Using Git with PL/SQL in a Shared Development Database

 

With Gitora 6, developers can work on the same code base (i.e. a package, procedure etc…) in the same database without blocking each other.

Many development teams use a single database for development. Many of them use the same database for testing, as well. They achieve this by using different schemas for different purposes. Gitora 6 enables these teams to create Git repos for these schemas and pull updates between them.

With Gitora 6, you can even create a different schema for every developer and have them merge their code using Git.

Gitora 6 enables you to implement any modern development workflow in a single database.

How does Gitora 6 work?

Gitora 6 introduces a new repo type called Single Schema Repo (SSR). As its name suggests an SSR manages database objects from a single schema. The DDL scripts in SSR’s don’t contain the schema prefixes so that Gitora can execute them in other schemas in the same database.

This enables developers to work on the same package, procedure, function, view etc… (i.e. anything that can be created with the CREATE OR REPLACE command) at the same time, in the same database in different schemas.

An Example

Let’s go through an example: Let’s assume that the team is working on a logistics software and have a schema named LOGISTICS that stores all their database objects. The team can create a schema (or already have one) called LOGISTICS_TEST in the same database. Here are the steps the team needs to follow so that they can pull their changes to LOGISTICS_TEST.

  1. Create a single schema Gitora repo that manages the LOGISTICS schema. Let’s call it the REPO_LOGISTICS.
  2. Add all relevant database objects in the LOGISTICS schema to the LOGISTICS_REPO.
  3. Create another single schema Gitora repo that manages the LOGISTICS_TEST schema. Let’s call it the REPO_LOGISTICS_TEST
  4. Pull from the REPO_LOGISTICS to REPO_LOGISTICS_TEST

That’s it. That’s all there is to it. From this point on, any change you make to the code in the LOGISTICS schema can be pulled to the LOGISTICS_TEST schema using Gitora (and vice versa).

Single Schema Repos can also be used to create separate development environments in the same database for every developer in the team.

Multiple Developers, Same Database

Assuming we already have the LOGISTICS schema and the REPO_LOGISTICS repo from the previous example, here is how that would work:

  1. Create a schema for each developer: LOGISTICS_JOE, LOGISTICS_MARY, LOGISTICS_PAUL.
  2. Create a single schema Gitora repo for each schema. Let’s call them REPO_LOGISTICS_JOE, REPO_LOGISTICS_MARY, REPO_LOGISTICS_PAUL respectively.
  3. Pull from the REPO_LOGISTICS to REPO_LOGISTICS_JOE, REPO_LOGISTICS_MARY and REPO_LOGISTICS_PAUL.

From this point on, all three developers can work in their own schema and edit any package, procedure, view etc… freely, without overwriting each other’s changes or being affected by them. Using Gitora, they can create new branches in their own repo, for features or bugs they work on without affecting anyone else.

When the time comes to put it all together, they can use Gitora to merge their changes.

We wrote a few articles to get you started:

Creating Single Schema Repos.

How to implement agile development in a single database

How to manage changes to tables with Single Schema Repos

Before Gitora 6, agile database development with Gitora required multiple databases to implement. With Gitora 6 you can achieve it in a single database.

We are incredibly excited to release Gitora 6. We believe it will bring a massive productivity boost to Oracle database development teams worldwide. We can’t wait for you to try it.

Gitora 6 is available now. You can download it from this link.

Comments

Popular posts from this blog

Agile PL/SQL Development with Git

Gitora 5 Source Control for Oracle is Out

Source Control for Oracle, a Customer Success Story