From: Dmitry Timoshkov [mailto:dmitry@codeweavers.com]
Greg Geldorp ggeldorp@vmware.com wrote:
I can easily imagine fonts for non-Western scripts not adhering to the hard-coded metrics in the test, which is why I thought skipping them on non-English systems might be ok. Of course, if you have an idea how to improve the test I'm all ears, after all the goal is to run as many tests as possible, not skip them.
Probably adding a separate set of data with FS_ARABIC for Small Fonts could help (similar to FS_JISJAPAN ones).
Turns out it isn't quite as easy as adding a separate set of test data, adding test data doesn't influence whether existing tests are run or what their expected values are. In other words, after adding FS_ARBIC test data the tests for FS_LATIN1 still fail. I've added a skip_lang_id member to the test data to skip some of the tests on Arabic machines. Not ideal (blacklisting instead of whitelisting), but I guess it will have to do for now.
Greg.