[XeTeX] Seeking short examples of complex renderings

Khaled Hosny khaledhosny at eglug.org
Fri Dec 6 12:18:50 CET 2013

On Fri, Dec 06, 2013 at 07:55:07PM +0900, Simon Cozens wrote:
> Hello XeTeXers,
> Sorry for a not-entirely-XeTeX-related request but I think there may be some
> merit in it for XeTeX in the future.
> I have recently been toying around implementing my own layout engine, and
> have started to check that it works nicely with non-Roman scripts. This has
> already thrown up a few bugs in pango (which I'm using to do the shaping),
> and that's just with scripts that I can read. I am sure there are other
> problems in scripts I can't read.
> So I thought it would be a useful thing for people like
> pango/xetex/graphite/harfbuzz/other layout and rendering tool developers to
> have a visual test suite, a collection of short strings which stress-test
> their engines in interesting ways: placement of composing characters,
> mandatory ligatures, mixed LTR/RTL, that sort of thing.

Both Pango and XeTeX use HarfBuzz which in turn can use Graphite, so I
think HarfBuzz and Graphite are the proper places for these tests (and
both already have test suites in place), so I’d forward this to the
respective developers.

> I have started putting a collection together but my own knowledge and
> experience is pretty limited. If you can contribute some short Unicode
> strings from a language you know which show an interesting rendering
> feature, I hope this will be something that can be beneficial to the text
> layout community as a whole.
> The test suite at the moment is at
> http://simoncozens.github.io/visual-testsuite/testsuite.html
> and you can contribute via github at
> http://github.com/simoncozens/visual-testsuite

Instead of comparing images, which can be affected by things unrelated
to layout like hinting, it would be better to compare glyph IDs with or
without glyph positioning, check HarfBuzz and Graphite test suites for


More information about the XeTeX mailing list