Some thoughts on beat writing

[John J. McDonough]

----- Original Message ----- 
From: "Karsten Wade" <kwade at redhat.com>
To: <fedora-docs-list at redhat.com>
Sent: Tuesday, December 09, 2008 3:18 PM
Subject: Re: Some thoughts on beat writing

> Not sure, sounds potentially useful.  If we can do it without much
> hassle, let's do it -- _into_a_special_namespace_.

Certainly willing to share, but it is kind of a PITA.  What I do is grab the 
repo data into a MySQL database so I can collect 'N' releases.  It's 
organized differently than the sqlite db, but no reason the sqlite couldn't 
be used.  It's just that I can make manual changes easier with phpMyAdmin 
than grokking SQL.  Then in a separate step I generate a WikiMedia XML 
import.  Actually, I could see this is something that might be useful for 
some other audiences, but I found it helpful to watch what packages were 
undergoing revision, and sometimes it is interesting to dig into those 
changes and find out what were the issues.

> Does this get more information than 'repodiff'?

Actually, less than repodiff, and that's the point.  It reads a list of 
packages in your beat and only reports differences in those packages. It 
isn't all that difficult to pick up changes, the challenge is picking up 
changes in a handful of specific packages when there are thousands to wade 
through.  And as we get close to release, many, many packages are changed.

--McD