All Downloads are FREE. Search and download functionalities are using the official Maven repository.

robotframework-2.7.7.doc.userguide.src.ExecutingTestCases.BasicUsage.rst Maven / Gradle / Ivy

The newest version!
Basic usage
-----------

Robot Framework test cases are executed from the command line, and the
end result is, by default, an `output file`_ in XML format and an HTML
report_ and log_. After the execution, output files can be combined and
otherwise `post-processed`__ with the :prog:`rebot` tool.

__ `Post-processing outputs`_

.. contents::
   :depth: 2
   :local:

Starting test execution
~~~~~~~~~~~~~~~~~~~~~~~

Synopsis
''''''''

::

    pybot|jybot|ipybot [options] data_sources
    python|jython|ipy -m robot.run [options] data_sources
    python|jython|ipy path/to/robot/run.py [options] data_sources
    java -jar robotframework.jar [options] data_sources

Test execution is normally started using :prog:`pybot`, :prog:`jybot`
or :prog:`ipybot` `runner script`_. These scripts are otherwise identical, but
the first one executes tests using Python_, the second using Jython_, and the
last one using IronPython_. Alternatively it is possible to use
:prog:`robot.run` `entry point`_ either as a module or a script using
any interpreter, or use the `standalone JAR distribution`_.

Regardless of execution approach, the path (or paths) to the test data to be
executed is given as an argument after the command. Additionally, different
command line options can be used to alter the test execution or generated
outputs in some way.

Specifying test data to be executed
'''''''''''''''''''''''''''''''''''

Robot Framework test cases are created in files__ and directories__,
and they are executed by giving the path to the file or directory in
question to the selected runner script. The path can be absolute or,
more commonly, relative to the directory where tests are executed
from. The given file or directory creates the top-level test suite,
which gets its name, unless overridden with the :opt:`--name` option__,
from the `file or directory name`__. Different execution possibilities
are illustrated in the examples below. Note that in these examples, as
well as in other examples in this section, only the :prog:`pybot` script
is used, but other execution approaches could be used similarly.

__ `Test case files`_
__ `Test suite directories`_
__ `Setting the name`_
__ `Test suite name and documentation`_

::

   pybot test_cases.html
   pybot path/to/my_tests/
   pybot c:\robot\tests.txt

It is also possible to give paths to several test case files or
directories at once, separated with spaces. In this case, Robot
Framework creates the top-level test suite automatically, and
the specified files and directories become its child test suites. The name
of the created test suite is got from child suite names by
catenating them together with an ampersand (&) and spaces. For example,
the name of the top-level suite in the first example below is
:name:`My Tests & Your Tests`. These automatically created names are
often quite long and complicated. In most cases, it is thus better to
use the :opt:`--name` option for overriding it, as in the second
example below::

   pybot my_tests.html your_tests.html
   pybot --name Example path/to/tests/pattern_*.html

Using command line options
~~~~~~~~~~~~~~~~~~~~~~~~~~

Robot Framework provides a number of command line options that can be
used to control how test cases are executed and what outputs are
generated. This section explains the option syntax, and what
options actually exist. How they can be used is discussed elsewhere
in this chapter.

Using options
'''''''''''''

When options are used, they must always be given between the runner
script and the data sources. For example::

   pybot -L debug my_tests.txt
   pybot --include smoke --variable HOST:10.0.0.42 path/to/tests/

Short and long options
''''''''''''''''''''''

Options always have a long name, such as :opt:`--name`, and the
most frequently needed options also have a short name, such as
:opt:`-N`. In addition to that, long options can be shortened as
long as they are unique. For example, :cli:`--logle DEBUG` works,
while :cli:`--lo log.html` does not, because the former matches only
:opt:`--loglevel`, but the latter matches several options. Short
and shortened options are practical when executing test cases
manually, but long options are recommended in `start-up scripts`__,
because they are easier to understand.

__ `Creating start-up scripts`_

The long option format is case-insensitive, which facilitates writing option
names in an easy-to-read format. For example, :opt:`--SuiteStatLevel`
is equivalent to, but easier to read than :opt:`--suitestatlevel`.

Setting option values
'''''''''''''''''''''

Most of the options require a value, which is given after the option
name. Both short and long options accept the value separated
from the option name with a space, as in :cli:`--include tag`
or :cli:`-i tag`. With long options, the separator can also be the
equals sign, for example :cli:`--include=tag`, and with short options the
separator can be omitted, as in :cli:`-itag`.

Some options can be specified several times. For example,
:cli:`--variable VAR1:value --variable VAR2:another` sets two
variables. If the options that take only one value are used several
times, the value given last is effective.

Option value as simple pattern
''''''''''''''''''''''''''''''

Many of the options take arguments as *simple patterns*. This means
that :code:`*` and :code:`?` can be used as special characters, so
that the former matches any string (even an empty string) and the
latter matches any single character. For example, :cli:`--include
prefix-\*` matches all tags starting with :code:`prefix-`, and
:cli:`--include a???` matches any tag that is four characters long and
starts with a character :code:`a`.

Test results
~~~~~~~~~~~~

Command line output
'''''''''''''''''''

The most visible output from test execution is the output displayed in
the command line. All executed test suites and test cases, as well as
their statuses, are shown there in real time. The example below shows the
output from executing a simple test suite with only two test cases::

   ==============================================================================
   Example test suite
   ==============================================================================
   First test :: Possible test documentation                             | PASS |
   ------------------------------------------------------------------------------
   Second test                                                           | FAIL |
   Error message is displayed here
   ==============================================================================
   Example test suite                                                    | FAIL |
   2 critical tests, 1 passed, 1 failed
   2 tests total, 1 passed, 1 failed
   ==============================================================================
   Output:  /path/to/output.xml
   Report:  /path/to/report.html
   Log:     /path/to/log.html

Starting from Robot Framework 2.7, there is also a notification on the console
whenever a top-level keyword in a test case ends. A green dot is used if
a keyword passes and a red F if it fails. These markers are written to the end
of line and they are overwritten by the test status when the test itself ends.
Writing the markers is disabled if console output is redirected to a file.

Generated output files
''''''''''''''''''''''

The command line output is very limited, and separate output files are
normally needed for investigating the test results. As the example
above shows, three output files are generated by default. The first
one is in XML format and contains all the information about test
execution. The second is a higher-level report and the third is a more
detailed log file. These files and other possible output files are
discussed in more detail in the section `Different output files`_.

Return codes
''''''''''''

Runner scripts communicate the overall test execution status to the
system running them using return codes. When the execution starts
successfully and no `critical test`_ fail, the return code is zero.
All possible return codes are explained in the table below.

.. table:: Possible return codes
   :class: tabular

   ========  ==========================================
      RC                    Explanation
   ========  ==========================================
   0         All critical tests passed.
   1-249     Returned number of critical tests failed.
   250       250 or more critical failures.
   251       Help or version information printed.
   252       Invalid test data or command line options.
   253       Test execution stopped by user.
   255       Unexpected internal error.
   ========  ==========================================

Return codes should always be easily available after the execution,
which makes it easy to automatically determine the overall execution
status. For example, in bash shell the return code is in special
variable :code:`$?`, and in Windows it is in :code:`%ERRORLEVEL%`
variable. If you use some external tool for running tests, consult its
documentation for how to get the return code.

Starting from Robot Framework 2.5.7, the return code can be set to 0 even if
there are critical failures using the :opt:`--NoStatusRC` command line option.
This might be useful, for example, in continuous integration servers where
post-processing of results is needed before the overall status of test
execution can be determined.

.. note:: Same return codes are also used with rebot_.

Errors and warnings during execution
''''''''''''''''''''''''''''''''''''

During the test execution there can be unexpected problems like
failing to import a library or a resource file or a keyword being
deprecated__. Depending on the severity such problems are categorized
as errors or warnings and they are written into the console (using the
standard error stream), shown on a separate *Test Execution Errors*
section in log files, and also written into Robot Framework's own
`system log`_. Normally these errors are generated by Robot Framework
core, but libraries can use `log level WARN`__ to write warnings.
Example below illustrates how errors and warnings look like in the log
file.

.. raw:: html

   
20090322 19:58:42.528 ERROR Error in file '/home/robot/tests.html' in table 'Setting' in element on row 2: Resource file 'resource.html' does not exist
20090322 19:58:43.931 WARN Keyword 'SomeLibrary.Example Keyword' is deprecated. Use keyword `Other Keyword` instead.
__ `Deprecating keywords`_ __ `Logging information`_ Escaping complicated characters ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Because spaces are used for separating options from each other, it is problematic to use them in option values. Some options, such as :opt:`--name`, automatically convert underscores to spaces, but with others spaces must be escaped. Additionally, many special characters are complicated to use on the command line. Because escaping complicated characters with a backslash or quoting the values does not always work too well, Robot Framework has its own generic escaping mechanism. Another possibility is using `argument files`_ where options can be specified in the plain text format. Both of these mechanisms work when executing tests and when post-processing outputs, and also some of the external supporting tools have the same or similar capabilities. In Robot Framework's command line escaping mechanism, problematic characters are escaped with freely selected text. The command line option to use is :opt:`--escape` (short version :opt:`-E`), which takes an argument in the format :opt:`what:with`, where :opt:`what` is the name of the character to escape and :opt:`with` is the string to escape it with. Characters that can be escaped are listed in the table below: .. table:: Available escapes :class: tabular ========= ============= ========= ============= Character Name to use Character Name to use ========= ============= ========= ============= & amp ( paren1 ' apos ) paren2 @ at % percent \\ blash \| pipe : colon ? quest , comma " quot { curly1 ; semic } curly2 / slash $ dollar \ space ! exclam [ square1 > gt ] square2 # hash \* star < lt \ \ ========= ============= ========= ============= The following examples make the syntax more clear. In the first example, the metadata :opt:`X` gets the value :code:`Value with spaces`, and in the second example variable :var:`${VAR}` is assigned to :code:`"Hello, world!"`:: --escape space:_ --metadata X:Value_with_spaces -E space:SP -E quot:QU -E comma:CO -E exclam:EX -v VAR:QUHelloCOSPworldEXQU Note that all the given command line arguments, including paths to test data, are escaped. Escape character sequences thus need to be selected carefully. Argument files ~~~~~~~~~~~~~~ Argument files allow placing all or some command line options and arguments into an external file where they will be read. This avoids the problems with characters that are problematic on the command line. If lot of options or arguments are needed, argument files also prevent the command that is used on the command line growing too long. Argument files are taken into use with :opt:`--argumentfile` option (short option :opt:`-A`) along with possible other command line options. Argument file syntax '''''''''''''''''''' Argument files can contain both command line options and paths to the test data, one option or data source per line. Both short and long options are supported, but the latter are recommended because they are easier to understand. Argument files can contain any characters without escaping, but spaces in the beginning and end of lines are ignored. Additionally, empty lines and lines starting with a hash mark (#) are ignored:: --doc This is an example (where "special characters" are ok!) --metadata X:Value with spaces --variable VAR:Hello, world! # This is a comment path/to/my/tests In the above example the separator between options and their values is a single space. In Robot Framework 2.7.6 and newer it is possible to use either an equal sign (=) or any number of spaces. As an example, the following three lines are identical:: --name An Example --name=An Example --name An Example If argument files contain non-ASCII characters, they must be saved using UTF-8 encoding. Using argument files '''''''''''''''''''' Argument files can be used either alone so that they contain all the options and paths to the test data, or along with other options and paths. When an argument file is used with other arguments, its contents are placed into the original list of arguments to the same place where the argument file option was. This means that options in argument files can override options before it, and its options can be overridden by options after it. It is possible to use :opt:`--argumentfile` option multiple times or even recursively:: pybot --argumentfile all_arguments.txt pybot --name Example --argumentfile other_options_and_paths.txt pybot --argumentfile default_options.txt --name Example my_tests.html pybot -A first.txt -A second.txt -A third.txt tests.txt Reading argument files from standard input '''''''''''''''''''''''''''''''''''''''''' Starting from Robot Framework 2.5.6, special argument file name :opt:`STDIN` can be used to read arguments from the standard input stream instead of a file. This can be useful when generating arguments with a script:: generate_arguments.sh | pybot --argumentfile STDIN generate_arguments.sh | pybot --name Example --argumentfile STDIN tests.txt Getting help and version information ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Both when executing test cases and when post-processing outputs, it is possible to get command line help with the option :opt:`--help` and its short version :opt:`-h`. These help texts have a short general overview and briefly explain the available command line options. All runner scripts also support getting the version information with the option :opt:`--version`. This information also contains Python or Jython version and the platform type:: $ pybot --version Robot Framework 2.7 (Python 2.6.6 on linux2) $ jybot --version Robot Framework 2.7 (Jython 2.5.2 on java1.6.0_21) C:\>rebot --version Rebot 2.7 (Python 2.7.1 on win32) Creating start-up scripts ~~~~~~~~~~~~~~~~~~~~~~~~~ Test cases are often executed automatically by a continuous integration system or some other mechanism. In such cases, there is a need to have a script for starting the test execution, and possibly also for post-processing outputs somehow. Similar scripts are also useful when running tests manually, especially if a large number of command line options are needed or setting up the test environment is complicated. In UNIX-like environments, shell scripts provide a simple but powerful mechanism for creating custom start-up scripts. Windows batch files can also be used, but they are more limited and often also more complicated. A platform-independent alternative is using Python or some other high-level programming language. Regardless of the language, it is recommended that long option names are used, because they are easier to understand than the short names. In the first examples, the same web tests are executed with different browsers and the results combined afterwards. This is easy with shell scripts, as practically you just list the needed commands one after another: .. sourcecode:: bash #!/bin/bash pybot --variable BROWSER:Firefox --name Firefox --log none --report none --output out/fx.xml login pybot --variable BROWSER:IE --name IE --log none --report none --output out/ie.xml login rebot --name Login --outputdir out --output login.xml out/fx.xml out/ie.xml Implementing the above example with Windows batch files is not very complicated, either. The most important thing to remember is that because :prog:`pybot` and :prog:`rebot` are implemented as batch files, :prog:`call` must be used when running them from another batch file. Otherwise execution would end when the first batch file is finished. .. sourcecode:: bat @echo off call pybot --variable BROWSER:Firefox --name Firefox --log none --report none --output out\fx.xml login call pybot --variable BROWSER:IE --name IE --log none --report none --output out\ie.xml login call rebot --name Login --outputdir out --output login.xml out\fx.xml out\ie.xml In the next examples, JAR files under the :path:`lib` directory are put into CLASSPATH before starting the test execution. In these examples, start-up scripts require that paths to the executed test data are given as arguments. It is also possible to use command line options freely, even though some options have already been set in the script. All this is relatively straight-forward using bash: .. sourcecode:: bash #!/bin/bash cp=. for jar in lib/*.jar; do cp=$cp:$jar done export CLASSPATH=$cp jybot --ouputdir /tmp/logs --suitestatlevel 2 $* Implementing this using Windows batch files is slightly more complicated. The difficult part is setting the variable containing the needed JARs inside a For loop, because, for some reason, that is not possible without a helper function. .. sourcecode:: bat @echo off set CP=. for %%jar in (lib\*.jar) do ( call :set_cp %%jar ) set CLASSPATH=%CP% jybot --ouputdir c:\temp\logs --suitestatlevel 2 %* goto :eof :: Helper for setting variables inside a for loop :set_cp set CP=%CP%;%1 goto :eof Modifying Java startup parameters ''''''''''''''''''''''''''''''''' Sometimes when using Jython there is need to alter the Java startup parameters. The most common use case is increasing the JVM maximum memory size as the default value may not be enough for creating reports and logs when outputs are very big. There are several ways to configure JVM options: 1. Modify Jython start-up script (:prog:`jython` shell script or :prog:`jython.bat` batch file) directly. This is a permanent configuration. 2. Set :var:`JYTHON_OPTS` environment variable. This can be done permanently in operating system level or per execution in a custom start-up script. 3. Pass the needed Java parameters wit :opt:`-J` option to Jython start-up script that will pass them forward to Java. This is especially easy when using `direct entry points`_:: jython -J-Xmx1024m -m robot.run some_tests.txt Debugging problems ~~~~~~~~~~~~~~~~~~ A test case can fail because the system under test does not work correctly, in which case the test has found a bug, or because the test itself is buggy. The error message explaining the failure is shown on the `command line output`_ and in the `report file`_, and sometimes the error message alone is enough to pinpoint the problem. More often that not, however, `log files`_ are needed because they have also other log messages and they show which keyword actually failed. When a failure is caused by the tested application, the error message and log messages ought to be enough to understand what caused it. If that is not the case, the test library does not provide `enough information`__ and needs to be enhanced. In this situation running the same test manually, if possible, may also reveal more information about the issue. Failures caused by test cases themselves or by keywords they use can sometimes be hard to debug. If the error message, for example, tells that a keyword is used with wrong number of arguments fixing the problem is obviously easy, but if a keyword is missing or fails in unexpected way finding the root cause can be harder. The first place to look for more information is the `execution errors`_ section in the log file. For example, an error about a failed test library import may well explain why a test has failed due to a missing keyword. If the log file does not provide enough information by default, it is possible to execute tests with a lower `log level`_. For example tracebacks showing where in the code the failure occurred are logged using the :msg:`DEBUG` level, and this information is invaluable when the problem is in an individual keyword. If the log file still does not have enough information, it is a good idea to enable the syslog_ and see what information it provides. It is also possible to add some keywords to the test cases to see what is going on. Especially `BuiltIn keywords`_ :name:`Log` and :name:`Log Variables` are useful. If nothing else works, it is always possible to search help from `mailing lists`_ or elsewhere. __ `Communicating with Robot Framework`_ Using the Python debugger (pdb) ''''''''''''''''''''''''''''''' It is also possible to use the pdb__ module from the Python standard library to set a break point and interactively debug a running test. The typical way of invoking pdb by inserting .. sourcecode:: python import pdb; pdb.set_trace() at the location you want to break into debugger will not work correctly with Robot Framework, though, as the standard output stream is redirected during keyword execution. Instead, you can use the following: .. sourcecode:: python import sys, pdb; pdb.Pdb(stdout=sys.__stdout__).set_trace() __ http://docs.python.org/2/library/pdb.html




© 2015 - 2024 Weber Informatics LLC | Privacy Policy