I'm happy that a new release has been possible.
This new release is almost a complete re-write and has many fixes and features.
For details, visit:
There's also a small HOWTO about pypt-offline
Cheers!
Ritesh
pypt-offline -- An Offline Package Manager
(C) 2005, 2006 Ritesh Raj Sarraf <rrs@researchut.com>
# INTRODUCTION
So you've decided to give this small piece of work a try.
Good ! Let's get it done faster.
pypt-offline is an offline package management tool written in the
Python Programming Language. This program, as of now, is intended for
people using Debian (And Debian based) systems.
This program allows leveraging the power of Debian (more precisely apt-get) onto a completely disconnected machine. Most of the people with slow or no internet connection (most of those in India/Nepal/Pakistan and nearby countries) have not considered using Debian (or Debian derived distributions) because Debian's real taste is experienced when it is connected to the internet.
This utility is an attempt in making that problem eradicate. I hope
this utility comes to use to you.
I'd be eager to hear your comments/suggestions. Feel free to drop an
email at rrs _AT_ researchut |DOT| com
#################################################
Let us assume you have a machine at home ( Hereby called Machine-A) with no or very expensive internet connection on which you've installed Debian (Hereby all Debian and Debian based systems will be called as Debian).
You or your friend works at a local city office where they have High Speed Internet Connection. The machine used there is Linux/Windows/Mac box. We'll call this Machine-B henceforth.
pypt-offline allows you to synchronize Machine-A's apt database with
the Debian archives.
It does so by
* Extracting the details of Machine-A's apt's database which needs to
updated
* Fetches the required database on Machine-B
* Synchronizes the fetched database back to Machine-A
With these 3 steps you can keep a disconnected Debian machine
up-to-date on a daily basis.
The rest of the document will describe the details along with the
commands.
STEP - 1
On Machine-A:
pypt-offline --set-update
With this command, pypt-offline extracts the details from apt's
database for the files that are required to update apt's package
database. The extracted information is stored in
pypt-offline --set-upgrade
With this command, pypt-offline extracts the details from apt's
database for the packages that are required to be upgraded. The
extraced information is stored in
There are 3 types of "upgrade" type. "upgrade", "dist-upgrade" and
"dselect-upgrade".
You can pass any one of them.
Note: --set-upgrade and --upgrade-type are mutually inclusive. You
cannot use one option without the other
pypt-offline --set-install
With this command, pypt-offline extracts the details from apt's
database for the packages (and its dependent packages) that are
required to be installed. The extracted information is stored in
Note: --set-install and --set-install-packages are mutually inclusive.
You cannot use one option without the other
The above mentioned options are executed on Machine-A. They extract the
details from apt's database.
Now the user needs to copy the extracted data file onto a removable
media and take it to Machine-B.
STEP - 2
On Machine-B:
With the extracted data file in hand, on Machine-B, execute the
following commands:
pypt-offline --fetch-update
With this command, pypt-offline fetches the required data from the
internet as directed by Machine-A
The options in square bracket are optional
-s
Also the freshly downloaded files are copied to this folder so that you
don't have to download it again the next time.
Note: This option is effective only for packages which are downloaded
and *NOT* for update files. Update files change almost daily in the
Debian repositories, hence keeping a local cache is useless.
--disable-md5check - It is *highly discouraged* to use this option.
By default, pypt-offline compares the md5checksum for every package it
downloads. If the md5checksum don't match, the file is deleted. This is
necessary to make sure that you don't end up having tampered packages.
Note: This option is effective only for packages which are downloaded
and *NOT* for update files. Update files currently don't provide
md5checksum details.
--zip - It is *highly encouraged* to use this option. Though
disabled by default, if enabled, pypt-offline creates a single zip file
of the downloaded files. This way it is easier to track the downloaded
files.
--zip-update-files - This option sets the zip file name. If this
option is not set, pypt-offline uses the default file name,
pypt-offline-update.zip
--zip-upgrade-files - This option sets the zip file name. If this
option is not set, pypt-offline uses the default file name,
pypt-offline-upgrade.zip
If you don't use the --zip option, the downloaded files are stored in
the folder you mentioned with the -d option. You'll need to copy the
files from the folder.
With this, once all the data has been downloaded, copy it to your
removable storage device and take it back to Machine-A.
STEP - 3
On Machine-A:
Once you come back to Machine-A:
pypt-offline --install-update
With this command, pypt-offline syncs the update files from
updates.zip. If a folder name is given as an argument to
--install-update, it searches for the required files and syncs them to
apt's package database.
pypt-offline --install-upgrade
With this command, pypt-offline syncs the package from updates.zip.
If a folder name is given as an argument to --install-update, it
searches for the requires files and syncs them to apt's package
database.
Note: Please keep in mind that this doesn't actually install/upgrade
the packages. It just makes the "To Be Downloaded" package available.
With it apt will not require to download any additional packages and
hence will do the installation as if it was downloaded from the
internet.
NOTE: If you use apt-listbugs to track package bugs while you
install, it'll fail because apt-listbugs will try to connect to the
internet.
Trackback link:
Please enable javascript to generate a trackback url