Since XPath doesn't have a way to escape quotes (or any other
character), we have to use a workaround by using concat() and switching
between quoting styles.
Before this change, adding a new logging message involving logging e.g. the
default duckduckgo setting value failed.
Now we basically use a black- instead of a whitelist and only fail if we get a
load status message for duckduckgo.
When a end-to-end test failed which would've marked an error message as
expected later in the test, seeing the teardown message about an unexpected
error being logged is really confusing.
In the long run, we should detect any accidental external accesses using
mitmproxy, as per #1282. In the meantime, we try to detect duckduckgo requests
being logged and fail the tests if that happens.
However, a duckduckgo URL is logged in fuzzy_url during startup/config init,
which is why we ignore it there.
This is a bit tricky since the test will actually run, but be marked as
skipped. The problem is we can't raise a pytest.skip.Exception during a test,
or it'll show up as an exception in a virtual Qt method.
Still this is better than nothing.
Using pytest.fail with pytrace=False hides the quteprocess output, which makes
it a lot harder to debug stuff.
This is because TestReport.longrepr is suddenly a string and we can't add infos
to it - see https://github.com/pytest-dev/pytest/issues/1316
The output was almost always useless without -v because the debug log wasn't
shown at all, only error/info.
Now we display a maximum of 50 lines (regardless of what level) without -v.
When using print and relying on pytest to capture it as stdout, we ran into
this pytest/pytest-qt issue:
https://github.com/pytest-dev/pytest-qt/issues/113
Now we use our own capturing mechanism instead, which also means we get nicer
output.
Fixes#1122.
webview uses QUrl to output the url and thus might convert some escape
sequences to their unicode characters. If we don't do the same
conversions here, we will "miss" the log message and fail the test.
