RE: Allowing Developers to request specific codepoints to load

I agree this would be a very useful feature [to allow requesting arbitrary / specific set of characters that may not be part of the actual content] but I think this is an orthogonal feature that should be independent from either patch subset or range request methods. I would envision it as a CSS feature that defines a specific set of codepoints/glyphs, adding it to what would be defined by clients.

 

I.e., I am not sure this should be defined or discussed as part of the PFE spec, but rather as a generic mechanism to instruct PFE clients to inject codepoints into incremental subset requests. The PFE mechanism would remain content independent, the request is generated by clients and what input/decisions they used to compose the request isn’t relevant for PFE.

 

Thanks,

Vlad

 

 

From: Garret Rieger [mailto:grieger@google.com] 
Sent: Friday, March 12, 2021 7:40 PM
To: w3c-webfonts-wg (public-webfonts-wg@w3.org) <public-webfonts-wg@w3.org>
Subject: Allowing Developers to request specific codepoints to load

 

A comment came up on the CSS issue for adding incremental transfer to supports asking if we had a mechanism to allow developers to request a specific set of characters to load via incremental transfer: https://github.com/w3c/csswg-drafts/issues/6063#issuecomment-797426308

 

This would likely be a useful use case to support. I think the easiest way to add this would be through the javascript font loading api. Currently in the load() <https://developer.mozilla.org/en-US/docs/Web/API/FontFaceSet/load>  call you can specify text. Which is used to pick the relevant unicode ranges. It's possible this could be extended to also pass the text down to the actual font loading call where the incxfer client would ensure those codepoints are in the set requested.

 

Thoughts?