Parsing Test Results

Robot Framework can create different output files after a test run.

• output.xml
• log.html
• report.html
• xunit.xml

Sometimes you want to parse your test results, e.g. because you want to transfer them to a different system (like a test management tool).

tip

Don't try to parse the .xml files using some python xml module.

Instead, use the Robot Framework API which allows you to get details about the executed tests, keywords, their data and the results much easier.
You can find the documentation for the results model here. Especially the ResultVisitor will allow you to access and modify your results at different points in time, e.g. when

• a test suite starts/ends
• a test case starts/ends
• a keyword starts/ends
• ...

Check the documentation for more possible entry points.
While visiting the results, you will have full access to the Robot Framework Model and its objects. You can easily read the status of a test or keyword, read its arguments or the elapsed execution time. Basically all data which is written into the output.xml can be accessed via the model.

Reading results using the RF API

Example which reads the execution time per test and sets the test status to failed, in case the execution took longer than a defined value max_seconds.

ExecutionTimeChecker.py

import sys
from robot.api import ExecutionResult, ResultVisitor


class ExecutionTimeChecker(ResultVisitor):

    def __init__(self, max_seconds):
        self.max_milliseconds = max_seconds * 1000

    def visit_test(self, test):
        if test.status == 'PASS' and test.elapsedtime > self.max_milliseconds:
            test.status = 'FAIL'
            test.message = 'Test execution took too long.'


def check_tests(seconds, inpath, outpath=None):
    result = ExecutionResult(inpath)
    result.visit(ExecutionTimeChecker(float(seconds)))
    result.save(outpath)


if __name__ == '__main__':
    try:
        check_tests(*sys.argv[1:])
    except TypeError:
        print(__doc__)

You can run the script from the terminal as shown below:

python ExecutionTimeChecker.py 10 output.xml checked_output.xml

Getting result statistics

Simple example to read number of passed and failed tests

from robot.api import ExecutionResult
result = ExecutionResult('output.xml')
stats = result.statistics
print(f"Number of Failed Tests: {stats.failed}")
print(f"Total number of Tests: {stats.passed}")

You can also create a table in a markdown file report.mdwith the results.

from robot.api import ExecutionResult, ResultVisitor
import sys

class MyResultVisitor(ResultVisitor):
    def __init__(self, markdown_file='report.md'):
        self.failed_tests = []
        self.passed_tests = []
        self.markdown_file = markdown_file

    def visit_test(self, test):
        if test.status == 'FAIL':
            self.failed_tests.append(test.name)
        elif test.status == 'PASS':
            self.passed_tests.append(test.name)

    def end_result(self, result):
        # Create a new markdown file
        with open(self.markdown_file, "w") as f:
            f.write("# Robot Framework Report\n")
            f.write("|Test|Status|\n")
            f.write("|---|---|\n")
            for test in self.passed_tests:
                f.write(f"|{test}|PASS|\n")
            for test in self.failed_tests:
                f.write(f"|{test}|FAIL|\n")
                
if __name__ == '__main__':
    try:
        output_file = sys.argv[1]
    except IndexError:
        output_file = "output.xml"
    try:
        markdown_file = sys.argv[2]
    except IndexError:
        markdown_file = "report.md"
    result = ExecutionResult(output_file)
    result.visit(MyResultVisitor())

That's how the result will look like:

Robot Framework Report

Test	Status
My First Test Case	PASS
My Second Test Case As Keyword	PASS
Some Other Test Case	PASS
A Failed Test Case	FAIL

More information

Check the documentation