Code freeze

from Wikipedia, the free encyclopedia

The code freeze describes the point in time within a software project from which the source code of the software is no longer to change until the current version is finally published. However, changes to remedy errors of greater relevance discovered in the test of the software are still permitted.

In the practice of software development, the code freeze is usually determined at the beginning of the system test phase, i. H. several weeks, etc. This can also be months before the planned release of a software version, so that there is still enough time to test the final version of the software. The goal is to minimize the number of bugs in the published software.

General

In a software project, changes are regularly made to the existing code during the implementation phase, i.e. during the creation of the source code, to add new features and to correct errors that have occurred. After the code freeze point, changes to add new features may no longer be made; the software was practically frozen at its current state . Changes to correct errors discovered in the test may generally still be made as long as the benefit that arises from eliminating the error is in a reasonable ratio to the risk arising from renewed changes to the source code and the associated possibility of inserting new ones Errors in the software inevitably arise. As a rule, there are special requirements for each change. It is common for a software developer not to be able to make the decision to change the code after the code freeze. Instead, the decision is usually made by a multi-member committee.

Practical implementation

Since the source code in software projects usually in a system for versioning located is from the code freeze introducing a new version of source files by "check" ( check in ) or "merge" (when using Branches) are not possible for the individual developer or only after certain conditions have been met.

Individual evidence

  1. Johannes Siedersleben: Software technology: practical knowledge for software engineers . Hanser Verlag, 2003, ISBN 978-3-446-21843-7 , p. 298 (accessed on August 25, 2011).