eclipse & subversion

Hi

I am using eclipse as an IDE and am setting up a subversion server.

In trying to add subeclipse I get the message:

"Subclipse Mylar Integration (1.0.1) requires plug-in "org.eclipse.mylar.tasks.core (0.9.2)", or later version."

Then I have difficulty finding how to install Mylar.

Then I find the Mylar project is now something different (Mylyn I believe) and Mylar has to be deinstalled.

Is anyone in a position to give me instructions on how to enable eclipse to be able to checkout/checkin from a subversion repository and be able to use it via Aptana?

Thanks

David Southwell

Hi,

"Subclipse Mylar Integration (1.0.1) requires plug-in "org.eclipse.mylar.tasks.core (0.9.2)", or later version."

Then I have difficulty finding how to install Mylar.   

So don't install it. You don't need Mylar at all for working with subversion.

I'm assuming you are installing through the "software update" menu option. If not, give it a try as it's the generally recommended way of installing plugins.

When installing subclipse you should see the available components to install. One of them will be the Mylar integration. Uncheck it, click on next and be happy :wink: You also have available the Mylyn integration for subclipse anyway in case you want it.

regards,

javier ramirez

I use Subclipse without Mylar Integration and I am happy :smiley:

Quoting David Southwell <david@atf4.com>, who spaketh thusly:

Is anyone in a position to give me instructions on how to enable eclipse to be able to checkout/checkin from a subversion repository and be able to use it via Aptana?

Here's something I wrote up this morning, as I just went through the install process on my new macbook :wink:

http://blog.spacemonkeylabs.com/articles/2007/07/17/installing-eclipse-php-ruby-subversion-and-rails

-- Mitch

The problem i am getting on our trials is that the databases seem to get screwed up for different users. Bear in mind we are using eclipse as our IDE along with subclipse as the subversion client. It makes a really great IDE but as soon as we began to make it work using the subversion server we got into trouble!!!

User A updates after User B has committed changes. Now user A then finds that his databaseis not in synch with the the update. It seems there need to be some rulesfor the subversion server to ensure that the repository is not updated with files that reflect a single developer's configuration.

For example after a modification by Developer A and a subsequent update by developer B, B gets rake terminating unexpectedly reporting it could not find the presence of a particular table in a database.

Obviously we do not know what we are doing here....I am sure we are making elementary mistakes. Does anyone have a set of rules and possibly some guidance to get me started!!!

I am a novice with subversion.

Thanks

If you aren't aware of it already, I would recommend taking a look at the free online Subversion book: http://svnbook.red-bean.com/

You can use the .ignore property to tell svn not to include certain files or directories in its operations.

Are you using migrations for modifying the database? That should make it pretty easy to keep everyone's db structure in sync.

-jc

David Southwell wrote:

If you aren't aware of it already, I would recommend taking a look at the free online Subversion book: http://svnbook.red-bean.com/

Yes that is a great resource

You can use the .ignore property to tell svn not to include certain files or directories in its operations.

Understood -- but am uncertain about which files and directories should be ignored

Are you using migrations for modifying the database? That should make it pretty easy to keep everyone's db structure in sync.

Yes but it seems that when the developer A get his files upgraded from developer B we find migrations by B not working. I do not know why.

`> > > The problem i am getting on our trials is that the databases seem to get

screwed up for different users.

Bear in mind we are using eclipse as our IDE along with subclipse as the

subversion client. It makes a really great IDE but as soon as we began to

make it work using the subversion server we got into trouble!!!

`

`> > > User A updates after User B has committed changes. Now user A then finds

that his databaseis not in synch with the the update. It seems there

need to be some rulesfor the subversion server to ensure that the

repository is not updated with files that reflect a single developer’s

configuration.`

`

For example after a modification by Developer A and a subsequent update

by developer B, B gets rake terminating unexpectedly reporting it could

not find the presence of a particular table in a database.

`

It sounds like your developers may be stepping on each others database.yml. Is it possible that after Developer-B updates, his database.yml is pointing to Developer-A's development db?

Either you have to share a development database and get in the habit of running rake db:migrate after an svn update to ensure you have the latest migrations from each developer. (and then the database.yml can be shared)

Or you need to create separate development databases and create different database.yml (set ignore property) for each developer.

Hope this helps,

Dave

I think you have hit the nail on the head. Each developer needs to use his/her own localhost database rather than a single network database, so we need to use individual database.yaml.

I must take a look at subversion and see if there is a way in which different files can be flagged for delivery to specific users. My guess is that I could do something smart using properties.

Thanks for your contribution David