How deeply does The Art of Unit Testing 3e cover constrained test order?

Explains the dependency, demonstrates its failure modes, and shows refactoring with reset and local setup.

Constrained test order

As explained in The Art of Unit Testing 3e

Constrained test order is an antipattern in which a test depends on another test having executed or not executed first.

Constrained test order occurs when one test depends on another test having run—or not run—because both rely on shared state. A test runner may change the order, exposing failures that are difficult to diagnose. section 8.1.3

A shared cache creates the dependency

One test adds user "a" to a shared UserCache; a later test expects login for that user to succeed. The first test fails if the cache already contains the user, while the third fails if the second has not run. Running only the third test also reveals the hidden setup dependency.

Remove order dependence by making each test establish its own state:

Treat tests as independently runnable rather than relying on runner order. section 8.1.3

How The Art of Unit Testing 3e covers it

liveBook manning.com

Read The Art of Unit Testing 3e on liveBook Get the book at manning.com

Related concepts

mutable shared stateuses · Mutable shared state is data or computational state that multiple components or execution contexts can access and whose value can change over time.
antipatternkind of · An antipattern is a recurring approach that appears reasonable but tends to produce undesirable or counterproductive results in a given context.
test isolationaddressed by · Test isolation is the property or practice of executing tests independently so their outcomes are not affected by other tests, shared state, dependencies, or external conditions.

This explainer was generated by Manning's knowledge graph from the text of The Art of Unit Testing 3e. The section links point to the parts of the book it draws on.