commit d46ecdfafd6883ac39bf24841af988d45d2fd64e Author: Damian Johnson atagar@torproject.org Date: Thu Apr 11 09:28:18 2013 -0700

Dropping the argument.* config options

Global state is icky, and in essence all stem.util.conf usage is global state. As mentioned in earlier commits the argument.* options were never used. Happily they were only consumed within run_tests.py so narrowing their scope is trivial. --- run_tests.py | 81 +++++++++++++++++++++++++++-------------------------- test/settings.cfg | 20 +------------ 2 files changed, 43 insertions(+), 58 deletions(-)

diff --git a/run_tests.py b/run_tests.py index 45c3e00..ca43539 100755 --- a/run_tests.py +++ b/run_tests.py @@ -30,14 +30,6 @@ OPT_EXPANDED = ["all", "unit", "integ", "style", "python3", "clean", "targets=", DIVIDER = "=" * 70

CONFIG = stem.util.conf.config_dict("test", { - "argument.unit": False, - "argument.integ": False, - "argument.style": False, - "argument.python3": False, - "argument.python3_clean": False, - "argument.test": "", - "argument.log": None, - "argument.tor": "tor", "msg.help": "", "target.config": {}, "target.description": {}, @@ -96,7 +88,7 @@ def _clean_orphaned_pyc(): os.remove(pyc_file)

-def _python3_setup(python3_destination): +def _python3_setup(python3_destination, clean): # Python 2.7.3 added some nice capabilities to 2to3, like '--output-dir'... # # http://docs.python.org/2/library/2to3.html @@ -106,7 +98,7 @@ def _python3_setup(python3_destination):

test.output.print_divider("EXPORTING TO PYTHON 3", True)

- if CONFIG["argument.python3_clean"]: + if clean: shutil.rmtree(python3_destination, ignore_errors = True)

if os.path.exists(python3_destination): @@ -144,7 +136,7 @@ def _python3_setup(python3_destination): return True

-def _print_style_issues(): +def _print_style_issues(run_unit, run_integ, run_style): base_path = os.path.sep.join(__file__.split(os.path.sep)[:-1]).lstrip("./") style_issues = test.static_checks.get_issues(os.path.join(base_path, "stem")) style_issues.update(test.static_checks.get_issues(os.path.join(base_path, "test"))) @@ -154,7 +146,7 @@ def _print_style_issues(): # available then use it. Its static checks are pretty quick so there's not # much overhead in including it with all tests.

- if CONFIG["argument.unit"] or CONFIG["argument.integ"]: + if run_unit or run_integ: if system.is_available("pyflakes"): style_issues.update(test.static_checks.pyflakes_issues(os.path.join(base_path, "stem"))) style_issues.update(test.static_checks.pyflakes_issues(os.path.join(base_path, "test"))) @@ -162,7 +154,7 @@ def _print_style_issues(): else: test.output.print_line("Static error checking requires pyflakes. Please install it from ...\n http://pypi.python.org/pypi/pyflakes%5Cn", *ERROR_ATTR)

- if CONFIG["argument.style"]: + if run_style: if system.is_available("pep8"): style_issues.update(test.static_checks.pep8_issues(os.path.join(base_path, "stem"))) style_issues.update(test.static_checks.pep8_issues(os.path.join(base_path, "test"))) @@ -212,21 +204,31 @@ if __name__ == '__main__': print "%s (for usage provide --help)" % exc sys.exit(1)

+ run_unit = False + run_integ = False + run_style = False + run_python3 = False + run_python3_clean = False + + test_prefix = None + logging_runlevel = None + tor_path = "tor" + for opt, arg in opts: if opt in ("-a", "--all"): - test_config.set("argument.unit", "true") - test_config.set("argument.integ", "true") - test_config.set("argument.style", "true") + run_unit = True + run_integ = True + run_style = True elif opt in ("-u", "--unit"): - test_config.set("argument.unit", "true") + run_unit = True elif opt in ("-i", "--integ"): - test_config.set("argument.integ", "true") + run_integ = True elif opt in ("-s", "--style"): - test_config.set("argument.style", "true") + run_style = True elif opt == "--python3": - test_config.set("argument.python3", "true") + run_python3 = True elif opt == "--clean": - test_config.set("argument.python3_clean", "true") + run_python3_clean = True elif opt in ("-t", "--targets"): integ_targets = arg.split(",")

@@ -245,11 +247,11 @@ if __name__ == '__main__': if target_config: test_config.set(target_config, "true") elif opt in ("-l", "--test"): - test_config.set("argument.test", arg) + test_prefix = arg elif opt in ("-l", "--log"): - test_config.set("argument.log", arg.upper()) + logging_runlevel = arg.upper() elif opt in ("--tor"): - test_config.set("argument.tor", arg) + tor_path = arg elif opt in ("-h", "--help"): # Prints usage information and quits. This includes a listing of the # valid integration targets. @@ -269,35 +271,34 @@ if __name__ == '__main__':

# basic validation on user input

- log_config = CONFIG["argument.log"] - if log_config and not log_config in log.LOG_VALUES: - print "'%s' isn't a logging runlevel, use one of the following instead:" % log_config + if logging_runlevel and not logging_runlevel in log.LOG_VALUES: + print "'%s' isn't a logging runlevel, use one of the following instead:" % logging_runlevel print " TRACE, DEBUG, INFO, NOTICE, WARN, ERROR" sys.exit(1)

# check that we have 2to3 and python3 available in our PATH - if CONFIG["argument.python3"]: + if run_python3: for required_cmd in ("2to3", "python3"): if not system.is_available(required_cmd): test.output.print_line("Unable to test python 3 because %s isn't in your path" % required_cmd, *test.runner.ERROR_ATTR) sys.exit(1)

- if CONFIG["argument.python3"] and sys.version_info[0] != 3: + if run_python3 and sys.version_info[0] != 3: python3_destination = os.path.join(CONFIG["integ.test_directory"], "python3")

- if _python3_setup(python3_destination): + if _python3_setup(python3_destination, run_python3_clean): python3_runner = os.path.join(python3_destination, "run_tests.py") exit_status = os.system("python3 %s %s" % (python3_runner, " ".join(sys.argv[1:]))) sys.exit(exit_status) else: sys.exit(1) # failed to do python3 setup

- if not CONFIG["argument.unit"] and not CONFIG["argument.integ"] and not CONFIG["argument.style"]: + if not run_unit and not run_integ and not run_style: test.output.print_line("Nothing to run (for usage provide --help)\n") sys.exit()

# if we have verbose logging then provide the testing config - our_level = stem.util.log.logging_level(CONFIG["argument.log"]) + our_level = stem.util.log.logging_level(logging_runlevel) info_level = stem.util.log.logging_level(stem.util.log.INFO)

if our_level <= info_level: @@ -312,7 +313,7 @@ if __name__ == '__main__': )

stem_logger = log.get_logger() - logging_buffer = log.LogBuffer(CONFIG["argument.log"]) + logging_buffer = log.LogBuffer(logging_runlevel) stem_logger.addHandler(logging_buffer)

test.output.print_divider("INITIALISING", True) @@ -322,11 +323,11 @@ if __name__ == '__main__':

print

- if CONFIG["argument.unit"]: + if run_unit: test.output.print_divider("UNIT TESTS", True) error_tracker.set_category("UNIT TEST")

- for test_class in test.runner.get_unit_tests(CONFIG["argument.test"]): + for test_class in test.runner.get_unit_tests(test_prefix): test.output.print_divider(test_class.__module__) suite = unittest.TestLoader().loadTestsFromTestCase(test_class) test_results = StringIO.StringIO() @@ -341,7 +342,7 @@ if __name__ == '__main__':

print

- if CONFIG["argument.integ"]: + if run_integ: test.output.print_divider("INTEGRATION TESTS", True) integ_runner = test.runner.get_runner()

@@ -375,7 +376,7 @@ if __name__ == '__main__': if target_prereq: # lazy loaded to skip system call if we don't have any prereqs if not our_version: - our_version = stem.version.get_system_tor_version(CONFIG["argument.tor"]) + our_version = stem.version.get_system_tor_version(tor_path)

if our_version < stem.version.Requirement[target_prereq]: skip_targets.append(target) @@ -401,12 +402,12 @@ if __name__ == '__main__': test.output.print_line("'%s' isn't a test.runner.Torrc enumeration" % opt) sys.exit(1)

- integ_runner.start(CONFIG["argument.tor"], extra_torrc_opts = torrc_opts) + integ_runner.start(tor_path, extra_torrc_opts = torrc_opts)

test.output.print_line("Running tests...", term.Color.BLUE, term.Attr.BOLD) print

- for test_class in test.runner.get_integ_tests(CONFIG["argument.test"]): + for test_class in test.runner.get_integ_tests(test_prefix): test.output.print_divider(test_class.__module__) suite = unittest.TestLoader().loadTestsFromTestCase(test_class) test_results = StringIO.StringIO() @@ -452,7 +453,7 @@ if __name__ == '__main__': # TODO: note unused config options afterward?

if not stem.prereq.is_python_3(): - _print_style_issues() + _print_style_issues(run_unit, run_integ, run_style)

runtime = time.time() - start_time

diff --git a/test/settings.cfg b/test/settings.cfg index 372ce33..81bb567 100644 --- a/test/settings.cfg +++ b/test/settings.cfg @@ -1,16 +1,7 @@ # Testing Configuration # -# The following are configurations used by stem's testing framework. You can -# overwrite any of these my making your own configuratino file and using the -# '--config' option of run_tests.py... -# -# run_tests.py --config my_testrc --unit -# -# The commonly customized options are... -# -# argument.* -# Default test runner arguments. You can set these to always run with -# particular arguments. +# The following contains customizable configuration information for stem's +# testing framework. # # integ.test_directory # Path used for our data directory and any temporary test resources. Relative @@ -36,13 +27,6 @@ # authentication configurations. If the 'all' option is set then the other # flags are ignored.

-argument.unit false -argument.integ false -argument.style false -argument.test -argument.log -argument.tor tor - integ.test_directory ./test/data integ.log ./test/data/log