Skip to content

Latest commit

 

History

History
229 lines (172 loc) · 9.77 KB

README.rst

File metadata and controls

229 lines (172 loc) · 9.77 KB
Apache 2.0 License PyPI packaged release Build status Documentation Code Coverage Contributions welcome

Documentation xunit2rst

This script can convert a JUnit/xUnit (.xml) file to a reStructuredText (.rst) file with traceable items.

This script allows you to connect your test reports to your test cases via the mlx.traceability Sphinx extension.

pip3 install mlx.xunit2rst
mlx.xunit2rst -i itest_report.xml -o my_lib_report.rst --prefix ITEST_MY_LIB-

mlx.xunit2rst --help

usage: xunit2rst [-h] -i INPUT_FILE -o RST_OUTPUT_FILE [--only EXPRESSION] [-s] [-p PREFIX]
                 [--trim-suffix] [--unit-or-integration UNIT_OR_INTEGRATION] [-t TYPE] [-f]
                 [-l LOG] [--links] [-v]


optional arguments:
  -h, --help            show this help message and exit
  -i INPUT_FILE, --input INPUT_FILE
                        The input XML file
  -o RST_OUTPUT_FILE, --output RST_OUTPUT_FILE
                        The output RST file
  --only EXPRESSION     Expression of tags for Sphinx' `only` directive that surrounds all
                        RST content. By default, no `only` directive is generated.
  -s, --itemize-suites  Flag to itemize testsuite elements instead of testcase
                        elements.
  -p PREFIX, --prefix PREFIX
                        Optional prefix to add to item IDs
  --trim-suffix         If the suffix of the --prefix argument ends with '_-' it gets
                        trimmed to '-'
  --unit-or-integration UNIT_OR_INTEGRATION
                        Deprecated alternative to --type; to be removed in version
                        2.0.0.
  -t TYPE, --type TYPE  Optional: give value starting with 'u', 'i' or 'q' to
                        explicitly define the type of test:
                        unit/integration/qualification test
  -f, --failure-message
                        Include the error message in case of test failure in the item's
                        body.
  -l LOG, --log LOG     Optional: path to the HTML log file, relative to where Sphinx
                        will put the --output, to create a link to.
  --links               Optional: inserts a link to the RobotFramework HTML log file
                        for each test case as ext_robotframeworklog link id.
  -v, --version         show program's version number and exit

If you use the --only input argument, you should also add 'sphinx_selective_exclude.eager_only' to the extensions list to prevent mlx.traceability from parsing the content and ignoring the effect of the only directive.

By default, all testcase elements from the input file are used to created treaceability items. This may not always be desired. The -s, --itemize-suites flag lets the script itemize testsuite elements instead. In this case, the testcase elements will still be parsed to determine whether the testsuite passed or failed or skipped.

Test Result of Itemized Suites
Pass (passes)
No test cases have failed. Some might be skipped.
Fail (fails)
At least one test case has failed.
Skip (skipped)
All test cases have been skipped.

The name attribute of the element to be itemized is used to build the item ID. Lower case letters get converted to upper case, whitespaces get converted to underscores, and & characters get converted to AND. A valid prefix must end with a hyphen to be recognized by the script. If there is an additional string prepended to this name by means of a dot, this string won't be taken into account. Example below:

MY_LIB.ITEST_MY_LIB-my function & keyword XML element name results in item ID ITEST_MY_LIB-MY_FUNCTION_AND_KEYWORD

Traceability item IDs have a prefix that is unique for the group they belong to, e.g. ITEST_MY_LIB-. The --prefix input argument lets you configure this prefix. It will be prepended to the item names found in the input file to build the item ID. By default, the script adds ITEST- or UTEST- for integration or unit test reports, unless the prefixes already exist in the input file.

Test reports that have a testsuites element as root in XML are treated as unit test reports. Otherwise the script treats the input file as an integration test report. This discerning behavior gets overridden when prefixes are found in the input file or the --prefix input argument is used. The script looks for a U or I as the first letter of the prefix. Lastly, you can explicitly define the type by using the --type input argument. Its value should start with u or i and gets parsed case-insensitively.

When the -f, --failure-message flag is set, the tool includes the messages of all failures and reasons for skipping of the item's test cases in its body.

Extra reStructuredText content for the generated test reports can be defined in a YAML file that contains a mapping, aka dictionary, of case insensitive test case names (or test suite names for -s, --itemize-suites) as keys and reStructuredText strings as values. Multiline strings are supported. This feature is used in the example documentation.

The path to the YAML file, an absolute path or relative to the input XML file, must be added as a metadata element to the XML content, with the string xunit2rst content file as name and the path as value.

Examples of valid metadata in XML:

<properties>
  <property name="xunit2rst content file" value="../extra_content.yml"/>
</properties>

<traits>
  <trait name="xUnit2rst Content File" value="/home/user/git/test_project/extra_content.yml"/>
</traits>

When using Robot Framework v5 or greater, this path can be provided as Free test suite metadata.

Use input parameter -l, --log PATH_TO_FILE to add a link to the HTML log file. This path should be relative to the location where Sphinx outputs the HTML version of the RST file generated by this tool. Be sure to put this log file in Sphinx' source directory and add it to html_extra_path in your conf.py. This way, the log file will end up in the root of Sphinx' output location.

The --links flag adds a link to each generated traceability item in the form of an external relationship. This requires you to configure the relationship ext_robotframeworklog in your conf.py like this:

traceability_relationships = {
    'ext_robotframeworklog': '',
}
traceability_relationship_to_string = {
    'ext_robotframeworklog': 'Details in log file',
}
traceability_external_relationship_to_url = {
    'ext_robotframeworklog': 'field1#field2'
}

You can highlight each test result (pass, fail and skip) in a different color so that they stand out more from the rest of the report. The element with the test result text contains two classes: xunit2rst and the test result itself. For HTML documentation, you can use the CSS stylesheet, provided with the Python package, by passing it to Sphinx with add_css_file, or by adding its contents manually to your own stylesheet.

This feature is demonstrated in the test reports of the example documentation.