Writing tests

A test is a value of type test_case, grouped into suites, and run via run:

open Alcobar

let identity =
  test_case "identity" [ int ] (fun x -> check_eq x x)

let () = run "my_lib" [ ("arith", [ identity ]) ]

This mirrors Alcotest.run: the first argument names the run, the second is a list of (suite_name, tests) pairs.

Running tests

Every alcobar test binary supports three modes, detected automatically:

• Alcotest mode (default): run as a regular test suite. Supports all of Alcotest's CLI flags (filtering, verbose, etc.) as well as:

  • --seed INT64 — fix the PRNG seed
  • --repeat N — iterations per test (default: 5000)
  • --timeout N — per-iteration timeout in seconds (default: 2; also reads ALCOBAR_TIMEOUT)
  • --budget SECONDS — total wall-clock budget per test (default: 2)
  • --infinite / -i — run until a failure is found
  • --alcobar-verbose — log each passing iteration
• Seed corpus mode: pass --gen-corpus DIR. Alcobar runs each test a few times and writes the exact bytes consumed by generators to DIR/seed_NNN. Use these to seed afl-fuzz.
• AFL mode: pass a file as the last argument. Alcobar reads from that file and drives a single test iteration — the form afl-fuzz expects when invoked as afl-fuzz -i in -o out -- ./my_test.exe @@.

To run under AFL you need a compiler with AFL instrumentation enabled (an opam switch tagged +afl). Build a native-code executable, not bytecode.

Relation to Crowbar

Alcobar is a fork of stedolan/crowbar. The generator API (int, map, choose, fix, dynamic_bind, ...) is unchanged. The testing API is different: add_test is replaced by test_case and an explicit run, and the runner embeds an Alcotest suite rather than invoking Cmdliner directly.