[RFC] The Future of the ALFS project

George Makrydakis gmakmail at gmail.com
Tue Feb 28 01:02:21 PST 2006


Hello Guys!

As I stated in my initial posts, bash indeed is extremely inefficient should you decide to make a killer meta-program like application (scripting is a support 
oriented programming tool, not a core production one). Nevertheless, it can be a handy tool _INSIDE_ stand - alone bash scripts especially since right now (see 
LAST post attachment) it "parses" the complete book in about a minute and a half tops, maintaining attribute - specific information while rewriting the file in 
such a way that it makes it "uninformed" enough to apply the =~ operator everywhere. My current development versions support entity dereferencing and in line 
parsing in ways I would say that bash could never have even made me think about till I started working on this project. :) For this, I am happy. Right now I 
have remodeled everything towards the direction indicated in the last post, for I believe it is the best way to go. I would prefer coding something like this in 
C, C++ or a C#/Java like language. Yet, bash offers itself to abuse beyond belief (with the disadvantage of slow exec times), which can be handy in certain 
situations.

I believe that I am close to completing this for the LFS/BLFS book based xml source (with tweaks the same applies for other book versions but I am now focused 
on the critical core). Close to after - the - weekend result to be feature complete and debugged.

Provided there is time on my side, I am available to start coding for your project in C/C++/C#/Java (preference for C++ if in binary), I will in any case 
complete the bash - based "parsing" approach, for it may prove handy as a tool elsewhere in the project (support scripts). Also with it goes a work for table 
structures in bash.

Thank you as always, will keep you updated. I am really looking forward to completing this; for the sake of imaginative bash scripting too!

George Makrydakis

gmak

PS: the coolest part about LFS is that it is like having the data define the way the OS is structured, you have documentation for it ready... So a single XML 
file is like short of having a "package management file".




More information about the alfs-discuss mailing list