feat: text-mode cedilla accent via \c (#3036)

* feat: text-mode cedilla accent via \c

Add text-mode accent function `\c` and corresponding Unicode support.

Part of #638

Co-authored-by: Kevin Barabash <kevinb@khanacademy.org>

* Update screenshots

Co-authored-by: Kevin Barabash <kevinb@khanacademy.org>

src/Parser.js

@@ -951,7 +951,8 @@ export default class Parser {
                 if (!unicodeAccents[accent]) {
                     throw new ParseError(`Unknown accent ' ${accent}'`, nucleus);
                 }
-                const command = unicodeAccents[accent][this.mode];
+                const command = unicodeAccents[accent][this.mode] ||
+                    unicodeAccents[accent].text;
                 if (!command) {
                     throw new ParseError(
                         `Accent ${accent} unsupported in ${this.mode} mode`,

src/fontMetrics.js

@@ -108,11 +108,9 @@ import metricMap from "./fontMetricsData";
 const extraCharacterMap = {
     // Latin-1
     'Å': 'A',
-    'Ç': 'C',
     'Ð': 'D',
     'Þ': 'o',
     'å': 'a',
-    'ç': 'c',
     'ð': 'd',
     'þ': 'o',
 

src/functions/accent.js

@@ -74,10 +74,14 @@ export const htmlBuilder: HtmlBuilderSupSub<"accent"> = (grp, options) => {
         // TODO(emily): Find a better way to get the skew
     }
 
+    const accentBelow = group.label === "\\c";
+
     // calculate the amount of space between the body and the accent
-    let clearance = Math.min(
-        body.height,
-        options.fontMetrics().xHeight);
+    let clearance = accentBelow
+        ? body.height + body.depth
+        : Math.min(
+            body.height,
+            options.fontMetrics().xHeight);
 
     // Build the accent
     let accentBody;
@@ -100,6 +104,9 @@ export const htmlBuilder: HtmlBuilderSupSub<"accent"> = (grp, options) => {
             // shift the accent over to a place we don't want.
             accent.italic = 0;
             width = accent.width;
+            if (accentBelow) {
+                clearance += accent.depth;
+            }
         }
 
         accentBody = buildCommon.makeSpan(["accent-body"], [accent]);
@@ -244,7 +251,7 @@ defineFunction({
     type: "accent",
     names: [
         "\\'", "\\`", "\\^", "\\~", "\\=", "\\u", "\\.", '\\"',
-        "\\r", "\\H", "\\v", "\\textcircled",
+        "\\c", "\\r", "\\H", "\\v", "\\textcircled",
     ],
     props: {
         numArgs: 1,

src/symbols.js

@@ -706,6 +706,7 @@ defineSymbol(text, main, accent, "\u02dc", "\\~"); // tilde
 defineSymbol(text, main, accent, "\u02c9", "\\="); // macron
 defineSymbol(text, main, accent, "\u02d8", "\\u"); // breve
 defineSymbol(text, main, accent, "\u02d9", "\\."); // dot above
+defineSymbol(text, main, accent, "\u00b8", "\\c"); // cedilla
 defineSymbol(text, main, accent, "\u02da", "\\r"); // ring above
 defineSymbol(text, main, accent, "\u02c7", "\\v"); // caron
 defineSymbol(text, main, accent, "\u00a8", '\\"'); // diaresis
@@ -875,7 +876,7 @@ for (let i = 0; i < 10; i++) {
 // but they are not actually in the font, nor are they supported by the
 // Unicode accent mechanism, so they fall back to Times font and look ugly.
 // TODO(edemaine): Fix this.
-export const extraLatin = "\u00c7\u00d0\u00de\u00e7\u00fe";
+export const extraLatin = "\u00d0\u00de\u00fe";
 for (let i = 0; i < extraLatin.length; i++) {
     const ch = extraLatin.charAt(i);
     defineSymbol(math, main, mathord, ch, ch);

src/unicodeAccents.js

@@ -14,4 +14,5 @@ module.exports = {
     '\u0307': {text: '\\.', math: '\\dot'},
     '\u030a': {text: '\\r', math: '\\mathring'},
     '\u030b': {text: '\\H'},
+    '\u0327': {text: '\\c'},
 };

test/unicode-spec.js

@@ -9,6 +9,12 @@ describe("unicode", function() {
             .toBuild();
     });
 
+    it("should build Latin-1 inside \\text{} like accent commands", function() {
+        expect`\text{ÀÁÂÃÄÅÈÉÊËÌÍÎÏÑÒÓÔÕÖÙÚÛÜÝàáâãäåèéêëìíîïñòóôõöùúûüýÿÇç}`
+            .toParseLike`\text{\`A\'A\^A\~A\"A\r A\`E\'E\^E\"E\`I\'I\^I\"I\~N\`O\'O\^O\~O\"O\`U\'U\^U\"U\'Y\`a\'a\^a\~a\"a\r a\`e\'e\^e\"e\`ı\'ı\^ı\"ı\~n\`o\'o\^o\~o\"o\`u\'u\^u\"u\'y\"y\c C\c c}`;
+        // TODO(edemaine): A few characters don't have analogs yet.
+    });
+
     it("should not parse Latin-1 outside \\text{} with strict", function() {
         const chars = 'ÀÁÂÃÄÅÈÉÊËÌÍÎÏÑÒÓÔÕÖÙÚÛÜÝàáâãäåèéêëìíîïñòóôõöùúûüýÿÇÐÞçþ';
         for (const ch of chars) {