Managing required packages in nALFS

Jamie Bennett jamie_bennett at pcpmicro.co.uk
Mon Sep 16 06:25:45 PDT 2002


> -----Original Message-----
> From: Vassili Dzuba [mailto:vassili.dzuba at wanadoo.fr]
>
> I thought we could use a single XML tree, regrouping all 
> the profiles
> fragment required to build the configuration we want.
> The profile for a single package is usually quite small (< 1K),
> so that a profile for a configuration containing 1000 packages
> translates into less than 1Mo of XML text, which is probably doable,
> even if the startup time wouldn't be blinding fast.

   Just thinking out aloud but how about 2 XML trees. One can be the
install tree (where the package(s) you want to install are kept) and
one can be a search tree. 

   The search tree could be loaded in every time by default and when 
a package dependency is found and not satisfied in the install tree 
we could easily just switch trees and search for the package we need. 
This can then be tagged on to the install tree and installed as normal.

   The advantage of this is that all profiles are loaded into memory
at run time (no need for dynamic loading of profiles) and all installable
packages are known at the very start. Switching tree's should be just 
a matter changing the element_t *el in functions to the search tree 
and adding a new profile should just be a do_add_profile(const char
*profile).

   This is just an idea and I'm still looking through the code to 
see if a) this is possible and b) if it is the right way of doing it.

> Vassili

-- http://www.linuxuk.org --------------------------------------
- Jamie Bennett     - 18 St Peters Terrace - jamie at linuxuk.org -
- Software Engineer - Lower Bristol Road   -                   -
- PCP Microproducts - Bath, England        -                   -
----------------------------------------------------------------
-- 
Unsubscribe: send email to listar at linuxfromscratch.org
and put 'unsubscribe alfs-discuss' in the subject header of the message



More information about the alfs-discuss mailing list