Before this, timeout only apply to repository caching.

Now timeout is used by all transfert functions.
A new parameter repo_timeout is available to set a custom timeout on repository

We now use system tar and gzip to compress/decompress payloads

Goals are:
- Speed-up extraction and creation of payloads.
- Allow to use non-python supported compressor like (pbzip2 and xz)
- Remove python2.7.1 tarfile.py module
- Remove python 2.7.1 gzip.py to avoid filename to be in gzip headers
- Remove gzipstream.py module used to stream gzip file

we need python-progressbar >= 2.3 to work with is.
We previously ship inside it because debian doesn't have 2.3 package available.
now we have built a debian package, we can remove code and add a dependency

old command line option are still present but no more in config file
this allow to override easily on command line config parameters

prepare and uprepare chroot functions now track non existant files
preparing are now done more carrefully about tricks

Image are not crawled anymore in all repositories by default
You needs to specify a list of repo to look in

* tricks /etc/mtab
* new output

this already more easy unchrooting after failure by chrooting again,
solving issue and exiting (which remove mounts again)

chroot failed when root is missing dirs before resolv.conf fil

This is done to be able to use preparation from scripts

this command is and helper to chroot inside payload mounting /{proc,sys,dev,dev/pts} fs
inside the chroot and disable daemon restart

We need installsystems is functional on squeeze systems...

So we needs to embed:
- argparse 2.7.1 (not in python2.6)
- progressbar 2.3 (not in debian squeeze)

We can now use installsystems without paramiko

Also fix some bug in PipeFile

This commit introduce remote repository and file accross ssh

We now use PipeFile to handle complex file type and compute md5 and filesize.

This commit add the following features in installsystems:
- A download progress bad
- check size before download
- check size after download
- check md5 after download

THis is very useful to be able to list it !

When huge payload was read, it could cause Out Of Memory error.
Use istools copyfileobj for read chunk by chunk.

Signed-off-by: Seblu <sebastien.luttringer@smartjog.com>

Signed-off-by: Seblu <sebastien.luttringer@smartjog.com>

Signed-off-by: Seblu <sebastien.luttringer@smartjog.com>

Script is now compiled before execution to have a better debugging features.
local directory is no more given for each scripts
global directory is new to each script execution
only namespace object must be used to share data across scripts