Closed camertron closed 6 years ago
Should we consider re-signing them with some TTFunk private key?
We'd have to distribute the private key. So what's the point?
Is there any use of an empty DSIG table? Can we skip it entirely?
We'd have to distribute the private key. So what's the point?
Not necessarily. It could be kept somewhere only committers have access to, although that's not really in the spirit of open-source. I'm in favor of not worrying about it.
Is there any use of an empty DSIG table? Can we skip it entirely?
Font validation tools like Font-Validator, certain versions of IE and I believe other browsers, and the Font Book application that comes with Mac OS will complain if the DSIG table isn't present.
It could be kept somewhere only committers have access to
Do you mean for tests?
Do you mean for tests?
Well, I meant in general but now that I give it more thought that doesn't make any sense lol. We'd have to distribute it with the gem which anyone can look at. So never mind, I'm crazy.
Some part of Microsoft GUI depends on the presence of DSIG (even a dummy empty one) to switch on opentype features. This is annoying and undocumented, but that's why it is needed.
This pull request is part of a larger effort to bring OTF support to TTFunk. See https://github.com/prawnpdf/ttfunk/issues/53 for details.
The only issue of note in this PR is that re-encoding the DSIG table will effectively remove all signatures from the font. Should we consider re-signing them with some TTFunk private key? My inclination is to not worry about it, at least for now.