[Xastir] Refactored/rewritten/split xastir

Tom Russo russo at bogodyn.org
Sun Jan 2 02:15:50 EST 2005


On Sun, Jan 02, 2005 at 12:32:36AM -0600, we recorded a bogon-computron collision of the <tbaggett at jump.net> flavor, containing:
> 
> 
> Forgive me if I am I'm getting involved here where I shouldn't, or if I don't understand all the circumstances of the project ...
> 
> > I think the right thing to do would be to have one or more
> > developers concentrate on splitting the current code base up into
> > multiple pieces, while other developers focus on bug fixing the
> > current stuff.  We'd want to do the splitting on a CVS branch so
> > that we wouldn't interrupt the bug-fixing on the main branch.
> > 
> > Once the split code was ready, it could be merged into the main
> > branch and everybody could try it out.  
> 
> FWIW, I suggest freezing the Xastir code and tagging it onto a branch of it's own. All bug fixes are made to the branch and merged into the trunk as needed. Leave the massive coding changes on the trunk.

Honestly, the things being discussed in this thread would warrant an entirely
separate repository, IMNSHO.  

> In my experience it is easier to merge multiple small changes (bug fixes) from a branch to a trunk than it is to merge massive developmental changes from a branch onto a trunk.

I agree -- this is how we handle releases where I work.  We freeze code,
create a "release branch" on which only bugfixes are allowed, and the main
trunk remains the primary development branch.  Bugs found during release
QA testing are fixed on the release branch, and merged back into the main trunk
after the code is released.  The release branches usually whither after the
release, although sometimes we have a second release of the branch code, usually
as a minor release with more bug fixes.  Turning the current version of
xastir into a branch off the main trunk, and the rebugged, split version
be the main trunk would be pretty much the same idea --- there could be
continued support of the monolithic code, but it would be the dead end branch.

The only experience I've had with creating a branch to do major development
hacks (on a different project, years ago) was a disaster.  I'd never want to 
have to deal with that hassle again.

-- 
Tom Russo    KM5VY     SAR502  DM64ux         http://www.swcp.com/~russo/
Tijeras, NM  QRPL#1592 K2#398  SOC#236 AHTB#1 http://www.qsl.net/~km5vy/
 "When life gives you lemons, find someone with a paper cut."



More information about the Xastir mailing list