Add remaining models

[jrhorn424]

We need to discuss this code. I'll make comments in the code, but a few general notes:

• Try to write more descriptive commit messages. When I rebase and I see you've changed a file without knowing why, it's hard to know which version is the "correct" one.
• Try to make more, smaller commits. For example, if you're adding rolify and it require taking out a join table definition, that's fine, but not it. If the changes you've made are unrelated to adding/removing something (like rolify), make a separate commit.
• It's especially important to mention why migrations are deleted.

We can't automatically merge this branch into master since there are so many conflicts. That largely unavoidable, but remember:

• When you start your days work, rebase against master to pull in any changes that have been committed there by others (just me at the moment). Also rebase against any published feature branches, since these may have changed as well.
• Smaller commits mean fewer conflicts. "Commit early and often".

[jrhorn424]

Let's do a code review together. You look at my code, and I'll look at yours. This is the best time to ask questions, since we can mark the code directly with comments. When we're both ready, we'll merge.

[jrhorn424]

I've extensively rewritten the history. I'll show you the DAG for the code later today. This means your local copy is probably broken, so we'll need to fix that. I'm not sure if any of these patches are correct, since I had no clue what the intent was for each commit.

One thing I noticed when resolving these conflicts is that we need to be consistent with the naming of our join tables. I prefer the format parents_childs_table. For example, experimenters_experiments_table. It's less clear what we should do in the case of many-to-many relationships. What do you think? I find users_sessions_table slightly more preferable than sessions_users_table, but I don't know why.

[mlfreer]

Names of joint tables are written consistency with ruby convention, that subjects should be in the alphabetic order. If it's necessary to change i can do it  

Понедельник, 7 октября 2013, 2:20 -07:00 от Jeffrey Horn notifications@github.com:

I've extensively rewritten the history. I'll show you the DAG for the code later today. This means your local copy is probably broken, so we'll need to fix that. I'm not sure if any of these patches are correct, since I had no clue what the intent was for each commit. One thing I noticed when resolving these conflicts is that we need to be consistent with the naming of our join tables. I prefer the format parents_childs_table. For example, experimenters_experiments_table. It's less clear what we should do in the case of many-to-many relationships. What do you think? I find users_sessions_table slightly more preferable than sessions_users_table, but I don't know why. — Reply to this email directly or view it on GitHub .

Миша Фреер

[jrhorn424]

Don't worry about changing the join table names, as long as we're following a convention.

[jrhorn424]

OK, you can pull --rebase to get my last commit, make any changes requested in this discussion, and push the commits to this branch. The pull request will be automatically updated. Once that's done, we can merge.