Add support for Latin-1, Cyrillic, and CJK characters inside \text{} (#508)

Summary: This diff provides support for Latin-1, Cyrillic, and CJK characters inside \text{} groups. For Latin-1 and Cyrillic characters we use glyph metrics from a glyph from Basic Latin that has roughly the same bounding box. We use the metrics for a capital 'M' to approximate the full-width CJK characters. Half-width characters are not supported yet. Test Plan: - make test - make screenshots Reviewers: emily
2025-10-08 12:38:39 +00:00 · 2016-08-01 17:51:40 -07:00
parent 92bbbffbc8
commit ec62ec39d8
15 changed files with 348 additions and 4 deletions
--- a/test/screenshotter/images/Unicode-chrome.png
+++ b/test/screenshotter/images/Unicode-chrome.png
--- a/test/screenshotter/images/Unicode-firefox.png
+++ b/test/screenshotter/images/Unicode-firefox.png
--- a/test/screenshotter/ss_data.yaml
+++ b/test/screenshotter/ss_data.yaml
@@ -114,6 +114,7 @@ Symbols1: |
    \maltese\degree\pounds\$
    \text{\maltese\degree}
 Text: \frac{a}{b}\text{c~ {ab} \ e}+fg
+Unicode: \begin{matrix}\text{ÀàÇçÉéÏïÖöÛû} \\ \text{БГДЖЗЙЛФЦШЫЮЯ} \\ \text{여보세요} \\ \text{私はバナナです} \end{matrix}
 UnsupportedCmds:
    tex: \err\,\frac\fracerr3\,2^\superr_\suberr\,\sqrt\sqrterr
    noThrow: 1
--- a/test/screenshotter/test.html
+++ b/test/screenshotter/test.html
@@ -11,6 +11,20 @@
      body {
        font-family: "DejaVu Serif",serif;
      }
+      @font-face {
+        font-family: "Mincho";
+        src: url("unicode-fonts/mincho/font_1_honokamin.ttf") format("truetype");
+      }
+      @font-face {
+        font-family: "Batang";
+        src: url("unicode-fonts/batang/batang.ttf") format("truetype");
+      }
+      .katex .cjk_fallback {
+          font-family: "Mincho",serif;
+      }
+      .katex .hangul_fallback {
+          font-family: "Batang",serif;
+      }
    </style>
  </head>
  <body>
--- a/test/unicode-spec.js
+++ b/test/unicode-spec.js
@@ -0,0 +1,103 @@
+/* eslint max-len:0 */
+/* global beforeEach: false */
+/* global jasmine: false */
+/* global expect: false */
+/* global it: false */
+/* global describe: false */
+var ParseError = require("../src/ParseError");
+var parseTree = require("../src/parseTree");
+var Settings = require("../src/Settings");
+
+var defaultSettings = new Settings({});
+
+var parseAndSetResult = function(expr, result, settings) {
+    try {
+        return parseTree(expr, settings || defaultSettings);
+    } catch (e) {
+        result.pass = false;
+        if (e instanceof ParseError) {
+            result.message = "'" + expr + "' failed " +
+                "parsing with error: " + e.message;
+        } else {
+            result.message = "'" + expr + "' failed " +
+                "parsing with unknown error: " + e.message;
+        }
+    }
+};
+
+describe("unicode", function() {
+    beforeEach(function() {
+        jasmine.addMatchers({
+
+            toParse: function() {
+                return {
+                    compare: function(actual, settings) {
+                        var usedSettings = settings ? settings : defaultSettings;
+
+                        var result = {
+                            pass: true,
+                            message: "'" + actual + "' succeeded parsing",
+                        };
+                        parseAndSetResult(actual, result, usedSettings);
+                        return result;
+                    },
+                };
+            },
+
+            toNotParse: function() {
+                return {
+                    compare: function(actual, settings) {
+                        var usedSettings = settings ? settings : defaultSettings;
+
+                        var result = {
+                            pass: false,
+                            message: "Expected '" + actual + "' to fail " +
+                                "parsing, but it succeeded",
+                        };
+
+                        try {
+                            parseTree(actual, usedSettings);
+                        } catch (e) {
+                            if (e instanceof ParseError) {
+                                result.pass = true;
+                                result.message = "'" + actual + "' correctly " +
+                                    "didn't parse with error: " + e.message;
+                            } else {
+                                result.message = "'" + actual + "' failed " +
+                                    "parsing with unknown error: " + e.message;
+                            }
+                        }
+
+                        return result;
+                    },
+                };
+            },
+        });
+    });
+
+    it("should parse Latin-1 inside \\text{}", function() {
+        expect('\\text{ÀàÇçÉéÏïÖöÛû}').toParse();
+    });
+
+    it("should not parse Latin-1 outside \\text{}", function() {
+        expect('ÀàÇçÉéÏïÖöÛû').toNotParse();
+    });
+
+    it("should parse Cyrillic inside \\text{}", function() {
+        expect('\\text{БГДЖЗЙЛФЦШЫЮЯ}').toParse();
+    });
+
+    it("should not parse Cyrillic outside \\text{}", function() {
+        expect('БГДЖЗЙЛФЦШЫЮЯ').toNotParse();
+    });
+
+    it("should parse CJK inside \\text{}", function() {
+        expect('\\text{私はバナナです}').toParse();
+        expect('\\text{여보세요}').toParse();
+    });
+
+    it("should not parse CJK outside \\text{}", function() {
+        expect('私はバナナです。').toNotParse();
+        expect('여보세요').toNotParse();
+    });
+});