Parsers¶

Sphinx-Test-Reports provides different parsers for test result files. One for XML-files, following the JUnit format, and one for generic JSON-files, which are not following any standard.

The needed parser is automatically selected by the file extensions xml or json.

Junit Parser¶

This parser reads xml files and handles the content as JUnit data. Other XML formats are not supported.

As JUnit format is not really standardized, different test framework produce slightly different JUnit xml files. Sphinx-Test-Reports supports the “dialects” of:

• pytest

• GoogleTest

• CasperJS

There is a high chance, that other tet frameworks are supported as well, as long as they provide a Junit file.

JSON Parser¶

The JSON parser can read any JSON file. But as the data structure is not following any standard, the user need to provide a mapping between the used JSON structure and the internal structure used by Sphinx-Test-Reports.

Even if this means some more configuration effort, the benefit is a completely customizable data structure.

For an example please take a look into JSON data example.

For mapping configuration please see tr_json_mapping.

Technical details¶

Each parser is realized by a class, which needs to provide specific functions. Also the internal object representation is the same. So each parser must map the external format to the internal representation.

Only the parser cares about the format. Other internal functions (like directives) just work with the common data representation and rely on it. So parsers are not allowed to rename or even extend this internal representation. If this is needed, all available parsers need to be updated as well.