[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: Can I have one project link to two subversion server?

From: David Weintraub <qazwart_at_gmail.com>
Date: Tue, 27 Apr 2010 13:39:14 -0400

On Tue, Apr 27, 2010 at 1:04 AM, frank wang <yixiaodafang_at_gmail.com> wrote:
> Anyway, here is reason I want to check in the code into a private
> resository before I merge the code into the office SVN.
> During the development, I added a lot of codes for debugging in many stages

There is no reason why you can't have debugging information in your
code and have displaying the debugging information as optional. If you
look at Maven and Ant, they both have debug settings. You use the
debug setting, and you get debug information. You don't use it, and
you see nothing.

You can always remove the debug information later on if you want, but
just because you'll have debug information doesn't mean that you have
to work offline from the rest of your team.

Staying connected with the rest of your team, doing regular commits to
the branch and repository place your fellow developers are working
helps produce better code. It forces you to communicate.

Most of the time, when a developer tells me they have to work
independently from the rest of the team and don't want to check in
their code into the repository until it is 100% complete, I find that
there is really another reason involved.

For example, the developer doesn't want to take time to do all those
tests because they want to work on finishing up the features. If they
check in their code, their fellow developers will bug them about this
or that not working. In reality, it's better and faster to look for
bugs while you code. Yes, debugging isn't fun, but it's why you get
paid the big bucks.

Sometimes, it is because the developer is biting off more than they
can reasonably handle. Restructuring what they want to do into more
manageable pieces usually helps.

Other developers tell me they don't want to commit incomplete code.
All code is incomplete. If it wasn't, there wouldn't be any changes
and we'd all be out of a job. Methods and classes can return faux data
for testing purposes. For example, if I was working on a method that
returns the price of an object, I could return either a fixed price or
a random value that would be good enough for other developers to use
for their coding while I work on the routines to actually connect to
the server and look up the price.

So, take a look at what you want to do and determine whether you
really should be working off to the side in your own repository.
Working off line like that means you miss the feedback of your fellow
developers and you miss the changes they're making which could end up
breaking everything you're doing.

-- 
David Weintraub
qazwart_at_gmail.com
Received on 2010-04-27 19:39:44 CEST

This is an archived mail posted to the Subversion Users mailing list.

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.