This is pretty neat!

Translated loosely from a Chinese blog post (http://drjimdiary.blogspot.com/2009/06/blog-post.html)

Before you join the company …

Boss: Welcome! Office without you sure will looks different!
Employee: If I am too tired working, I may just quit.
Boss: Don’t worry about that, I won’t let it happens.
Employee: Can I rest on weekends?
Boss: Of coz! That’s the bottom line of our company policy.
Employee: Do we need to OT till midnight?
Boss: No way, Who told you that?
Employee: Do we have meal allowance?
Boss: Needless to say, its definitely higher than other companies.
Employee: Is it possible that I will work till death?
Boss: No, why are you thinking in that way?
Employee: Will the company organize overseas trip for us?
Boss: It’s part of our company policy!
Employee: Do I need to come to work on time?
Boss: No, it depends.
Employee: How about salary, always paid on time?
Boss: Always!
Employee: Will the new hire got to do all the jobs?
Boss: How can that be possible? There are many seniors staff above you.
Employee: Will I get a chance if there is vacancy for management position?
Boss: No question about it, that’s how the company survived.
Employee: You are not lying to me are you?

After you join the company, just read in reverse order …

So I put up a simple script to be ran as a cron job to backup my database and auto email it to gmail for good.

NOTE: This should be feasible only if you have relatively small database.

Requirements: You will need to have mysqldump, tar, and nail programs installed.

Create this simple shell script in your home directory, I call it backup_db.sh

#!/bin/bash
mysqldump --user=your_db_username --password=your_db_password --all-databases > mysql.db.backup.sql
tar zcvf mysql.db.backup.tar.gz mysql.db.backup.sql
rm mysql.db.backup.sql
nail -s "Database backup" -a ./mysql.db.backup.tar.gz dummy@gmail.com < /dev/null
rm mysql.db.backup.tar.gz

Then in your cron job, add this line:

01 23 * * 0 /home/your_username/backup_db.sh 2>&1

This will run the back up script once a week on Sunday, 1101pm. You can modify the timing and frequency to your preference.

In this way, you will see an email sitting in your gmail inbox with your database backup, on every monday morning

Assuming your mysql database is relatively small, say about 10MB in size, and you have 7G gmail account; then 7G / 10MB / 54 weeks = 12.9. In another word, with a gmail account, you can backup your data for almost 13 years. Heh.

Then again, this solution is probably only good for small web application, say personal blog kind of thing. But shall be a good way to utilize your gmail storage (seriously, anyone really using > 1G of gmail space?).

Have fun backup-ing.

I am trying this in Windows, so I used cygwin’s ssh. You should have ssh already if you are on a linux desktop.

To create a proxy server, simply run this command:

$ssh username@linode-ip-address -D 9999

Enter the password to login when prompted.

This will create a proxy server in your own computer (localhost) at port 9999.

So in you browser, say Firefox, just need to configure the proxy setting to localhost:9999 and that’s it, you have a secure proxy as all web traffic will be tunneled through this ssh connection.

NOTE: I found that you need to keep all other proxy setting blank, only set the proxy for SOCKS, else it may not work.

With that, have fun browsing, securely!

References:

http://embraceubuntu.com/2006/12/08/ssh-tunnel-socks-proxy-forwarding-secure-browsing/

http://keystoneit.wordpress.com/2010/01/22/ssh-tunneling-with-firefox/

Result is 1 cool and fast and responsive web server. Yay~

All I can say is I am soooo happy that I made the switch.

The web page loads much faster, I got direct ssh access, dedicated RAM/HD/Bandwidth for my host. How cool can it be. Php is more responsive too.

Don’t have to talk about other things, just this blog is loading so much faster, yay !

That aside, I have something brewing under the table…. shall talk more about it when it’s more complete. It’s quite usable already actually, but need to wrap up some loose ends.

Stay tuned.

I am recently reading off comics from 动漫屋. But frankly I do not like to read comics online. It’s slow and it annoys me when I have wait a while to read the next page.

So I studied the site structure and comes up a way to download the comics first and read them locally.

Requirements:

• Firefox
• DownThemAll Plugin

You see, there are quite a few of such (Chinese) comics portal around, mostly based in (guess what) China. Almost all of them put in some form of protection so you can not right click->view source and find out how to hard link to their pictures. Maybe they do not know such thing called Web Developer Extension for Firefox.

So after studying the web site for awhile with the help of Web Developer Ex, I figured a way to re-generate the links to the pictures directly using PHP.

Next I use DownThemAll (install it here if you have not done so:http://www.downthemall.net/) to capture the links and start downloading them. I do not want to download them directly using PHP as it’s is not meant to be a file downloader, and on a windows machine, it doesn’t support multi-thread (no fork). Even it can be done, it shall needs some coding effort. Why not use an existing tool with excellent UI ?

The PHP code is as below. For those want to try it out I host it here with instructions: http://www.xieer.com/comicparser

If I have the time I may add in support for other comic portal.

NOTE 1: The script might not work for comics with many volumes / chapters due to my hosting server seems to be pretty slow and get time out when paring

NOTE 2: As above, the parsing might take some time so please be patient.

PHP Source:

<?
date_default_timezone_set ("Asia/Singapore");

if (empty($_POST['comicurl']))
{
display_form ();
die;
}

$html_filename = urlencode ($_POST['comicurl']);
if (file_exists ($html_filename))
{
readfile ($html_filename);
display_link ();
die;
}

$raw_html = @file_get_contents ($_POST['comicurl']);
$raw_tokens = explode ("\n", $raw_html);
foreach ($raw_tokens as $token)
{
if (strstr($token, "showcomic") && !strstr($token, "span"))
{
$pagetoken = explode ("showcomic", $token);
$page = $pagetoken[1];
$page = "http://www.dm5.com/showcomic" . str_replace (strstr ($page, "\""),"", $page);
$comic_url[] = $page;
}
}

set_time_limit (0);
if (isset($comic_url) && count($comic_url))
{
$url_i=1;
foreach ($comic_url as $curl)
{
$d = str_replace ("http://www.dm5.com/showcomic", "", $curl);
$next_url = "http://www.dm5.com/display.aspx?id={$d}";
//echo "$next_url\n"; flush ();
echo "{$url_i}\n"; flush ();
for ($i=1; $i<=5; $i++)
{
$n = file_get_contents ($next_url);
$m = explode ("\n", $n);
if (!empty($n) && count($m)>10 )
{
foreach ($m as $nl)
{
if (strstr($nl,"array_img[") && strstr($nl, "] ="))
{
$tokens = explode ("=", $nl);
$url = $tokens[1];
$url = str_replace ("'", "", $url);
$url = trim(str_replace (";", "", $url));
$array_img[] = $url;
}
}
echo " okay, \n";
sleep(1);
break;
}
else echo " retrying {$i}...\n";
}
$url_i++;
}
}
else
{
echo "No comic url found for processing.";
die;
}
echo "<br/>\n";

// the img url is coded in %uhhhh%uhhhh.... format
// so we need to break it into an array, each of hhhh, then covert it back to a unicode char
// join them as a string and we have the actual url
$html_link = "";
if (isset($array_img) && count($array_img))
{
foreach ($array_img as $img)
{
//echo $img;
$img_tokens = explode ('%u', $img);
$url = '';
foreach ($img_tokens as $token)
{
$token = trim($token);
if (!empty($token))
{
$char = unichr(('0x' . $token)+0);
$url .= $char;
}
}
//echo $url . "<br/>\n";
$html_link .= "<a href='{$url}'>link</a>, \n";
}
}
else
{
echo "No image url found for processing.";
die;
}

echo "<br/>Parsing done. Right Click, and select DownThemAll to start downloading the link now<br/><br/>\n";
echo $html_link;
display_link ();

file_put_contents ($html_filename, $html_link);

// ------------------------------------------------------------------------------
function unichr($u)
{
return mb_convert_encoding('&#' . intval($u) . ';', 'UTF-8', 'HTML-ENTITIES');
}

function display_link ()
{
echo "<br/><a href='http://www.xieer.com/comicparser/index.php'>Back to parser main page</a>";
}

function display_form ()
{ ?>
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
"http://www.w3.org/TR/html4/strict.dtd">

<html>
<head>
<title>Comic URL Parser</title>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<script src="http://ajax.googleapis.com/ajax/libs/jquery/1.3.2/jquery.min.js" type="text/javascript"></script>
</head>
<body style='font-family:arial;'>
<fieldset style='margin:20px; padding:10px; border:2px solid #f0f0f0;'>
<legend style='font-weight:bold; font-size:1.5em; padding:5px; color:brown;'>
<img src='googlesms_s.png' alt='Comic Parser'  style="border:none;">
</legend>

<form method='post' action='<?= $_SERVER['PHP_SELF']?>'>

<div style='margin:20px; margin-top:10px;'>
<div style='font-weight:bold; float:left; width:90px; height:25px; background:#DFF0F2; padding:10px; margin:0px; border:none;'>Comic URL</div>
<input type='text' name='comicurl' style='padding:6px; margin-left:-3px; font-size:1.5em; width:450px; height:33px; background:#EBF6F7; border:none;'><input type='submit' value='Parse Comic Link' style='margin:0px; width:150px; padding:12px; background:#eee; border:none; border-top:2px solid #eee;'>

<p>
The URL should look like http://www.dm5.com/Type.aspx?id=abcd
</p>
</div>

<div style='margin:20px; width:600px;'>
<p></p>
</div>

</form>
</fieldset>

<fieldset style='margin:20px; padding:20px; border:2px solid #f0f0f0;'>
<legend style='font-weight:bold; font-size:1.5em; padding:5px; color:brown;'>
<b>How To Use</b>
</legend>
<p>This application is meant to be used together with
<b><a href='http://www.mozilla.com/en-US/products/download.html'>Firefox</a> +
<a href='https://addons.mozilla.org/en-US/firefox/addon/201'>DownThemAll</a></b> Download Manager Addon.
</p>
<p>(Or with your own choice of download manager) So please install them first before using this parser.</p>
<ul>
<li style='line-height:1.4em;'>Goto <a href='http://www.dm5.com'>动漫屋 DM5</a> web page to browse thru the comics. <b>(NOTE:Obviously, It's in Chinese  </b></li>
<li style='line-height:1.4em;'>Find out the URL of the comic that you are interested, it should looks like http://www.dm5.com/Type.aspx?id=abcd.</li>
<li style='line-height:1.4em;'>Enter the URL above and click on the button. That's it.</li>
<li style='line-height:1.4em;'>After finishing parsing, links will be shown. Use your download manager to capture the links and start downloading.</li>
</ul>
</fieldset>

<fieldset style='margin:20px; padding:20px; border:2px solid #f0f0f0;'>
<legend style='font-weight:bold; font-size:1.5em; padding:5px; color:brown;'>
<b>Tips</b>
</legend>
<ul>
<li style='line-height:1.4em;'>If you do not know how to use DownThemAll or any download manager, please do your own <a href='http://is.gd/BUO0'>googling</a>.</li>
<li style='line-height:1.4em;'>Or check out this <a href='http://is.gd/BUQy'>quick tutorial</a> at LifeHacker.</li>
<li style='line-height:1.4em;'>You may be doing multiple chapters download, to avoid overwriting/confusion, set up the renaming mask. I use <b>*subdirs*.*name*.*ext*</b></li>
<li style='line-height:1.4em;'>If you are lucky, some comic page might had been parsed and cached. Then the processing will be much faster.</li>
</ul>
</fieldset>

<fieldset style='margin:20px; padding:20px; border:2px solid #f0f0f0;'>
<legend style='font-weight:bold; font-size:1.5em; padding:5px; color:brown;'>
<b>About</b>
</legend>
<p style='border:none; float:right;'>
<a href="http://validator.w3.org/check?uri=referer"><img
src="http://www.w3.org/Icons/valid-html401"
alt="Valid HTML 4.01 Strict" height="31" width="88" style="border:none;"></a>
</p>
<p style='color:red; font-weight:bold;'>To know more about how this works, refer to my blog post <a href='http://blog.xieer.com/2009/01/how-to-let-people-to-send-free-sms-to-you/'>here</a>.</p>
<div style='margin:5px;'><b>NOTE</b> :
<ul>
<li style='line-height:1.4em;'>Unless cached, the time needed to parse the links depends on how big is the comic, more links more time. Start the parsing and go make a coffee.</li>
<li style='line-height:1.4em;'>The parser doesn't DOWNLOAD the comic for you, it just digs out the links so you can download easily/automatically.</li>
<li style='line-height:1.4em;'>That said, <b>Please be considerate</b> and try not to overload the comic server. Keep simultaneous download to about 2 threads.</li>
<li style='line-height:1.4em;'>If you stress out the server so much that its down, its bad for everyone including yourself.</li>
<li style='line-height:1.4em;'>The parser only support 动漫屋 DM5 web page for now. More might be supported in the future if I have the time.</li>
</ul>
</div>

</fieldset>
</body>
</html>
<? } ?>

It was only a few of them. Now I have thousands of them. And WP doesn’t seem to have a select all / delete all interface for such comments.

So I go straight to the mysql database and remove them, with simple SQL command.

And next I installed a simple WP plugin called (obviously) simpleCaptcha. And (maybe not) amazingly, I don’t have a single spam comment since then (2,3 days so far).

Hope this keep the site clean for the time being. Good luck to myself.

So I had been doing quite a bit of reading and now started some hands on to get myself familar with the IDE, build process, language conventions etc.

I managed to code up a simple console program, no GUI, and ran well in the Emulator. But I had been trying hard to deploy it on an actual symbian phone, a Nokia N95 to be exact.

So after two days of googling and trial and error, I finally managed to get it installed. But still unable to launch as I can’t see it under Application folder. I do see it appears under App Manager tho. May because I did not defined any icon ?

Anyway the points below are tips I found over the net for deploying self-signed Symbian App. I won’t explain why Symbian App need to be signed, you can google that yourself. I will just tell you that Signing Symbian (even signing it yourself) is a requirement for S60 3rd Edition onwards.

1) In the project mmp file, do not set the Vendor ID. Setting this will cause error saying something like “Unable to install protected app from untrusted source”

2) Set the UID1 to 0 and UID2 to unprotected range:

3) In PKG file, use unprotected range of value for UID, and shall be the same one used in mmp file:

4) Put in the correct platform / product ID else the phone might prompt that the application is not compatible:

Next I am gonna try out GUI stuff and also do developer signed application instead of self-signed.

也许以后我可以用它来写一些文章或小说哩。

很久没有写中文了。因为不太会其它的中文输入法，用笔又好像太麻烦了。

不管怎麽样，以后我有什么好点子的话我就可以在这里发表了。呵。