Merge 36b3fb5 into 46a9b2e

.eslintignore

@@ -1,3 +1,4 @@
-testdata
-node_modules
-coverage
+/testdata/
+/node_modules/
+/coverage/
+/vendor/

lib/parseCommandLineOptions.js

@@ -79,6 +79,12 @@ module.exports = function parseCommandLineOptions(argv) {
       type: 'boolean',
       default: false,
     })
+    .options('harfbuzz', {
+      type: 'boolean',
+      describe:
+        'Experimental: Use the harfbuzz subsetter instead of pyftsubset. Requires node.js 10+ for wasm support',
+      default: false,
+    })
     .options('silent', {
       alias: 's',
       describe: `Do not write anything to stdout`,
@@ -97,6 +103,13 @@ module.exports = function parseCommandLineOptions(argv) {
       type: 'boolean',
       default: false,
     })
+    .check(({ harfbuzz }) => {
+      if (harfbuzz && parseInt(process.versions.node) < 10) {
+        return 'The --harfbuzz option requires node.js 10 or above';
+      } else {
+        return true;
+      }
+    })
     .wrap(require('yargs').terminalWidth());
 
   const { _: inputFiles, ...rest } = yargs.argv;

lib/subfont.js

@@ -23,6 +23,7 @@ module.exports = async function subfont(
     recursive = false,
     fallbacks = true,
     dynamic = false,
+    harfbuzz = false,
   },
   console
 ) {
@@ -185,6 +186,7 @@ module.exports = async function subfont(
     subsetPerPage,
     formats,
     omitFallbacks: !fallbacks,
+    harfbuzz,
     dynamic,
     console,
   });

lib/subsetFonts.js

@@ -278,13 +278,18 @@ function getSubsetPromiseId(fontUsage, format) {
 async function getSubsetsForFontUsage(
   assetGraph,
   htmlAssetTextsWithProps,
-  formats
+  formats,
+  harfbuzz
 ) {
   let subsetLocalFont;
 
-  try {
-    subsetLocalFont = require('./subsetLocalFont');
-  } catch (err) {}
+  if (harfbuzz) {
+    subsetLocalFont = require('./subsetLocalFontWithHarfbuzz');
+  } else {
+    try {
+      subsetLocalFont = require('./subsetLocalFont');
+    } catch (err) {}
+  }
 
   const allFonts = [];
 
@@ -582,6 +587,7 @@ async function subsetFonts(
     onlyInfo,
     dynamic,
     console = global.console,
+    harfbuzz,
   } = {}
 ) {
   if (!validFontDisplayValues.includes(fontDisplay)) {
@@ -830,7 +836,12 @@ async function subsetFonts(
   }
 
   // Generate subsets:
-  await getSubsetsForFontUsage(assetGraph, htmlAssetTextsWithProps, formats);
+  await getSubsetsForFontUsage(
+    assetGraph,
+    htmlAssetTextsWithProps,
+    formats,
+    harfbuzz
+  );
 
   // Warn about missing glyphs
   const missingGlyphsErrors = [];

lib/subsetLocalFontWithHarfbuzz.js

@@ -0,0 +1,103 @@
+/* global WebAssembly */
+const fs = require('fs');
+const readFileAsync = require('util').promisify(fs.readFile);
+const _ = require('lodash');
+const wawoff2 = require('wawoff2');
+const woffTool = require('woff2sfnt-sfnt2woff');
+
+function HB_TAG(chunkName) {
+  return chunkName.split('').reduce(function (a, ch) {
+    return (a << 8) + ch.charCodeAt(0);
+  }, 0);
+}
+
+const loadAndInitializeHarfbuzz = _.once(async () => {
+  const {
+    instance: { exports },
+  } = await WebAssembly.instantiate(
+    await readFileAsync(require.resolve('harfbuzzjs/subset/hb-subset.wasm'))
+  );
+  exports.memory.grow(400); // each page is 64kb in size
+
+  const heapu8 = new Uint8Array(exports.memory.buffer);
+  return [exports, heapu8];
+});
+
+async function subset(originalFont, targetFormat, text) {
+  const [exports, heapu8] = await loadAndInitializeHarfbuzz();
+
+  const signature = originalFont.slice(0, 4).toString();
+  if (signature === 'wOFF') {
+    originalFont = woffTool.toSfnt(originalFont);
+  } else if (signature === 'wOF2') {
+    originalFont = await wawoff2.decompress(originalFont);
+  }
+
+  const fontBuffer = exports.malloc(originalFont.byteLength);
+  heapu8.set(new Uint8Array(originalFont), fontBuffer);
+
+  // Create the face
+  const blob = exports.hb_blob_create(
+    fontBuffer,
+    originalFont.byteLength,
+    2, // HB_MEMORY_MODE_WRITABLE
+    0,
+    0
+  );
+  const face = exports.hb_face_create(blob, 0);
+  exports.hb_blob_destroy(blob);
+
+  // Add glyph indices and subset
+  const glyphs = exports.hb_set_create();
+
+  for (let i = 0; i < text.length; i += 1) {
+    exports.hb_set_add(glyphs, text.charCodeAt(i));
+  }
+
+  const input = exports.hb_subset_input_create_or_fail();
+  const inputGlyphs = exports.hb_subset_input_unicode_set(input);
+  exports.hb_set_del(
+    exports.hb_subset_input_drop_tables_set(input),
+    HB_TAG('GSUB')
+  );
+  exports.hb_set_del(
+    exports.hb_subset_input_drop_tables_set(input),
+    HB_TAG('GPOS')
+  );
+  exports.hb_set_del(
+    exports.hb_subset_input_drop_tables_set(input),
+    HB_TAG('GDEF')
+  );
+
+  exports.hb_set_union(inputGlyphs, glyphs);
+  const subset = exports.hb_subset(face, input);
+
+  // Clean up
+  exports.hb_subset_input_destroy(input);
+
+  // Get result blob
+  const result = exports.hb_face_reference_blob(subset);
+
+  const offset = exports.hb_blob_get_data(result, 0);
+  let subsetFont = heapu8.slice(
+    offset,
+    offset + exports.hb_blob_get_length(result)
+  );
+
+  // Clean up
+  exports.hb_blob_destroy(result);
+  exports.hb_face_destroy(subset);
+
+  if (targetFormat === 'woff2') {
+    subsetFont = Buffer.from(await wawoff2.compress(subsetFont));
+  } else if (targetFormat === 'woff') {
+    subsetFont = woffTool.toWoff(subsetFont);
+  } else {
+    // targetFormat === 'truetype'
+    subsetFont = Buffer.from(subsetFont);
+  }
+  return subsetFont;
+}
+
+const limiter = require('p-limit')(1);
+module.exports = (...args) => limiter(() => subset(...args));

package.json

@@ -56,15 +56,19 @@
     "font-tracer": "^2.0.1",
     "fontkit": "^1.8.0",
     "gettemporaryfilepath": "^1.0.1",
+    "harfbuzzjs": "^0.1.3",
     "lines-and-columns": "^1.1.6",
     "lodash": "^4.17.15",
     "memoizesync": "^1.1.1",
+    "p-limit": "^3.0.2",
     "postcss": "^7.0.32",
     "postcss-value-parser": "^4.0.2",
     "pretty-bytes": "^5.1.0",
     "puppeteer-core": "^3.1.0",
     "specificity": "^0.4.1",
     "urltools": "^0.4.1",
+    "wawoff2": "^1.0.2",
+    "woff2sfnt-sfnt2woff": "^1.0.0",
     "yargs": "^15.4.0"
   },
   "devDependencies": {

test/parseCommandLineOptions.js

@@ -26,6 +26,7 @@ describe('parseCommandLineOptions', function () {
         recursive: true,
         fallbacks: false,
         dynamic: false,
+        harfbuzz: false,
       }
     );
   });

test/referenceImages.js

@@ -7,6 +7,7 @@ describe('reference images', function () {
     inlineSubsets: [false, true],
     omitFallbacks: [false, true],
     dynamic: [false, true],
+    harfbuzz: [false, true],
   })) {
     describe(`with ${Object.keys(options)
       .map((key) => `${key}: ${options[key]}`)