<div dir="ltr">On Thu, Jun 29, 2017 at 11:18 PM, Matthias Clasen <span dir="ltr"><<a href="mailto:matthias.clasen@gmail.com" target="_blank">matthias.clasen@gmail.com</a>></span> wrote:<br><div class="gmail_extra"><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr">I had another go at this here: <a href="https://github.com/matthiasclasen/cairo/tree/emoji-again" target="_blank">https://github.com/<wbr>matthiasclasen/cairo/tree/<wbr>emoji-again</a><br></div>
</blockquote></div><br></div><div class="gmail_extra">I've spent some more time on this branch. It now uses paint only for clusters that consists<br>purely of color glyphs, rewriting the inputs to remove handled clusters, and then falls through<br></div><div class="gmail_extra">the show_glyphs code paths.<br><br></div><div class="gmail_extra">It seems to work ok, at least as far as gedit / pango let me test easily. I'm less confident that<br>the CAIRO_TEXT_CLUSTER_FLAG_BACKWARD case is entirely correct, that is hard for me<br></div><div class="gmail_extra">to judge.<br><br></div><div class="gmail_extra">An unrelated observation: pango treats the gap between an emoji modifier base and a variant selector<br></div><div class="gmail_extra">as a cursor position. That looks like a bug.<br></div><div class="gmail_extra"> <br></div></div>