Skip to content

Feature/ccmp #688

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 22 commits into from
Apr 10, 2024
Merged
Show file tree
Hide file tree
Changes from 16 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
19 changes: 19 additions & 0 deletions src/bidi.js
Original file line number Diff line number Diff line change
Expand Up @@ -9,6 +9,8 @@ import arabicWordCheck from './features/arab/contextCheck/arabicWord.js';
import arabicSentenceCheck from './features/arab/contextCheck/arabicSentence.js';
import arabicPresentationForms from './features/arab/arabicPresentationForms.js';
import arabicRequiredLigatures from './features/arab/arabicRequiredLigatures.js';
import ccmpReplacementCheck from './features/ccmp/contextCheck/ccmpReplacement.js';
import ccmpReplacement from './features/ccmp/ccmpReplacementLigatures.js';
import latinWordCheck from './features/latn/contextCheck/latinWord.js';
import latinLigature from './features/latn/latinLigatures.js';
import thaiWordCheck from './features/thai/contextCheck/thaiWord.js';
Expand Down Expand Up @@ -42,6 +44,7 @@ Bidi.prototype.setText = function (text) {
* arabic sentence check for adjusting arabic layout
*/
Bidi.prototype.contextChecks = ({
ccmpReplacementCheck,
latinWordCheck,
arabicWordCheck,
arabicSentenceCheck,
Expand All @@ -64,6 +67,7 @@ function registerContextChecker(checkId) {
* tokenize text input
*/
function tokenizeText() {
registerContextChecker.call(this, 'ccmpReplacement');
registerContextChecker.call(this, 'latinWord');
registerContextChecker.call(this, 'arabicWord');
registerContextChecker.call(this, 'arabicSentence');
Expand Down Expand Up @@ -160,6 +164,18 @@ function applyArabicPresentationForms() {
}
}

/**
* Apply ccmp replacement
*/
function applyCcmpReplacement() {
checkGlyphIndexStatus.call(this);
const ranges = this.tokenizer.getContextRanges('ccmpReplacement');
for(let i = 0; i < ranges.length; i++) {
const range = ranges[i];
ccmpReplacement.call(this, range);
}
}

/**
* Apply required arabic ligatures
*/
Expand Down Expand Up @@ -223,6 +239,9 @@ Bidi.prototype.checkContextReady = function (contextId) {
* https://learn.microsoft.com/en-us/typography/opentype/spec/features_ae#tag-ccmp
*/
Bidi.prototype.applyFeaturesToContexts = function () {
if (this.checkContextReady('ccmpReplacement')) {
applyCcmpReplacement.call(this);
}
if (this.checkContextReady('arabicWord')) {
applyArabicPresentationForms.call(this);
applyArabicRequireLigatures.call(this);
Expand Down
14 changes: 13 additions & 1 deletion src/features/applySubstitution.js
Original file line number Diff line number Diff line change
Expand Up @@ -30,6 +30,15 @@ function chainingSubstitutionFormat3(action, tokens, index) {
for(let i = 0; i < action.substitution.length; i++) {
const subst = action.substitution[i];
const token = tokens[index + i];
if (Array.isArray(subst)) {
if (subst.length){
// TODO: replace one glyph with multiple glyphs
token.setState(action.tag, subst[0]);
} else {
token.setState('deleted', true);
}
continue;
}
token.setState(action.tag, subst);
}
}
Expand All @@ -50,14 +59,17 @@ function ligatureSubstitutionFormat1(action, tokens, index) {
}
}


/**
* Supported substitutions
*/
const SUBSTITUTIONS = {
11: singleSubstitutionFormat1,
12: singleSubstitutionFormat2,
63: chainingSubstitutionFormat3,
41: ligatureSubstitutionFormat1
41: ligatureSubstitutionFormat1,
51: chainingSubstitutionFormat3,
53: chainingSubstitutionFormat3
};

/**
Expand Down
47 changes: 47 additions & 0 deletions src/features/ccmp/ccmpReplacementLigatures.js
Original file line number Diff line number Diff line change
@@ -0,0 +1,47 @@
import { ContextParams } from '../../tokenizer.js';
import applySubstitution from '../applySubstitution.js';

// @TODO: use commonFeatureUtils.js for reduction of code duplication
// once #564 has been merged.

/**
* Update context params
* @param {any} tokens a list of tokens
* @param {number} index current item index
*/
function getContextParams(tokens, index) {
const context = tokens.map(token => token.activeState.value);
return new ContextParams(context, index || 0);
}

/**
* Apply Arabic required ligatures to a context range
* @param {ContextRange} range a range of tokens
*/
function ccmpReplacementLigatures(range) {
const script = 'delf';
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I guess this should be 'DFLT'?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is the only place in the codebase with 'delf'.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Sorry this made it through the review process, I didn't follow up because it was marked as resolved. Good catch, thanks!

@TonyJR could you comment on this please? I think it should be DFLT for the default script. And for the future, please let reviewers resolve the conversations. :)

const tag = 'ccmp';
let tokens = this.tokenizer.getRangeTokens(range);
let contextParams = getContextParams(tokens);
for(let index = 0; index < contextParams.context.length; index++) {
if (!this.query.getFeature({tag, script, contextParams})){
continue;
}
contextParams.setCurrentIndex(index);
let substitutions = this.query.lookupFeature({
tag, script, contextParams
});
if (substitutions.length) {
for(let i = 0; i < substitutions.length; i++) {
const action = substitutions[i];
applySubstitution(action, tokens, index);
}
contextParams = getContextParams(tokens);
}
}
}

export default ccmpReplacementLigatures;



12 changes: 12 additions & 0 deletions src/features/ccmp/contextCheck/ccmpReplacement.js
Original file line number Diff line number Diff line change
@@ -0,0 +1,12 @@
function ccmpReplacementStartCheck(contextParams) {
return contextParams.index === 0 && contextParams.context.length > 1;
}

function ccmpReplacementEndCheck(contextParams) {
return contextParams.index === contextParams.context.length - 1;
}

export default {
startCheck: ccmpReplacementStartCheck,
endCheck: ccmpReplacementEndCheck
};
117 changes: 117 additions & 0 deletions src/features/featureQuery.js
Original file line number Diff line number Diff line change
Expand Up @@ -203,6 +203,94 @@ function ligatureSubstitutionFormat1(contextParams, subtable) {
return null;
}

/**
* Handle context substitution - format 1
* @param {ContextParams} contextParams context params to lookup
*/
function contextSubstitutionFormat1(contextParams, subtable) {
let glyphId = contextParams.current;
let ligSetIndex = lookupCoverage(glyphId, subtable.coverage);
if (ligSetIndex === -1)
return null;
for (const ruleSet of subtable.ruleSets) {
for (const rule of ruleSet) {
let matched = true;
for (let i = 0; i < rule.input.length; i++) {
if (contextParams.lookahead[i] !== rule.input[i]){
matched = false;
break;
}
}
if (matched) {
let substitutions = [];
substitutions.push(glyphId);
for (let i = 0; i < rule.input.length; i++) {
substitutions.push(rule.input[i]);
}
const parser = (substitutions, lookupRecord)=>{
const {lookupListIndex,sequenceIndex} = lookupRecord;
const {subtables} = this.getLookupByIndex(lookupListIndex);
for (const subtable of subtables){
let ligSetIndex = lookupCoverage(substitutions[sequenceIndex], subtable.coverage);
if (ligSetIndex !== -1){
substitutions[sequenceIndex] = subtable.deltaGlyphId;
}
}
};

for (let i = 0; i < rule.lookupRecords.length; i++) {
const lookupRecord = rule.lookupRecords[i];
parser(substitutions, lookupRecord);
}

return substitutions;
}
}
}
return null;
}

/**
* Handle context substitution - format 3
* @param {ContextParams} contextParams context params to lookup
*/
function contextSubstitutionFormat3(contextParams, subtable) {
let substitutions = [];

for (let i = 0; i < subtable.coverages.length; i++){
const lookupRecord = subtable.lookupRecords[i];
const coverage = subtable.coverages[i];

let glyphIndex = contextParams.context[contextParams.index + lookupRecord.sequenceIndex];
let ligSetIndex = lookupCoverage(glyphIndex, coverage);
if (ligSetIndex === -1){
return null;
}
let lookUp = this.font.tables.gsub.lookups[lookupRecord.lookupListIndex];
for (let i = 0; i < lookUp.subtables.length; i++){
let subtable = lookUp.subtables[i];
let ligSetIndex = lookupCoverage(glyphIndex, subtable.coverage);
if (ligSetIndex === -1)
return null;
switch (lookUp.lookupType) {
case 1:{
let ligature = subtable.substitute[ligSetIndex];
substitutions.push(ligature);
break;
}
case 2:{
let ligatureSet = subtable.sequences[ligSetIndex];
substitutions.push(ligatureSet);
break;
}
default:
break;
}
}
}
return substitutions;
}

/**
* Handle decomposition substitution - format 1
* @param {number} glyphIndex glyph index
Expand Down Expand Up @@ -327,8 +415,17 @@ FeatureQuery.prototype.getLookupMethod = function(lookupTable, subtable) {
return glyphIndex => decompositionSubstitutionFormat1.apply(
this, [glyphIndex, subtable]
);
case '51':
return contextParams => contextSubstitutionFormat1.apply(
this, [contextParams, subtable]
);
case '53':
return contextParams => contextSubstitutionFormat3.apply(
this, [contextParams, subtable]
);
default:
throw new Error(
`substitutionType : ${substitutionType} ` +
`lookupType: ${lookupTable.lookupType} - ` +
`substFormat: ${subtable.substFormat} ` +
'is not yet supported'
Expand Down Expand Up @@ -435,6 +532,26 @@ FeatureQuery.prototype.lookupFeature = function (query) {
}));
}
break;
case '51':
substitution = lookup(contextParams);
if (Array.isArray(substitution) && substitution.length) {
substitutions.splice(currentIndex, 1, new SubstitutionAction({
id: 51,
tag: query.tag,
substitution
}));
}
break;
case '53':
substitution = lookup(contextParams);
if (Array.isArray(substitution) && substitution.length) {
substitutions.splice(currentIndex, 1, new SubstitutionAction({
id: 53,
tag: query.tag,
substitution: substitution,
}));
}
break;
}
contextParams = new ContextParams(substitutions, currentIndex);
if (Array.isArray(substitution) && !substitution.length) continue;
Expand Down
39 changes: 39 additions & 0 deletions test/emoji.js
Original file line number Diff line number Diff line change
@@ -0,0 +1,39 @@
import assert from 'assert';
import { parse } from '../src/opentype.js';
import { readFileSync } from 'fs';
const loadSync = (url, opt) => parse(readFileSync(url), opt);

describe('noto emoji with ccmp', () => {
let notoEmojiFont;
let substitution;
before(()=> {
notoEmojiFont = loadSync('./test/fonts/noto-emoji.ttf');
});

describe('ccmp features', () => {

it('shape emoji with sub_0', () => {
let options = {
kerning: true,
language: 'dflt',
features: [
{ script: 'DFLT', tags: ['ccmp'] },
]
};
let glyphIndexes = notoEmojiFont.stringToGlyphIndexes('👨‍👩‍👧‍👦👨‍👩‍👧',options);
assert.deepEqual(glyphIndexes, [1463,1462]);
});

it('shape emoji with sub_5', () => {
let options = {
kerning: true,
language: 'dflt',
features: [
{ script: 'DFLT', tags: ['ccmp'] },
]
};
let glyphIndexes = notoEmojiFont.stringToGlyphIndexes('🇺🇺',options);
assert.deepEqual(glyphIndexes, [1850]);
});
});
});
29 changes: 29 additions & 0 deletions test/featureQuery.js
Original file line number Diff line number Diff line change
Expand Up @@ -9,6 +9,8 @@ describe('featureQuery.js', function() {
let arabicFont;
let arabicFontChanga;
let latinFont;
let sub5Font;

let query = {};
before(function () {
/**
Expand All @@ -23,6 +25,11 @@ describe('featureQuery.js', function() {
*/
latinFont = loadSync('./test/fonts/FiraSansMedium.woff');
query.latin = new FeatureQuery(latinFont);
/**
* default
*/
sub5Font = loadSync('./test/fonts/sub5.ttf');
query.sub5 = new FeatureQuery(sub5Font);
});
describe('getScriptFeature', function () {
it('should return features indexes of a given script tag', function () {
Expand Down Expand Up @@ -144,6 +151,28 @@ describe('featureQuery.js', function() {
const substitutions = lookup(contextParams);
assert.deepEqual(substitutions, { ligGlyph: 1145, components: [76]});
});
it('should parse multiple glyphs -ligature substitution format 1 (51)', function () {
const feature = query.sub5.getFeature({tag: 'ccmp', script: 'DFLT'});
const featureLookups = query.sub5.getFeatureLookups(feature);
const lookupSubtables = query.sub5.getLookupSubtables(featureLookups[0]);
const substitutionType = query.sub5.getSubstitutionType(featureLookups[0], lookupSubtables[0]);
assert.equal(substitutionType, 51);
const lookup = query.sub5.getLookupMethod(featureLookups[0], lookupSubtables[0]);
let contextParams = new ContextParams([1, 88, 1], 0);
const substitutions = lookup(contextParams);
assert.deepEqual(substitutions, [85, 88, 85]);
});
it('should parse multiple glyphs -ligature substitution format 3 (53)', function () {
const feature = query.sub5.getFeature({tag: 'ccmp', script: 'DFLT'});
const featureLookups = query.sub5.getFeatureLookups(feature);
const lookupSubtables = query.sub5.getLookupSubtables(featureLookups[1]);
const substitutionType = query.sub5.getSubstitutionType(featureLookups[1], lookupSubtables[0]);
assert.equal(substitutionType, 53);
const lookup = query.sub5.getLookupMethod(featureLookups[0], lookupSubtables[0]);
let contextParams = new ContextParams([2, 3], 0);
const substitutions = lookup(contextParams);
assert.deepEqual(substitutions, [54, 54]);
});
it('should decompose a glyph - multiple substitution format 1 (21)', function () {
const feature = query.arabic.getFeature({tag: 'ccmp', script: 'arab'});
const featureLookups = query.arabic.getFeatureLookups(feature);
Expand Down
Loading