Cram: It's test time

Cram is a functional testing framework for command line applications. Cram tests look like snippets of interactive shell sessions. Cram runs each command and compares the command output in the test with the command's actual output.

Here's a snippet from Cram's own test suite:

The $PYTHON environment variable should be set when running this test
from Python.

  $ [ -n "$PYTHON" ] || PYTHON="`which python`"
  $ if [ -n "$COVERAGE" ]; then
  >   coverage erase
  >   alias cram="`which coverage` run --branch -a $TESTDIR/../scripts/cram"
  > else
  >   alias cram="$PYTHON $TESTDIR/../scripts/cram"
  > fi
  $ command -v md5 > /dev/null || alias md5=md5sum


  $ cram -h
  [Uu]sage: cram \[OPTIONS\] TESTS\.\.\. (re)

  [Oo]ptions: (re)
    -h, --help          show this help message and exit
    -V, --version       show version information and exit
    -q, --quiet         don't print diffs
    -v, --verbose       show filenames and test status
    -i, --interactive   interactively merge changed test output
    -d, --debug         write script output directly to the terminal
    -y, --yes           answer yes to all questions
    -n, --no            answer no to all questions
    -E, --preserve-env  don't reset common environment variables
    --keep-tmpdir       keep temporary directories
    --shell=PATH        shell to use for running tests (default: /bin/sh)
    --shell-opts=OPTS   arguments to invoke shell with
    --indent=NUM        number of spaces to use for indentation (default: 2)
    --xunit-file=PATH   path to write xUnit XML output

The format in a nutshell:



Install Cram using make:

$ wget
$ tar zxvf cram-0.7.tar.gz
$ cd cram-0.7
$ make install


Cram will print a dot for each passing test. If a test fails, a unified context diff is printed showing the test's expected output and the actual output. Skipped tests (empty tests and tests that exit with return code 80) are marked with s instead of a dot.

For example, if we run Cram on its own example tests:

--- examples/fail.t
+++ examples/fail.t.err
@@ -3,21 +3,22 @@
   $ echo 1
   $ echo 1
-  2
+  1
   $ echo 1

 Invalid regex:

   $ echo 1
-  +++ (re)
+  1

 Offset regular expression:

   $ printf 'foo\nbar\nbaz\n\n1\nA\n@\n'
+  bar

   \d (re)
   [A-Z] (re)
-  #
+  @
# Ran 6 tests, 2 skipped, 1 failed.

Cram will also write the test with its actual output to examples/fail.t.err, allowing you to use other diff tools. This file is automatically removed the next time the test passes.

When you're first writing a test, you might just write the commands and run the test to see what happens. If you run Cram with -i or --interactive, you'll be prompted to merge the actual output back into the test. This makes it easy to quickly prototype new tests.

You can specify a default set of options by creating a .cramrc file. For example:

verbose = True
indent = 4

Is the same as invoking Cram with --verbose and --indent=4.

To change what configuration file Cram loads, you can set the CRAMRC environment variable. You can also specify command line options in the CRAM environment variable.

Note that the following environment variables are reset before tests are run:

Cram also provides the following environment variables to tests:

Also note that care should be taken with commands that close the test shell's stdin. For example, if you're trying to invoke ssh in a test, try adding the -n option to prevent it from closing stdin. Similarly, if you invoke a daemon process that inherits stdout and fails to close it, it may cause Cram to hang while waiting for the test shell's stdout to be fully closed.


Download the official development repository using Mercurial:

hg clone

Or Git:

git clone

Test Cram using Cram:

pip install -r requirements.txt
make test

Visit Bitbucket or GitHub if you'd like to fork the project, watch for new changes, or report issues.