[Koha-devel] Ready to Hack

[Chris Cormack]

Hiya Nathan

Nathan Gray wrote:
> I've been watching Koha for a little while now, and finally my wife told
> me to just install it, because she wants to use it.  As she reads books,
> she wants to keep track of which ones she likes, so she can check them
> out when our two-year old gets to that reading level.
> 
> So now we've got a working Koha system on her computer.  I've noticed
> already a few things I'd like to tweak, so I can use it more
> comfortably.
> 
Excellent

> I am downloading from CVS right now.  Is there a way to run Koha
> directly from the CVS tree?  Any quick start pointers?  I figure I can
> leave plain Koha on my wife's machine, and put the CVS version on mine,
> both running off her database.
> 
There is a little tutorial on the wiki about running Koha from cvs.
http://www.saas.nsw.edu.au/koha_wiki/index.php?page=CVSSymLinkInstallation

> I'm a Perl programmer by trade, with a lot of my time spent re-writing
> modules to increase stability, efficiency, and readability (at least, I
> understand it better when I'm done).  That's mostly what I'd like to do
> here.
> 
Sounds good to me. Code cleanup is always appreciated. What would also be
highly useful is, while you tidy modules. Write some unit tests for them. 

> First off, I'd like to get Z39.50 searching a little nicer.  Part of
> that might just be clearer documentation.  (I'm having problems finding
> some ISBN numbers in the Library of Congress, even when they have a LC
> call number, so it would be nice to be able to search on that as well,
> or know which other servers I should be looking at.)  And so when I
> search by title and get lots of records with the same title, it would be
> nice to have a few more columns to differentiate between them, like date
> published, or LC call number if there is no ISBN.
> 
Sounds like a good plan to me.

> I would also like to allow the use of Net::SMTP, rather than
> Mail::Sendmail, and perhaps the DateTime suite rather than Date::Manip.
> I would also like to be able to use DBD::Pg, rather than DBD::mysql, 
> though I am not volunteering for that quite yet.
> 
> How should I go about getting changes back into CVS?  Is there a contact
> person I need to go through?
> 

The thing to bear in mind, is that in CVS there is the HEAD where
development for 2.2.x is going on.
Paul Polain is the release manager for 2.2  so talking to him about what you
plan to work on is a good idea. Just to remove duplicated effort etc.
Then there is the rel_2_0 branch, which is where the 2.0.x releases are
drawn from.
So bugfixing etc should be done in that branch, new features added in the
HEAD.
M J Ray is looking after the 2.0.x releases. So he's your best contact for
work in that branch.

When you've had a chat with them, (im sure they will see this message and
reply). You can let myself or Pat Eyler know your sourceforge username and
we'll give you write access to the cvs repository.

Chris
-- 
Chris Cormack                                                     Programmer
027 4500 789                                       Katipo Communications Ltd
chris at katipo.co.nz                                          www.katipo.co.nz