JUnitparser

4.0.2 · active · verified Sat Mar 28

JUnitparser is a Python library designed to parse and manipulate JUnit/xUnit Result XML files. It allows users to read and modify existing XML reports or create new ones from scratch. The library is actively maintained, with its current version being 4.0.2, and has a consistent release cadence addressing new features and bug fixes.

Warnings

breaking In version 4.0.0, several arguments for writing methods were renamed or changed from positional to keyword. `filepath` in `write_xml`, `TestSuite.write`, and `JUnitXml.write` is now `file_or_filename`. `pretty` became a keyword-only argument. `to_console` was removed; use `sys.stdout` for console output. `JUnitXml.fromfile`'s `filepath` argument is now `file`.
Fix: Update method calls to use new argument names (`file_or_filename`, `file`) and pass `pretty` as a keyword argument. Replace `to_console=True` with `obj.write(sys.stdout)`.
breaking As of version 4.0.0, `JUnitXml.fromfile`, `JUnitXml.fromstring`, and `JUnitXml.fromroot` methods consistently return a `JUnitXml` instance. Previously, if the root element was `<testsuite>`, it might return a `TestSuite` instance directly.
Fix: Always expect a `JUnitXml` object as the return type and iterate over its `testsuites` attribute if you need to access individual suites: `for suite in xml: ...`
breaking In version 3.0.0, support for Python 2 was completely dropped.
Fix: Ensure your project is running on Python 3.6 or later.
breaking In version 2.0.0, the `TestCase.result` attribute was changed from a single object to a list of result objects (e.g., `Failure`, `Skipped`, `Error`, `Success`). This allows for multiple outcomes, particularly relevant for pytest reports.
Fix: Update code that accesses `TestCase.result` to expect and iterate over a list. For example, `case.result[0]` to get the first result or `for res in case.result: ...`.
gotcha The `TestCase.result` setter in version 4.0.0 and above now throws a `ValueError` if an invalid type is assigned, whereas earlier versions might have silently ignored it.
Fix: Ensure that any value assigned to `TestCase.result` is an instance of a valid result class (e.g., `Success`, `Failure`, `Error`, `Skipped`) or a list of such instances.

Install

pip install junitparser Install stable version

Imports

JUnitXml
```
from junitparser import JUnitXml
```
TestCase
```
from junitparser import TestCase
```
TestSuite
```
from junitparser import TestSuite
```
Skipped
```
from junitparser import Skipped
```
Error
```
from junitparser import Error
```

Quickstart

This quickstart demonstrates how to create a JUnit XML report from scratch, defining test cases with different results (skipped, successful, error), grouping them into a test suite, and then writing the complete report to an XML file. The `pretty=True` argument ensures human-readable formatting.

import os
from junitparser import TestCase, TestSuite, JUnitXml, Skipped, Error

# Create some test cases
case1 = TestCase('test_feature_a', 'module.submodule', 1.23)
case1.result = Skipped('Not implemented yet')

case2 = TestCase('test_feature_b', 'module.submodule', 0.45)
# By default, a test case is considered successful if no result is explicitly set or if it's an instance of Success

case3 = TestCase('test_feature_c', 'another_module', 2.00)
case3.result = Error('Connection refused', 'NetworkError')

# Create a test suite and add cases
suite = TestSuite('MyTestSuite')
suite.add_testcase(case1)
suite.add_testcase(case2)
suite.add_testcase(case3)

# Create a JUnitXml object and add the suite
xml = JUnitXml()
xml.add_testsuite(suite)

# Define output path
output_filename = os.environ.get('JUNIT_OUTPUT_PATH', 'report.xml')

# Write the XML to a file
xml.write(output_filename, pretty=True)

print(f"Generated JUnit XML report at: {output_filename}")

# Example of reading and modifying an existing XML (requires a dummy file)
# with open('existing_report.xml', 'w') as f:
#     f.write('<testsuites><testsuite name="Existing" tests="1" failures="0"><testcase name="passing_test" /></testsuite></testsuites>')
# existing_xml = JUnitXml.fromfile('existing_report.xml')
# for ts in existing_xml:
#     ts.name = "Modified Existing Suite"
# existing_xml.write('modified_report.xml', pretty=True)

view raw JSON →