JSON Archive file balloons with repeated Test Runs

[realulim]

The following test case shows that if I run a test 5 times in a row, there will be 31 tests in the report instead of 5.

    @Test
    public void countAppendedTests() throws IOException {
        ExtentReports extent = new ExtentReports();
        int expectedTestCount = 5;
        String testName = "Unique Test Name " + UUID.randomUUID().toString();
        for (int i = 0; i < expectedTestCount; i++) {
            String reportPath = "target/append/index.html";
            String jsonPath = "target/append/index.json";
            ExtentSparkReporter sparkReporter = new ExtentSparkReporter(reportPath);
            JsonFormatter jsonReporter = new JsonFormatter(jsonPath);
            extent.createDomainFromJsonArchive(jsonPath);
            extent.attachReporter(jsonReporter, sparkReporter);
            extent.createTest(testName).info("some text");
            extent.flush();
        }
        List<com.aventstack.extentreports.model.Test> appendedTests = null;
        appendedTests = extent.getReport().getTestList().stream().filter(t -> {
            return t.getName().equals(testName);
        }).collect(Collectors.toList());
        Assert.assertEquals(appendedTests.size(), expectedTestCount);
    }

[anshooarora]

This is expected:

• 1st run: 1 test as there is nothing to create a domain from external file
• 2nd run: 3 tests
  • 1 test in memory from previous run
  • 1 test parsed from external file
  • 1 new test created
• 3rd run: 7 tests
  • 3 tests from previous run (in memory)
  • 3 tests parsed from file
  • 1 new test created
    ...
• 4th run: 15 tests
  ...
• 5th run: 31 tests

createDomainFromJsonArchive will read from an external file and update the existing collection of tests, so the above test isn't valid. expectedTestCount should be 31.

[realulim]

The javadocs say that this is the equivalent of the former "appendExisting" option. IIRC that option appended subsequent test runs to the report. This is what I expected here as well. I don't understand the use case you described.