1. DataFox : BSNL DataOne Broadband fuels my Internet. And it has a 500GB ul/dl limit per month. That’s why /me can’t do without the tiny statusbar applet this nice extension provides to keep a tab on monthly usage.
2. Download Helper : The easy way to download YouTube videos to hard disk (when /me is not using youtube-dl script from the CLI i.e.!)
3. Download Statusbar : “View and manage downloads from a tidy statusbar - without the download window getting in the way of your web browsing”. Yeh…works just as advertised, esp. that ‘getting in the way’ bit.
4. DownThemAll : The only download manager /me ever uses, be it on or off the web browser. /me just loves its filtering options.
5. FireFTP : /me’s favourite FTP client.
6. Fire Gesture : The extensions /me uses all the time. Real handy, to the point of making /me kinda lazy!
7. Flashblock : /me hates those silly flash ads eating into my monthly data usage limit, not to mention the elevated page loading times.
8. Identifox / Twitterfox : Ever since /me discovered microblogging, these extensions are his constant companions!
9. InFormEnter : /me is too lazy to type all those form information over and over again.
10. Orkut Toolbar : /me spends lotsa time on Orkut, and this is the BEST orkut related extension for Firefox. Its great that it has recently been updated to spport Firefox 3.5
11. UrlLink : Very useful for converting those non-hyperlinks into hyperlinks and open them in new/same tabs.

This guide should be helpful to those as well who wants to install Ubuntu on more than one machines. they will be able to install the same software in all of those machines, but download them from net only once, thereby saving precious bandwidth.

Backing up the deb packages from your current Ubuntu installation

This step needs to be performed in your existing Ubuntu installtion

You must regularly back up your deb files from your current (old) installation in some place which is not affected by a reinstall (e.g on a removable device like a pen drive). Just copy all the deb-s from /var/cache/apt/archives to some such location. Depending on the number of deb-s you have accumulated, this may take some time. For example,

sudo cp /var/cache/apt/archives/*.deb /path/to/backups/

Obviously, replace /path/to/backup with your actual backup directory.

At this stage, it is probably a good idea to take a rough note of the names of the ‘extra’ software you have installed.

Install dpkg-dev package in your new installation

From this step onwards, everything is to be done on your new Ubuntu installation.

Before being able to create a local repository, you need to install the dpkg-dev package. Don’t worry, it is included in your Ubuntu live CD package pool, so no need for internet! Just make sure that the Ubuntu CD repository is enabled in your Software Sources and the CD is in your drive, and do

sudo apt-get install dpkg-dev

Creating the local Repository

After the reinstall, in your new Ubuntu, create a directory and put all the .debs from your backup in it . For example:

mkdir ~/debs
cp /path/to/backup/*.deb ~/debs
cd

While still in your home directory (i.e. ~/) do:

dpkg-scanpackages debs /dev/null | gzip > debs/Packages.gz

In the above line,  the command generates a file Packages.gz that contains various information about the packages, which are used by APT.

Add the local repository to sources.lst

To use the newly created local repo, you have to add it to your sources.lst file.

First of all, take a backup:

sudo cp /etc/apt/sources.list /etc/apt/sources.list.orig

Now open the file in your text editor :

sudo gedit /etc/apt/sources.list

Delete everything except the line which looks like:

deb cdrom:[Ubuntu 9.04 _Jaunty Jackalope_ - Release i386 (20090420.1)]/ jaunty main restricted

(Depending on the version of Ubuntu you are using, the line will change accordingly. This line represents the package repo on your Live CD)

Add the following as a new line after the above:

deb file:/home/<your_username> debs/

Make sure that you replace <your_username> with your actual username.

Save and exit the text editor.

Update the apt database to use the local repo:

sudo apt-get update

Install the software as usual. This is where the rough note containing the names of previously installed applications will come in handy.

sudo apt-get install <package_name>

You’d probably be able to use the GUI tools like Synaptic Package Manager normally to install packages as well.

Source: http://www.debian.org/doc/manuals/apt-howto/ch-basico.en.html#s-dpkg-scanpackages

Reverting back to “natural” state

Right now, the only software repo you are able to use is your own local one. That is not really a good thing, because you have no access to the huge number of other software available online, nor can you update your packages to the latest versions! So, after you have finished with installing all the old packages from your backups, it is time to revert the system back to its “natural” state like it was before. To do so:

sudo cp /etc/apt/sources.list.orig /etc/apt/sources.list

sudo apt-get update

This will sync your local apt database with the online servers. This step requires active internet, though the downloaded data would be relatively small in size.

At this stage, you’d probably want to delete the files in your local repo, as it will no longer be needed (assuming of course that you are completely done with installing packages from your old backups). This could save a lot of hard disk space:

rm -rf ~/debs

Important Notes

• In your old installation, make sure you back up each and every deb file that had ever been downloaded to your cache. If, at any point of time in your old install,  you had run #apt-get clean or had manually deleted any of the existing deb-s  from your cache, this method is liable to fail, complaining of missing dependencies for some packages.
• Running #apt-get autoclean (which purges only the outdated packages from cache) on your old installation should be safe, though. In fact, it is probably a good idea to do so, just to keep you backup size to minimal.
• The packages you install from your local repository will generally not be cached again in your new installation. So, if you are looking for another reinstall sometime in near future, do not delete the backup deb-s as yet!
• The main disadvantage of this method is that your installed packages will be at an older version than what is available on the online repos. If the version difference is too much, it may sometimes create problems while installing newer packages from online repos.
• This guide will, in all probabilities. not work for cross-version reinstalltions of Ubuntu. I.e. if your backups are from Ubuntu Intrepid, and your new reinstall is Ubuntu Jaunty, the method won’t work, excpet on isolated cases.

Here’s the script:

#!/usr/bin/env bash

if [ "$#" -ne 2 -o `echo "$1" | awk -F ":" '{print NF}'` -ne 3 -o "$1" == "-h" -o "$1" == "--help" ];then
	## Print uasge instructions
	echo -e "Usage:	 mycron  \n	(Both arguments are mandatory)\nRuns specified command at specified time\n"
	echo -e "Note: You must specify the time for execution EXACTLY as HH:MM:SS\n"
	echo -e " E.g.: 09:00:00 instead of 9:00:00 or 09:00 "
	echo -e "\nAlso, 24 Hour clock format is expected,\nso use 21:00:00 instead of 09:00:00 if you mean PM\n\n"
	echo -e "mycron -h, --help : Display this help message and exit\n"
	exit
fi

while true;do
    ctime=`date | tr -s ' ' | cut -d' ' -f4 `
    if [ "$ctime" == "$1" ];then
	"$2" &
	echo -e "Process $2 with PID $! has been run at $ctime\nAborting script $0"
	exit
    fi
done

Of course, this script can be made to do things a lot better, but right now, it does exactly what I need!

While being extremely funny, this is quite inconvenient. Now I cannot easily go to a page that comes up on google search since on clicking the link, a separate warnign page opens up! I really hope that this bug (or whatever you may call it) gets resolved at the earliest. /me cannot survive for long in the virtual world without google!

The idea of the project I have in mind is to develop some kind of a Ebook Library Manager. For long, I myself have felt the need of such an application to manage my reasonably large collection (nearly 2 gigs) of ebooks I have on my hard disk. None of the existing collection managers (like GCstar, Alexandria,etc) catered to my needs, so I needed to write an application of my own. Originally, my plan was to do it in Python with an Sqlite3 database backend and WxPython for the GUI. Now, Php would replace Python, MySql would replace Sqlite, and of course, there would be no need for a GUI toolkit as the frontend will be handled on the web browser itself by using HTML/JS. This is a rather good thing because I won’t have to worry about the WxPython codes, which is quite difficult & time consuming to perfect(esp. the layouts & alignments) in absence of a proper RAD tool. Also, the program will now be web-based & hence platform independent with no need for a specific GUI toolkit to be installed. On the downside, using PHP and MySql makes running a webserver like Apache mandatory, which sounds a bit overkill & out-of-context since the purpose of the program is to manage local ebooks. And I have to learn PHP, HTML and Javascript fairly well before starting the project.

I hope that this idea qualifies as a valid A Level Project, esp. with the aprox. 350 man hours involvement requirement posed my DOEACC. Formal allocation of guide and commencement of projects will begin from 3rd week of February. I intend to have a working draft of the project by then to show. Till then, onwards to w3schhols and the PHP Manual!

#!/usr/bin/env python

from pyPdf import PdfFileWriter, PdfFileReader
import sys

opfile=sys.argv[-1]
output = PdfFileWriter()

for ipfile in sys.argv[1:-1]:
	input=PdfFileReader(file(ipfile, "rb"))
	for pnum in range(0,input.getNumPages()):
		output.addPage(input.getPage(pnum))
outputStream = file(opfile, "wb")
output.write(outputStream)
outputStream.close()

Its really easy to use it. Save it as pypdfmerge.py, and run it as:
python pypdfmerge.py  <input_file1> <input_file2> <input_file3> .... <output_file>
The last argument is always taken as the output file.
Note that it will throw some Deprecation warnings, but they are generally harmless, so ignore them.

??????-