Re: less dependencies in Debian packages ?
On 2007-04-10 17:00 +0200, firstname.lastname@example.org wrote:
> I tried to look if somebody already made a proposal similar to this one,
> but couldn't find anything. Eventually I was directed to this list, so
> this could not be the right place for such a post. Thanks to direct me
> if any of you know of a better one.
> It is often said that (some) debian packages carry too much dependencies to
> be useful for embedded projects or sometimes drag in annoying libraries.
And it is true.
> Making these dependencies optional would help people who try to minimize
> their system image and still enable the people who need them to use them.
> How to do it ? The dependencies I'm discussing are decided at compile time.
> If one could build a program with every possible 'interesting' combination
> of compile-time (enable or disable library) options (and using ccache to
> speed things up), he could then create an archive of binary patches against
> the most stripped down version of the binary. For example the utility
> EDelta (http://www.diku.dk/~jacobg/edelta/) efficiently produces patches
> for executable binaries (but isn't sufficient since it doesn't support
> n-way patches).
> Then, the binary the user calls would be some sort of rebuilder script
> which would pick the correct patches depending on what is really available
> on the system. Many hard dependencies could be moved to 'recommends' or
> 'suggests'. To prevent rebuilding the real binary every time, it could be
> built every time a depending package is installed.
> I'd like to know if you think this idea could work or not (and if not, why)
It is an interesting idea, although I worry about the 'some sort of
rebuilder script' part. Is it really possible to have sensible
'patches' from a base image to versions with extra libraries, or do
you in fact end up with just a load of different binaries (because the
differences do not localise well?)
It seems to me that if all the versions are available then the
package-size could be large. It depends how well this patching scheme
works. If it is too larget then you might as well just install the
deps, as is done now, and still get a smaller system overall.
Emdebian is working on choosing our own fairly minimal useful set.
This is OK until you want something that is not enabled by default in
Do you have any proof-of-concept numbers which would show that the
binary patches idea actually makes sense numerically?
Principal hats: Balloonz - Toby Churchill - Aleph One - Debian