Where does the build come from?

The build (compilation) of software is created as a result of the process of compiling, linking, and packaging the project's source code together with necessary resources (libraries, configuration files, images, etc.) into a single executable or distributable package.

Manual build involves the developer performing these steps manually or with scripts on their local computer. Automated builds use continuous integration (CI) systems such as Jenkins, GitLab CI, GitHub Actions, TeamCity, and others.

The automated build process generally looks like this:

1. Trigger: Changes in the repository (e.g., commit to a specific branch) or scheduled build execution trigger the build pipeline.
2. Code retrieval: The CI system fetches the latest version of the source code from the version control system (Git, SVN, etc.).
3. Build: Build commands are executed according to the project configuration (e.g., mvn clean package, npm build, dotnet publish). This stage includes compiling the source code, running unit tests, and creating build artifacts.
4. Testing: Various types of automated tests (integration, e2e, etc.) may be run.
5. Packaging: The build artifacts are packaged into the appropriate format (WAR/JAR for Java, Docker image, NPM package, MSI, etc.).
6. Publishing: The final artifact (build) is published to an artifact repository (Nexus, Artifactory, Docker Registry), from where it can be retrieved for deployment or further testing.

Thus, the build is taken from the automated or manual process of compiling, linking, and packaging the source code, most often performed by the CI system after changes in the repository.