Bram Stein
Bram Stein
I guess :smile: You would need a couple more tools to complete the package though: something for subsetting and a tool to generate EOT files (for IE6-8).
There is ttf2eot, but it doesn't do mtx compression. For subsetting there is Google's [sfntly](https://code.google.com/p/sfntly/), but I still need to figure out how to compile Java in a homebrew formula.
I've added both to the webfonttools repository now.
Awesome, that would work for us. I'll leave the bike shedding to others, but I'd be happy to open a PR to implement it.
@m4rc1e Thanks for letting us know. We'll give it a try (currently upgrading to the latest gftools), and then we can test.
You could do it without the canvas tag by choosing a fallback font with known dimensions. You can construct a special font that supports the whole unicode range and falls...
I think the performance argument is not a very good one. It'll get slow with very large paragraphs, but there are ways around that (splitting the paragraph, falling back to...
This is on purpose. I needed to transfer the JSON representation over the wire and left out anything that wasn't defined. (Assuming the receiving end knows about this convention.)
I prefer not to introduce a dependency on jQuery. Wouldn't modifying the event matcher to test for `window.event ? event.which : event.keyCode` be enough? I hate to fix this problem...
That indeed sounds like the right approach. @papandreou could you do a pull request for those failing tests. Would be nice to have the tests before we start working on...