Skip to content

Design predicates

A software-integration testing model that classifies inter-unit call relationships and assigns complexity increments used to bound independent integration paths.

Version
v2 · 2026-09-19 · History
Domain-specific #
4123
Domain group
Applied Sciences & Engineering
Origin domain
Computer Science & Software Engineering
Subdomain
Software Testing → Computer Science & Software Engineering

Core Idea

McCabe’s unconditional, conditional, mutually exclusive and iterative call predicates describe control coupling between units; their scores combine only under the method’s feasibility assumptions. Calls between units are typed by their control condition, each predicate contributes a fixed integration complexity and the resulting graph measure guides a set of tests intended to exercise independent call behavior. The abstraction is therefore identified by a declared carrier, a transformation or constraint over that carrier, and an invariant that tells an analyst whether the named structure is genuinely present.

How would you explain it like I'm…

Four Ways to Ask for Help

In a computer program, one part can ask another part for help. Sometimes it always asks, sometimes only if something is true, sometimes it picks one helper or another, and sometimes it asks again and again. Design predicates are names for these kinds of asking, used to figure out how many tests you need to try them all.

Kinds of Calls Between Program Parts

Programs are split into units, like functions, that call each other. Design predicates, from McCabe's work on testing, sort those calls into four kinds: calls that always happen (unconditional), calls that happen only if a condition is true (conditional), calls where exactly one of several options is chosen (mutually exclusive), and calls inside a loop that may repeat (iterative). Each kind adds a set amount to a score for how complicated the connections between units are. That score helps decide how many tests are needed to try out the different ways units call each other.

Call-Type Integration Complexity

Design predicates are McCabe's classification of calls between software units by the control condition under which they occur: unconditional, conditional, mutually exclusive conditional, and iterative. Each type contributes a fixed amount of integration complexity, a measure of control coupling between units. Adding these up over a module's call structure yields a graph-based complexity number that guides how many integration tests to design so that independent call behaviors are exercised. The approach only works under its assumptions: combining the scores presumes the call combinations are actually feasible, so infeasible combinations can make the count misleading. Using the method properly requires stating the units, the call graph, how predicate types are recognized, and what coverage the tests achieved.

 

Design predicates are McCabe's typology of inter-unit calls used to quantify control coupling in a software design: unconditional, conditional, mutually exclusive, and iterative call predicates, distinguished by the control condition (none, a branch, a multi-way choice, or a loop) under which one unit invokes another. Each predicate type contributes a fixed integration complexity, and combining these contributions over the call graph yields a graph-based measure that bounds and guides a set of integration tests intended to exercise independent call behavior. Applying it requires explicit software units and call graph, recognition rules for each predicate type, the underlying branch and loop conditions, per-predicate complexity values, and the combination rule. The combined scores are valid only under the method's feasibility assumptions: infeasible combinations of calls, for instance where conditions are correlated, can make the count overstate the tests actually needed or achievable. Coverage evidence should confirm that the derived test paths were exercised. It is a design-level integration measure rather than a general account of software testing.

Scope of Application

Design predicates belongs to software testing and is useful where the analyst can specify the typed software testing carrier, including objects, relations, parameters, conventions, evidence, boundaries, and comparison targets, then evaluate the software units and call graph, predicate types and recognition rules, branch and loop conditions, per-predicate complexity values, combination rule, infeasible combinations, test-path bound and coverage evidence are explicit. The scope is broad within that domain but bounded by the need for the software units and call graph, predicate types and recognition rules, branch and loop conditions, per-predicate complexity values, combination rule, infeasible combinations, test-path bound and coverage evidence are explicit. Descriptive software-testing method only; safety-critical assurance requires governing standards and independent validation.

Clarity

The abstraction clarifies a crowded vocabulary by making the software units and call graph, predicate types and recognition rules, branch and loop conditions, per-predicate complexity values, combination rule, infeasible combinations, test-path bound and coverage evidence are explicit the center of the account. A claim should name the carrier, the governing operation or relation, the applicable assumptions, and the recognition test.

Manages Complexity

Without the abstraction, an analyst must reason directly over many local details: the carrier roles, admissibility assumptions, competing conventions, derived invariants, boundary cases, and proof or validation obligations specific to Design predicates. Design predicates compresses them into the roles in the structural signature. That compression permits comparison across instances without erasing the variables that determine validity. It also exposes which details may be varied safely and which are constitutive.

Abstract Reasoning

  1. Identify the carrier. State what the elements, states, objects, or observations are: the typed software testing carrier, including objects, relations, parameters, conventions, evidence, boundaries, and comparison targets. Reject examples whose alleged carrier belongs to a different problem.
  2. Lock the constitutive rule. Express the software units and call graph, predicate types and recognition rules, branch and loop conditions, per-predicate complexity values, combination rule, infeasible combinations, test-path bound and coverage evidence are explicit independently of one notation or implementation.

Knowledge Transfer

Knowledge transfers strongly among subfields of software testing because they reuse the typed software testing carrier, including objects, relations, parameters, conventions, evidence, boundaries, and comparison targets, Calls between units are typed by their control condition, each predicate contributes a fixed integration complexity and the resulting graph measure guides a set of tests intended to exercise independent call behavior., and type the carrier, state every parameter and convention in the definition, test that the software units and call graph, predicate types and recognition rules, branch and loop conditions, per-predicate complexity values, combination rule, infeasible combinations, test-path bound and coverage evidence are explicit, compare the nearest accepted identity, and report counterexamples, uncertainty, and limiting cases.

Relationships to Other Abstractions

Local relationship map for Design predicatesParents appear above the current abstraction, mutual partners to the right, and children below. Node labels state whether each abstraction is prime or domain-specific; colors identify relation types.Design predicatesDOMAINPrime abstraction: Classification — is a kind ofClassificationPRIME

Current abstraction Design predicates Domain-specific

Parents (1) — outgoing typed relations

  • Design predicates is a kind of Classification Prime

    It remains its own entry because its identity is fixed by the software units and call graph, predicate types and recognition rules, branch and loop conditions, per-predicate complexity values, combination rule, infeasible combinations, test-path bound and coverage evidence.

Hierarchy path (1) — routes to 1 parentless root

Neighborhood in Abstraction Space

Design predicates sits in a crowded region of the domain-specific corpus (15th percentile for distinctiveness): several abstractions share nearly its structure, so a description that fits it tends to fit its neighbors too.

Family — Software Architecture & Modeling Methods (23 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-10-25