cerxes/babel
3
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Change token structure of template literals

Browse Source 
Fix various template parsing issues, makes tokenizer useable from outside
the parser.

Closes #169
Closes #173
This commit is contained in:
Marijn Haverbeke 2014-12-11 14:11:39 +01:00
parent 53d98f1175
commit 2cb3dbcb41
2 changed files with 200 additions and 86 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

123
acorn.js
View File

@ -302,11 +302,12 @@
var metParenL;
// This is used by parser for detecting if it's inside ES6
// Template String. If it is, it should treat '$' as prefix before
// '{expression}' and everything else as string literals.
// This is used by the tokenizer to track the template strings it is
// inside, and count the amount of open braces seen inside them, to
// be able to switch back to a template token when the } to match ${
// is encountered. It will hold an array of integers.
var inTemplate;
var templates;
function initParserState() {
lastStart = lastEnd = tokPos;
@ -409,7 +410,7 @@
var _braceR = {type: "}"}, _parenL = {type: "(", beforeExpr: true}, _parenR = {type: ")"};
var _comma = {type: ",", beforeExpr: true}, _semi = {type: ";", beforeExpr: true};
var _colon = {type: ":", beforeExpr: true}, _dot = {type: "."}, _question = {type: "?", beforeExpr: true};
var _arrow = {type: "=>", beforeExpr: true}, _bquote = {type: "`"}, _dollarBraceL = {type: "${", beforeExpr: true};
var _arrow = {type: "=>", beforeExpr: true}, _template = {type: "template"}, _templateContinued = {type: "templateContinued"};
var _ellipsis = {type: "...", prefix: true, beforeExpr: true};
// Operators. These carry several kinds of properties to help the
@ -452,8 +453,8 @@
parenL: _parenL, parenR: _parenR, comma: _comma, semi: _semi, colon: _colon,
dot: _dot, ellipsis: _ellipsis, question: _question, slash: _slash, eq: _eq,
name: _name, eof: _eof, num: _num, regexp: _regexp, string: _string,
arrow: _arrow, bquote: _bquote, dollarBraceL: _dollarBraceL, star: _star,
assign: _assign};
arrow: _arrow, template: _template, templateContinued: _templateContinued, star: _star,
assign: _assign};
for (var kw in keywordTypes) exports.tokTypes["_" + kw] = keywordTypes[kw];
// This is a trick taken from Esprima. It turns out that, on
@ -548,6 +549,10 @@
var newline = /[\n\r\u2028\u2029]/;
function isNewLine(code) {
return code === 10 || code === 13 || code === 0x2028 || code == 0x2029;
}
// Matches a whole line break (where CRLF is considered a single
// line break). Used to count lines.
@ -598,7 +603,7 @@
}
tokRegexpAllowed = true;
metParenL = 0;
inTemplate = false;
templates = [];
}
// Called at the end of every token. Sets `tokEnd`, `tokVal`, and
@ -788,25 +793,6 @@
return finishOp(code === 61 ? _eq : _prefix, 1);
}
// Get token inside ES6 template (special rules work there).
function getTemplateToken(code) {
// '`' and '${' have special meanings, but they should follow
// string (can be empty)
if (tokType === _string) {
if (code === 96) { // '`'
++tokPos;
return finishToken(_bquote);
} else
if (code === 36 && input.charCodeAt(tokPos + 1) === 123) { // '${'
tokPos += 2;
return finishToken(_dollarBraceL);
}
}
// anything else is considered string literal
return readTmplString();
}
function getTokenFromCode(code) {
switch (code) {
// The interpretation of a dot depends on whether it is followed
@ -821,15 +807,23 @@
case 44: ++tokPos; return finishToken(_comma);
case 91: ++tokPos; return finishToken(_bracketL);
case 93: ++tokPos; return finishToken(_bracketR);
case 123: ++tokPos; return finishToken(_braceL);
case 125: ++tokPos; return finishToken(_braceR);
case 123:
++tokPos;
if (templates.length) ++templates[templates.length - 1];
return finishToken(_braceL);
case 125:
++tokPos;
if (templates.length && --templates[templates.length - 1] === 0)
return readTemplateString(_templateContinued);
else
return finishToken(_braceR);
case 58: ++tokPos; return finishToken(_colon);
case 63: ++tokPos; return finishToken(_question);
case 96: // '`'
if (options.ecmaVersion >= 6) {
++tokPos;
return finishToken(_bquote, undefined, false);
return readTemplateString(_template);
}
case 48: // '0'
@ -890,8 +884,6 @@
var code = input.charCodeAt(tokPos);
if (inTemplate) return getTemplateToken(code);
// Identifier or keyword. '\uXXXX' sequences are allowed in
// identifiers, so '\' also dispatches to that.
if (isIdentifierStart(code) || code === 92 /* '\' */) return readWord();
@ -1074,28 +1066,34 @@
}
}
function readTmplString() {
var out = "";
function readTemplateString(type) {
if (type == _templateContinued) templates.pop();
var out = "", start = tokPos;;
for (;;) {
if (tokPos >= inputLen) raise(tokStart, "Unterminated string constant");
var ch = input.charCodeAt(tokPos);
if (ch === 96 || ch === 36 && input.charCodeAt(tokPos + 1) === 123) // '`', '${'
return finishToken(_string, out);
if (ch === 92) { // '\'
if (tokPos >= inputLen) raise(tokStart, "Unterminated template");
var ch = input.charAt(tokPos);
if (ch === "`" || ch === "$" && input.charCodeAt(tokPos + 1) === 123) { // '`', '${'
var raw = input.slice(start, tokPos);
++tokPos;
if (ch == "$") { ++tokPos; templates.push(1); }
return finishToken(type, {cooked: out, raw: raw});
}
if (ch === "\\") { // '\'
out += readEscapedChar();
} else {
++tokPos;
if (newline.test(String.fromCharCode(ch))) {
if (ch === 13 && input.charCodeAt(tokPos) === 10) {
if (newline.test(ch)) {
if (ch === "\r" && input.charCodeAt(tokPos) === 10) {
++tokPos;
ch = 10;
ch = "\n";
}
if (options.locations) {
++tokCurLine;
tokLineStart = tokPos;
}
}
out += String.fromCharCode(ch); // '\'
out += ch;
}
}
}
@ -2010,7 +2008,7 @@
node.callee = base;
node.arguments = parseExprList(_parenR, false);
return parseSubscripts(finishNode(node, "CallExpression"), start, noCalls);
} else if (tokType === _bquote) {
} else if (tokType === _template) {
var node = startNodeAt(start);
node.tag = base;
node.quasi = parseTemplate();
@ -2125,7 +2123,7 @@
case _new:
return parseNew();
case _bquote:
case _template:
return parseTemplate();
default:
@ -2149,33 +2147,24 @@
// Parse template expression.
function parseTemplateElement() {
var elem = startNode();
elem.value = tokVal;
elem.tail = input.charCodeAt(tokEnd - 1) !== 123; // '{'
next();
return finishNode(elem, "TemplateElement");
}
function parseTemplate() {
var oldInTemplate = inTemplate;
inTemplate = true;
var node = startNode();
node.expressions = [];
node.quasis = [];
next();
for (;;) {
var elem = startNode();
elem.value = {cooked: tokVal, raw: input.slice(tokStart, tokEnd)};
elem.tail = false;
next();
node.quasis.push(finishNode(elem, "TemplateElement"));
if (tokType === _bquote) { // '`', end of template
elem.tail = true;
break;
}
inTemplate = false;
expect(_dollarBraceL);
var curElt = parseTemplateElement();
node.quasis = [curElt];
while (!curElt.tail) {
node.expressions.push(parseExpression());
inTemplate = true;
// hack to include previously skipped space
tokPos = tokEnd;
expect(_braceR);
if (tokType !== _templateContinued) unexpected();
node.quasis.push(curElt = parseTemplateElement());
}
inTemplate = oldInTemplate;
next();
return finishNode(node, "TemplateLiteral");
}

163
test/tests-harmony.js
View File

@ -628,8 +628,8 @@ test("`42`", {
value: {raw: "42", cooked: "42"},
tail: true,
loc: {
start: {line: 1, column: 1},
end: {line: 1, column: 3}
start: {line: 1, column: 0},
end: {line: 1, column:4}
}
}],
expressions: [],
@ -674,8 +674,8 @@ test("raw`42`", {
value: {raw: "42", cooked: "42"},
tail: true,
loc: {
start: {line: 1, column: 4},
end: {line: 1, column: 6}
start: {line: 1, column: 3},
end: {line: 1, column: 7}
}
}],
expressions: [],
@ -726,8 +726,8 @@ test("raw`hello ${name}`", {
value: {raw: "hello ", cooked: "hello "},
tail: false,
loc: {
start: {line: 1, column: 4},
end: {line: 1, column: 10}
start: {line: 1, column: 3},
end: {line: 1, column: 12}
}
},
{
@ -735,8 +735,8 @@ test("raw`hello ${name}`", {
value: {raw: "", cooked: ""},
tail: true,
loc: {
start: {line: 1, column: 17},
end: {line: 1, column: 17}
start: {line: 1, column: 16},
end: {line: 1, column: 18}
}
}
],
@ -784,8 +784,8 @@ test("`$`", {
value: {raw: "$", cooked: "$"},
tail: true,
loc: {
start: {line: 1, column: 1},
end: {line: 1, column: 2}
start: {line: 1, column: 0},
end: {line: 1, column: 3}
}
}],
expressions: [],
@ -820,8 +820,8 @@ test("`\\n\\r\\b\\v\\t\\f\\\n\\\r\n`", {
value: {raw: "\\n\\r\\b\\v\\t\\f\\\n\\\r\n", cooked: "\n\r\b\u000b\t\f"},
tail: true,
loc: {
start: {line: 1, column: 1},
end: {line: 3, column: 0}
start: {line: 1, column: 0},
end: {line: 3, column: 1}
}
}],
expressions: [],
@ -856,8 +856,8 @@ test("`\n\r\n`", {
value: {raw: "\n\r\n", cooked: "\n\n"},
tail: true,
loc: {
start: {line: 1, column: 1},
end: {line: 3, column: 0}
start: {line: 1, column: 0},
end: {line: 3, column: 1}
}
}],
expressions: [],
@ -892,8 +892,8 @@ test("`\\u{000042}\\u0042\\x42u0\\102\\A`", {
value: {raw: "\\u{000042}\\u0042\\x42u0\\102\\A", cooked: "BBBu0BA"},
tail: true,
loc: {
start: {line: 1, column: 1},
end: {line: 1, column: 29}
start: {line: 1, column: 0},
end: {line: 1, column: 30}
}
}],
expressions: [],
@ -940,8 +940,8 @@ test("new raw`42`", {
value: {raw: "42", cooked: "42"},
tail: true,
loc: {
start: {line: 1, column: 8},
end: {line: 1, column: 10}
start: {line: 1, column: 7},
end: {line: 1, column: 11}
}
}],
expressions: [],
@ -976,6 +976,131 @@ test("new raw`42`", {
locations: true
});
test("`outer${{x: {y: 10}}}bar${`nested${function(){return 1;}}endnest`}end`",{
type: "Program",
body: [
{
type: "ExpressionStatement",
expression: {
type: "TemplateLiteral",
expressions: [
{
type: "ObjectExpression",
properties: [
{
type: "Property",
method: false,
shorthand: false,
computed: false,
key: {
type: "Identifier",
name: "x"
},
value: {
type: "ObjectExpression",
properties: [
{
type: "Property",
method: false,
shorthand: false,
computed: false,
key: {
type: "Identifier",
name: "y"
},
value: {
type: "Literal",
value: 10,
raw: "10"
},
kind: "init"
}
]
},
kind: "init"
}
]
},
{
type: "TemplateLiteral",
expressions: [
{
type: "FunctionExpression",
id: null,
params: [],
defaults: [],
rest: null,
generator: false,
body: {
type: "BlockStatement",
body: [
{
type: "ReturnStatement",
argument: {
type: "Literal",
value: 1,
raw: "1"
}
}
]
},
expression: false
}
],
quasis: [
{
type: "TemplateElement",
value: {
cooked: "nested",
raw: "nested"
},
tail: false
},
{
type: "TemplateElement",
value: {
cooked: "endnest",
raw: "endnest"
},
tail: true
}
]
}
],
quasis: [
{
type: "TemplateElement",
value: {
cooked: "outer",
raw: "outer"
},
tail: false
},
{
type: "TemplateElement",
value: {
cooked: "bar",
raw: "bar"
},
tail: false
},
{
type: "TemplateElement",
value: {
cooked: "end",
raw: "end"
},
tail: true
}
]
}
}
]
}, {
ecmaVersion: 6
});
// ES6: Switch Case Declaration
test("switch (answer) { case 42: let t = 42; break; }", {
@ -13959,7 +14084,7 @@ testFail("class A extends yield B { }", "Unexpected token (1:22)", {ecmaVersion:
testFail("class default", "Unexpected token (1:6)", {ecmaVersion: 6});
testFail("`test", "Unterminated string constant (1:1)", {ecmaVersion: 6});
testFail("`test", "Unterminated template (1:0)", {ecmaVersion: 6});
testFail("switch `test`", "Unexpected token (1:7)", {ecmaVersion: 6});