harmony-v3.0.26

2017-07-23 16:54:46 +08:00
parent 91f8b57b3e 98b850580b
commit 1e0c7d2bc5
5 changed files with 123 additions and 64 deletions
--- a/lib/output.js
+++ b/lib/output.js
@@ -54,7 +54,6 @@ function OutputStream(options) {

    options = defaults(options, {
        ascii_only       : false,
-        ascii_identifiers: undefined,
        beautify         : false,
        bracketize       : false,
        comments         : false,
@@ -78,9 +77,6 @@ function OutputStream(options) {
        wrap_iife        : false,
    }, true);

-    if (typeof options.ascii_identifiers === 'undefined')
-        options.ascii_identifiers = options.ascii_only;
-
    if (options.shorthand === undefined)
        options.shorthand = options.ecma > 5;

@@ -118,20 +114,16 @@ function OutputStream(options) {
    var current_pos = 0;
    var OUTPUT = "";

-    function to_ascii(str, identifier) {
-        return str.replace(/[\ud800-\udbff][\udc00-\udfff]|[\u0000-\u001f\u007f-\uffff]/g, function(ch) {
-            var code = get_full_char_code(ch, 0).toString(16);
-
-            if ((identifier && code.length === 1 && options.ecma >= 6) || code.length > 4) {
-                if (options.ecma < 6) {
-                    if (identifier) {
-                        return ch; // no \u{} support
-                    }
-                    return "\\u" + ch.charCodeAt(0).toString(16) + "\\u"
-                        + ch.charCodeAt(1).toString(16);
-                }
+    var to_utf8 = options.ascii_only ? function(str, identifier) {
+        if (options.ecma >= 6) {
+            str = str.replace(/[\ud800-\udbff][\udc00-\udfff]/g, function(ch) {
+                var code = get_full_char_code(ch, 0).toString(16);
                return "\\u{" + code + "}";
-            } else if (code.length <= 2 && !identifier) {
+            });
+        }
+        return str.replace(/[\u0000-\u001f\u007f-\uffff]/g, function(ch) {
+            var code = ch.charCodeAt(0).toString(16);
+            if (code.length <= 2 && !identifier) {
                while (code.length < 2) code = "0" + code;
                return "\\x" + code;
            } else {
@@ -139,6 +131,12 @@ function OutputStream(options) {
                return "\\u" + code;
            }
        });
+    } : function(str) {
+        return str.replace(/[\ud800-\udbff](?![\udc00-\udfff])/g, function(ch) {
+            return "\\u" + ch.charCodeAt(0).toString(16);
+        }).replace(/(^|[^\ud800-\udbff])([\udc00-\udfff])/g, function(match, prefix, ch) {
+            return prefix + "\\u" + ch.charCodeAt(0).toString(16);
+        });
    };

    function make_string(str, quote) {
@@ -172,7 +170,7 @@ function OutputStream(options) {
        function quote_template() {
            return '`' + str.replace(/`/g, '\\`') + '`';
        }
-        if (options.ascii_only) str = to_ascii(str);
+        str = to_utf8(str);
        if (quote === "`") return quote_template();
        switch (options.quote_style) {
          case 1:
@@ -198,8 +196,7 @@ function OutputStream(options) {

    function make_name(name) {
        name = name.toString();
-        if (options.ascii_identifiers)
-            name = to_ascii(name, true);
+        name = to_utf8(name, true);
        return name;
    };

@@ -461,7 +458,7 @@ function OutputStream(options) {
        last            : function() { return last },
        semicolon       : semicolon,
        force_semicolon : force_semicolon,
-        to_ascii        : to_ascii,
+        to_utf8         : to_utf8,
        print_name      : function(name) { print(make_name(name)) },
        print_string    : function(str, quote, escape_directive) {
            var encoded = encode_string(str, quote);
@@ -1713,9 +1710,7 @@ function OutputStream(options) {
        if (regexp.raw_source) {
            str = "/" + regexp.raw_source + str.slice(str.lastIndexOf("/"));
        }
-        if (output.option("ascii_only")) {
-            str = output.to_ascii(str);
-        }
+        str = output.to_utf8(str);
        output.print(str);
        var p = output.parent();
        if (p instanceof AST_Binary && /^in/.test(p.operator) && p.left === self)
--- a/package.json
+++ b/package.json
@@ -4,7 +4,7 @@
  "homepage": "https://github.com/mishoo/UglifyJS2/tree/harmony",
  "author": "Mihai Bazon <mihai.bazon@gmail.com> (http://lisperator.net/)",
  "license": "BSD-2-Clause",
-  "version": "3.0.25",
+  "version": "3.0.26",
  "engines": {
    "node": ">=0.8.0"
  },
@@ -26,12 +26,12 @@
    "LICENSE"
  ],
  "dependencies": {
-    "commander": "~2.9.0",
+    "commander": "~2.11.0",
    "source-map": "~0.5.1"
  },
  "devDependencies": {
-    "acorn": "~5.0.3",
-    "mocha": "~2.3.4",
+    "acorn": "~5.1.1",
+    "mocha": "~3.4.2",
    "semver": "~5.3.0"
  },
  "scripts": {
--- a/test/compress/unicode.js
+++ b/test/compress/unicode.js
@@ -43,15 +43,6 @@ unicode_string_literals: {
    expect_exact: 'var a="6 length unicode character: \\u{101111}";'
 }

-// Don't escape identifiers below es6 (or in this case double escaped in expect_exact)
-unicode_output_es5_surrogates: {
-    beautify = {ascii_only: true, ecma: 5}
-    input: {
-        var \u{10000} = "6 length unicode character: \u{10FFFF}";
-    }
-    expect_exact: 'var \u{10000}="6 length unicode character: \\udbff\\udfff";'
-}
-
 check_escape_style: {
    beautify = {ascii_only: true, ecma: 6}
    input: {
@@ -65,20 +56,6 @@ check_escape_style: {
    expect_exact: 'var a="\\x01";var \\ua0081="\\x10";var \\u0100="\\u0100";var \\u1000="\\u1000";var \\u{10000}="\\u{10000}";var \\u{2f800}="\\u{100000}";'
 }

-// Don't escape identifiers below es6, no escaped identifiers support and no \u{} syntax
-check_escape_style_es5: {
-    beautify = {ascii_only: true, ecma: 5}
-    input: {
-        var a = "\x01";
-        var \ua0081 = "\x10"; // \u0081 only in ID_Continue
-        var \u0100 = "\u0100";
-        var \u1000 = "\u1000";
-        var \u{10000} = "\u{10000}"; // Identifier won't be escaped in es 5.1
-        var \u{2f800} = "\u{100000}"; // Same
-    }
-    expect_exact: 'var a="\\x01";var \\ua0081="\\x10";var \\u0100="\\u0100";var \\u1000="\\u1000";var \ud800\udc00="\\ud800\\udc00";var \ud87e\udc00="\\udbc0\\udc00";'
-}
-
 ID_continue_with_surrogate_pair: {
    beautify = {ascii_only: true, ecma: 6}
    input: {
@@ -103,18 +80,42 @@ non_escape_2_non_escape: {
    expect_exact: 'var µþ="µþ";'
 }

-non_escape_2_half_escape1: {
-    beautify = {ascii_only: false, ascii_identifiers: true, ecma: 6}
-    input: {
-        var µþ = "µþ";
+issue_2242_1: {
+    beautify = {
+        ascii_only: false,
    }
-    expect_exact: 'var \\u00b5\\u00fe="µþ";'
+    input: {
+        console.log("\ud83d", "\ude00", "\ud83d\ude00", "\ud83d@\ude00");
+    }
+    expect_exact: 'console.log("\\ud83d","\\ude00","\ud83d\ude00","\\ud83d@\\ude00");'
 }

-non_escape_2_half_escape2: {
-    beautify = {ascii_only: true, ascii_identifiers: false, ecma: 6}
-    input: {
-        var µþ = "µþ";
+issue_2242_2: {
+    beautify = {
+        ascii_only: true,
    }
-    expect_exact: 'var µþ="\\xb5\\xfe";'
-}
+    input: {
+        console.log("\ud83d", "\ude00", "\ud83d\ude00", "\ud83d@\ude00");
+    }
+    expect_exact: 'console.log("\\ud83d","\\ude00","\\ud83d\\ude00","\\ud83d@\\ude00");'
+}
+
+issue_2242_3: {
+    options = {
+        evaluate: false,
+    }
+    input: {
+        console.log("\ud83d" + "\ude00", "\ud83d" + "@" + "\ude00");
+    }
+    expect_exact: 'console.log("\\ud83d"+"\\ude00","\\ud83d"+"@"+"\\ude00");'
+}
+
+issue_2242_4: {
+    options = {
+        evaluate: true,
+    }
+    input: {
+        console.log("\ud83d" + "\ude00", "\ud83d" + "@" + "\ude00");
+    }
+    expect_exact: 'console.log("\ud83d\ude00","\\ud83d@\\ude00");'
+}
--- a/test/mocha/string-literal.js
+++ b/test/mocha/string-literal.js
@@ -78,4 +78,41 @@ describe("String literals", function() {
        assert.equal(UglifyJS.parse('"use strict";"\\08"').print_to_string(), '"use strict";"\\08";');
        assert.equal(UglifyJS.parse('"use strict";"\\09"').print_to_string(), '"use strict";"\\09";');
    });
+
+    it("Should not unescape unpaired surrogates", function() {
+        var code = [];
+        for (var i = 0; i <= 0xF; i++) {
+            code.push("\\u000" + i.toString(16));
+        }
+        for (;i <= 0xFF; i++) {
+            code.push("\\u00" + i.toString(16));
+        }
+        for (;i <= 0xFFF; i++) {
+            code.push("\\u0" + i.toString(16));
+        }
+        for (; i <= 0xFFFF; i++) {
+            code.push("\\u" + i.toString(16));
+        }
+        code = '"' + code.join() + '"';
+        var normal = UglifyJS.minify(code, {
+            compress: false,
+            mangle: false,
+            output: {
+                ascii_only: false
+            }
+        });
+        if (normal.error) throw normal.error;
+        assert.ok(code.length > normal.code.length);
+        assert.strictEqual(eval(code), eval(normal.code));
+        var ascii = UglifyJS.minify(code, {
+            compress: false,
+            mangle: false,
+            output: {
+                ascii_only: false
+            }
+        });
+        if (ascii.error) throw ascii.error;
+        assert.ok(code.length > ascii.code.length);
+        assert.strictEqual(eval(code), eval(ascii.code));
+    });
 });
--- a/test/mocha/unicode.js
+++ b/test/mocha/unicode.js
@@ -1,4 +1,5 @@
 var assert = require("assert");
+var semver = require("semver");
 var uglify = require("../node");

 describe("Unicode", function() {
@@ -138,8 +139,33 @@ describe("Unicode", function() {

    it("Should parse raw characters correctly", function() {
        var ast = uglify.parse('console.log("\\udbff");');
-        assert.strictEqual(ast.print_to_string(), 'console.log("\udbff");');
+        assert.strictEqual(ast.print_to_string(), 'console.log("\\udbff");');
        ast = uglify.parse(ast.print_to_string());
-        assert.strictEqual(ast.print_to_string(), 'console.log("\udbff");');
+        assert.strictEqual(ast.print_to_string(), 'console.log("\\udbff");');
    });
+
+    if (semver.satisfies(process.version, ">=4")) {
+        it("Should not unescape unpaired surrogates", function() {
+            var code = [];
+            for (var i = 0; i <= 0xFFFF; i++) {
+                code.push("\\u{" + i.toString(16) + "}");
+            }
+            code = '"' + code.join() + '"';
+            [true, false].forEach(function(ascii_only) {
+                [5, 6].forEach(function(ecma) {
+                    var result = uglify.minify(code, {
+                        compress: false,
+                        mangle: false,
+                        output: {
+                            ascii_only: ascii_only
+                        },
+                        ecma: ecma
+                    });
+                    if (result.error) throw result.error;
+                    assert.ok(code.length > result.code.length);
+                    assert.strictEqual(eval(code), eval(result.code));
+                });
+            });
+        });
+    }
 });