diff options
Diffstat (limited to 'std/encoding')
| -rw-r--r-- | std/encoding/testdata/string.toml | 2 | ||||
| -rw-r--r-- | std/encoding/toml.ts | 148 | ||||
| -rw-r--r-- | std/encoding/toml_test.ts | 11 |
3 files changed, 126 insertions, 35 deletions
diff --git a/std/encoding/testdata/string.toml b/std/encoding/testdata/string.toml index 05f425f11..640717d0e 100644 --- a/std/encoding/testdata/string.toml +++ b/std/encoding/testdata/string.toml @@ -32,3 +32,5 @@ trimmed in raw strings. withApostrophe = "What if it's not?" withSemicolon = "const message = 'hello world';" withHexNumberLiteral = "Prevent bug from stripping string here ->0xabcdef" +withUnicodeChar1 = "\u3042" +withUnicodeChar2 = "Deno\U01F995" diff --git a/std/encoding/toml.ts b/std/encoding/toml.ts index add9880ce..a397a7bd9 100644 --- a/std/encoding/toml.ts +++ b/std/encoding/toml.ts @@ -264,42 +264,43 @@ class Parser { const value = this._parseData(line.slice(idx + 1)); return new KeyValuePair(key, value); } - // TODO (zekth) Need refactor using ACC _parseData(dataString: string): unknown { dataString = dataString.trim(); + switch (dataString[0]) { + case '"': + case "'": + return this._parseString(dataString); + case "[": + case "{": + return this._parseInlineTableOrArray(dataString); + default: { + // Strip a comment. + const match = /#.*$/.exec(dataString); + if (match) { + dataString = dataString.slice(0, match.index).trim(); + } - if (this._isDate(dataString)) { - return new Date(dataString.split("#")[0].trim()); - } - - if (this._isLocalTime(dataString)) { - return eval(`"${dataString.split("#")[0].trim()}"`); - } - - const cut3 = dataString.substring(0, 3).toLowerCase(); - const cut4 = dataString.substring(0, 4).toLowerCase(); - if (cut3 === "inf" || cut4 === "+inf") { - return Infinity; - } - if (cut4 === "-inf") { - return -Infinity; - } - - if (cut3 === "nan" || cut4 === "+nan" || cut4 === "-nan") { - return NaN; - } - - // If binary / octal / hex - const hex = /^(0(?:x|o|b)[0-9a-f_]*)[^#]/gi.exec(dataString); - if (hex && hex[0]) { - return hex[0].trim(); - } - - const testNumber = this._isParsableNumber(dataString); - if (testNumber && !isNaN(testNumber as number)) { - return testNumber; + switch (dataString) { + case "true": + return true; + case "false": + return false; + case "inf": + case "+inf": + return Infinity; + case "-inf": + return -Infinity; + case "nan": + case "+nan": + case "-nan": + return NaN; + default: + return this._parseNumberOrDate(dataString); + } + } } - + } + _parseInlineTableOrArray(dataString: string): unknown { const invalidArr = /,\]/g.exec(dataString); if (invalidArr) { dataString = dataString.replace(/,]/g, "]"); @@ -320,7 +321,10 @@ class Parser { } return JSON.parse(dataString); } - + throw new TOMLError("Malformed inline table or array literal"); + } + _parseString(dataString: string): string { + const quote = dataString[0]; // Handle First and last EOL for multiline strings if (dataString.startsWith(`"\\n`)) { dataString = dataString.replace(`"\\n`, `"`); @@ -332,14 +336,88 @@ class Parser { } else if (dataString.endsWith(`\\n'`)) { dataString = dataString.replace(`\\n'`, `'`); } - return eval(dataString); + let value = ""; + for (let i = 1; i < dataString.length; i++) { + switch (dataString[i]) { + case "\\": + i++; + // See https://toml.io/en/v1.0.0-rc.3#string + switch (dataString[i]) { + case "b": + value += "\b"; + break; + case "t": + value += "\t"; + break; + case "n": + value += "\n"; + break; + case "f": + value += "\f"; + break; + case "r": + value += "\r"; + break; + case "u": + case "U": { + // Unicode character + const codePointLen = dataString[i] === "u" ? 4 : 6; + const codePoint = parseInt( + "0x" + dataString.slice(i + 1, i + 1 + codePointLen), + 16, + ); + value += String.fromCodePoint(codePoint); + i += codePointLen; + break; + } + case "\\": + value += "\\"; + break; + default: + value += dataString[i]; + break; + } + break; + case quote: + if (dataString[i - 1] !== "\\") { + return value; + } + break; + default: + value += dataString[i]; + break; + } + } + throw new TOMLError("Incomplete string literal"); + } + _parseNumberOrDate(dataString: string): unknown { + if (this._isDate(dataString)) { + return new Date(dataString); + } + + if (this._isLocalTime(dataString)) { + return dataString; + } + + // If binary / octal / hex + const hex = /^(0(?:x|o|b)[0-9a-f_]*)/gi.exec(dataString); + if (hex && hex[0]) { + return hex[0].trim(); + } + + const testNumber = this._isParsableNumber(dataString); + if (testNumber !== false && !isNaN(testNumber as number)) { + return testNumber; + } + + return String(dataString); } _isLocalTime(str: string): boolean { const reg = /(\d{2}):(\d{2}):(\d{2})/; return reg.test(str); } _isParsableNumber(dataString: string): number | boolean { - const m = /((?:\+|-|)[0-9_\.e+\-]*)[^#]/i.exec(dataString.trim()); + const m = /((?:\+|-|)[0-9_\.e+\-]*)[^#]/i.exec(dataString); if (!m) { return false; } else { diff --git a/std/encoding/toml_test.ts b/std/encoding/toml_test.ts index 8e25d61ea..870793844 100644 --- a/std/encoding/toml_test.ts +++ b/std/encoding/toml_test.ts @@ -32,6 +32,8 @@ Deno.test({ withSemicolon: `const message = 'hello world';`, withHexNumberLiteral: "Prevent bug from stripping string here ->0xabcdef", + withUnicodeChar1: "あ", + withUnicodeChar2: "Deno🦕", }, }; const actual = parseFile(path.join(testdataDir, "string.toml")); @@ -466,3 +468,12 @@ Deno.test({ assertEquals(actual, expected); }, }); + +Deno.test({ + name: "[TOML] Parse malformed local time as String (#8433)", + fn(): void { + const expected = { sign: "2020-01-01x" }; + const actual = parse(`sign='2020-01-01x'`); + assertEquals(actual, expected); + }, +}); |