前往小程序,Get更优阅读体验!
立即前往
首页
学习
活动
专区
工具
TVP
发布
社区首页 >专栏 >Robot Framework源码解析(2) - 执行测试的入口点

Robot Framework源码解析(2) - 执行测试的入口点

作者头像
yuanyuan
发布2019-09-02 18:04:13
1.3K0
发布2019-09-02 18:04:13
举报
文章被收录于专栏:小满小满

我们再来看 src/robot/run.py 的工作原理。摘录部分代码:

代码语言:javascript
复制
 1 from robot.conf import RobotSettings
 2 from robot.model import ModelModifier
 3 from robot.output import LOGGER, pyloggingconf
 4 from robot.reporting import ResultWriter
 5 from robot.running import TestSuiteBuilder
 6 from robot.utils import Application, unic, text
 7 
 8 class RobotFramework(Application):
 9 
10     def __init__(self):
11         Application.__init__(self, USAGE, arg_limits=(1,),
12                              env_options='ROBOT_OPTIONS', logger=LOGGER)
13 
14     def main(self, datasources, **options):
15         ......
16 
17 def run_cli(arguments=None, exit=True):
18     if arguments is None:
19         arguments = sys.argv[1:]
20     return RobotFramework().execute_cli(arguments, exit=exit)
21 
22 
23 def run(*tests, **options):
24    return RobotFramework().execute(*tests, **options)
25 
26 
27 if __name__ == '__main__':
28     run_cli(sys.argv[1:])

在上一章我们提到Java的命令行入口其实最终还是转到了其它入口点,例如robot.run的run_cli(mytests.robot)

这里就先看第51行的run_cli方法 ,方法很简单,只是调用了RobotFramework类中的execute_cli方法。RobotFramework是run.py的一个内部类,也是Application的子类。通过第6行的 from robot.utils import Application可查看Application是做什么的。

src/robot/utils/application.py

摘录部分代码:

代码语言:javascript
复制
 1 class Application(object):
 2 
 3     def __init__(self, usage, name=None, version=None, arg_limits=None,
 4                  env_options=None, logger=None, **auto_options):
 5         self._ap = ArgumentParser(usage, name, version, arg_limits,
 6                                   self.validate, env_options, **auto_options)
 7         self._logger = logger or DefaultLogger()
 8 
 9     def main(self, arguments, **options):
10         raise NotImplementedError
11 
12 ......
13 
14     def execute_cli(self, cli_arguments, exit=True):
15         with self._logger:
16             self._logger.info('%s %s' % (self._ap.name, self._ap.version))
17             options, arguments = self._parse_arguments(cli_arguments)
18             rc = self._execute(arguments, options)
19         if exit:
20             self._exit(rc)
21         return rc
22     def _parse_arguments(self, cli_args):
23         try:
24             options, arguments = self.parse_arguments(cli_args)
25         except Information as msg:
26             self._report_info(msg.message)
27         except DataError as err:
28             self._report_error(err.message, help=True, exit=True)
29         else:
30             self._logger.info('Arguments: %s' % ','.join(arguments))
31             return options, arguments
32 
33     def parse_arguments(self, cli_args):
34         """Public interface for parsing command line arguments.
35 
36         :param    cli_args: Command line arguments as a list
37         :returns: options (dict), arguments (list)
38         :raises:  :class:`~robot.errors.Information` when --help or --version used
39         :raises:  :class:`~robot.errors.DataError` when parsing fails
40         """
41         return self._ap.parse_args(cli_args)
42 
43     def execute(self, *arguments, **options):
44         with self._logger:
45             self._logger.info('%s %s' % (self._ap.name, self._ap.version))
46             return self._execute(list(arguments), options)
47 
48     def _execute(self, arguments, options):
49         try:
50             rc = self.main(arguments, **options)
51         except DataError as err:
52             return self._report_error(err.message, help=True)
53         except (KeyboardInterrupt, SystemExit):
54             return self._report_error('Execution stopped by user.',
55                                       rc=STOPPED_BY_USER)
56         except:
57             error, details = get_error_details(exclude_robot_traces=False)
58             return self._report_error('Unexpected error: %s' % error,
59                                       details, rc=FRAMEWORK_ERROR)
60         else:
61             return rc or 0

Application的execute_cli方法,其实也只是做了参数的解析工作(请看第17行 和 第18行的方法调用),具体的任务如何执行交给了本实例的main方法(第50行)。那么仍然回到 src/robot/run.py 看RobotFramework的main方法:

代码语言:javascript
复制
 1     def main(self, datasources, **options):
 2         settings = RobotSettings(options)
 3         LOGGER.register_console_logger(**settings.console_output_config)
 4         LOGGER.info('Settings:\n%s' % unic(settings))
 5         builder = TestSuiteBuilder(settings['SuiteNames'],
 6                                    extension=settings.extension,
 7                                    rpa=settings.rpa)
 8         suite = builder.build(*datasources)
 9         settings.rpa = builder.rpa
10         suite.configure(**settings.suite_config)
11         if settings.pre_run_modifiers:
12             suite.visit(ModelModifier(settings.pre_run_modifiers,
13                                       settings.run_empty_suite, LOGGER))
14         with pyloggingconf.robot_handler_enabled(settings.log_level):
15             old_max_error_lines = text.MAX_ERROR_LINES
16             text.MAX_ERROR_LINES = settings.max_error_lines
17             try:
18                 result = suite.run(settings)
19             finally:
20                 text.MAX_ERROR_LINES = old_max_error_lines
21             LOGGER.info("Tests execution ended. Statistics:\n%s"
22                         % result.suite.stat_message)
23             if settings.log or settings.report or settings.xunit:
24                 writer = ResultWriter(settings.output if settings.log
25                                       else result)
26                 writer.write_results(settings.get_rebot_settings())
27         return result.return_code

在这个方法里,进行了设置项的赋值(第2行),真正执行测试并输出测试结果(第18行)。通过第5,8,18行可以看到测试的执行过程首先是通过TestSuiteBuilder构建了一个suite,然后执行该suite的run方法。那么我们来看看TestSuiteBuilder是如何构建一个suite的。

通过from robot.running import TestSuiteBuilder可以知道TestSuiteBuilder是在robot.running路径下,我们先看看这个包的__init__.py

代码语言:javascript
复制
1 from .builder import TestSuiteBuilder, ResourceFileBuilder
2 from .context import EXECUTION_CONTEXTS
3 from .model import Keyword, TestCase, TestSuite
4 from .testlibraries import TestLibrary
5 from .usererrorhandler import UserErrorHandler
6 from .userkeyword import UserLibrary
7 from .runkwregister import RUN_KW_REGISTER

由第1行可以看出TestSuiteBuilder在 src/robot/running/builder.py:摘录的部分代码:

代码语言:javascript
复制
 1 class TestSuiteBuilder(object):
 2 
 3     def __init__(self, include_suites=None, warn_on_skipped='DEPRECATED',
 4                  extension=None, rpa=None):
 5         self.include_suites = include_suites
 6         self.extensions = self._get_extensions(extension)
 7         builder = StepBuilder()
 8         self._build_steps = builder.build_steps
 9         self._build_step = builder.build_step
10         self.rpa = rpa
11         self._rpa_not_given = rpa is None
12         # TODO: Remove in RF 3.2.
13         if warn_on_skipped != 'DEPRECATED':
14             warnings.warn("Option 'warn_on_skipped' is deprecated and has no "
15                           "effect.", DeprecationWarning)
16 ......
17 
18     def build(self, *paths):
19         """
20         :param paths: Paths to test data files or directories.
21         :return: :class:`~robot.running.model.TestSuite` instance.
22         """
23         if not paths:
24             raise DataError('One or more source paths required.')
25         if len(paths) == 1:
26             return self._parse_and_build(paths[0])
27         root = TestSuite()
28         for path in paths:
29             root.suites.append(self._parse_and_build(path))
30         root.rpa = self.rpa
31         return root
32 ......

build方法的最后返回了一个TestSuite对象。走到这里好像有点太快了,为了更好的理解这个TestSuite,我们回过头来,顺藤摸瓜看看这个build的参数paths是什么: def build(self, *paths)(builder.py) <-- builder.build(*datasources) (run.py)<-- def main(self, datasources, **options): <-- self.main(arguments, **options)(Application.py) <-- def _execute(self, arguments, options): <-- self._execute(arguments, options) <-- def execute_cli(self, cli_arguments, exit=True):(Application.py) <--RobotFramework().execute_cli(arguments, exit=exit)(run.py) <-- def run_cli(arguments=None, exit=True):(run.py)

原来这个paths是命令后选项参数或者是方法调用时传递过来的参数。例如

代码语言:javascript
复制
1 from robot import run_cli
2 
3 # Run tests and return the return code.
4 rc = run_cli(['--name', 'Example', 'tests.robot'], exit=False)
5 
6 # Run tests and exit to the system automatically.
7 run_cli(['--name', 'Example', 'tests.robot'])

或者 像第一篇文章中 java -jar robotframework.jar run mytests.robot这个命令,经过JarRunner解析会最终调用robot.run的run_cli("mytests.robot")这个方法

所以这个TestSuiteBuilder的目的是通过解析datasource来构建一个TestSuite ,接着回到builder.py的 build方法最后的TestSuite对象上,来看看TestSuite什么。

通过robot.running的_init_.py :from .model import Keyword, TestCase, TestSuite,可以看出TestSuite在 src/robot/running/model.py,摘录有关TestSuite的代码:

代码语言:javascript
复制
 1 class TestSuite(model.TestSuite):
 2     """Represents a single executable test suite.
 3 
 4     See the base class for documentation of attributes not documented here.
 5     """
 6     __slots__ = ['resource']
 7     test_class = TestCase    #: Internal usage only.
 8     keyword_class = Keyword  #: Internal usage only.
 9 
10     def __init__(self,  name='', doc='', metadata=None, source=None, rpa=False):
11         model.TestSuite.__init__(self, name, doc, metadata, source, rpa)
12         #: :class:`ResourceFile` instance containing imports, variables and
13         #: keywords the suite owns. When data is parsed from the file system,
14         #: this data comes from the same test case file that creates the suite.
15         self.resource = ResourceFile(source=source)
16 。。。。。。
17     def run(self, settings=None, **options):
18         from .namespace import IMPORTER
19         from .signalhandler import STOP_SIGNAL_MONITOR
20         from .runner import Runner
21 
22         with LOGGER:
23             if not settings:
24                 settings = RobotSettings(options)
25                 LOGGER.register_console_logger(**settings.console_output_config)
26             with pyloggingconf.robot_handler_enabled(settings.log_level):
27                 with STOP_SIGNAL_MONITOR:
28                     IMPORTER.reset()
29                     output = Output(settings)
30                     runner = Runner(output, settings)
31                     self.visit(runner)
32                 output.close(runner.result)
33         return runner.result

通过第10行的__init__方法可以看到,TestSuite初始化的时候包括name,doc,metadata,import,Variabel等数据。通过同一个图片我想大家应该就可以更 好的理解这里封装的信息了:

是的,就是这个可视化工具RIDE里的信息.当然这个类里面封装的信息并不全,因为它是model.TestSuite的子类,在父类中封装了更多的信息。

仍然回到 src/robot/run.py 的main方法,suite构建后会调用suite.run方法收集result。看 TestSuite类的第31行 self.visit(runner),这个visit方法都做了写什么?参数runner有时什么呢? 我们先在父类中看visit方法

/src/robot/model/testsuite.py

代码语言:javascript
复制
1     def visit(self, visitor):
2         """:mod:`Visitor interface <robot.model.visitor>` entry-point."""
3         visitor.visit_suite(self)

方法很简单,只是去调用这个runner参数的visit_suite()方法。我们通过TestSuite类run方法中的from .runner import Runner可以知道 这个runner参数是:

src/robot/running/runner.py

代码语言:javascript
复制
 1 class Runner(SuiteVisitor):
 2 
 3 。。。。。。

Runner是SuiteVisitor的子类,里面并没有visit_suite 方法。去看看SuiteVisitor。 通过model包__init__.py的 from .visitor import SuiteVisitor 可知,SuiteVisitor在 /src/robot/model/visitor.py,摘录部分代码:

代码语言:javascript
复制
 1 class SuiteVisitor(object):
 2     """Abstract class to ease traversing through the test suite structure.
 3 
 4     See the :mod:`module level <robot.model.visitor>` documentation for more
 5     information and an example.
 6     """
 7 
 8     def visit_suite(self, suite):
 9         """Implements traversing through the suite and its direct children.
10 
11         Can be overridden to allow modifying the passed in ``suite`` without
12         calling :func:`start_suite` or :func:`end_suite` nor visiting child
13         suites, tests or keywords (setup and teardown) at all.
14         """
15         if self.start_suite(suite) is not False:
16             suite.keywords.visit(self)
17             suite.suites.visit(self)
18             suite.tests.visit(self)
19             self.end_suite(suite)
20 
21     def start_suite(self, suite):
22         """Called when suite starts. Default implementation does nothing.
23 
24         Can return explicit ``False`` to stop visiting.
25         """
26         pass
27 
28     def end_suite(self, suite):
29         """Called when suite ends. Default implementation does nothing."""
30         pass

在visit_suite方法中,开始了测试的执行,start_suite,end_suite 都在Runner具体实现. 今天先写到这里,下一章再接着分析visit_suite()里调用的各个方法的具体实现.

如果喜欢作者的文章,请关注"写代码的猿"订阅号以便第一时间获得最新内容。本文版权归作者所有,欢迎转载.

本文参与 腾讯云自媒体分享计划,分享自作者个人站点/博客。
原始发表:2019-04-08 ,如有侵权请联系 cloudcommunity@tencent.com 删除

本文分享自 作者个人站点/博客 前往查看

如有侵权,请联系 cloudcommunity@tencent.com 删除。

本文参与 腾讯云自媒体分享计划  ,欢迎热爱写作的你一起参与!

评论
登录后参与评论
0 条评论
热度
最新
推荐阅读
领券
问题归档专栏文章快讯文章归档关键词归档开发者手册归档开发者手册 Section 归档