1 Let's pick test/settings/TestSettings.py as our example. First, notice the file 2 name "TestSettings.py", the Test*.py pattern is the default mechanism that the 3 test driver uses for discovery of tests. As to TestSettings.py, it defines a 4 class: 5 6 class SettingsCommandTestCase(TestBase): 7 8 derived from TestBase, which is defined in test/lldbtest.py and is itself 9 derived from Python's unittest framework's TestCase class. See also 10 http://docs.python.org/library/unittest.html for more details. 11 12 To just run the TestSettings.py test, chdir to the lldb test directory, and then 13 type the following command: 14 15 /Volumes/data/lldb/svn/trunk/test $ ./dotest.py settings 16 ---------------------------------------------------------------------- 17 Collected 6 tests 18 19 ---------------------------------------------------------------------- 20 Ran 6 tests in 8.699s 21 22 OK (expected failures=1) 23 /Volumes/data/lldb/svn/trunk/test $ 24 25 Pass '-v' option to the test driver to also output verbose descriptions of the 26 individual test cases and their test status: 27 28 /Volumes/data/lldb/svn/trunk/test $ ./dotest.py -v settings 29 ---------------------------------------------------------------------- 30 Collected 6 tests 31 32 test_set_auto_confirm (TestSettings.SettingsCommandTestCase) 33 Test that after 'set auto-confirm true', manual confirmation should not kick in. ... ok 34 test_set_output_path (TestSettings.SettingsCommandTestCase) 35 Test that setting target.process.output-path for the launched process works. ... expected failure 36 test_set_prompt (TestSettings.SettingsCommandTestCase) 37 Test that 'set prompt' actually changes the prompt. ... ok 38 test_set_term_width (TestSettings.SettingsCommandTestCase) 39 Test that 'set term-width' actually changes the term-width. ... ok 40 test_with_dsym (TestSettings.SettingsCommandTestCase) 41 Test that run-args and env-vars are passed to the launched process. ... ok 42 test_with_dwarf (TestSettings.SettingsCommandTestCase) 43 Test that run-args and env-vars are passed to the launched process. ... ok 44 45 ---------------------------------------------------------------------- 46 Ran 6 tests in 5.735s 47 48 OK (expected failures=1) 49 /Volumes/data/lldb/svn/trunk/test $ 50 51 Underneath, the '-v' option passes keyword argument verbosity=2 to the 52 Python's unittest.TextTestRunner (see also 53 http://docs.python.org/library/unittest.html#unittest.TextTestRunner). For very 54 detailed descriptions about what's going on during the test, pass '-t' to the 55 test driver, which asks the test driver to trace the commands executed and to 56 display their output. For brevity, the '-t' output is not included here. 57 58 Notice the 'expected failures=1' message at the end of the run. This is because 59 of a bug currently in lldb such that setting target.process.output-path to 60 'stdout.txt' does not have any effect on the redirection of the standard output 61 of the subsequent launched process. We are using unittest2 (a backport of new 62 unittest features for Python 2.4-2.6) to decorate (mark) the particular test 63 method as such: 64 65 @unittest2.expectedFailure 66 # rdar://problem/8435794 67 # settings set target.process.output-path does not seem to work 68 def test_set_output_path(self): 69 70 See http://pypi.python.org/pypi/unittest2 for more details. 71 72 Now let's look inside the test method: 73 74 def test_set_output_path(self): 75 """Test that setting target.process.output-path for the launched process works.""" 76 self.buildDefault() 77 78 exe = os.path.join(os.getcwd(), "a.out") 79 self.runCmd("file " + exe, CURRENT_EXECUTABLE_SET) 80 81 # Set the output-path and verify it is set. 82 self.runCmd("settings set target.process.output-path 'stdout.txt'") 83 self.expect("settings show target.process.output-path", 84 startstr = "target.process.output-path (string) = 'stdout.txt'") 85 86 self.runCmd("run", RUN_SUCCEEDED) 87 88 # The 'stdout.txt' file should now exist. 89 self.assertTrue(os.path.isfile("stdout.txt"), 90 "'stdout.txt' exists due to target.process.output-path.") 91 92 # Read the output file produced by running the program. 93 with open('stdout.txt', 'r') as f: 94 output = f.read() 95 96 self.expect(output, exe=False, 97 startstr = "This message should go to standard out.") 98 99 The self.buildDefault() statement is used to build a default binary for this 100 test instance. For this particular test case, since we don't really care what 101 debugging format is used, we instruct the build subsystem to build the default 102 binary for us. The base class TestBase has defined three instance methods: 103 104 def buildDefault(self, architecture=None, compiler=None, dictionary=None): 105 """Platform specific way to build the default binaries.""" 106 module = __import__(sys.platform) 107 if not module.buildDefault(self, architecture, compiler, dictionary): 108 raise Exception("Don't know how to build default binary") 109 110 def buildDsym(self, architecture=None, compiler=None, dictionary=None): 111 """Platform specific way to build binaries with dsym info.""" 112 module = __import__(sys.platform) 113 if not module.buildDsym(self, architecture, compiler, dictionary): 114 raise Exception("Don't know how to build binary with dsym") 115 116 def buildDwarf(self, architecture=None, compiler=None, dictionary=None): 117 """Platform specific way to build binaries with dwarf maps.""" 118 module = __import__(sys.platform) 119 if not module.buildDwarf(self, architecture, compiler, dictionary): 120 raise Exception("Don't know how to build binary with dwarf") 121 122 And the test/plugins/darwin.py provides the implementation for all three build 123 methods using the makefile mechanism. We envision that linux plugin can use a 124 similar approach to accomplish the task of building the binaries. 125 126 Mac OS X provides an additional way to manipulate archived DWARF debug symbol 127 files and produces dSYM files. The buildDsym() instance method is used by the 128 test method to build the binary with dsym info. For an example of this, 129 see test/array_types/TestArrayTypes.py: 130 131 @unittest2.skipUnless(sys.platform.startswith("darwin"), "requires Darwin") 132 def test_with_dsym_and_run_command(self): 133 """Test 'frame variable var_name' on some variables with array types.""" 134 self.buildDsym() 135 self.array_types() 136 137 This method is decorated with a skipUnless decorator so that it will only gets 138 included into the test suite if the platform it is running on is 'darwin', aka 139 Mac OS X. 140 141 Type 'man dsymutil' for more details. 142 143 After the binary is built, it is time to specify the file to be used as the main 144 executable by lldb: 145 146 exe = os.path.join(os.getcwd(), "a.out") 147 self.runCmd("file " + exe, CURRENT_EXECUTABLE_SET) 148 149 This is where the attribute assignment: 150 151 class SettingsCommandTestCase(TestBase): 152 153 mydir = "settings" 154 155 which happens right after the SettingsCommandTestCase class declaration comes 156 into place. It specifies the relative directory to the top level 'test' so that 157 the test harness can change its working directory in order to find the 158 executable as well as the source code files. The runCmd() method is defined in 159 the TestBase base class (within test/lldbtest.py) and its purpose is to pass the 160 specified command to the lldb command interpreter. It's like you're typing the 161 command within an interactive lldb session. 162 163 The CURRENT_EXECUTABLE_SET is an assert message defined in the lldbtest module 164 so that it can be reused from other test modules. 165 166 By default, the runCmd() is going to check the return status of the command 167 execution and fails the test if it is not a success. The assert message, in our 168 case CURRENT_EXECUTABLE_SET, is used in the exception printout if this happens. 169 170 There are cases when we don't care about the return status from the command 171 execution. This can be accomplished by passing the keyword argument pair 172 'check=False' to the method. 173 174 After the current executable is set, we'll then execute two more commands: 175 176 # Set the output-path and verify it is set. 177 self.runCmd("settings set target.process.output-path 'stdout.txt'") 178 self.expect("settings show target.process.output-path", 179 SETTING_MSG("target.process.output-path"), 180 startstr = "target.process.output-path (string) = 'stdout.txt'") 181 182 The first uses the 'settings set' command to set the static setting 183 target.process.output-path to be 'stdout.txt', instead of the default 184 '/dev/stdout'. We then immediately issue a 'settings show' command to check 185 that, indeed, the setting did take place. Notice that we use a new method 186 expect() to accomplish the task, which in effect issues a runCmd() behind the 187 door and grabs the output from the command execution and expects to match the 188 start string of the output against what we pass in as the value of the keyword 189 argument pair: 190 191 startstr = "target.process.output-path (string) = 'stdout.txt'" 192 193 Take a look at TestBase.expect() within lldbtest.py for more details. Among 194 other things, it can also match against a list of regexp patterns as well as a 195 list of sub strings. And it can also perform negative matching, i.e., instead 196 of expecting something from the output of command execution, it can perform the 197 action of 'not expecting' something. 198 199 This will launch/run the program: 200 201 self.runCmd("run", RUN_SUCCEEDED) 202 203 And this asserts that the file 'stdout.txt' should be present after running the 204 program. 205 206 # The 'stdout.txt' file should now exist. 207 self.assertTrue(os.path.isfile("stdout.txt"), 208 "'stdout.txt' exists due to target.process.output-path.") 209 210 Also take a look at main.cpp which emits some message to the stdout. Now, if we 211 pass this assertion, it's time to examine the contents of the file to make sure 212 it contains the same message as programmed in main.cpp: 213 214 # Read the output file produced by running the program. 215 with open('stdout.txt', 'r') as f: 216 output = f.read() 217 218 self.expect(output, exe=False, 219 startstr = "This message should go to standard out.") 220 221 We open the file and read its contents into output, then issue an expect() 222 method. The 'exe=False' keyword argument pair tells expect() that don't try to 223 execute the first arg as a command at all. Instead, treat it as a string to 224 match against whatever is thrown in as keyword argument pairs! 225 226 There are also other test methods present in the TestSettings.py mode: 227 test_set_prompt(), test_set_term_width(), test_set_auto_confirm(), 228 test_with_dsym(), and test_with_dwarf(). We are using the default test loader 229 from unittest framework, which uses the 'test' method name prefix to identify 230 test methods automatically. 231 232 This finishes the walkthrough of the test method test_set_output_path(self). 233 Before we say goodbye, notice the little method definition at the top of the 234 file: 235 236 @classmethod 237 def classCleanup(cls): 238 system(["/bin/sh", "-c", "rm -f output.txt"]) 239 system(["/bin/sh", "-c", "rm -f stdout.txt"]) 240 241 This is a classmethod (as shown by the @classmethod decorator) which allows the 242 individual test class to perform cleanup actions after the test harness finishes 243 with the particular test class. This is part of the so-called test fixture in 244 the unittest framework. From http://docs.python.org/library/unittest.html: 245 246 A test fixture represents the preparation needed to perform one or more tests, 247 and any associate cleanup actions. This may involve, for example, creating 248 temporary or proxy databases, directories, or starting a server process. 249 250 The TestBase class uses such fixture with setUp(self), tearDown(self), 251 setUpClass(cls), and tearDownClass(cls). And within teraDownClass(cls), it 252 checks whether the current class has an attribute named 'classCleanup', and 253 executes as a method if present. In this particular case, the classCleanup() 254 calls a utility function system() defined in lldbtest.py in order to remove the 255 files created by running the program as the tests are executed. 256 257 This system() function uses the Python subprocess module to spawn the process 258 and to retrieve its results. If the test instance passes the keyword argument 259 pair 'sender=self', the detailed command execution through the operating system 260 also gets recorded in a session object. If the test instance fails or errors, 261 the session info automatically gets dumped to a file grouped under a directory 262 named after the timestamp of the particular test suite run. 263 264 For simple cases, look for the timestamp directory in the same directory of the 265 test driver program dotest.py. For example, if we comment out the 266 @expectedFailure decorator for TestSettings.py, and then run the test module: 267 268 /Volumes/data/lldb/svn/trunk/test $ ./dotest.py -v settings 269 ---------------------------------------------------------------------- 270 Collected 6 tests 271 272 test_set_auto_confirm (TestSettings.SettingsCommandTestCase) 273 Test that after 'set auto-confirm true', manual confirmation should not kick in. ... ok 274 test_set_output_path (TestSettings.SettingsCommandTestCase) 275 Test that setting target.process.output-path for the launched process works. ... FAIL 276 test_set_prompt (TestSettings.SettingsCommandTestCase) 277 Test that 'set prompt' actually changes the prompt. ... ok 278 test_set_term_width (TestSettings.SettingsCommandTestCase) 279 Test that 'set term-width' actually changes the term-width. ... ok 280 test_with_dsym (TestSettings.SettingsCommandTestCase) 281 Test that run-args and env-vars are passed to the launched process. ... ok 282 test_with_dwarf (TestSettings.SettingsCommandTestCase) 283 Test that run-args and env-vars are passed to the launched process. ... ok 284 285 ====================================================================== 286 FAIL: test_set_output_path (TestSettings.SettingsCommandTestCase) 287 Test that setting target.process.output-path for the launched process works. 288 ---------------------------------------------------------------------- 289 Traceback (most recent call last): 290 File "/Volumes/data/lldb/svn/trunk/test/settings/TestSettings.py", line 125, in test_set_output_path 291 "'stdout.txt' exists due to target.process.output-path.") 292 AssertionError: False is not True : 'stdout.txt' exists due to target.process.output-path. 293 294 ---------------------------------------------------------------------- 295 Ran 6 tests in 8.219s 296 297 FAILED (failures=1) 298 /Volumes/data/lldb/svn/trunk/test $ ls 2010-10-19-14:10:49.059609 299 300 NOTE: This directory name has been changed to not contain the ':' character 301 which is not allowed in windows platforms. We'll change the ':' to '_' 302 and get rid of the microsecond resolution by modifying the test driver. 303 304 TestSettings.SettingsCommandTestCase.test_set_output_path.log 305 /Volumes/data/lldb/svn/trunk/test $ 306 307 We get one failure and a timestamp directory 2010-10-19-14:10:49.059609. 308 For education purposes, the directory and its contents are reproduced here in 309 the same directory as the current file. 310
