win32 iptest: Use subprocess.Popen() instead of os.system().

IPython/testing/iptest.py

@@ -1,578 +1,571 b''
 # -*- coding: utf-8 -*-
 """IPython Test Suite Runner.
 
 This module provides a main entry point to a user script to test IPython
 itself from the command line. There are two ways of running this script:
 
 1. With the syntax `iptest all`.  This runs our entire test suite by
    calling this script (with different arguments) recursively.  This
    causes modules and package to be tested in different processes, using nose
    or trial where appropriate.
 2. With the regular nose syntax, like `iptest -vvs IPython`.  In this form
    the script simply calls nose, but with special command line flags and
    plugins loaded.
 
 """
 
 #-----------------------------------------------------------------------------
 #  Copyright (C) 2009-2011  The IPython Development Team
 #
 #  Distributed under the terms of the BSD License.  The full license is in
 #  the file COPYING, distributed as part of this software.
 #-----------------------------------------------------------------------------
 
 #-----------------------------------------------------------------------------
 # Imports
 #-----------------------------------------------------------------------------
 from __future__ import print_function
 
 # Stdlib
 import glob
 import os
 import os.path as path
 import signal
 import sys
 import subprocess
 import tempfile
 import time
 import warnings
 
 # Note: monkeypatch!
 # We need to monkeypatch a small problem in nose itself first, before importing
 # it for actual use.  This should get into nose upstream, but its release cycle
 # is slow and we need it for our parametric tests to work correctly.
 from IPython.testing import nosepatch
 
 # Monkeypatch extra assert methods into nose.tools if they're not already there.
 # This can be dropped once we no longer test on Python 2.6
 from IPython.testing import nose_assert_methods
 
 # Now, proceed to import nose itself
 import nose.plugins.builtin
 from nose.plugins.xunit import Xunit
 from nose import SkipTest
 from nose.core import TestProgram
 
 # Our own imports
 from IPython.utils import py3compat
 from IPython.utils.importstring import import_item
 from IPython.utils.path import get_ipython_module_path, get_ipython_package_dir
 from IPython.utils.process import find_cmd, pycmd2argv
 from IPython.utils.sysinfo import sys_info
 from IPython.utils.warn import warn
 
 from IPython.testing import globalipapp
 from IPython.testing.plugin.ipdoctest import IPythonDoctest
 from IPython.external.decorators import KnownFailure, knownfailureif
 
 pjoin = path.join
 
 
 #-----------------------------------------------------------------------------
 # Globals
 #-----------------------------------------------------------------------------
 
 
 #-----------------------------------------------------------------------------
 # Warnings control
 #-----------------------------------------------------------------------------
 
 # Twisted generates annoying warnings with Python 2.6, as will do other code
 # that imports 'sets' as of today
 warnings.filterwarnings('ignore', 'the sets module is deprecated',
                         DeprecationWarning )
 
 # This one also comes from Twisted
 warnings.filterwarnings('ignore', 'the sha module is deprecated',
                         DeprecationWarning)
 
 # Wx on Fedora11 spits these out
 warnings.filterwarnings('ignore', 'wxPython/wxWidgets release number mismatch',
                         UserWarning)
 
 # ------------------------------------------------------------------------------
 # Monkeypatch Xunit to count known failures as skipped.
 # ------------------------------------------------------------------------------
 def monkeypatch_xunit():
     try:
         knownfailureif(True)(lambda: None)()
     except Exception as e:
         KnownFailureTest = type(e)
 
     def addError(self, test, err, capt=None):
         if issubclass(err[0], KnownFailureTest):
             err = (SkipTest,) + err[1:]
         return self.orig_addError(test, err, capt)
 
     Xunit.orig_addError = Xunit.addError
     Xunit.addError = addError
 
 #-----------------------------------------------------------------------------
 # Logic for skipping doctests
 #-----------------------------------------------------------------------------
 def extract_version(mod):
     return mod.__version__
 
 def test_for(item, min_version=None, callback=extract_version):
     """Test to see if item is importable, and optionally check against a minimum
     version.
 
     If min_version is given, the default behavior is to check against the
     `__version__` attribute of the item, but specifying `callback` allows you to
     extract the value you are interested in. e.g::
 
         In [1]: import sys
 
         In [2]: from IPython.testing.iptest import test_for
 
         In [3]: test_for('sys', (2,6), callback=lambda sys: sys.version_info)
         Out[3]: True
 
     """
     try:
         check = import_item(item)
     except (ImportError, RuntimeError):
         # GTK reports Runtime error if it can't be initialized even if it's
         # importable.
         return False
     else:
         if min_version:
             if callback:
                 # extra processing step to get version to compare
                 check = callback(check)
 
             return check >= min_version
         else:
             return True
 
 # Global dict where we can store information on what we have and what we don't
 # have available at test run time
 have = {}
 
 have['curses'] = test_for('_curses')
 have['matplotlib'] = test_for('matplotlib')
 have['numpy'] = test_for('numpy')
 have['pexpect'] = test_for('IPython.external.pexpect')
 have['pymongo'] = test_for('pymongo')
 have['pygments'] = test_for('pygments')
 have['qt'] = test_for('IPython.external.qt')
 have['rpy2'] = test_for('rpy2')
 have['sqlite3'] = test_for('sqlite3')
 have['cython'] = test_for('Cython')
 have['oct2py'] = test_for('oct2py')
 have['tornado'] = test_for('tornado.version_info', (2,1,0), callback=None)
 have['wx'] = test_for('wx')
 have['wx.aui'] = test_for('wx.aui')
 
 if os.name == 'nt':
     min_zmq = (2,1,7)
 else:
     min_zmq = (2,1,4)
 
 def version_tuple(mod):
     "turn '2.1.9' into (2,1,9), and '2.1dev' into (2,1,999)"
     # turn 'dev' into 999, because Python3 rejects str-int comparisons
     vs = mod.__version__.replace('dev', '.999')
     tup = tuple([int(v) for v in vs.split('.') ])
     return tup
 
 have['zmq'] = test_for('zmq', min_zmq, version_tuple)
 
 #-----------------------------------------------------------------------------
 # Functions and classes
 #-----------------------------------------------------------------------------
 
 def report():
     """Return a string with a summary report of test-related variables."""
 
     out = [ sys_info(), '\n']
 
     avail = []
     not_avail = []
 
     for k, is_avail in have.items():
         if is_avail:
             avail.append(k)
         else:
             not_avail.append(k)
 
     if avail:
         out.append('\nTools and libraries available at test time:\n')
         avail.sort()
         out.append('   ' + ' '.join(avail)+'\n')
 
     if not_avail:
         out.append('\nTools and libraries NOT available at test time:\n')
         not_avail.sort()
         out.append('   ' + ' '.join(not_avail)+'\n')
 
     return ''.join(out)
 
 
 def make_exclude():
     """Make patterns of modules and packages to exclude from testing.
 
     For the IPythonDoctest plugin, we need to exclude certain patterns that
     cause testing problems.  We should strive to minimize the number of
     skipped modules, since this means untested code.
 
     These modules and packages will NOT get scanned by nose at all for tests.
     """
     # Simple utility to make IPython paths more readably, we need a lot of
     # these below
     ipjoin = lambda *paths: pjoin('IPython', *paths)
 
     exclusions = [ipjoin('external'),
                   ipjoin('quarantine'),
                   ipjoin('deathrow'),
                   # This guy is probably attic material
                   ipjoin('testing', 'mkdoctests'),
                   # Testing inputhook will need a lot of thought, to figure out
                   # how to have tests that don't lock up with the gui event
                   # loops in the picture
                   ipjoin('lib', 'inputhook'),
                   # Config files aren't really importable stand-alone
                   ipjoin('config', 'profile'),
                   # The notebook 'static' directory contains JS, css and other
                   # files for web serving.  Occasionally projects may put a .py
                   # file in there (MathJax ships a conf.py), so we might as
                   # well play it safe and skip the whole thing.
                   ipjoin('frontend', 'html', 'notebook', 'static')
                   ]
     if not have['sqlite3']:
         exclusions.append(ipjoin('core', 'tests', 'test_history'))
         exclusions.append(ipjoin('core', 'history'))
     if not have['wx']:
         exclusions.append(ipjoin('lib', 'inputhookwx'))
     
     # FIXME: temporarily disable autoreload tests, as they can produce
     # spurious failures in subsequent tests (cythonmagic).
     exclusions.append(ipjoin('extensions', 'autoreload'))
     exclusions.append(ipjoin('extensions', 'tests', 'test_autoreload'))
 
     # We do this unconditionally, so that the test suite doesn't import
     # gtk, changing the default encoding and masking some unicode bugs.
     exclusions.append(ipjoin('lib', 'inputhookgtk'))
     exclusions.append(ipjoin('zmq', 'gui', 'gtkembed'))
 
     # These have to be skipped on win32 because the use echo, rm, cd, etc.
     # See ticket https://github.com/ipython/ipython/issues/87
     if sys.platform == 'win32':
         exclusions.append(ipjoin('testing', 'plugin', 'test_exampleip'))
         exclusions.append(ipjoin('testing', 'plugin', 'dtexample'))
 
     if not have['pexpect']:
         exclusions.extend([ipjoin('lib', 'irunner'),
                            ipjoin('lib', 'tests', 'test_irunner'),
                            ipjoin('frontend', 'terminal', 'console'),
                            ])
 
     if not have['zmq']:
         exclusions.append(ipjoin('zmq'))
         exclusions.append(ipjoin('frontend', 'qt'))
         exclusions.append(ipjoin('frontend', 'html'))
         exclusions.append(ipjoin('frontend', 'consoleapp.py'))
         exclusions.append(ipjoin('frontend', 'terminal', 'console'))
         exclusions.append(ipjoin('parallel'))
     elif not have['qt'] or not have['pygments']:
         exclusions.append(ipjoin('frontend', 'qt'))
 
     if not have['pymongo']:
         exclusions.append(ipjoin('parallel', 'controller', 'mongodb'))
         exclusions.append(ipjoin('parallel', 'tests', 'test_mongodb'))
 
     if not have['matplotlib']:
         exclusions.extend([ipjoin('core', 'pylabtools'),
                            ipjoin('core', 'tests', 'test_pylabtools'),
                            ipjoin('zmq', 'pylab'),
         ])
 
     if not have['cython']:
         exclusions.extend([ipjoin('extensions', 'cythonmagic')])
         exclusions.extend([ipjoin('extensions', 'tests', 'test_cythonmagic')])
 
     if not have['oct2py']:
         exclusions.extend([ipjoin('extensions', 'octavemagic')])
         exclusions.extend([ipjoin('extensions', 'tests', 'test_octavemagic')])
 
     if not have['tornado']:
         exclusions.append(ipjoin('frontend', 'html'))
 
     if not have['rpy2'] or not have['numpy']:
         exclusions.append(ipjoin('extensions', 'rmagic'))
         exclusions.append(ipjoin('extensions', 'tests', 'test_rmagic'))
 
     # This is needed for the reg-exp to match on win32 in the ipdoctest plugin.
     if sys.platform == 'win32':
         exclusions = [s.replace('\\','\\\\') for s in exclusions]
     
     # check for any exclusions that don't seem to exist:
     parent, _ = os.path.split(get_ipython_package_dir())
     for exclusion in exclusions:
         if exclusion.endswith(('deathrow', 'quarantine')):
             # ignore deathrow/quarantine, which exist in dev, but not install
             continue
         fullpath = pjoin(parent, exclusion)
         if not os.path.exists(fullpath) and not glob.glob(fullpath + '.*'):
             warn("Excluding nonexistent file: %r\n" % exclusion)
 
     return exclusions
 
 
 class IPTester(object):
     """Call that calls iptest or trial in a subprocess.
     """
     #: string, name of test runner that will be called
     runner = None
     #: list, parameters for test runner
     params = None
     #: list, arguments of system call to be made to call test runner
     call_args = None
     #: list, process ids of subprocesses we start (for cleanup)
     pids = None
     #: str, coverage xml output file
     coverage_xml = None
 
     def __init__(self, runner='iptest', params=None):
         """Create new test runner."""
         p = os.path
         if runner == 'iptest':
             iptest_app = get_ipython_module_path('IPython.testing.iptest')
             self.runner = pycmd2argv(iptest_app) + sys.argv[1:]
         else:
             raise Exception('Not a valid test runner: %s' % repr(runner))
         if params is None:
             params = []
         if isinstance(params, str):
             params = [params]
         self.params = params
 
         # Assemble call
         self.call_args = self.runner+self.params
         
         # Find the section we're testing (IPython.foo)
         for sect in self.params:
             if sect.startswith('IPython'): break
         else:
             raise ValueError("Section not found", self.params)
         
         if '--with-xunit' in self.call_args:
             
             self.call_args.append('--xunit-file')
             # FIXME: when Windows uses subprocess.call, these extra quotes are unnecessary:
             xunit_file = path.abspath(sect+'.xunit.xml')
             if sys.platform == 'win32':
                 xunit_file = '"%s"' % xunit_file
             self.call_args.append(xunit_file)
         
         if '--with-xml-coverage' in self.call_args:
             self.coverage_xml = path.abspath(sect+".coverage.xml")
             self.call_args.remove('--with-xml-coverage')
             self.call_args = ["coverage", "run", "--source="+sect] + self.call_args[1:]
 
         # Store pids of anything we start to clean up on deletion, if possible
         # (on posix only, since win32 has no os.kill)
         self.pids = []
 
-    if sys.platform == 'win32':
-        def _run_cmd(self):
-            # On Windows, use os.system instead of subprocess.call, because I
-            # was having problems with subprocess and I just don't know enough
-            # about win32 to debug this reliably.  Os.system may be the 'old
-            # fashioned' way to do it, but it works just fine.  If someone
-            # later can clean this up that's fine, as long as the tests run
-            # reliably in win32.
-            # What types of problems are you having. They may be related to
-            # running Python in unboffered mode. BG.
-            for ndx, arg in enumerate(self.call_args):
-                # Enclose in quotes if necessary and legal
-                if ' ' in arg and os.path.isfile(arg) and arg[0] != '"':
-                    self.call_args[ndx] = '"%s"' % arg
-            call_args = [py3compat.cast_unicode(x) for x in self.call_args]
-            cmd = py3compat.unicode_to_str(u' '.join(call_args))
-            return os.system(cmd)
-    else:
-        def _run_cmd(self):
-            # print >> sys.stderr, '*** CMD:', ' '.join(self.call_args) # dbg
-            subp = subprocess.Popen(self.call_args)
-            self.pids.append(subp.pid)
-            # If this fails, the pid will be left in self.pids and cleaned up
-            # later, but if the wait call succeeds, then we can clear the
-            # stored pid.
-            retcode = subp.wait()
-            self.pids.pop()
-            return retcode
+    def _run_cmd(self):
+        # print >> sys.stderr, '*** CMD:', ' '.join(self.call_args) # dbg
+        subp = subprocess.Popen(self.call_args)
+        self.pids.append(subp.pid)
+        # If this fails, the pid will be left in self.pids and cleaned up
+        # later, but if the wait call succeeds, then we can clear the
+        # stored pid.
+        retcode = subp.wait()
+        self.pids.pop()
+        return retcode
 
     def run(self):
         """Run the stored commands"""
         try:
             retcode = self._run_cmd()
         except:
             import traceback
             traceback.print_exc()
             return 1  # signal failure
         
         if self.coverage_xml:
             subprocess.call(["coverage", "xml", "-o", self.coverage_xml])
         return retcode
 
+    @staticmethod
+    def _kill(pid):
+        """Attempt to kill the process on a best effort basis."""
+        if hasattr(os, 'kill'):
+            os.kill(pid, signal.SIGKILL)
+        elif sys.platform == 'win32':
+            # Python 2.6 on Windows doesn't have os.kill.
+            from ctypes import windll
+            PROCESS_TERMINATE = 1
+            handle = windll.kernel32.OpenProcess(PROCESS_TERMINATE, False, pid)
+            windll.kernel32.TerminateProcess(handle, -1)
+            windll.kernel32.CloseHandle(handle)
+        else:
+            raise NotImplementedError
+
     def __del__(self):
         """Cleanup on exit by killing any leftover processes."""
-
-        if not hasattr(os, 'kill'):
-            return
-
         for pid in self.pids:
             try:
                 print('Cleaning stale PID:', pid)
-                os.kill(pid, signal.SIGKILL)
-            except OSError:
+                self._kill(pid)
+            except (OSError, NotImplementedError):
                 # This is just a best effort, if we fail or the process was
                 # really gone, ignore it.
                 pass
 
 
 def make_runners():
     """Define the top-level packages that need to be tested.
     """
 
     # Packages to be tested via nose, that only depend on the stdlib
     nose_pkg_names = ['config', 'core', 'extensions', 'frontend', 'lib',
                      'testing', 'utils', 'nbformat' ]
 
     if have['zmq']:
         nose_pkg_names.append('zmq')
         nose_pkg_names.append('parallel')
 
     # For debugging this code, only load quick stuff
     #nose_pkg_names = ['core', 'extensions']  # dbg
 
     # Make fully qualified package names prepending 'IPython.' to our name lists
     nose_packages = ['IPython.%s' % m for m in nose_pkg_names ]
 
     # Make runners
     runners = [ (v, IPTester('iptest', params=v)) for v in nose_packages ]
 
     return runners
 
 
 def run_iptest():
     """Run the IPython test suite using nose.
 
     This function is called when this script is **not** called with the form
     `iptest all`.  It simply calls nose with appropriate command line flags
     and accepts all of the standard nose arguments.
     """
     # Apply our monkeypatch to Xunit
     if '--with-xunit' in sys.argv and not hasattr(Xunit, 'orig_addError'):
         monkeypatch_xunit()
 
     warnings.filterwarnings('ignore',
         'This will be removed soon.  Use IPython.testing.util instead')
 
     argv = sys.argv + [ '--detailed-errors',  # extra info in tracebacks
 
                         '--with-ipdoctest',
                         '--ipdoctest-tests','--ipdoctest-extension=txt',
 
                         # We add --exe because of setuptools' imbecility (it
                         # blindly does chmod +x on ALL files).  Nose does the
                         # right thing and it tries to avoid executables,
                         # setuptools unfortunately forces our hand here.  This
                         # has been discussed on the distutils list and the
                         # setuptools devs refuse to fix this problem!
                         '--exe',
                         ]
 
     if nose.__version__ >= '0.11':
         # I don't fully understand why we need this one, but depending on what
         # directory the test suite is run from, if we don't give it, 0 tests
         # get run.  Specifically, if the test suite is run from the source dir
         # with an argument (like 'iptest.py IPython.core', 0 tests are run,
         # even if the same call done in this directory works fine).  It appears
         # that if the requested package is in the current dir, nose bails early
         # by default.  Since it's otherwise harmless, leave it in by default
         # for nose >= 0.11, though unfortunately nose 0.10 doesn't support it.
         argv.append('--traverse-namespace')
 
     # use our plugin for doctesting.  It will remove the standard doctest plugin
     # if it finds it enabled
     plugins = [IPythonDoctest(make_exclude()), KnownFailure()]
     # We need a global ipython running in this process
     globalipapp.start_ipython()
     # Now nose can run
     TestProgram(argv=argv, addplugins=plugins)
 
 
 def run_iptestall():
     """Run the entire IPython test suite by calling nose and trial.
 
     This function constructs :class:`IPTester` instances for all IPython
     modules and package and then runs each of them.  This causes the modules
     and packages of IPython to be tested each in their own subprocess using
     nose.
     """
 
     runners = make_runners()
 
     # Run the test runners in a temporary dir so we can nuke it when finished
     # to clean up any junk files left over by accident.  This also makes it
     # robust against being run in non-writeable directories by mistake, as the
     # temp dir will always be user-writeable.
     curdir = os.getcwdu()
     testdir = tempfile.gettempdir()
     os.chdir(testdir)
 
     # Run all test runners, tracking execution time
     failed = []
     t_start = time.time()
     try:
         for (name, runner) in runners:
             print('*'*70)
             print('IPython test group:',name)
             res = runner.run()
             if res:
                 failed.append( (name, runner) )
     finally:
         os.chdir(curdir)
     t_end = time.time()
     t_tests = t_end - t_start
     nrunners = len(runners)
     nfail = len(failed)
     # summarize results
     print()
     print('*'*70)
     print('Test suite completed for system with the following information:')
     print(report())
     print('Ran %s test groups in %.3fs' % (nrunners, t_tests))
     print()
     print('Status:')
     if not failed:
         print('OK')
     else:
         # If anything went wrong, point out what command to rerun manually to
         # see the actual errors and individual summary
         print('ERROR - %s out of %s test groups failed.' % (nfail, nrunners))
         for name, failed_runner in failed:
             print('-'*40)
             print('Runner failed:',name)
             print('You may wish to rerun this one individually, with:')
             failed_call_args = [py3compat.cast_unicode(x) for x in failed_runner.call_args]
             print(u' '.join(failed_call_args))
             print()
         # Ensure that our exit code indicates failure
         sys.exit(1)
 
 
 def main():
     for arg in sys.argv[1:]:
         if arg.startswith('IPython'):
             # This is in-process
             run_iptest()
     else:
         # This starts subprocesses
         run_iptestall()
 
 
 if __name__ == '__main__':
     main()