ParsertTest.ioReaderTest() depends system settings, leads to failure

[xzel23]

Steps to reproduce the problem (provide example input):

• clone the commonmark repository on a machine with a platform dependent standard charset that is not UTF-8. Note that the default charset used by InputStreamReader depends on different settings and so this may not be reproducible without effort. The test passed on my Mac (english locale) and failed on Windows 7 (english version with german locale).

• compile and run tests

Expected behavior:

• tests should pass

Actual behavior:

• Failure in ParserTest

  org.commonmark.test.ParserTest > ioReaderTest FAILED
      org.junit.ComparisonFailure at ParserTest.java:45
  
  1357 tests completed, 1 failed, 1 skipped
  

Cause/Fix: ParserTest.ioReaderTest() creates an InputStreamReader that uses platform standard encoding and compares to results obtained when using UTF-8 charset (in TestResources.readAsString(URL)):

    @Test
    public void ioReaderTest() throws IOException {
        ...
        try (InputStreamReader reader = new InputStreamReader(input1)) {

should be (requires Java 7, or use Charset.forName("UTF-8") as in TestResources.java):

    @Test
    public void ioReaderTest() throws IOException {
        ...
        try (InputStreamReader reader = new InputStreamReader(input1, StandardCharsets.UTF_8)) {

Caveat: this happened while I was working on my own fork where I am working towards solving #125, not on master. However I only changed build files so far and this is quite obvious.

[robinst]

Thanks, fixed!