Code repository or SCM (Source Code Management) software terminology can be confusing. This page provides definitions for some of the most commonly used terms.
On this page:
FishEye and its documentation uses the following terms:
A branch is an independent stream of work in a repository. For example, you might copy a set of files in the repository into a new branch, where you can carry on with a separate stream of work without cluttering up the main production area on trunk.
Different SCMs handle branching and merging in different ways. The common aspects allow users to create a branch in which to make changes which do not affect the files in other branches and the trunk development stream. These changes can then be merged into the trunk in a controlled fashion when a development unit of work is complete. Branches can also be used for experimental changes so that these do not affect the main development.
A changeset is a collection of changes to files in a repository which are committed to the repository in a single operation with a single commit message. Not all SCMs support atomic commit operations. For these SCMs, FishEye will determine the file revisions which make up the changeset using a reliable heuristic (set of rules).
Different SCMs use different terms for the concept of a changeset, such as "changelist", which is generally interchangeable with changeset.
A commit is a single entry of content (usually source code) into a repository. It can be a single file or comprised of multiple file versions.
A committer is a user of an SCM repository who is adding content to the repository (where it will be permanently archived). Typically, a committer is a programmer who is committing source code but SCMs can also store other files such as documents, images and schematics.
An abbreviation for 'Changeset ID', this is a code that is used to reference every set of files that is committed to a repository. For example, if you commit three files to a repository, they are collectively a changeset, and will share the one CSID. CSIDs normally appear as a number, for example '31905'. In FishEye, CSIDs appear as links that you can click to visit the 'Changeset View', which shows a list of the files in the left column, and the file contents or diffs in the right hand pane. In some circumstances you can hover your mouse over the CSID to see the 'Changeset Hover' dialog, which displays the commit message, author, timestamp and files in the changeset.
The "head" revision is the latest change to be made to a file in either a trunk or a branch part of a repository.
FishEye can render issue IDs or Bug IDs that appear in commit messages or comments as hyperlinks to the relevant issue/bug in your issue/bug tracker. The "linker" patterns used to detect the ID substrings can be configured separately for each repository. See Linkers.
A repository is a area managed by an SCM where you store a set of related files, typically from a project or set of projects. The SCM is responsible for version controlling the files in the repository and maintaining their change history. A repository will contain the trunk and all branches for the files of the various projects. A single SCM instance can typically house multiple repositories.
SCM (Source Code Management) software is a category of computer software that archives complex sets of files, for example, all the source code comprised in a large multi-part software project. SCM systems keep copies of every version of every file, allowing you to completely restore and build any version of the software from any point in time.
Committers typically add new versions of code to the SCM once it is tested and approved by peer code review or quality assurance.
Each instance of an SCM can host multiple repositories.
"Slurping" is a term that is synonymous with "repository scanning". FishEye must do intensive scans of the contents of repositories (SCM systems) in order to provide its lightning-fast web-based browsing of their contents. This can be referred to as a slurp, or slurping.
In SCM terminology, a "tag" is a label that is added to a number of files, to capture their collective state at a particular moment in time. A typical tag would be a specific software version number, that could be referenced to see all the files that belong to a specific version build of a software project.
In SCM terminology, the "trunk" is the central part of the version control "tree". For example, you might copy a set of files in the repository into a new branch, where you can do new experimental work without cluttering up the main production area on trunk.
Was this helpful?
Thanks for your feedback!