qemu image by this way.

dd of=flash.img bs=128k count=128 if=/dev/zerodd of=flash.img bs=128k conv=notrunc if=u-boot-connex-400-r1604.bindd of=flash.img bs=128k conv=notrunc seek=2 if=gumstix-basic-image-gumstix-custom-connex.jffs2dd of=flash.img bs=128k conv=notrunc seek=120 if=uImage-2.6.21-r1-gumstix-custom-connex.bin

The italic have to be changed according to your files. The bold are those difference from the reference tutorial. Because of memory, the verdex has 32 MB flash memory but the connex has only 16MB. So 256*128k changes to 128*128k. The kernel is on the end of the memory (Size -1MB) so (256-8)*128k changes to (128-8)*128k.

Now there will be a "flash.img". This is your qemu image file.
You can run it by

qemu-system-arm -M connex -m 289 -nographic -monitor null -pflash flash.img

And see it booted. But my problem now is I can't exit the qemu. -_-" Ctrl+A x doesn't work.

I think buildroot platfrom can do this method too. Because it builds to the 3 files as above and then use the same method to emulate it.

How to make BCM4318 (with Acer wireless button) works in Debian!

2008-10-16T19:36:00.000+07:00

My laptop model is Acer 5021. I have a problem with the wireless , in particular the acer wireless button that's in front of the body can't lid up to enable the wireless just like in Windows. In the past I got it work by installing ndiswrapper and acerhk. But after the Edgy came out , it haven't worked anymore.

Now I switched from Ubuntu to Debian and found the same problem. But I see the suggestion message while it's booting. Something likes

[ 59.267579] b43-phy0 ERROR: Firmware file "b43/ucode5.fw" not found or load failed.
[ 59.267588] b43-phy0 ERROR: You must go to http://linuxwireless.org/en/users/Drivers/b43#devicefirmware and download the correct firmware (version 4).

So I went to the url in the message. It told me to do this

sudo apt-get install b43-fwcutter

Now I reboot and the error message disappeared. But I still can't use the wireless because of the Acer wireless button. I googling more , go to acerhk website. And found that acerhk is stopped and got into the main kernel development (names acer_acpi) that's the good news. and now acer_acpi is moved to acer_wmi.

if you have the kernel>= 2.6.25 , it might have the acer_wmi in the kernel already and be detected to load automatically. You can check this by

lsmod | grep acer

you would found acer_wmi.

so the rest is easy. Due to this useful (and hidden) document , you can

To read the status of the wireless radio (0=off, 1=on):cat /sys/devices/platform/acer-wmi/wirelessTo enable the wireless radio:echo 1 > /sys/devices/platform/acer-wmi/wirelessTo disable the wireless radio:echo 0 > /sys/devices/platform/acer-wmi/wirelessTo set the state of the wireless radio when loading acer-wmi, pass:wireless=X (where X is 0 or 1)

and It works!!!

I wonder this method also works with the other Acer models which have the problem with the wireless button in front of it.

good luck.

How to Fix Firefox always keeps showing "Restore Previous Session".

2008-08-10T04:29:00.000+07:00

Today, my ubuntu crashed while I was using firefox. I rebooted and everything seem fine except one thing. Firefox show "Restore Previous Session" dialog everytime I open firefox.
Even though I've shut it down properly , the dialog keeps showing.

I googled around but I had no luck. I think that the problem is firefox config files may be some corrupted because of the crashing. So I went around in ~/.mozilla and try out some way and PONG! It's fixed! :)

Here is the solution.

1. $ cd ~/.mozilla/firefox/fwegtctl.defaults (you may change this to your firefox profile directory)
2. $ mv sessionstore.js sessionstore.js.bak
3. Restart the firefox.

Just simple. I guess that sessionstore.js saves the data about session (its filename tells that. HAHA) and firefox on opening will see whether this file exists or not.
Good Luck!
UPDATE : I've just found that there is another problem. The problem is the location bar doesn't update anyway. It's still stay blank or some url and when you change the tab it's doesn't effect. The location bar search feature also off. I don't know this problem and previous problem if it's an consequence. I spent a hour finding out the solution in the .mozilla/ (yes, again!)
at last, the solution is right down.

1. $ cd ~/.mozilla/firefox/fwegtctl.defaults (you may change this to your firefox profile directory)2. $ mv places.sqlite places.sqlite.bak
3. $ mv places.sqlite-journal places.sqlite-journal.bak

I hope that these soulutions will work for you.

Practice English By a Song.

2008-05-31T09:35:00.000+07:00

It's my little project. I came up with this idea while I was on the car. In Thailand, English teacher sometimes comes in the class with the radio. Then she will give students a paper. The paper is a lyric but it's not a complete lyric. There will be many blank field on the paper. She will turn on a song and She has students to fill in the blank to complete the lyric. I think it's a good idea for practice listening by a song.

So I will do it in a same way but on the internet. :)

check out this.
http://www.teerapap.net/EnglishByMySong/start.html

Thanks for "Yael Naim - New Soul" that's a sample song because I like this song very much. Actually I use this song on the website arbitrarily.

Firefox 3 beta 5 lags on ubuntu 8.04

2008-05-04T05:24:00.000+07:00

I upgraded from Gutsy to Hardy last week. It comes with the new Firefox 3 beta 5. After I had used for a few days, I found that firefox lags very much. When I switch the tab or scroll the page or just leave it , it start to lag the system and then system sometime freeze after that. In the system monitor, firefox process was interruptible. I noticed the harddisk light always show or blink fast but I'm not sure it's the same case.

At first I though that it resulted from upgrade process that it's not a fresh install. Maybe some configs file are not upgraded correctly. After few days passed, I felt very annoy. I started to find out what it is.

I found this bug. It's the same. In the bug page they give a method to solve this bug temporarily
There is an easy workaround.

This workaround is successful:
1. Open Firefox 3
2. Disable phishing/attack detection in the security section of the Preferences
3. Close Firefox
4. Make hidden files visible (Locations - Personal folder - View - show hidden files)
5. Delete /home/user/.mozilla/firefox/blahblahblah.default/urlclassifier*.sqlite
6. Open Firefox.
7. There is no next step. You're done. :-)
Greetz, Pjotr.

I follow this instruction. It works!!

I like the bug report system too much. haha

I hope that the patch will be published very soon. I wait for it.