#34 Better CSS in OT test

[Christoph Päper]

The OpenType test should preferably use the high-level CSS properties where available. It still doesn’t work anywhere, though.

  @font-feature-values Junicode {
               @styleset {
                 nordic: 1;
                 insular: 2;
                 romanhigh: 4;
                 romanlow: 5;
                 enlarged: 6;
                 deleted: 7;
                 alt-yogh: 8;
                 retired: 9;
                 mufi-hook: 14;
                 e-caudata: 15;
                 old-style-punct: 18;
                 latin2gothic: 19;
                 gothic2latin: 20;
                 }
               }

  .insular   { -moz-font-feature-settings: "ss02=1";
               -webkit-font-feature-settings: "ss02=1";
               font-feature-settings: "ss02=1";
               font-variant-alternates: styleset(insular);
               }
  .enlarged  { -moz-font-feature-settings: "ss06=1";
               -webkit-font-feature-settings: "ss06=1";
               font-feature-settings: "ss06=1";
               font-variant-alternates: styleset(enlarged);
               }
  .romanhigh { -moz-font-feature-settings: "ss04=1";
               -webkit-font-feature-settings: "ss04=1";
               font-feature-settings: "ss04=1";
               font-variant-alternates: styleset(romanhigh);
               }
  .romanlow  { -moz-font-feature-settings: "ss05=1";
               -webkit-font-feature-settings: "ss05=1";
               font-feature-settings: "ss05=1";
               font-variant-alternates: styleset(romanlow);
               }
  .smallcap  { -moz-font-feature-settings: "smcp=1";
               -webkit-font-feature-settings: "smcp=1";
               font-feature-settings: "smcp=1";
               //font-variant: small-caps;
               font-variant-caps: small-caps;
               }
  .disclig   { -moz-font-feature-settings: "dlig=1";
               -webkit-font-feature-settings: "dlig=1";
               font-feature-settings: "dlig=1";
               font-variant-ligatures: discretionary-ligatures;
               }
  .histlig   { -moz-font-feature-settings: "hlig=1";
               -webkit-font-feature-settings: "hlig=1";
               font-feature-settings: "hlig=1"; 
               font-variant-ligatures: historical-ligatures;
               }
  .underdot  { -moz-font-feature-settings: "ss07=1";
               -webkit-font-feature-settings: "ss07=1";
               font-feature-settings: "ss07=1"; 
               font-variant-alternates: styleset(deleted);
               }
  .oldnum    { -moz-font-feature-settings: "onum=1";
               -webkit-font-feature-settings: "onum=1";
               font-feature-settings: "onum=1"; 
               font-variant-numeric: oldstyle-nums;
               }
  .mirrored   { -moz-font-feature-settings: "rtlm=1";
               -webkit-font-feature-settings: "rtlm=1";
               font-feature-settings: "rtlm=1"; 
               /* that’s one OT feature that is not directly accessible through CSS Fonts 3 */
               }
  .historic  { -moz-font-feature-settings: "hist=1";
               -webkit-font-feature-settings: "hist=1";
               font-feature-settings: "hist=1"; 
               font-variant-alternates: historical-forms;
               }
  .rtl       {direction: rtl; unicode-bidi:bidi-override; }

[Peter Baker]

This test used to work. I've found several bugs in Junicode's OT, which I hope will fix the problem. Will test before next release.

[Christoph Päper]

The test page ist still using only font-feature-settings (also prefixed with -webkit- and -moz-) instead of the appropriate high-level properties which are now (finally) supported by Safari.